6 ani în urmă · 7e185f73be
--- a/luajit.mod/lib/win32/libluajit_x64.a
+++ b/luajit.mod/lib/win32/libluajit_x64.a
--- a/luajit.mod/lib/win32/libluajit_x86.a
+++ b/luajit.mod/lib/win32/libluajit_x86.a
--- a/luajit.mod/luajit.bmx
+++ b/luajit.mod/luajit.bmx
@@ -5,9 +5,13 @@ bbdoc: LuaJIT
 
				 end rem

			
 
				 Module zeke.luajit

			
 
				 

			
 
				-ModuleInfo "Version: 1.14"

			
 
				+ModuleInfo "Version: 1.16"

			
 
				 ModuleInfo "Author: Zeke"

			
 
				 

			
 
				+ModuleInfo "History: 1.16"

			
 
				+ModuleInfo "History: Updated to LuaJIT 2.1.0-beta3.f0e865d."

			
 
				+ModuleInfo "History: 1.15"

			
 
				+ModuleInfo "History: Fixed lua_integer size for 64-bit."

			
 
				 ModuleInfo "History: 1.14"

			
 
				 ModuleInfo "History: Fixed reflection issues."

			
 
				 ModuleInfo "History: 1.13"

			
--- a/luajit.mod/luajit/.gitignore
+++ b/luajit.mod/luajit/.gitignore
@@ -0,0 +1,11 @@
 
				+*.[oa]
			
 
				+*.so
			
 
				+*.obj
			
 
				+*.lib
			
 
				+*.exp
			
 
				+*.dll
			
 
				+*.exe
			
 
				+*.manifest
			
 
				+*.dmp
			
 
				+*.swp
			
 
				+.tags
			
--- a/luajit.mod/luajit/Makefile
+++ b/luajit.mod/luajit/Makefile
@@ -14,9 +14,10 @@
 
				 ##############################################################################
			
 
				 
			
 
				 MAJVER=  2
			
 
				-MINVER=  0
			
 
				-RELVER=  5
			
 
				-VERSION= $(MAJVER).$(MINVER).$(RELVER)
			
 
				+MINVER=  1
			
 
				+RELVER=  0
			
 
				+PREREL=  -beta3
			
 
				+VERSION= $(MAJVER).$(MINVER).$(RELVER)$(PREREL)
			
 
				 ABIVER=  5.1
			
 
				 
			
 
				 ##############################################################################
			
@@ -84,8 +85,10 @@ FILE_SO= libluajit.so
 
				 FILE_MAN= luajit.1
			
 
				 FILE_PC= luajit.pc
			
 
				 FILES_INC= lua.h lualib.h lauxlib.h luaconf.h lua.hpp luajit.h
			
 
				-FILES_JITLIB= bc.lua v.lua dump.lua dis_x86.lua dis_x64.lua dis_arm.lua \
			
 
				-	      dis_ppc.lua dis_mips.lua dis_mipsel.lua bcsave.lua vmdef.lua
			
 
				+FILES_JITLIB= bc.lua bcsave.lua dump.lua p.lua v.lua zone.lua \
			
 
				+	      dis_x86.lua dis_x64.lua dis_arm.lua dis_arm64.lua \
			
 
				+	      dis_arm64be.lua dis_ppc.lua dis_mips.lua dis_mipsel.lua \
			
 
				+	      dis_mips64.lua dis_mips64el.lua vmdef.lua
			
 
				 
			
 
				 ifeq (,$(findstring Windows,$(OS)))
			
 
				   HOST_SYS:= $(shell uname -s)
			
@@ -115,7 +118,7 @@ install: $(INSTALL_DEP)
 
				 	$(MKDIR) $(INSTALL_DIRS)
			
 
				 	cd src && $(INSTALL_X) $(FILE_T) $(INSTALL_T)
			
 
				 	cd src && test -f $(FILE_A) && $(INSTALL_F) $(FILE_A) $(INSTALL_STATIC) || :
			
 
				-	$(RM) $(INSTALL_TSYM) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2)
			
 
				+	$(RM) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2)
			
 
				 	cd src && test -f $(FILE_SO) && \
			
 
				 	  $(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) && \
			
 
				 	  $(LDCONFIG) $(INSTALL_LIB) && \
			
@@ -127,12 +130,18 @@ install: $(INSTALL_DEP)
 
				 	  $(RM) $(FILE_PC).tmp
			
 
				 	cd src && $(INSTALL_F) $(FILES_INC) $(INSTALL_INC)
			
 
				 	cd src/jit && $(INSTALL_F) $(FILES_JITLIB) $(INSTALL_JITLIB)
			
 
				-	$(SYMLINK) $(INSTALL_TNAME) $(INSTALL_TSYM)
			
 
				 	@echo "==== Successfully installed LuaJIT $(VERSION) to $(PREFIX) ===="
			
 
				+	@echo ""
			
 
				+	@echo "Note: the development releases deliberately do NOT install a symlink for luajit"
			
 
				+	@echo "You can do this now by running this command (with sudo):"
			
 
				+	@echo ""
			
 
				+	@echo "  $(SYMLINK) $(INSTALL_TNAME) $(INSTALL_TSYM)"
			
 
				+	@echo ""
			
 
				+
			
 
				 
			
 
				 uninstall:
			
 
				 	@echo "==== Uninstalling LuaJIT $(VERSION) from $(PREFIX) ===="
			
 
				-	$(UNINSTALL) $(INSTALL_TSYM) $(INSTALL_T) $(INSTALL_STATIC) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2) $(INSTALL_MAN)/$(FILE_MAN) $(INSTALL_PC)
			
 
				+	$(UNINSTALL) $(INSTALL_T) $(INSTALL_STATIC) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2) $(INSTALL_MAN)/$(FILE_MAN) $(INSTALL_PC)
			
 
				 	for file in $(FILES_JITLIB); do \
			
 
				 	  $(UNINSTALL) $(INSTALL_JITLIB)/$$file; \
			
 
				 	  done
			
--- a/luajit.mod/luajit/README
+++ b/luajit.mod/luajit/README
@@ -1,5 +1,5 @@
 
				-README for LuaJIT 2.0.5
			
 
				------------------------
			
 
				+README for LuaJIT 2.1.0-beta3
			
 
				+-----------------------------
			
 
				 
			
 
				 LuaJIT is a Just-In-Time (JIT) compiler for the Lua programming language.
			
 
				 
			
--- a/luajit.mod/luajit/doc/bluequad-print.css
+++ b/luajit.mod/luajit/doc/bluequad-print.css
@@ -1,4 +1,4 @@
 
				-/* Copyright (C) 2004-2017 Mike Pall.
			
 
				+/* Copyright (C) 2004-2018 Mike Pall.
			
 
				  *
			
 
				  * You are welcome to use the general ideas of this design for your own sites.
			
 
				  * But please do not steal the stylesheet, the layout or the color scheme.
			
--- a/luajit.mod/luajit/doc/bluequad.css
+++ b/luajit.mod/luajit/doc/bluequad.css
@@ -1,4 +1,4 @@
 
				-/* Copyright (C) 2004-2017 Mike Pall.
			
 
				+/* Copyright (C) 2004-2018 Mike Pall.
			
 
				  *
			
 
				  * You are welcome to use the general ideas of this design for your own sites.
			
 
				  * But please do not steal the stylesheet, the layout or the color scheme.
			
--- a/luajit.mod/luajit/doc/changes.html
+++ b/luajit.mod/luajit/doc/changes.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>LuaJIT Change History</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -44,6 +43,8 @@ div.major { max-width: 600px; padding: 1em; margin: 1em 0 1em 0; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -72,6 +73,96 @@ to see whether newer versions are available.
 
				 </p>
			
 
				 
			
 
				 <div class="major" style="background: #d0d0ff;">
			
 
				+<h2 id="LuaJIT-2.1.0-beta3">LuaJIT 2.1.0-beta3 &mdash; 2017-05-01</h2>
			
 
				+<ul>
			
 
				+<li>Rewrite memory block allocator.</li>
			
 
				+<li>Add various extension from Lua 5.2/5.3.</li>
			
 
				+<li>Remove old Lua 5.0 compatibility defines.</li>
			
 
				+<li>Set arg table before evaluating <tt>LUA_INIT</tt> and <tt>-e</tt> chunks.</li>
			
 
				+<li>Fix FOLD rules for <tt>math.abs()</tt> and FP negation.</li>
			
 
				+<li>Fix soft-float <tt>math.abs()</tt> and negation.</li>
			
 
				+<li>Fix formatting of some small denormals at low precision.</li>
			
 
				+<li>LJ_GC64: Add JIT compiler support.</li>
			
 
				+<li>x64/LJ_GC64: Add JIT compiler backend.</li>
			
 
				+<li>x86/x64: Generate BMI2 shifts and rotates, if available.</li>
			
 
				+<li>Windows/x86: Add full exception interoperability.</li>
			
 
				+<li>ARM64: Add big-endian support.</li>
			
 
				+<li>ARM64: Add JIT compiler backend.</li>
			
 
				+<li>MIPS: Fix <tt>TSETR</tt> barrier.</li>
			
 
				+<li>MIPS: Support MIPS16 interlinking.</li>
			
 
				+<li>MIPS soft-float: Fix code generation for <tt>HREF</tt>.</li>
			
 
				+<li>MIPS64: Add MIPS64 hard-float JIT compiler backend.</li>
			
 
				+<li>MIPS64: Add MIPS64 hard-float/soft-float support to interpreter.</li>
			
 
				+<li>FFI: Compile bitfield loads/stores.</li>
			
 
				+<li>Various fixes common with the 2.0 branch.</li>
			
 
				+</ul>
			
 
				+
			
 
				+<h2 id="LuaJIT-2.1.0-beta2">LuaJIT 2.1.0-beta2 &mdash; 2016-03-03</h2>
			
 
				+<ul>
			
 
				+<li>Enable trace stitching.</li>
			
 
				+<li>Use internal implementation for converting FP numbers to strings.</li>
			
 
				+<li>Parse Unicode escape <tt>'\u{XX...}'</tt> in string literals.</li>
			
 
				+<li>Add MIPS soft-float support.</li>
			
 
				+<li>Switch MIPS port to dual-number mode.</li>
			
 
				+<li>x86/x64: Add support for AES-NI, AVX and AVX2 to DynASM.</li>
			
 
				+<li>FFI: Add <tt>ssize_t</tt> declaration.</li>
			
 
				+<li>FFI: Parse <tt>#line NN</tt> and <tt>#NN</tt>.</li>
			
 
				+<li>Various minor fixes.</li>
			
 
				+</ul>
			
 
				+
			
 
				+<h2 id="LuaJIT-2.1.0-beta1">LuaJIT 2.1.0-beta1 &mdash; 2015-08-25</h2>
			
 
				+<p>
			
 
				+This is a brief summary of the major changes in LuaJIT 2.1 compared to 2.0.
			
 
				+Please take a look at the commit history for more details.
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li>Changes to the VM core:
			
 
				+<ul>
			
 
				+<li>Add low-overhead profiler (<tt>-jp</tt>).</li>
			
 
				+<li>Add <tt>LJ_GC64</tt> mode: 64 bit GC object references (really: 47 bit). Interpreter-only for now.</li>
			
 
				+<li>Add <tt>LJ_FR2</tt> mode: Two-slot frame info. Required by <tt>LJ_GC64</tt> mode.</li>
			
 
				+<li>Add <tt>table.new()</tt> and <tt>table.clear()</tt>.</li>
			
 
				+<li>Parse binary number literals (<tt>0bxxx</tt>).</li>
			
 
				+</ul></li>
			
 
				+<li>Improvements to the JIT compiler:
			
 
				+<ul>
			
 
				+<li>Add trace stitching (disabled for now).</li>
			
 
				+<li>Compile various builtins: <tt>string.char()</tt>, <tt>string.reverse()</tt>, <tt>string.lower()</tt>, <tt>string.upper()</tt>, <tt>string.rep()</tt>, <tt>string.format()</tt>, <tt>table.concat()</tt>, <tt>bit.tohex()</tt>, <tt>getfenv(0)</tt>, <tt>debug.getmetatable()</tt>.</li>
			
 
				+<li>Compile <tt>string.find()</tt> for fixed string searches (no patterns).</li>
			
 
				+<li>Compile <tt>BC_TSETM</tt>, e.g. <tt>{1,2,3,f()}</tt>.</li>
			
 
				+<li>Compile string concatenations (<tt>BC_CAT</tt>).</li>
			
 
				+<li>Compile <tt>__concat</tt> metamethod.</li>
			
 
				+<li>Various minor optimizations.</li>
			
 
				+</ul></li>
			
 
				+<li>Internal Changes:
			
 
				+<ul>
			
 
				+<li>Add support for embedding LuaJIT bytecode for builtins.</li>
			
 
				+<li>Replace various builtins with embedded bytecode.</li>
			
 
				+<li>Refactor string buffers and string formatting.</li>
			
 
				+<li>Remove obsolete non-truncating number to integer conversions.</li>
			
 
				+</ul></li>
			
 
				+<li>Ports:
			
 
				+<ul>
			
 
				+<li>Add Xbox One port (<tt>LJ_GC64</tt> mode).</li>
			
 
				+<li>ARM64: Add port of the interpreter (<tt>LJ_GC64</tt> mode).</li>
			
 
				+<li>x64: Add separate port of the interpreter to <tt>LJ_GC64</tt> mode.</li>
			
 
				+<li>x86/x64: Drop internal x87 math functions. Use libm functions.</li>
			
 
				+<li>x86: Remove x87 support from interpreter. SSE2 is mandatory now.</li>
			
 
				+<li>PPC/e500: Drop support for this architecture.</li>
			
 
				+</ul></li>
			
 
				+<li>FFI library:
			
 
				+<ul>
			
 
				+<li>FFI: Add 64 bit bitwise operations.</li>
			
 
				+<li>FFI: Compile VLA/VLS and large cdata allocations with default initialization.</li>
			
 
				+<li>FFI: Compile conversions from functions to function pointers.</li>
			
 
				+<li>FFI: Compile lightuserdata to <tt>void *</tt> conversion.</li>
			
 
				+<li>FFI: Compile <tt>ffi.gc(cdata, nil)</tt>, too.</li>
			
 
				+<li>FFI: Add <tt>ffi.typeinfo()</tt>.</li>
			
 
				+</ul></li>
			
 
				+</ul>
			
 
				+</div>
			
 
				+
			
 
				+<div class="major" style="background: #ffffd0;">
			
 
				 <h2 id="LuaJIT-2.0.5">LuaJIT 2.0.5 &mdash; 2017-05-01</h2>
			
 
				 <ul>
			
 
				 <li>Add workaround for MSVC 2015 stdio changes.</li>
			
@@ -81,7 +172,7 @@ to see whether newer versions are available.
 
				 <li>Remove internal <tt>__mode = "K"</tt> and replace with safe check.</li>
			
 
				 <li>Add "proto" field to <tt>jit.util.funcinfo()</tt>.</li>
			
 
				 <li>Fix GC step size calculation.</li>
			
 
				-<li>Initialize <tt>uv->immutable</tt> for upvalues of loaded chunks.</li>
			
 
				+<li>Initialize <tt>uv-&gt;immutable</tt> for upvalues of loaded chunks.</li>
			
 
				 <li>Fix for cdata vs. non-cdata arithmetics/comparisons.</li>
			
 
				 <li>Drop leftover regs in 'for' iterator assignment, too.</li>
			
 
				 <li>Fix PHI remarking in SINK pass.</li>
			
@@ -777,240 +868,11 @@ This matches the behavior of Lua 5.1, but not the specification.</li>
 
				 no point in listing differences over earlier versions.</li>
			
 
				 </ul>
			
 
				 </div>
			
 
				-
			
 
				-<div class="major" style="background: #ffff80;">
			
 
				-<h2 id="LuaJIT-1.1.8">LuaJIT 1.1.8 &mdash; 2012-04-16</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1.5. Also integrated fixes for all
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.5"><span class="ext">&raquo;</span>&nbsp;<span class="ext">&raquo;</span>&nbsp;currently known bugs in Lua 5.1.5</a>.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.7">LuaJIT 1.1.7 &mdash; 2011-05-05</h2>
			
 
				-<ul>
			
 
				-<li>Added fixes for the
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.4"><span class="ext">&raquo;</span>&nbsp;currently known bugs in Lua 5.1.4</a>.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.6">LuaJIT 1.1.6 &mdash; 2010-03-28</h2>
			
 
				-<ul>
			
 
				-<li>Added fixes for the
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.4"><span class="ext">&raquo;</span>&nbsp;currently known bugs in Lua 5.1.4</a>.</li>
			
 
				-<li>Removed wrong GC check in <tt>jit_createstate()</tt>.
			
 
				-Thanks to Tim Mensch.</li>
			
 
				-<li>Fixed bad assertions while compiling <tt>table.insert()</tt> and
			
 
				-<tt>table.remove()</tt>.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.5">LuaJIT 1.1.5 &mdash; 2008-10-25</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1.4. Fixes all
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.3"><span class="ext">&raquo;</span>&nbsp;known bugs in Lua 5.1.3</a>.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.4">LuaJIT 1.1.4 &mdash; 2008-02-05</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1.3. Fixes all
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.2"><span class="ext">&raquo;</span>&nbsp;known bugs in Lua 5.1.2</a>.</li>
			
 
				-<li>Fixed possible (but unlikely) stack corruption while compiling
			
 
				-<tt>k^x</tt> expressions.</li>
			
 
				-<li>Fixed DynASM template for cmpss instruction.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.3">LuaJIT 1.1.3 &mdash; 2007-05-24</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1.2. Fixes all
			
 
				-<a href="http://www.lua.org/bugs.html#5.1.1"><span class="ext">&raquo;</span>&nbsp;known bugs in Lua 5.1.1</a>.</li>
			
 
				-<li>Merged pending Lua 5.1.x fixes: "return -nil" bug, spurious count hook call.</li>
			
 
				-<li>Remove a (sometimes) wrong assertion in <tt>luaJIT_findpc()</tt>.</li>
			
 
				-<li>DynASM now allows labels for displacements and <tt>.aword</tt>.</li>
			
 
				-<li>Fix some compiler warnings for DynASM glue (internal API change).</li>
			
 
				-<li>Correct naming for SSSE3 (temporarily known as SSE4) in DynASM and x86 disassembler.</li>
			
 
				-<li>The loadable debug modules now handle redirection to stdout
			
 
				-(e.g. <tt>-j&nbsp;trace=-</tt>).</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.2">LuaJIT 1.1.2 &mdash; 2006-06-24</h2>
			
 
				-<ul>
			
 
				-<li>Fix MSVC inline assembly: use only local variables with
			
 
				-<tt>lua_number2int()</tt>.</li>
			
 
				-<li>Fix "attempt to call a thread value" bug on Mac OS X:
			
 
				-make values of consts used as lightuserdata keys unique
			
 
				-to avoid joining by the compiler/linker.</li>
			
 
				-</ul>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.1">LuaJIT 1.1.1 &mdash; 2006-06-20</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1.1. Fixes all
			
 
				-<a href="http://www.lua.org/bugs.html#5.1"><span class="ext">&raquo;</span>&nbsp;known bugs in Lua 5.1</a>.</li>
			
 
				-<li>Enforce (dynamic) linker error for EXE/DLL version mismatches.</li>
			
 
				-<li>Minor changes to DynASM: faster pre-processing, smaller encoding
			
 
				-for some immediates.</li>
			
 
				-</ul>
			
 
				-<p>
			
 
				-This release is in sync with Coco 1.1.1 (see the
			
 
				-<a href="http://coco.luajit.org/changes.html"><span class="ext">&raquo;</span>&nbsp;Coco Change History</a>).
			
 
				-</p>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.1.0">LuaJIT 1.1.0 &mdash; 2006-03-13</h2>
			
 
				-<ul>
			
 
				-<li>Merged with Lua 5.1 (final).</li>
			
 
				-
			
 
				-<li>New JIT call frame setup:
			
 
				-<ul>
			
 
				-<li>The C stack is kept 16 byte aligned (faster).
			
 
				-Mandatory for Mac OS X on Intel, too.</li>
			
 
				-<li>Faster calling conventions for internal C helper functions.</li>
			
 
				-<li>Better instruction scheduling for function prologue, OP_CALL and
			
 
				-OP_RETURN.</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Miscellaneous optimizations:
			
 
				-<ul>
			
 
				-<li>Faster loads of FP constants. Remove narrow-to-wide store-to-load
			
 
				-forwarding stalls.</li>
			
 
				-<li>Use (scalar) SSE2 ops (if the CPU supports it) to speed up slot moves
			
 
				-and FP to integer conversions.</li>
			
 
				-<li>Optimized the two-argument form of <tt>OP_CONCAT</tt> (<tt>a..b</tt>).</li>
			
 
				-<li>Inlined <tt>OP_MOD</tt> (<tt>a%b</tt>).
			
 
				-With better accuracy than the C variant, too.</li>
			
 
				-<li>Inlined <tt>OP_POW</tt> (<tt>a^b</tt>). Unroll <tt>x^k</tt> or
			
 
				-use <tt>k^x = 2^(log2(k)*x)</tt> or call <tt>pow()</tt>.</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Changes in the optimizer:
			
 
				-<ul>
			
 
				-<li>Improved hinting for table keys derived from table values
			
 
				-(<tt>t1[t2[x]]</tt>).</li>
			
 
				-<li>Lookup hinting now works with arbitrary object types and
			
 
				-supports index chains, too.</li>
			
 
				-<li>Generate type hints for arithmetic and comparison operators,
			
 
				-OP_LEN, OP_CONCAT and OP_FORPREP.</li>
			
 
				-<li>Remove several hint definitions in favour of a generic COMBINE hint.</li>
			
 
				-<li>Complete rewrite of <tt>jit.opt_inline</tt> module
			
 
				-(ex <tt>jit.opt_lib</tt>).</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Use adaptive deoptimization:
			
 
				-<ul>
			
 
				-<li>If runtime verification of a contract fails, the affected
			
 
				-instruction is recompiled and patched on-the-fly.
			
 
				-Regular programs will trigger deoptimization only occasionally.</li>
			
 
				-<li>This avoids generating code for uncommon fallback cases
			
 
				-most of the time. Generated code is up to 30% smaller compared to
			
 
				-LuaJIT&nbsp;1.0.3.</li>
			
 
				-<li>Deoptimization is used for many opcodes and contracts:
			
 
				-<ul>
			
 
				-<li>OP_CALL, OP_TAILCALL: type mismatch for callable.</li>
			
 
				-<li>Inlined calls: closure mismatch, parameter number and type mismatches.</li>
			
 
				-<li>OP_GETTABLE, OP_SETTABLE: table or key type and range mismatches.</li>
			
 
				-<li>All arithmetic and comparison operators, OP_LEN, OP_CONCAT,
			
 
				-OP_FORPREP: operand type and range mismatches.</li>
			
 
				-</ul></li>
			
 
				-<li>Complete redesign of the debug and traceback info
			
 
				-(bytecode &harr; mcode) to support deoptimization.
			
 
				-Much more flexible and needs only 50% of the space.</li>
			
 
				-<li>The modules <tt>jit.trace</tt>, <tt>jit.dumphints</tt> and
			
 
				-<tt>jit.dump</tt> handle deoptimization.</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Inlined many popular library functions
			
 
				-(for commonly used arguments only):
			
 
				-<ul>
			
 
				-<li>Most <tt>math.*</tt> functions (the 18 most used ones)
			
 
				-[2x-10x faster].</li>
			
 
				-<li><tt>string.len</tt>, <tt>string.sub</tt> and <tt>string.char</tt>
			
 
				-[2x-10x faster].</li>
			
 
				-<li><tt>table.insert</tt>, <tt>table.remove</tt> and <tt>table.getn</tt>
			
 
				-[3x-5x faster].</li>
			
 
				-<li><tt>coroutine.yield</tt> and <tt>coroutine.resume</tt>
			
 
				-[3x-5x faster].</li>
			
 
				-<li><tt>pairs</tt>, <tt>ipairs</tt> and the corresponding iterators
			
 
				-[8x-15x faster].</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Changes in the core and loadable modules and the stand-alone executable:
			
 
				-<ul>
			
 
				-<li>Added <tt>jit.version</tt>, <tt>jit.version_num</tt>
			
 
				-and <tt>jit.arch</tt>.</li>
			
 
				-<li>Reorganized some internal API functions (<tt>jit.util.*mcode*</tt>).</li>
			
 
				-<li>The <tt>-j dump</tt> output now shows JSUB names, too.</li>
			
 
				-<li>New x86 disassembler module written in pure Lua. No dependency
			
 
				-on ndisasm anymore. Flexible API, very compact (500 lines)
			
 
				-and complete (x87, MMX, SSE, SSE2, SSE3, SSSE3, privileged instructions).</li>
			
 
				-<li><tt>luajit -v</tt> prints the LuaJIT version and copyright
			
 
				-on a separate line.</li>
			
 
				-</ul></li>
			
 
				-
			
 
				-<li>Added SSE, SSE2, SSE3 and SSSE3 support to DynASM.</li>
			
 
				-<li>Miscellaneous doc changes. Added a section about
			
 
				-<a href="install.html#embedding">embedding LuaJIT</a>.</li>
			
 
				-</ul>
			
 
				-<p>
			
 
				-This release is in sync with Coco 1.1.0 (see the
			
 
				-<a href="http://coco.luajit.org/changes.html"><span class="ext">&raquo;</span>&nbsp;Coco Change History</a>).
			
 
				-</p>
			
 
				-</div>
			
 
				-
			
 
				-<div class="major" style="background: #ffffd0;">
			
 
				-<h2 id="LuaJIT-1.0.3">LuaJIT 1.0.3 &mdash; 2005-09-08</h2>
			
 
				-<ul>
			
 
				-<li>Even more docs.</li>
			
 
				-<li>Unified closure checks in <tt>jit.*</tt>.</li>
			
 
				-<li>Fixed some range checks in <tt>jit.util.*</tt>.</li>
			
 
				-<li>Fixed __newindex call originating from <tt>jit_settable_str()</tt>.</li>
			
 
				-<li>Merged with Lua 5.1 alpha (including early bug fixes).</li>
			
 
				-</ul>
			
 
				-<p>
			
 
				-This is the first public release of LuaJIT.
			
 
				-</p>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.0.2">LuaJIT 1.0.2 &mdash; 2005-09-02</h2>
			
 
				-<ul>
			
 
				-<li>Add support for flushing the Valgrind translation cache <br>
			
 
				-(<tt>MYCFLAGS= -DUSE_VALGRIND</tt>).</li>
			
 
				-<li>Add support for freeing executable mcode memory to the <tt>mmap()</tt>-based
			
 
				-variant for POSIX systems.</li>
			
 
				-<li>Reorganized the C&nbsp;function signature handling in
			
 
				-<tt>jit.opt_lib</tt>.</li>
			
 
				-<li>Changed to index-based hints for inlining C&nbsp;functions.
			
 
				-Still no support in the backend for inlining.</li>
			
 
				-<li>Hardcode <tt>HEAP_CREATE_ENABLE_EXECUTE</tt> value if undefined.</li>
			
 
				-<li>Misc. changes to the <tt>jit.*</tt> modules.</li>
			
 
				-<li>Misc. changes to the Makefiles.</li>
			
 
				-<li>Lots of new docs.</li>
			
 
				-<li>Complete doc reorg.</li>
			
 
				-</ul>
			
 
				-<p>
			
 
				-Not released because Lua 5.1 alpha came out today.
			
 
				-</p>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.0.1">LuaJIT 1.0.1 &mdash; 2005-08-31</h2>
			
 
				-<ul>
			
 
				-<li>Missing GC step in <tt>OP_CONCAT</tt>.</li>
			
 
				-<li>Fix result handling for C &ndash;> JIT calls.</li>
			
 
				-<li>Detect CPU feature bits.</li>
			
 
				-<li>Encode conditional moves (<tt>fucomip</tt>) only when supported.</li>
			
 
				-<li>Add fallback instructions for FP compares.</li>
			
 
				-<li>Add support for <tt>LUA_COMPAT_VARARG</tt>. Still disabled by default.</li>
			
 
				-<li>MSVC needs a specific place for the <tt>CALLBACK</tt> attribute
			
 
				-(David Burgess).</li>
			
 
				-<li>Misc. doc updates.</li>
			
 
				-</ul>
			
 
				-<p>
			
 
				-Interim non-public release.
			
 
				-Special thanks to Adam D. Moss for reporting most of the bugs.
			
 
				-</p>
			
 
				-
			
 
				-<h2 id="LuaJIT-1.0.0">LuaJIT 1.0.0 &mdash; 2005-08-29</h2>
			
 
				-<p>
			
 
				-This is the initial non-public release of LuaJIT.
			
 
				-</p>
			
 
				-</div>
			
 
				 <br class="flush">
			
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/contact.html
+++ b/luajit.mod/luajit/doc/contact.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Contact</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -41,6 +40,8 @@
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -91,7 +92,7 @@ xD("fyZKB8xv\"FJytmz8.KAB0u52D")
 
				 <h2>Copyright</h2>
			
 
				 <p>
			
 
				 All documentation is
			
 
				-Copyright &copy; 2005-2017 Mike Pall.
			
 
				+Copyright &copy; 2005-2018 Mike Pall.
			
 
				 </p>
			
 
				 
			
 
				 
			
@@ -99,7 +100,7 @@ Copyright &copy; 2005-2017 Mike Pall.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_c_api.html
+++ b/luajit.mod/luajit/doc/ext_c_api.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Lua/C API Extensions</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -41,6 +40,8 @@
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a class="current" href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -89,8 +90,8 @@ other Lua/C API functions).
 
				 </p>
			
 
				 <p>
			
 
				 The third argument specifies the mode, which is 'or'ed with a flag.
			
 
				-The flag can be <tt>LUAJIT_MODE_OFF</tt> to turn a feature on,
			
 
				-<tt>LUAJIT_MODE_ON</tt> to turn a feature off, or
			
 
				+The flag can be <tt>LUAJIT_MODE_OFF</tt> to turn a feature off,
			
 
				+<tt>LUAJIT_MODE_ON</tt> to turn a feature on, or
			
 
				 <tt>LUAJIT_MODE_FLUSH</tt> to flush cached code.
			
 
				 </p>
			
 
				 <p>
			
@@ -177,7 +178,7 @@ Also note that this mechanism is not without overhead.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_ffi.html
+++ b/luajit.mod/luajit/doc/ext_ffi.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>FFI Library</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -41,6 +40,8 @@
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -320,7 +321,7 @@ without undue conversion penalties.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_ffi_api.html
+++ b/luajit.mod/luajit/doc/ext_ffi_api.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>ffi.* API Functions</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -46,6 +45,8 @@ td.abiparam { font-weight: bold; width: 6em; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -466,6 +467,10 @@ otherwise. The following parameters are currently defined:
 
				 <td class="abiparam">eabi</td><td class="abidesc">EABI variant of the standard ABI</td></tr>
			
 
				 <tr class="odd">
			
 
				 <td class="abiparam">win</td><td class="abidesc">Windows variant of the standard ABI</td></tr>
			
 
				+<tr class="even">
			
 
				+<td class="abiparam">uwp</td><td class="abidesc">Universal Windows Platform</td></tr>
			
 
				+<tr class="odd">
			
 
				+<td class="abiparam">gc64</td><td class="abidesc">64 bit GC references</td></tr>
			
 
				 </table>
			
 
				 
			
 
				 <h3 id="ffi_os"><tt>ffi.os</tt></h3>
			
@@ -542,8 +547,8 @@ corresponding ctype.
 
				 The parser for Lua source code treats numeric literals with the
			
 
				 suffixes <tt>LL</tt> or <tt>ULL</tt> as signed or unsigned 64&nbsp;bit
			
 
				 integers. Case doesn't matter, but uppercase is recommended for
			
 
				-readability. It handles both decimal (<tt>42LL</tt>) and hexadecimal
			
 
				-(<tt>0x2aLL</tt>) literals.
			
 
				+readability. It handles decimal (<tt>42LL</tt>), hexadecimal
			
 
				+(<tt>0x2aLL</tt>) and binary (<tt>0b101010LL</tt>) literals.
			
 
				 </p>
			
 
				 <p>
			
 
				 The imaginary part of complex numbers can be specified by suffixing
			
@@ -556,7 +561,7 @@ named <tt>i</tt>.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_ffi_semantics.html
+++ b/luajit.mod/luajit/doc/ext_ffi_semantics.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>FFI Semantics</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -46,6 +45,8 @@ td.convop { font-style: italic; width: 40%; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -183,6 +184,8 @@ a <tt>typedef</tt>, except re-declarations will be ignored):
 
				 <tt>uint16_t</tt>, <tt>uint32_t</tt>, <tt>uint64_t</tt>,
			
 
				 <tt>intptr_t</tt>, <tt>uintptr_t</tt>.</li>
			
 
				 
			
 
				+<li>From <tt>&lt;unistd.h&gt;</tt> (POSIX): <tt>ssize_t</tt>.</li>
			
 
				+
			
 
				 </ul>
			
 
				 <p>
			
 
				 You're encouraged to use these types in preference to
			
@@ -730,6 +733,22 @@ You'll have to explicitly convert a 64&nbsp;bit integer to a Lua
 
				 number (e.g. for regular floating-point calculations) with
			
 
				 <tt>tonumber()</tt>. But note this may incur a precision loss.</li>
			
 
				 
			
 
				+<li><b>64&nbsp;bit bitwise operations</b>: the rules for 64&nbsp;bit
			
 
				+arithmetic operators apply analogously.<br>
			
 
				+
			
 
				+Unlike the other <tt>bit.*</tt> operations, <tt>bit.tobit()</tt>
			
 
				+converts a cdata number via <tt>int64_t</tt> to <tt>int32_t</tt> and
			
 
				+returns a Lua number.<br>
			
 
				+
			
 
				+For <tt>bit.band()</tt>, <tt>bit.bor()</tt> and <tt>bit.bxor()</tt>, the
			
 
				+conversion to <tt>int64_t</tt> or <tt>uint64_t</tt> applies to
			
 
				+<em>all</em> arguments, if <em>any</em> argument is a cdata number.<br>
			
 
				+
			
 
				+For all other operations, only the first argument is used to determine
			
 
				+the output type. This implies that a cdata number as a shift count for
			
 
				+shifts and rotates is accepted, but that alone does <em>not</em> cause
			
 
				+a cdata number output.
			
 
				+
			
 
				 </ul>
			
 
				 
			
 
				 <h3 id="cdata_comp">Comparisons of cdata objects</h3>
			
@@ -844,7 +863,7 @@ place of a type, you'd need to use <tt>ffi.typeof("int")</tt> instead.
 
				 <p>
			
 
				 The main use for parameterized types are libraries implementing abstract
			
 
				 data types
			
 
				-(<a href="http://www.freelists.org/post/luajit/ffi-type-of-pointer-to,8"><span class="ext">&raquo;</span>&nbsp;example</a>),
			
 
				+(<a href="https://www.freelists.org/post/luajit/ffi-type-of-pointer-to,8">example</a>),
			
 
				 similar to what can be achieved with C++ template metaprogramming.
			
 
				 Another use case are derived types of anonymous structs, which avoids
			
 
				 pollution of the global struct namespace.
			
@@ -1201,14 +1220,12 @@ The following operations are currently not compiled and may exhibit
 
				 suboptimal performance, especially when used in inner loops:
			
 
				 </p>
			
 
				 <ul>
			
 
				-<li>Bitfield accesses and initializations.</li>
			
 
				 <li>Vector operations.</li>
			
 
				 <li>Table initializers.</li>
			
 
				 <li>Initialization of nested <tt>struct</tt>/<tt>union</tt> types.</li>
			
 
				-<li>Allocations of variable-length arrays or structs.</li>
			
 
				-<li>Allocations of C&nbsp;types with a size &gt; 128&nbsp;bytes or an
			
 
				-alignment &gt; 8&nbsp;bytes.</li>
			
 
				-<li>Conversions from lightuserdata to <tt>void&nbsp;*</tt>.</li>
			
 
				+<li>Non-default initialization of VLA/VLS or large C&nbsp;types
			
 
				+(&gt; 128&nbsp;bytes or &gt; 16 array elements.</li>
			
 
				+<li>Bitfield initializations.</li>
			
 
				 <li>Pointer differences for element sizes that are not a power of
			
 
				 two.</li>
			
 
				 <li>Calls to C&nbsp;functions with aggregates passed or returned by
			
@@ -1224,7 +1241,6 @@ value.</li>
 
				 Other missing features:
			
 
				 </p>
			
 
				 <ul>
			
 
				-<li>Bit operations for 64&nbsp;bit types.</li>
			
 
				 <li>Arithmetic for <tt>complex</tt> numbers.</li>
			
 
				 <li>Passing structs by value to vararg C&nbsp;functions.</li>
			
 
				 <li><a href="extensions.html#exceptions">C++ exception interoperability</a>
			
@@ -1235,7 +1251,7 @@ compiled.</li>
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_ffi_tutorial.html
+++ b/luajit.mod/luajit/doc/ext_ffi_tutorial.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>FFI Tutorial</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -48,6 +47,8 @@ td.idiomlua b { font-weight: normal; color: #2142bf; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -591,7 +592,7 @@ it to a local variable in the function scope is unnecessary.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_jit.html
+++ b/luajit.mod/luajit/doc/ext_jit.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>jit.* Library</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -41,6 +40,8 @@
 
				 <a class="current" href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -151,7 +152,7 @@ Contains the target OS name:
 
				 <h3 id="jit_arch"><tt>jit.arch</tt></h3>
			
 
				 <p>
			
 
				 Contains the target architecture name:
			
 
				-"x86", "x64", "arm", "ppc", "ppcspe", or "mips".
			
 
				+"x86", "x64", "arm", "arm64", "ppc", "mips" or "mips64".
			
 
				 </p>
			
 
				 
			
 
				 <h2 id="jit_opt"><tt>jit.opt.*</tt> &mdash; JIT compiler optimization control</h2>
			
@@ -189,7 +190,7 @@ if you want to know more.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/ext_profiler.html
+++ b/luajit.mod/luajit/doc/ext_profiler.html
@@ -0,0 +1,364 @@
 
				+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
			
 
				+<html>
			
 
				+<head>
			
 
				+<title>Profiler</title>
			
 
				+<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				+<meta name="Language" content="en">
			
 
				+<link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				+<link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
 
				+</head>
			
 
				+<body>
			
 
				+<div id="site">
			
 
				+<a href="http://luajit.org"><span>Lua<span id="logo">JIT</span></span></a>
			
 
				+</div>
			
 
				+<div id="head">
			
 
				+<h1>Profiler</h1>
			
 
				+</div>
			
 
				+<div id="nav">
			
 
				+<ul><li>
			
 
				+<a href="luajit.html">LuaJIT</a>
			
 
				+<ul><li>
			
 
				+<a href="http://luajit.org/download.html">Download <span class="ext">&raquo;</span></a>
			
 
				+</li><li>
			
 
				+<a href="install.html">Installation</a>
			
 
				+</li><li>
			
 
				+<a href="running.html">Running</a>
			
 
				+</li></ul>
			
 
				+</li><li>
			
 
				+<a href="extensions.html">Extensions</a>
			
 
				+<ul><li>
			
 
				+<a href="ext_ffi.html">FFI Library</a>
			
 
				+<ul><li>
			
 
				+<a href="ext_ffi_tutorial.html">FFI Tutorial</a>
			
 
				+</li><li>
			
 
				+<a href="ext_ffi_api.html">ffi.* API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_ffi_semantics.html">FFI Semantics</a>
			
 
				+</li></ul>
			
 
				+</li><li>
			
 
				+<a href="ext_jit.html">jit.* Library</a>
			
 
				+</li><li>
			
 
				+<a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a class="current" href="ext_profiler.html">Profiler</a>
			
 
				+</li></ul>
			
 
				+</li><li>
			
 
				+<a href="status.html">Status</a>
			
 
				+<ul><li>
			
 
				+<a href="changes.html">Changes</a>
			
 
				+</li></ul>
			
 
				+</li><li>
			
 
				+<a href="faq.html">FAQ</a>
			
 
				+</li><li>
			
 
				+<a href="http://luajit.org/performance.html">Performance <span class="ext">&raquo;</span></a>
			
 
				+</li><li>
			
 
				+<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
			
 
				+</li><li>
			
 
				+<a href="http://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
			
 
				+</li></ul>
			
 
				+</div>
			
 
				+<div id="main">
			
 
				+<p>
			
 
				+LuaJIT has an integrated statistical profiler with very low overhead. It
			
 
				+allows sampling the currently executing stack and other parameters in
			
 
				+regular intervals.
			
 
				+</p>
			
 
				+<p>
			
 
				+The integrated profiler can be accessed from three levels:
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li>The <a href="#hl_profiler">bundled high-level profiler</a>, invoked by the
			
 
				+<a href="#j_p"><tt>-jp</tt></a> command line option.</li>
			
 
				+<li>A <a href="#ll_lua_api">low-level Lua API</a> to control the profiler.</li>
			
 
				+<li>A <a href="#ll_c_api">low-level C API</a> to control the profiler.</li>
			
 
				+</ul>
			
 
				+
			
 
				+<h2 id="hl_profiler">High-Level Profiler</h2>
			
 
				+<p>
			
 
				+The bundled high-level profiler offers basic profiling functionality. It
			
 
				+generates simple textual summaries or source code annotations. It can be
			
 
				+accessed with the <a href="#j_p"><tt>-jp</tt></a> command line option
			
 
				+or from Lua code by loading the underlying <tt>jit.p</tt> module.
			
 
				+</p>
			
 
				+<p>
			
 
				+To cut to the chase &mdash; run this to get a CPU usage profile by
			
 
				+function name:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+luajit -jp myapp.lua
			
 
				+</pre>
			
 
				+<p>
			
 
				+It's <em>not</em> a stated goal of the bundled profiler to add every
			
 
				+possible option or to cater for special profiling needs. The low-level
			
 
				+profiler APIs are documented below. They may be used by third-party
			
 
				+authors to implement advanced functionality, e.g. IDE integration or
			
 
				+graphical profilers.
			
 
				+</p>
			
 
				+<p>
			
 
				+Note: Sampling works for both interpreted and JIT-compiled code. The
			
 
				+results for JIT-compiled code may sometimes be surprising. LuaJIT
			
 
				+heavily optimizes and inlines Lua code &mdash; there's no simple
			
 
				+one-to-one correspondence between source code lines and the sampled
			
 
				+machine code.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="j_p"><tt>-jp=[options[,output]]</tt></h3>
			
 
				+<p>
			
 
				+The <tt>-jp</tt> command line option starts the high-level profiler.
			
 
				+When the application run by the command line terminates, the profiler
			
 
				+stops and writes the results to <tt>stdout</tt> or to the specified
			
 
				+<tt>output</tt> file.
			
 
				+</p>
			
 
				+<p>
			
 
				+The <tt>options</tt> argument specifies how the profiling is to be
			
 
				+performed:
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li><tt>f</tt> &mdash; Stack dump: function name, otherwise module:line.
			
 
				+This is the default mode.</li>
			
 
				+<li><tt>F</tt> &mdash; Stack dump: ditto, but dump module:name.</li>
			
 
				+<li><tt>l</tt> &mdash; Stack dump: module:line.</li>
			
 
				+<li><tt>&lt;number&gt;</tt> &mdash; stack dump depth (callee &larr;
			
 
				+caller). Default: 1.</li>
			
 
				+<li><tt>-&lt;number&gt;</tt> &mdash; Inverse stack dump depth (caller
			
 
				+&rarr; callee).</li>
			
 
				+<li><tt>s</tt> &mdash; Split stack dump after first stack level. Implies
			
 
				+depth&nbsp;&ge;&nbsp;2 or depth&nbsp;&le;&nbsp;-2.</li>
			
 
				+<li><tt>p</tt> &mdash; Show full path for module names.</li>
			
 
				+<li><tt>v</tt> &mdash; Show VM states.</li>
			
 
				+<li><tt>z</tt> &mdash; Show <a href="#jit_zone">zones</a>.</li>
			
 
				+<li><tt>r</tt> &mdash; Show raw sample counts. Default: show percentages.</li>
			
 
				+<li><tt>a</tt> &mdash; Annotate excerpts from source code files.</li>
			
 
				+<li><tt>A</tt> &mdash; Annotate complete source code files.</li>
			
 
				+<li><tt>G</tt> &mdash; Produce raw output suitable for graphical tools.</li>
			
 
				+<li><tt>m&lt;number&gt;</tt> &mdash; Minimum sample percentage to be shown.
			
 
				+Default: 3%.</li>
			
 
				+<li><tt>i&lt;number&gt;</tt> &mdash; Sampling interval in milliseconds.
			
 
				+Default: 10ms.<br>
			
 
				+Note: The actual sampling precision is OS-dependent.</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				+The default output for <tt>-jp</tt> is a list of the most CPU consuming
			
 
				+spots in the application. Increasing the stack dump depth with (say)
			
 
				+<tt>-jp=2</tt> may help to point out the main callers or callees of
			
 
				+hotspots. But sample aggregation is still flat per unique stack dump.
			
 
				+</p>
			
 
				+<p>
			
 
				+To get a two-level view (split view) of callers/callees, use
			
 
				+<tt>-jp=s</tt> or <tt>-jp=-s</tt>. The percentages shown for the second
			
 
				+level are relative to the first level.
			
 
				+</p>
			
 
				+<p>
			
 
				+To see how much time is spent in each line relative to a function, use
			
 
				+<tt>-jp=fl</tt>.
			
 
				+</p>
			
 
				+<p>
			
 
				+To see how much time is spent in different VM states or
			
 
				+<a href="#jit_zone">zones</a>, use <tt>-jp=v</tt> or <tt>-jp=z</tt>.
			
 
				+</p>
			
 
				+<p>
			
 
				+Combinations of <tt>v/z</tt> with <tt>f/F/l</tt> produce two-level
			
 
				+views, e.g. <tt>-jp=vf</tt> or <tt>-jp=fv</tt>. This shows the time
			
 
				+spent in a VM state or zone vs. hotspots. This can be used to answer
			
 
				+questions like "Which time consuming functions are only interpreted?" or
			
 
				+"What's the garbage collector overhead for a specific function?".
			
 
				+</p>
			
 
				+<p>
			
 
				+Multiple options can be combined &mdash; but not all combinations make
			
 
				+sense, see above. E.g. <tt>-jp=3si4m1</tt> samples three stack levels
			
 
				+deep in 4ms intervals and shows a split view of the CPU consuming
			
 
				+functions and their callers with a 1% threshold.
			
 
				+</p>
			
 
				+<p>
			
 
				+Source code annotations produced by <tt>-jp=a</tt> or <tt>-jp=A</tt> are
			
 
				+always flat and at the line level. Obviously, the source code files need
			
 
				+to be readable by the profiler script.
			
 
				+</p>
			
 
				+<p>
			
 
				+The high-level profiler can also be started and stopped from Lua code with:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+require("jit.p").start(options, output)
			
 
				+...
			
 
				+require("jit.p").stop()
			
 
				+</pre>
			
 
				+
			
 
				+<h3 id="jit_zone"><tt>jit.zone</tt> &mdash; Zones</h3>
			
 
				+<p>
			
 
				+Zones can be used to provide information about different parts of an
			
 
				+application to the high-level profiler. E.g. a game could make use of an
			
 
				+<tt>"AI"</tt> zone, a <tt>"PHYS"</tt> zone, etc. Zones are hierarchical,
			
 
				+organized as a stack.
			
 
				+</p>
			
 
				+<p>
			
 
				+The <tt>jit.zone</tt> module needs to be loaded explicitly:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+local zone = require("jit.zone")
			
 
				+</pre>
			
 
				+<ul>
			
 
				+<li><tt>zone("name")</tt> pushes a named zone to the zone stack.</li>
			
 
				+<li><tt>zone()</tt> pops the current zone from the zone stack and
			
 
				+returns its name.</li>
			
 
				+<li><tt>zone:get()</tt> returns the current zone name or <tt>nil</tt>.</li>
			
 
				+<li><tt>zone:flush()</tt> flushes the zone stack.</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				+To show the time spent in each zone use <tt>-jp=z</tt>. To show the time
			
 
				+spent relative to hotspots use e.g. <tt>-jp=zf</tt> or <tt>-jp=fz</tt>.
			
 
				+</p>
			
 
				+
			
 
				+<h2 id="ll_lua_api">Low-level Lua API</h2>
			
 
				+<p>
			
 
				+The <tt>jit.profile</tt> module gives access to the low-level API of the
			
 
				+profiler from Lua code. This module needs to be loaded explicitly:
			
 
				+<pre class="code">
			
 
				+local profile = require("jit.profile")
			
 
				+</pre>
			
 
				+<p>
			
 
				+This module can be used to implement your own higher-level profiler.
			
 
				+A typical profiling run starts the profiler, captures stack dumps in
			
 
				+the profiler callback, adds them to a hash table to aggregate the number
			
 
				+of samples, stops the profiler and then analyzes all of the captured
			
 
				+stack dumps. Other parameters can be sampled in the profiler callback,
			
 
				+too. But it's important not to spend too much time in the callback,
			
 
				+since this may skew the statistics.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="profile_start"><tt>profile.start(mode, cb)</tt>
			
 
				+&mdash; Start profiler</h3>
			
 
				+<p>
			
 
				+This function starts the profiler. The <tt>mode</tt> argument is a
			
 
				+string holding options:
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li><tt>f</tt> &mdash; Profile with precision down to the function level.</li>
			
 
				+<li><tt>l</tt> &mdash; Profile with precision down to the line level.</li>
			
 
				+<li><tt>i&lt;number&gt;</tt> &mdash; Sampling interval in milliseconds (default
			
 
				+10ms).</br>
			
 
				+Note: The actual sampling precision is OS-dependent.
			
 
				+</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				+The <tt>cb</tt> argument is a callback function which is called with
			
 
				+three arguments: <tt>(thread, samples, vmstate)</tt>. The callback is
			
 
				+called on a separate coroutine, the <tt>thread</tt> argument is the
			
 
				+state that holds the stack to sample for profiling. Note: do
			
 
				+<em>not</em> modify the stack of that state or call functions on it.
			
 
				+</p>
			
 
				+<p>
			
 
				+<tt>samples</tt> gives the number of accumulated samples since the last
			
 
				+callback (usually 1).
			
 
				+</p>
			
 
				+<p>
			
 
				+<tt>vmstate</tt> holds the VM state at the time the profiling timer
			
 
				+triggered. This may or may not correspond to the state of the VM when
			
 
				+the profiling callback is called. The state is either <tt>'N'</tt>
			
 
				+native (compiled) code, <tt>'I'</tt> interpreted code, <tt>'C'</tt>
			
 
				+C&nbsp;code, <tt>'G'</tt> the garbage collector, or <tt>'J'</tt> the JIT
			
 
				+compiler.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="profile_stop"><tt>profile.stop()</tt>
			
 
				+&mdash; Stop profiler</h3>
			
 
				+<p>
			
 
				+This function stops the profiler.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="profile_dump"><tt>dump = profile.dumpstack([thread,] fmt, depth)</tt>
			
 
				+&mdash; Dump stack </h3>
			
 
				+<p>
			
 
				+This function allows taking stack dumps in an efficient manner. It
			
 
				+returns a string with a stack dump for the <tt>thread</tt> (coroutine),
			
 
				+formatted according to the <tt>fmt</tt> argument:
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li><tt>p</tt> &mdash; Preserve the full path for module names. Otherwise
			
 
				+only the file name is used.</li>
			
 
				+<li><tt>f</tt> &mdash; Dump the function name if it can be derived. Otherwise
			
 
				+use module:line.</li>
			
 
				+<li><tt>F</tt> &mdash; Ditto, but dump module:name.</li>
			
 
				+<li><tt>l</tt> &mdash; Dump module:line.</li>
			
 
				+<li><tt>Z</tt> &mdash; Zap the following characters for the last dumped
			
 
				+frame.</li>
			
 
				+<li>All other characters are added verbatim to the output string.</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				+The <tt>depth</tt> argument gives the number of frames to dump, starting
			
 
				+at the topmost frame of the thread. A negative number dumps the frames in
			
 
				+inverse order.
			
 
				+</p>
			
 
				+<p>
			
 
				+The first example prints a list of the current module names and line
			
 
				+numbers of up to 10 frames in separate lines. The second example prints
			
 
				+semicolon-separated function names for all frames (up to 100) in inverse
			
 
				+order:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+print(profile.dumpstack(thread, "l\n", 10))
			
 
				+print(profile.dumpstack(thread, "lZ;", -100))
			
 
				+</pre>
			
 
				+
			
 
				+<h2 id="ll_c_api">Low-level C API</h2>
			
 
				+<p>
			
 
				+The profiler can be controlled directly from C&nbsp;code, e.g. for
			
 
				+use by IDEs. The declarations are in <tt>"luajit.h"</tt> (see
			
 
				+<a href="ext_c_api.html">Lua/C API</a> extensions).
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="luaJIT_profile_start"><tt>luaJIT_profile_start(L, mode, cb, data)</tt>
			
 
				+&mdash; Start profiler</h3>
			
 
				+<p>
			
 
				+This function starts the profiler. <a href="#profile_start">See
			
 
				+above</a> for a description of the <tt>mode</tt> argument.
			
 
				+</p>
			
 
				+<p>
			
 
				+The <tt>cb</tt> argument is a callback function with the following
			
 
				+declaration:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+typedef void (*luaJIT_profile_callback)(void *data, lua_State *L,
			
 
				+                                        int samples, int vmstate);
			
 
				+</pre>
			
 
				+<p>
			
 
				+<tt>data</tt> is available for use by the callback. <tt>L</tt> is the
			
 
				+state that holds the stack to sample for profiling. Note: do
			
 
				+<em>not</em> modify this stack or call functions on this stack &mdash;
			
 
				+use a separate coroutine for this purpose. <a href="#profile_start">See
			
 
				+above</a> for a description of <tt>samples</tt> and <tt>vmstate</tt>.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="luaJIT_profile_stop"><tt>luaJIT_profile_stop(L)</tt>
			
 
				+&mdash; Stop profiler</h3>
			
 
				+<p>
			
 
				+This function stops the profiler.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="luaJIT_profile_dumpstack"><tt>p = luaJIT_profile_dumpstack(L, fmt, depth, len)</tt>
			
 
				+&mdash; Dump stack </h3>
			
 
				+<p>
			
 
				+This function allows taking stack dumps in an efficient manner.
			
 
				+<a href="#profile_dump">See above</a> for a description of <tt>fmt</tt>
			
 
				+and <tt>depth</tt>.
			
 
				+</p>
			
 
				+<p>
			
 
				+This function returns a <tt>const&nbsp;char&nbsp;*</tt> pointing to a
			
 
				+private string buffer of the profiler. The <tt>int&nbsp;*len</tt>
			
 
				+argument returns the length of the output string. The buffer is
			
 
				+overwritten on the next call and deallocated when the profiler stops.
			
 
				+You either need to consume the content immediately or copy it for later
			
 
				+use.
			
 
				+</p>
			
 
				+<br class="flush">
			
 
				+</div>
			
 
				+<div id="foot">
			
 
				+<hr class="hide">
			
 
				+Copyright &copy; 2005-2018
			
 
				+<span class="noprint">
			
 
				+&middot;
			
 
				+<a href="contact.html">Contact</a>
			
 
				+</span>
			
 
				+</div>
			
 
				+</body>
			
 
				+</html>
			
--- a/luajit.mod/luajit/doc/extensions.html
+++ b/luajit.mod/luajit/doc/extensions.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Extensions</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -58,6 +57,8 @@ td.excinterop {
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -113,6 +114,9 @@ bit.lshift bit.rshift bit.arshift bit.rol  bit.ror  bit.bswap
 
				 This module is a LuaJIT built-in &mdash; you don't need to download or
			
 
				 install Lua BitOp. The Lua BitOp site has full documentation for all
			
 
				 <a href="http://bitop.luajit.org/api.html"><span class="ext">&raquo;</span>&nbsp;Lua BitOp API functions</a>.
			
 
				+The FFI adds support for
			
 
				+<a href="ext_ffi_semantics.html#cdata_arith">64&nbsp;bit bitwise operations</a>,
			
 
				+using the same API functions.
			
 
				 </p>
			
 
				 <p>
			
 
				 Please make sure to <tt>require</tt> the module before using any of
			
@@ -146,6 +150,11 @@ LuaJIT adds some
 
				 <a href="ext_c_api.html">extra functions to the Lua/C API</a>.
			
 
				 </p>
			
 
				 
			
 
				+<h3 id="profiler">Profiler</h3>
			
 
				+<p>
			
 
				+LuaJIT has an <a href="ext_profiler.html">integrated profiler</a>.
			
 
				+</p>
			
 
				+
			
 
				 <h2 id="library">Enhanced Standard Library Functions</h2>
			
 
				 
			
 
				 <h3 id="xpcall"><tt>xpcall(f, err [,args...])</tt> passes arguments</h3>
			
@@ -173,7 +182,7 @@ in <tt>"-inf"</tt>.
 
				 <h3 id="tonumber"><tt>tonumber()</tt> etc. use builtin string to number conversion</h3>
			
 
				 <p>
			
 
				 All string-to-number conversions consistently convert integer and
			
 
				-floating-point inputs in decimal and hexadecimal on all platforms.
			
 
				+floating-point inputs in decimal, hexadecimal and binary on all platforms.
			
 
				 <tt>strtod()</tt> is <em>not</em> used anymore, which avoids numerous
			
 
				 problems with poor C library implementations. The builtin conversion
			
 
				 function provides full precision according to the IEEE-754 standard, it
			
@@ -197,6 +206,37 @@ for dot releases (x.y.0 &rarr; x.y.1), but may change with major or
 
				 minor releases (2.0 &rarr; 2.1) or between any beta release. Foreign
			
 
				 bytecode (e.g. from Lua 5.1) is incompatible and cannot be loaded.
			
 
				 </p>
			
 
				+<p>
			
 
				+Note: <tt>LJ_GC64</tt> mode requires a different frame layout, which implies
			
 
				+a different, incompatible bytecode format for ports that use this mode (e.g.
			
 
				+ARM64 or MIPS64) or when explicitly enabled for x64. This may be rectified
			
 
				+in the future.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="table_new"><tt>table.new(narray, nhash)</tt> allocates a pre-sized table</h3>
			
 
				+<p>
			
 
				+An extra library function <tt>table.new()</tt> can be made available via
			
 
				+<tt>require("table.new")</tt>. This creates a pre-sized table, just like
			
 
				+the C API equivalent <tt>lua_createtable()</tt>. This is useful for big
			
 
				+tables if the final table size is known and automatic table resizing is
			
 
				+too expensive.
			
 
				+</p>
			
 
				+
			
 
				+<h3 id="table_clear"><tt>table.clear(tab)</tt> clears a table</h3>
			
 
				+<p>
			
 
				+An extra library function <tt>table.clear()</tt> can be made available
			
 
				+via <tt>require("table.clear")</tt>. This clears all keys and values
			
 
				+from a table, but preserves the allocated array/hash sizes. This is
			
 
				+useful when a table, which is linked from multiple places, needs to be
			
 
				+cleared and/or when recycling a table for use by the same context. This
			
 
				+avoids managing backlinks, saves an allocation and the overhead of
			
 
				+incremental array/hash part growth.
			
 
				+</p>
			
 
				+<p>
			
 
				+Please note this function is meant for very specific situations. In most
			
 
				+cases it's better to replace the (usually single) link with a new table
			
 
				+and let the GC do its work.
			
 
				+</p>
			
 
				 
			
 
				 <h3 id="math_random">Enhanced PRNG for <tt>math.random()</tt></h3>
			
 
				 <p>
			
@@ -271,6 +311,26 @@ indexes for varargs.</li>
 
				 <li><tt>debug.getupvalue()</tt> and <tt>debug.setupvalue()</tt> handle
			
 
				 C&nbsp;functions.</li>
			
 
				 <li><tt>debug.upvalueid()</tt> and <tt>debug.upvaluejoin()</tt>.</li>
			
 
				+<li>Lua/C API extensions:
			
 
				+<tt>lua_version()</tt>
			
 
				+<tt>lua_upvalueid()</tt>
			
 
				+<tt>lua_upvaluejoin()</tt>
			
 
				+<tt>lua_loadx()</tt>
			
 
				+<tt>lua_copy()</tt>
			
 
				+<tt>lua_tonumberx()</tt>
			
 
				+<tt>lua_tointegerx()</tt>
			
 
				+<tt>luaL_fileresult()</tt>
			
 
				+<tt>luaL_execresult()</tt>
			
 
				+<tt>luaL_loadfilex()</tt>
			
 
				+<tt>luaL_loadbufferx()</tt>
			
 
				+<tt>luaL_traceback()</tt>
			
 
				+<tt>luaL_setfuncs()</tt>
			
 
				+<tt>luaL_pushmodule()</tt>
			
 
				+<tt>luaL_newlibtable()</tt>
			
 
				+<tt>luaL_newlib()</tt>
			
 
				+<tt>luaL_testudata()</tt>
			
 
				+<tt>luaL_setmetatable()</tt>
			
 
				+</li>
			
 
				 <li>Command line option <tt>-E</tt>.</li>
			
 
				 <li>Command line checks <tt>__tostring</tt> for errors.</li>
			
 
				 </ul>
			
@@ -296,6 +356,8 @@ exit status.</li>
 
				 <li><tt>debug.setmetatable()</tt> returns object.</li>
			
 
				 <li><tt>debug.getuservalue()</tt> and <tt>debug.setuservalue()</tt>.</li>
			
 
				 <li>Remove <tt>math.mod()</tt>, <tt>string.gfind()</tt>.</li>
			
 
				+<li><tt>package.searchers</tt>.</li>
			
 
				+<li><tt>module()</tt> returns the module table.</li>
			
 
				 </ul>
			
 
				 <p>
			
 
				 Note: this provides only partial compatibility with Lua 5.2 at the
			
@@ -304,6 +366,21 @@ Lua&nbsp;5.1, which prevents implementing features that would otherwise
 
				 break the Lua/C API and ABI (e.g. <tt>_ENV</tt>).
			
 
				 </p>
			
 
				 
			
 
				+<h2 id="lua53">Extensions from Lua 5.3</h2>
			
 
				+<p>
			
 
				+LuaJIT supports some extensions from Lua&nbsp;5.3:
			
 
				+<ul>
			
 
				+<li>Unicode escape <tt>'\u{XX...}'</tt> embeds the UTF-8 encoding in string literals.</li>
			
 
				+<li>The argument table <tt>arg</tt> can be read (and modified) by <tt>LUA_INIT</tt> and <tt>-e</tt> chunks.</li>
			
 
				+<li><tt>io.read()</tt> and <tt>file:read()</tt> accept formats with or without a leading <tt>*</tt>.</li>
			
 
				+<li><tt>assert()</tt> accepts any type of error object.</li>
			
 
				+<li><tt>table.move(a1, f, e, t [,a2])</tt>.</li>
			
 
				+<li><tt>coroutine.isyieldable()</tt>.</li>
			
 
				+<li>Lua/C API extensions:
			
 
				+<tt>lua_isyieldable()</tt>
			
 
				+</li>
			
 
				+</ul>
			
 
				+
			
 
				 <h2 id="exceptions">C++ Exception Interoperability</h2>
			
 
				 <p>
			
 
				 LuaJIT has built-in support for interoperating with C++&nbsp;exceptions.
			
@@ -318,25 +395,30 @@ the toolchain used to compile LuaJIT:
 
				 </tr>
			
 
				 <tr class="odd separate">
			
 
				 <td class="excplatform">POSIX/x64, DWARF2 unwinding</td>
			
 
				-<td class="exccompiler">GCC 4.3+</td>
			
 
				+<td class="exccompiler">GCC 4.3+, Clang</td>
			
 
				 <td class="excinterop"><b style="color: #00a000;">Full</b></td>
			
 
				 </tr>
			
 
				 <tr class="even">
			
 
				+<td class="excplatform">ARM <tt>-DLUAJIT_UNWIND_EXTERNAL</tt></td>
			
 
				+<td class="exccompiler">GCC, Clang</td>
			
 
				+<td class="excinterop"><b style="color: #00a000;">Full</b></td>
			
 
				+</tr>
			
 
				+<tr class="odd">
			
 
				 <td class="excplatform">Other platforms, DWARF2 unwinding</td>
			
 
				-<td class="exccompiler">GCC</td>
			
 
				+<td class="exccompiler">GCC, Clang</td>
			
 
				 <td class="excinterop"><b style="color: #c06000;">Limited</b></td>
			
 
				 </tr>
			
 
				-<tr class="odd">
			
 
				+<tr class="even">
			
 
				 <td class="excplatform">Windows/x64</td>
			
 
				 <td class="exccompiler">MSVC or WinSDK</td>
			
 
				 <td class="excinterop"><b style="color: #00a000;">Full</b></td>
			
 
				 </tr>
			
 
				-<tr class="even">
			
 
				+<tr class="odd">
			
 
				 <td class="excplatform">Windows/x86</td>
			
 
				 <td class="exccompiler">Any</td>
			
 
				-<td class="excinterop"><b style="color: #a00000;">No</b></td>
			
 
				+<td class="excinterop"><b style="color: #00a000;">Full</b></td>
			
 
				 </tr>
			
 
				-<tr class="odd">
			
 
				+<tr class="even">
			
 
				 <td class="excplatform">Other platforms</td>
			
 
				 <td class="exccompiler">Other compilers</td>
			
 
				 <td class="excinterop"><b style="color: #a00000;">No</b></td>
			
@@ -385,20 +467,12 @@ C++ destructors.</li>
 
				 <li>Lua errors <b>cannot</b> be caught on the C++ side.</li>
			
 
				 <li>Throwing Lua errors across C++ frames will <b>not</b> call
			
 
				 C++ destructors.</li>
			
 
				-<li>Additionally, on Windows/x86 with SEH-based C++&nbsp;exceptions:
			
 
				-it's <b>not</b> safe to throw a Lua error across any frames containing
			
 
				-a C++ function with any try/catch construct or using variables with
			
 
				-(implicit) destructors. This also applies to any functions which may be
			
 
				-inlined in such a function. It doesn't matter whether <tt>lua_error()</tt>
			
 
				-is called inside or outside of a try/catch or whether any object actually
			
 
				-needs to be destroyed: the SEH chain is corrupted and this will eventually
			
 
				-lead to the termination of the process.</li>
			
 
				 </ul>
			
 
				 <br class="flush">
			
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/faq.html
+++ b/luajit.mod/luajit/doc/faq.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Frequently Asked Questions (FAQ)</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -44,6 +43,8 @@ dd { margin-left: 1.5em; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -174,7 +175,7 @@ the development of certain features, if they are important to you.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/install.html
+++ b/luajit.mod/luajit/doc/install.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Installation</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -69,6 +68,8 @@ td.compatno {
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -112,17 +113,17 @@ operating systems, CPUs and compilers:
 
				 </tr>
			
 
				 <tr class="odd separate">
			
 
				 <td class="compatcpu">x86 (32 bit)</td>
			
 
				-<td class="compatos">GCC 4.x+<br>GCC 3.4</td>
			
 
				-<td class="compatos">GCC 4.x+<br>GCC 3.4</td>
			
 
				+<td class="compatos">GCC 4.2+</td>
			
 
				+<td class="compatos">GCC 4.2+</td>
			
 
				 <td class="compatos">XCode 5.0+<br>Clang</td>
			
 
				 <td class="compatos">MSVC, MSVC/EE<br>WinSDK<br>MinGW, Cygwin</td>
			
 
				 </tr>
			
 
				 <tr class="even">
			
 
				 <td class="compatcpu">x64 (64 bit)</td>
			
 
				-<td class="compatos">GCC 4.x+</td>
			
 
				-<td class="compatos">ORBIS (<a href="#ps4">PS4</a>)</td>
			
 
				+<td class="compatos">GCC 4.2+</td>
			
 
				+<td class="compatos">GCC 4.2+<br>ORBIS (<a href="#ps4">PS4</a>)</td>
			
 
				 <td class="compatos">XCode 5.0+<br>Clang</td>
			
 
				-<td class="compatos">MSVC + SDK v7.0<br>WinSDK v7.0</td>
			
 
				+<td class="compatos">MSVC + SDK v7.0<br>WinSDK v7.0<br>Durango (<a href="#xboxone">Xbox One</a>)</td>
			
 
				 </tr>
			
 
				 <tr class="odd">
			
 
				 <td class="compatcpu"><a href="#cross2">ARMv5+<br>ARM9E+</a></td>
			
@@ -132,21 +133,21 @@ operating systems, CPUs and compilers:
 
				 <td class="compatos compatno">&nbsp;</td>
			
 
				 </tr>
			
 
				 <tr class="even">
			
 
				-<td class="compatcpu"><a href="#cross2">PPC</a></td>
			
 
				-<td class="compatos">GCC 4.3+</td>
			
 
				-<td class="compatos">GCC 4.3+<br>GCC 4.1 (<a href="#ps3">PS3</a>)</td>
			
 
				+<td class="compatcpu"><a href="#cross2">ARM64</a></td>
			
 
				+<td class="compatos">GCC 4.8+</td>
			
 
				+<td class="compatos compatno">&nbsp;</td>
			
 
				+<td class="compatos">XCode 6.0+<br>Clang 3.5+</td>
			
 
				 <td class="compatos compatno">&nbsp;</td>
			
 
				-<td class="compatos">XEDK (<a href="#xbox360">Xbox 360</a>)</td>
			
 
				 </tr>
			
 
				 <tr class="odd">
			
 
				-<td class="compatcpu"><a href="#cross2">PPC/e500v2</a></td>
			
 
				-<td class="compatos">GCC 4.3+</td>
			
 
				+<td class="compatcpu"><a href="#cross2">PPC</a></td>
			
 
				 <td class="compatos">GCC 4.3+</td>
			
 
				+<td class="compatos">GCC 4.3+<br>GCC 4.1 (<a href="#ps3">PS3</a>)</td>
			
 
				 <td class="compatos compatno">&nbsp;</td>
			
 
				-<td class="compatos compatno">&nbsp;</td>
			
 
				+<td class="compatos">XEDK (<a href="#xbox360">Xbox 360</a>)</td>
			
 
				 </tr>
			
 
				 <tr class="even">
			
 
				-<td class="compatcpu"><a href="#cross2">MIPS</a></td>
			
 
				+<td class="compatcpu"><a href="#cross2">MIPS32<br>MIPS64</a></td>
			
 
				 <td class="compatos">GCC 4.3+</td>
			
 
				 <td class="compatos">GCC 4.3+</td>
			
 
				 <td class="compatos compatno">&nbsp;</td>
			
@@ -173,6 +174,14 @@ MSVC or WinSDK.</li>
 
				 Please read the instructions given in these files, before changing
			
 
				 any settings.
			
 
				 </p>
			
 
				+<p>
			
 
				+LuaJIT on x64 currently uses 32 bit GC objects by default.
			
 
				+<tt>LJ_GC64</tt> mode may be explicitly enabled:
			
 
				+add <tt>XCFLAGS=-DLUAJIT_ENABLE_GC64</tt> to the make command or run
			
 
				+<tt>msvcbuild gc64</tt> for MSVC/WinSDK. Please check the note
			
 
				+about the <a href="extensions.html#string_dump">bytecode format</a>
			
 
				+differences, too.
			
 
				+</p>
			
 
				 
			
 
				 <h2 id="posix">POSIX Systems (Linux, OSX, *BSD etc.)</h2>
			
 
				 <h3>Prerequisites</h3>
			
@@ -200,7 +209,7 @@ which is probably the default on your system, anyway. Simply run:
 
				 make
			
 
				 </pre>
			
 
				 <p>
			
 
				-This always builds a native x86, x64 or PPC binary, depending on the host OS
			
 
				+This always builds a native binary, depending on the host OS
			
 
				 you're running this command on. Check the section on
			
 
				 <a href="#cross">cross-compilation</a> for more options.
			
 
				 </p>
			
@@ -331,25 +340,36 @@ directory where <tt>luajit.exe</tt> is installed
 
				 
			
 
				 <h2 id="cross">Cross-compiling LuaJIT</h2>
			
 
				 <p>
			
 
				+First, let's clear up some terminology:
			
 
				+</p>
			
 
				+<ul>
			
 
				+<li>Host: This is your development system, usually based on a x64 or x86 CPU.</li>
			
 
				+<li>Target: This is the target system you want LuaJIT to run on, e.g. Android/ARM.</li>
			
 
				+<li>Toolchain: This comprises a C compiler, linker, assembler and a matching C library.</li>
			
 
				+<li>Host (or system) toolchain: This is the toolchain used to build native binaries for your host system.</li>
			
 
				+<li>Cross-compile toolchain: This is the toolchain used to build binaries for the target system. They can only be run on the target system.</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				 The GNU Makefile-based build system allows cross-compiling on any host
			
 
				-for any supported target, as long as both architectures have the same
			
 
				-pointer size. If you want to cross-compile to any 32 bit target on an
			
 
				-x64 OS, you need to install the multilib development package (e.g.
			
 
				-<tt>libc6-dev-i386</tt> on Debian/Ubuntu) and build a 32 bit host part
			
 
				-(<tt>HOST_CC="gcc -m32"</tt>).
			
 
				+for any supported target:
			
 
				 </p>
			
 
				+<ul>
			
 
				+<li>Yes, you need a toolchain for both your host <em>and</em> your target!</li>
			
 
				+<li>Both host and target architectures must have the same pointer size.</li>
			
 
				+<li>E.g. if you want to cross-compile to a 32 bit target on a 64 bit host, you need to install the multilib development package (e.g. <tt>libc6-dev-i386</tt> on Debian/Ubuntu) and build a 32 bit host part (<tt>HOST_CC="gcc -m32"</tt>).</li>
			
 
				+<li>64 bit targets always require compilation on a 64 bit host.</li>
			
 
				+</ul>
			
 
				 <p>
			
 
				 You need to specify <tt>TARGET_SYS</tt> whenever the host OS and the
			
 
				-target OS differ, or you'll get assembler or linker errors. E.g. if
			
 
				-you're compiling on a Windows or OSX host for embedded Linux or Android,
			
 
				-you need to add <tt>TARGET_SYS=Linux</tt> to the examples below. For a
			
 
				-minimal target OS, you may need to disable the built-in allocator in
			
 
				-<tt>src/Makefile</tt> and use <tt>TARGET_SYS=Other</tt>. Don't forget to
			
 
				-specify the same <tt>TARGET_SYS</tt> for the install step, too.
			
 
				+target OS differ, or you'll get assembler or linker errors:
			
 
				 </p>
			
 
				+<ul>
			
 
				+<li>E.g. if you're compiling on a Windows or OSX host for embedded Linux or Android, you need to add <tt>TARGET_SYS=Linux</tt> to the examples below.</li>
			
 
				+<li>For a minimal target OS, you may need to disable the built-in allocator in <tt>src/Makefile</tt> and use <tt>TARGET_SYS=Other</tt>.</li>
			
 
				+<li>Don't forget to specify the same <tt>TARGET_SYS</tt> for the install step, too.</li>
			
 
				+</ul>
			
 
				 <p>
			
 
				-The examples below only show some popular targets &mdash; please check
			
 
				-the comments in <tt>src/Makefile</tt> for more details.
			
 
				+Here are some examples where host and target have the same CPU:
			
 
				 </p>
			
 
				 <pre class="code">
			
 
				 # Cross-compile to a 32 bit binary on a multilib x64 OS
			
@@ -367,37 +387,47 @@ use the canonical toolchain triplets for Linux.
 
				 </p>
			
 
				 <p>
			
 
				 Since there's often no easy way to detect CPU features at runtime, it's
			
 
				-important to compile with the proper CPU or architecture settings. You
			
 
				-can specify these when building the toolchain yourself. Or add
			
 
				-<tt>-mcpu=...</tt> or <tt>-march=...</tt> to <tt>TARGET_CFLAGS</tt>. For
			
 
				-ARM it's important to have the correct <tt>-mfloat-abi=...</tt> setting,
			
 
				-too. Otherwise LuaJIT may not run at the full performance of your target
			
 
				-CPU.
			
 
				+important to compile with the proper CPU or architecture settings:
			
 
				+</o>
			
 
				+<ul>
			
 
				+<li>The best way to get consistent results is to specify the correct settings when building the toolchain yourself.</li>
			
 
				+<li>For a pre-built, generic toolchain add <tt>-mcpu=...</tt> or <tt>-march=...</tt> and other necessary flags to <tt>TARGET_CFLAGS</tt>.</li>
			
 
				+<li>For ARM it's important to have the correct <tt>-mfloat-abi=...</tt> setting, too. Otherwise LuaJIT may not run at the full performance of your target CPU.</li>
			
 
				+<li>For MIPS it's important to select a supported ABI (o32 on MIPS32, n64 on MIPS64) and consistently compile your project either with hard-float or soft-float compiler settings.</li>
			
 
				+</ul>
			
 
				+<p>
			
 
				+Here are some examples for targets with a different CPU than the host:
			
 
				 </p>
			
 
				 <pre class="code">
			
 
				 # ARM soft-float
			
 
				 make HOST_CC="gcc -m32" CROSS=arm-linux-gnueabi- \
			
 
				      TARGET_CFLAGS="-mfloat-abi=soft"
			
 
				 
			
 
				-# ARM soft-float ABI with VFP (example for Cortex-A8)
			
 
				+# ARM soft-float ABI with VFP (example for Cortex-A9)
			
 
				 make HOST_CC="gcc -m32" CROSS=arm-linux-gnueabi- \
			
 
				-     TARGET_CFLAGS="-mcpu=cortex-a8 -mfloat-abi=softfp"
			
 
				+     TARGET_CFLAGS="-mcpu=cortex-a9 -mfloat-abi=softfp"
			
 
				 
			
 
				-# ARM hard-float ABI with VFP (armhf, requires recent toolchain)
			
 
				+# ARM hard-float ABI with VFP (armhf, most modern toolchains)
			
 
				 make HOST_CC="gcc -m32" CROSS=arm-linux-gnueabihf-
			
 
				 
			
 
				+# ARM64
			
 
				+make CROSS=aarch64-linux-
			
 
				+
			
 
				 # PPC
			
 
				 make HOST_CC="gcc -m32" CROSS=powerpc-linux-gnu-
			
 
				-# PPC/e500v2 (fast interpreter only)
			
 
				-make HOST_CC="gcc -m32" CROSS=powerpc-e500v2-linux-gnuspe-
			
 
				 
			
 
				-# MIPS big-endian
			
 
				+# MIPS32 big-endian
			
 
				 make HOST_CC="gcc -m32" CROSS=mips-linux-
			
 
				-# MIPS little-endian
			
 
				+# MIPS32 little-endian
			
 
				 make HOST_CC="gcc -m32" CROSS=mipsel-linux-
			
 
				+
			
 
				+# MIPS64 big-endian
			
 
				+make CROSS=mips-linux- TARGET_CFLAGS="-mips64r2 -mabi=64"
			
 
				+# MIPS64 little-endian
			
 
				+make CROSS=mipsel-linux- TARGET_CFLAGS="-mips64r2 -mabi=64"
			
 
				 </pre>
			
 
				 <p>
			
 
				-You can cross-compile for <b id="android">Android</b> using the <a href="http://developer.android.com/sdk/ndk/index.html"><span class="ext">&raquo;</span>&nbsp;Android NDK</a>.
			
 
				+You can cross-compile for <b id="android">Android</b> using the <a href="https://developer.android.com/ndk/index.html">Android NDK</a>.
			
 
				 The environment variables need to match the install locations and the
			
 
				 desired target platform. E.g. Android&nbsp;4.0 corresponds to ABI level&nbsp;14.
			
 
				 For details check the folder <tt>docs</tt> in the NDK directory.
			
@@ -411,7 +441,7 @@ to build/deploy or which lowest common denominator you want to pick:
 
				 # Android/ARM, armeabi (ARMv5TE soft-float), Android 2.2+ (Froyo)
			
 
				 NDK=/opt/android/ndk
			
 
				 NDKABI=8
			
 
				-NDKVER=$NDK/toolchains/arm-linux-androideabi-4.6
			
 
				+NDKVER=$NDK/toolchains/arm-linux-androideabi-4.9
			
 
				 NDKP=$NDKVER/prebuilt/linux-x86/bin/arm-linux-androideabi-
			
 
				 NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-arm"
			
 
				 make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
			
@@ -419,16 +449,16 @@ make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
 
				 # Android/ARM, armeabi-v7a (ARMv7 VFP), Android 4.0+ (ICS)
			
 
				 NDK=/opt/android/ndk
			
 
				 NDKABI=14
			
 
				-NDKVER=$NDK/toolchains/arm-linux-androideabi-4.6
			
 
				+NDKVER=$NDK/toolchains/arm-linux-androideabi-4.9
			
 
				 NDKP=$NDKVER/prebuilt/linux-x86/bin/arm-linux-androideabi-
			
 
				 NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-arm"
			
 
				 NDKARCH="-march=armv7-a -mfloat-abi=softfp -Wl,--fix-cortex-a8"
			
 
				 make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF $NDKARCH"
			
 
				 
			
 
				-# Android/MIPS, mips (MIPS32R1 hard-float), Android 4.0+ (ICS)
			
 
				+# Android/MIPS, mipsel (MIPS32R1 hard-float), Android 4.0+ (ICS)
			
 
				 NDK=/opt/android/ndk
			
 
				 NDKABI=14
			
 
				-NDKVER=$NDK/toolchains/mipsel-linux-android-4.6
			
 
				+NDKVER=$NDK/toolchains/mipsel-linux-android-4.9
			
 
				 NDKP=$NDKVER/prebuilt/linux-x86/bin/mipsel-linux-android-
			
 
				 NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-mips"
			
 
				 make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
			
@@ -436,7 +466,7 @@ make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
 
				 # Android/x86, x86 (i686 SSE3), Android 4.0+ (ICS)
			
 
				 NDK=/opt/android/ndk
			
 
				 NDKABI=14
			
 
				-NDKVER=$NDK/toolchains/x86-4.6
			
 
				+NDKVER=$NDK/toolchains/x86-4.9
			
 
				 NDKP=$NDKVER/prebuilt/linux-x86/bin/i686-linux-android-
			
 
				 NDKF="--sysroot $NDK/platforms/android-$NDKABI/arch-x86"
			
 
				 make HOST_CC="gcc -m32" CROSS=$NDKP TARGET_FLAGS="$NDKF"
			
@@ -452,11 +482,19 @@ much slower than the JIT compiler. Please complain to Apple, not me.
 
				 Or use Android. :-p
			
 
				 </p>
			
 
				 <pre class="code">
			
 
				+# iOS/ARM (32 bit)
			
 
				 ISDKP=$(xcrun --sdk iphoneos --show-sdk-path)
			
 
				 ICC=$(xcrun --sdk iphoneos --find clang)
			
 
				 ISDKF="-arch armv7 -isysroot $ISDKP"
			
 
				 make DEFAULT_CC=clang HOST_CC="clang -m32 -arch i386" \
			
 
				      CROSS="$(dirname $ICC)/" TARGET_FLAGS="$ISDKF" TARGET_SYS=iOS
			
 
				+
			
 
				+# iOS/ARM64
			
 
				+ISDKP=$(xcrun --sdk iphoneos --show-sdk-path)
			
 
				+ICC=$(xcrun --sdk iphoneos --find clang)
			
 
				+ISDKF="-arch arm64 -isysroot $ISDKP"
			
 
				+make DEFAULT_CC=clang CROSS="$(dirname $ICC)/" \
			
 
				+     TARGET_FLAGS="$ISDKF" TARGET_SYS=iOS
			
 
				 </pre>
			
 
				 
			
 
				 <h3 id="consoles">Cross-compiling for consoles</h3>
			
@@ -513,6 +551,16 @@ the following commands:
 
				 cd src
			
 
				 xedkbuild
			
 
				 </pre>
			
 
				+<p>
			
 
				+To cross-compile for <b id="xboxone">Xbox One</b> from a Windows host,
			
 
				+open a "Visual Studio .NET Command Prompt" (64&nbsp;bit host compiler),
			
 
				+<tt>cd</tt> to the directory where you've unpacked the sources and run
			
 
				+the following commands:
			
 
				+</p>
			
 
				+<pre class="code">
			
 
				+cd src
			
 
				+xb1build
			
 
				+</pre>
			
 
				 
			
 
				 <h2 id="embed">Embedding LuaJIT</h2>
			
 
				 <p>
			
@@ -543,14 +591,11 @@ intend to load Lua/C modules at runtime.
 
				 </li>
			
 
				 <li>
			
 
				 If you're building a 64 bit application on OSX which links directly or
			
 
				-indirectly against LuaJIT, you need to link your main executable
			
 
				-with these flags:
			
 
				+indirectly against LuaJIT which is not built for <tt>LJ_GC64</tt> mode,
			
 
				+you need to link your main executable with these flags:
			
 
				 <pre class="code">
			
 
				 -pagezero_size 10000 -image_base 100000000
			
 
				 </pre>
			
 
				-Also, it's recommended to <tt>rebase</tt> all (self-compiled) shared libraries
			
 
				-which are loaded at runtime on OSX/x64 (e.g. C extension modules for Lua).
			
 
				-See: <tt>man rebase</tt>
			
 
				 </li>
			
 
				 </ul>
			
 
				 <p>Additional hints for initializing LuaJIT using the C API functions:</p>
			
@@ -636,7 +681,7 @@ to me (the upstream) and not you (the package maintainer), anyway.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/luajit.html
+++ b/luajit.mod/luajit/doc/luajit.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>LuaJIT</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -126,6 +125,8 @@ table.feature small {
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -150,7 +151,7 @@ Lua is a powerful, dynamic and light-weight programming language.
 
				 It may be embedded or used as a general-purpose, stand-alone language.
			
 
				 </p>
			
 
				 <p>
			
 
				-LuaJIT is Copyright &copy; 2005-2017 Mike Pall, released under the
			
 
				+LuaJIT is Copyright &copy; 2005-2018 Mike Pall, released under the
			
 
				 <a href="http://www.opensource.org/licenses/mit-license.php"><span class="ext">&raquo;</span>&nbsp;MIT open source license</a>.
			
 
				 </p>
			
 
				 <p>
			
@@ -164,13 +165,13 @@ LuaJIT is Copyright &copy; 2005-2017 Mike Pall, released under the
 
				 <tr><td><span style="font-size:90%;">Embedded</span></td><td>Android</td><td>iOS</td></tr>
			
 
				 </table>
			
 
				 <table class="feature os os3">
			
 
				-<tr><td>PS3</td><td>PS4</td><td>PS Vita</td><td>Xbox 360</td></tr>
			
 
				+<tr><td>PS3</td><td>PS4</td><td>PS Vita</td><td>Xbox 360</td><td>Xbox One</td></tr>
			
 
				 </table>
			
 
				 <table class="feature compiler">
			
 
				-<tr><td>GCC</td><td>CLANG<br>LLVM</td><td>MSVC</td></tr>
			
 
				+<tr><td>GCC</td><td>Clang<br>LLVM</td><td>MSVC</td></tr>
			
 
				 </table>
			
 
				 <table class="feature cpu">
			
 
				-<tr><td>x86</td><td>x64</td><td>ARM</td><td>PPC</td><td>e500</td><td>MIPS</td></tr>
			
 
				+<tr><td>x86<br>x64</td><td>ARM<br>ARM64</td><td>PPC</td><td>MIPS32<br>MIPS64</td></tr>
			
 
				 </table>
			
 
				 <table class="feature fcompat">
			
 
				 <tr><td>Lua&nbsp;5.1<br>API+ABI</td><td>+&nbsp;JIT</td><td>+&nbsp;BitOp</td><td>+&nbsp;FFI</td><td>Drop-in<br>DLL/.so</td></tr>
			
@@ -224,7 +225,7 @@ Please select a sub-topic in the navigation bar to learn more about LuaJIT.
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/running.html
+++ b/luajit.mod/luajit/doc/running.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Running LuaJIT</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -63,6 +62,8 @@ td.param_default {
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a href="status.html">Status</a>
			
@@ -178,6 +179,7 @@ Here are the available LuaJIT control commands:
 
				 <li id="j_flush"><tt>-jflush</tt> &mdash; Flushes the whole cache of compiled code.</li>
			
 
				 <li id="j_v"><tt>-jv</tt> &mdash; Shows verbose information about the progress of the JIT compiler.</li>
			
 
				 <li id="j_dump"><tt>-jdump</tt> &mdash; Dumps the code and structures used in various compiler stages.</li>
			
 
				+<li id="j_p"><tt>-jp</tt> &mdash; Start the <a href="ext_profiler.html">integrated profiler</a>.</li>
			
 
				 </ul>
			
 
				 <p>
			
 
				 The <tt>-jv</tt> and <tt>-jdump</tt> commands are extension modules
			
@@ -296,7 +298,7 @@ Here are the parameters and their default settings:
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/doc/status.html
+++ b/luajit.mod/luajit/doc/status.html
@@ -3,8 +3,7 @@
 
				 <head>
			
 
				 <title>Status</title>
			
 
				 <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
			
 
				-<meta name="Author" content="Mike Pall">
			
 
				-<meta name="Copyright" content="Copyright (C) 2005-2017, Mike Pall">
			
 
				+<meta name="Copyright" content="Copyright (C) 2005-2018">
			
 
				 <meta name="Language" content="en">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad.css" media="screen">
			
 
				 <link rel="stylesheet" type="text/css" href="bluequad-print.css" media="print">
			
@@ -44,6 +43,8 @@ ul li { padding-bottom: 0.3em; }
 
				 <a href="ext_jit.html">jit.* Library</a>
			
 
				 </li><li>
			
 
				 <a href="ext_c_api.html">Lua/C API</a>
			
 
				+</li><li>
			
 
				+<a href="ext_profiler.html">Profiler</a>
			
 
				 </li></ul>
			
 
				 </li><li>
			
 
				 <a class="current" href="status.html">Status</a>
			
@@ -95,12 +96,23 @@ handled correctly. The error may fall through an on-trace
 
				 <tt>lua_atpanic</tt> on x64. This issue will be fixed with the new
			
 
				 garbage collector.
			
 
				 </li>
			
 
				+<li>
			
 
				+LuaJIT on 64 bit systems provides a <b>limited range</b> of 47 bits for the
			
 
				+<b>legacy <tt>lightuserdata</tt></b> data type.
			
 
				+This is only relevant on x64 systems which use the negative part of the
			
 
				+virtual address space in user mode, e.g. Solaris/x64, and on ARM64 systems
			
 
				+configured with a 48 bit or 52 bit VA.
			
 
				+Avoid using <tt>lightuserdata</tt> to hold pointers that may point outside
			
 
				+of that range, e.g. variables on the stack. In general, avoid this data
			
 
				+type for new code and replace it with (much more performant) FFI bindings.
			
 
				+FFI cdata pointers can address the full 64 bit range.
			
 
				+</li>
			
 
				 </ul>
			
 
				 <br class="flush">
			
 
				 </div>
			
 
				 <div id="foot">
			
 
				 <hr class="hide">
			
 
				-Copyright &copy; 2005-2017 Mike Pall
			
 
				+Copyright &copy; 2005-2018
			
 
				 <span class="noprint">
			
 
				 &middot;
			
 
				 <a href="contact.html">Contact</a>
			
--- a/luajit.mod/luajit/dynasm/dasm_arm.h
+++ b/luajit.mod/luajit/dynasm/dasm_arm.h
@@ -254,6 +254,7 @@ void dasm_put(Dst_DECL, int start, ...)
 
				       case DASM_IMMV8:
			
 
				 	CK((n & 3) == 0, RANGE_I);
			
 
				 	n >>= 2;
			
 
				+	/* fallthrough */
			
 
				       case DASM_IMML8:
			
 
				       case DASM_IMML12:
			
 
				 	CK(n >= 0 ? ((n>>((ins>>5)&31)) == 0) :
			
@@ -371,6 +372,7 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	  break;
			
 
				 	case DASM_REL_LG:
			
 
				 	  CK(n >= 0, UNDEF_LG);
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_REL_PC:
			
 
				 	  CK(n >= 0, UNDEF_PC);
			
 
				 	  n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base) - 4;
			
--- a/luajit.mod/luajit/dynasm/dasm_arm.lua
+++ b/luajit.mod/luajit/dynasm/dasm_arm.lua
@@ -9,9 +9,9 @@
 
				 local _info = {
			
 
				   arch =	"arm",
			
 
				   description =	"DynASM ARM module",
			
 
				-  version =	"1.3.0",
			
 
				-  vernum =	 10300,
			
 
				-  release =	"2011-05-05",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2015-10-18",
			
 
				   author =	"Mike Pall",
			
 
				   license =	"MIT",
			
 
				 }
			
--- a/luajit.mod/luajit/dynasm/dasm_arm64.h
+++ b/luajit.mod/luajit/dynasm/dasm_arm64.h
@@ -0,0 +1,519 @@
 
				+/*
			
 
				+** DynASM ARM64 encoding engine.
			
 
				+** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+** Released under the MIT license. See dynasm.lua for full copyright notice.
			
 
				+*/
			
 
				+
			
 
				+#include <stddef.h>
			
 
				+#include <stdarg.h>
			
 
				+#include <string.h>
			
 
				+#include <stdlib.h>
			
 
				+
			
 
				+#define DASM_ARCH		"arm64"
			
 
				+
			
 
				+#ifndef DASM_EXTERN
			
 
				+#define DASM_EXTERN(a,b,c,d)	0
			
 
				+#endif
			
 
				+
			
 
				+/* Action definitions. */
			
 
				+enum {
			
 
				+  DASM_STOP, DASM_SECTION, DASM_ESC, DASM_REL_EXT,
			
 
				+  /* The following actions need a buffer position. */
			
 
				+  DASM_ALIGN, DASM_REL_LG, DASM_LABEL_LG,
			
 
				+  /* The following actions also have an argument. */
			
 
				+  DASM_REL_PC, DASM_LABEL_PC,
			
 
				+  DASM_IMM, DASM_IMM6, DASM_IMM12, DASM_IMM13W, DASM_IMM13X, DASM_IMML,
			
 
				+  DASM__MAX
			
 
				+};
			
 
				+
			
 
				+/* Maximum number of section buffer positions for a single dasm_put() call. */
			
 
				+#define DASM_MAXSECPOS		25
			
 
				+
			
 
				+/* DynASM encoder status codes. Action list offset or number are or'ed in. */
			
 
				+#define DASM_S_OK		0x00000000
			
 
				+#define DASM_S_NOMEM		0x01000000
			
 
				+#define DASM_S_PHASE		0x02000000
			
 
				+#define DASM_S_MATCH_SEC	0x03000000
			
 
				+#define DASM_S_RANGE_I		0x11000000
			
 
				+#define DASM_S_RANGE_SEC	0x12000000
			
 
				+#define DASM_S_RANGE_LG		0x13000000
			
 
				+#define DASM_S_RANGE_PC		0x14000000
			
 
				+#define DASM_S_RANGE_REL	0x15000000
			
 
				+#define DASM_S_UNDEF_LG		0x21000000
			
 
				+#define DASM_S_UNDEF_PC		0x22000000
			
 
				+
			
 
				+/* Macros to convert positions (8 bit section + 24 bit index). */
			
 
				+#define DASM_POS2IDX(pos)	((pos)&0x00ffffff)
			
 
				+#define DASM_POS2BIAS(pos)	((pos)&0xff000000)
			
 
				+#define DASM_SEC2POS(sec)	((sec)<<24)
			
 
				+#define DASM_POS2SEC(pos)	((pos)>>24)
			
 
				+#define DASM_POS2PTR(D, pos)	(D->sections[DASM_POS2SEC(pos)].rbuf + (pos))
			
 
				+
			
 
				+/* Action list type. */
			
 
				+typedef const unsigned int *dasm_ActList;
			
 
				+
			
 
				+/* Per-section structure. */
			
 
				+typedef struct dasm_Section {
			
 
				+  int *rbuf;		/* Biased buffer pointer (negative section bias). */
			
 
				+  int *buf;		/* True buffer pointer. */
			
 
				+  size_t bsize;		/* Buffer size in bytes. */
			
 
				+  int pos;		/* Biased buffer position. */
			
 
				+  int epos;		/* End of biased buffer position - max single put. */
			
 
				+  int ofs;		/* Byte offset into section. */
			
 
				+} dasm_Section;
			
 
				+
			
 
				+/* Core structure holding the DynASM encoding state. */
			
 
				+struct dasm_State {
			
 
				+  size_t psize;			/* Allocated size of this structure. */
			
 
				+  dasm_ActList actionlist;	/* Current actionlist pointer. */
			
 
				+  int *lglabels;		/* Local/global chain/pos ptrs. */
			
 
				+  size_t lgsize;
			
 
				+  int *pclabels;		/* PC label chains/pos ptrs. */
			
 
				+  size_t pcsize;
			
 
				+  void **globals;		/* Array of globals (bias -10). */
			
 
				+  dasm_Section *section;	/* Pointer to active section. */
			
 
				+  size_t codesize;		/* Total size of all code sections. */
			
 
				+  int maxsection;		/* 0 <= sectionidx < maxsection. */
			
 
				+  int status;			/* Status code. */
			
 
				+  dasm_Section sections[1];	/* All sections. Alloc-extended. */
			
 
				+};
			
 
				+
			
 
				+/* The size of the core structure depends on the max. number of sections. */
			
 
				+#define DASM_PSZ(ms)	(sizeof(dasm_State)+(ms-1)*sizeof(dasm_Section))
			
 
				+
			
 
				+
			
 
				+/* Initialize DynASM state. */
			
 
				+void dasm_init(Dst_DECL, int maxsection)
			
 
				+{
			
 
				+  dasm_State *D;
			
 
				+  size_t psz = 0;
			
 
				+  int i;
			
 
				+  Dst_REF = NULL;
			
 
				+  DASM_M_GROW(Dst, struct dasm_State, Dst_REF, psz, DASM_PSZ(maxsection));
			
 
				+  D = Dst_REF;
			
 
				+  D->psize = psz;
			
 
				+  D->lglabels = NULL;
			
 
				+  D->lgsize = 0;
			
 
				+  D->pclabels = NULL;
			
 
				+  D->pcsize = 0;
			
 
				+  D->globals = NULL;
			
 
				+  D->maxsection = maxsection;
			
 
				+  for (i = 0; i < maxsection; i++) {
			
 
				+    D->sections[i].buf = NULL;  /* Need this for pass3. */
			
 
				+    D->sections[i].rbuf = D->sections[i].buf - DASM_SEC2POS(i);
			
 
				+    D->sections[i].bsize = 0;
			
 
				+    D->sections[i].epos = 0;  /* Wrong, but is recalculated after resize. */
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* Free DynASM state. */
			
 
				+void dasm_free(Dst_DECL)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  int i;
			
 
				+  for (i = 0; i < D->maxsection; i++)
			
 
				+    if (D->sections[i].buf)
			
 
				+      DASM_M_FREE(Dst, D->sections[i].buf, D->sections[i].bsize);
			
 
				+  if (D->pclabels) DASM_M_FREE(Dst, D->pclabels, D->pcsize);
			
 
				+  if (D->lglabels) DASM_M_FREE(Dst, D->lglabels, D->lgsize);
			
 
				+  DASM_M_FREE(Dst, D, D->psize);
			
 
				+}
			
 
				+
			
 
				+/* Setup global label array. Must be called before dasm_setup(). */
			
 
				+void dasm_setupglobal(Dst_DECL, void **gl, unsigned int maxgl)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  D->globals = gl - 10;  /* Negative bias to compensate for locals. */
			
 
				+  DASM_M_GROW(Dst, int, D->lglabels, D->lgsize, (10+maxgl)*sizeof(int));
			
 
				+}
			
 
				+
			
 
				+/* Grow PC label array. Can be called after dasm_setup(), too. */
			
 
				+void dasm_growpc(Dst_DECL, unsigned int maxpc)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  size_t osz = D->pcsize;
			
 
				+  DASM_M_GROW(Dst, int, D->pclabels, D->pcsize, maxpc*sizeof(int));
			
 
				+  memset((void *)(((unsigned char *)D->pclabels)+osz), 0, D->pcsize-osz);
			
 
				+}
			
 
				+
			
 
				+/* Setup encoder. */
			
 
				+void dasm_setup(Dst_DECL, const void *actionlist)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  int i;
			
 
				+  D->actionlist = (dasm_ActList)actionlist;
			
 
				+  D->status = DASM_S_OK;
			
 
				+  D->section = &D->sections[0];
			
 
				+  memset((void *)D->lglabels, 0, D->lgsize);
			
 
				+  if (D->pclabels) memset((void *)D->pclabels, 0, D->pcsize);
			
 
				+  for (i = 0; i < D->maxsection; i++) {
			
 
				+    D->sections[i].pos = DASM_SEC2POS(i);
			
 
				+    D->sections[i].ofs = 0;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+
			
 
				+#ifdef DASM_CHECKS
			
 
				+#define CK(x, st) \
			
 
				+  do { if (!(x)) { \
			
 
				+    D->status = DASM_S_##st|(p-D->actionlist-1); return; } } while (0)
			
 
				+#define CKPL(kind, st) \
			
 
				+  do { if ((size_t)((char *)pl-(char *)D->kind##labels) >= D->kind##size) { \
			
 
				+    D->status = DASM_S_RANGE_##st|(p-D->actionlist-1); return; } } while (0)
			
 
				+#else
			
 
				+#define CK(x, st)	((void)0)
			
 
				+#define CKPL(kind, st)	((void)0)
			
 
				+#endif
			
 
				+
			
 
				+static int dasm_imm12(unsigned int n)
			
 
				+{
			
 
				+  if ((n >> 12) == 0)
			
 
				+    return n;
			
 
				+  else if ((n & 0xff000fff) == 0)
			
 
				+    return (n >> 12) | 0x1000;
			
 
				+  else
			
 
				+    return -1;
			
 
				+}
			
 
				+
			
 
				+static int dasm_ffs(unsigned long long x)
			
 
				+{
			
 
				+  int n = -1;
			
 
				+  while (x) { x >>= 1; n++; }
			
 
				+  return n;
			
 
				+}
			
 
				+
			
 
				+static int dasm_imm13(int lo, int hi)
			
 
				+{
			
 
				+  int inv = 0, w = 64, s = 0xfff, xa, xb;
			
 
				+  unsigned long long n = (((unsigned long long)hi) << 32) | (unsigned int)lo;
			
 
				+  unsigned long long m = 1ULL, a, b, c;
			
 
				+  if (n & 1) { n = ~n; inv = 1; }
			
 
				+  a = n & -n; b = (n+a)&-(n+a); c = (n+a-b)&-(n+a-b);
			
 
				+  xa = dasm_ffs(a); xb = dasm_ffs(b);
			
 
				+  if (c) {
			
 
				+    w = dasm_ffs(c) - xa;
			
 
				+    if (w == 32) m = 0x0000000100000001UL;
			
 
				+    else if (w == 16) m = 0x0001000100010001UL;
			
 
				+    else if (w == 8) m = 0x0101010101010101UL;
			
 
				+    else if (w == 4) m = 0x1111111111111111UL;
			
 
				+    else if (w == 2) m = 0x5555555555555555UL;
			
 
				+    else return -1;
			
 
				+    s = (-2*w & 0x3f) - 1;
			
 
				+  } else if (!a) {
			
 
				+    return -1;
			
 
				+  } else if (xb == -1) {
			
 
				+    xb = 64;
			
 
				+  }
			
 
				+  if ((b-a) * m != n) return -1;
			
 
				+  if (inv) {
			
 
				+    return ((w - xb) << 6) | (s+w+xa-xb);
			
 
				+  } else {
			
 
				+    return ((w - xa) << 6) | (s+xb-xa);
			
 
				+  }
			
 
				+  return -1;
			
 
				+}
			
 
				+
			
 
				+/* Pass 1: Store actions and args, link branches/labels, estimate offsets. */
			
 
				+void dasm_put(Dst_DECL, int start, ...)
			
 
				+{
			
 
				+  va_list ap;
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  dasm_ActList p = D->actionlist + start;
			
 
				+  dasm_Section *sec = D->section;
			
 
				+  int pos = sec->pos, ofs = sec->ofs;
			
 
				+  int *b;
			
 
				+
			
 
				+  if (pos >= sec->epos) {
			
 
				+    DASM_M_GROW(Dst, int, sec->buf, sec->bsize,
			
 
				+      sec->bsize + 2*DASM_MAXSECPOS*sizeof(int));
			
 
				+    sec->rbuf = sec->buf - DASM_POS2BIAS(pos);
			
 
				+    sec->epos = (int)sec->bsize/sizeof(int) - DASM_MAXSECPOS+DASM_POS2BIAS(pos);
			
 
				+  }
			
 
				+
			
 
				+  b = sec->rbuf;
			
 
				+  b[pos++] = start;
			
 
				+
			
 
				+  va_start(ap, start);
			
 
				+  while (1) {
			
 
				+    unsigned int ins = *p++;
			
 
				+    unsigned int action = (ins >> 16);
			
 
				+    if (action >= DASM__MAX) {
			
 
				+      ofs += 4;
			
 
				+    } else {
			
 
				+      int *pl, n = action >= DASM_REL_PC ? va_arg(ap, int) : 0;
			
 
				+      switch (action) {
			
 
				+      case DASM_STOP: goto stop;
			
 
				+      case DASM_SECTION:
			
 
				+	n = (ins & 255); CK(n < D->maxsection, RANGE_SEC);
			
 
				+	D->section = &D->sections[n]; goto stop;
			
 
				+      case DASM_ESC: p++; ofs += 4; break;
			
 
				+      case DASM_REL_EXT: break;
			
 
				+      case DASM_ALIGN: ofs += (ins & 255); b[pos++] = ofs; break;
			
 
				+      case DASM_REL_LG:
			
 
				+	n = (ins & 2047) - 10; pl = D->lglabels + n;
			
 
				+	/* Bkwd rel or global. */
			
 
				+	if (n >= 0) { CK(n>=10||*pl<0, RANGE_LG); CKPL(lg, LG); goto putrel; }
			
 
				+	pl += 10; n = *pl;
			
 
				+	if (n < 0) n = 0;  /* Start new chain for fwd rel if label exists. */
			
 
				+	goto linkrel;
			
 
				+      case DASM_REL_PC:
			
 
				+	pl = D->pclabels + n; CKPL(pc, PC);
			
 
				+      putrel:
			
 
				+	n = *pl;
			
 
				+	if (n < 0) {  /* Label exists. Get label pos and store it. */
			
 
				+	  b[pos] = -n;
			
 
				+	} else {
			
 
				+      linkrel:
			
 
				+	  b[pos] = n;  /* Else link to rel chain, anchored at label. */
			
 
				+	  *pl = pos;
			
 
				+	}
			
 
				+	pos++;
			
 
				+	break;
			
 
				+      case DASM_LABEL_LG:
			
 
				+	pl = D->lglabels + (ins & 2047) - 10; CKPL(lg, LG); goto putlabel;
			
 
				+      case DASM_LABEL_PC:
			
 
				+	pl = D->pclabels + n; CKPL(pc, PC);
			
 
				+      putlabel:
			
 
				+	n = *pl;  /* n > 0: Collapse rel chain and replace with label pos. */
			
 
				+	while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = pos;
			
 
				+	}
			
 
				+	*pl = -pos;  /* Label exists now. */
			
 
				+	b[pos++] = ofs;  /* Store pass1 offset estimate. */
			
 
				+	break;
			
 
				+      case DASM_IMM:
			
 
				+	CK((n & ((1<<((ins>>10)&31))-1)) == 0, RANGE_I);
			
 
				+	n >>= ((ins>>10)&31);
			
 
				+#ifdef DASM_CHECKS
			
 
				+	if ((ins & 0x8000))
			
 
				+	  CK(((n + (1<<(((ins>>5)&31)-1)))>>((ins>>5)&31)) == 0, RANGE_I);
			
 
				+	else
			
 
				+	  CK((n>>((ins>>5)&31)) == 0, RANGE_I);
			
 
				+#endif
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				+      case DASM_IMM6:
			
 
				+	CK((n >> 6) == 0, RANGE_I);
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				+      case DASM_IMM12:
			
 
				+	CK(dasm_imm12((unsigned int)n) != -1, RANGE_I);
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				+      case DASM_IMM13W:
			
 
				+	CK(dasm_imm13(n, n) != -1, RANGE_I);
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				+      case DASM_IMM13X: {
			
 
				+	int m = va_arg(ap, int);
			
 
				+	CK(dasm_imm13(n, m) != -1, RANGE_I);
			
 
				+	b[pos++] = n;
			
 
				+	b[pos++] = m;
			
 
				+	break;
			
 
				+	}
			
 
				+      case DASM_IMML: {
			
 
				+#ifdef DASM_CHECKS
			
 
				+	int scale = (p[-2] >> 30);
			
 
				+	CK((!(n & ((1<<scale)-1)) && (unsigned int)(n>>scale) < 4096) ||
			
 
				+	   (unsigned int)(n+256) < 512, RANGE_I);
			
 
				+#endif
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+stop:
			
 
				+  va_end(ap);
			
 
				+  sec->pos = pos;
			
 
				+  sec->ofs = ofs;
			
 
				+}
			
 
				+#undef CK
			
 
				+
			
 
				+/* Pass 2: Link sections, shrink aligns, fix label offsets. */
			
 
				+int dasm_link(Dst_DECL, size_t *szp)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  int secnum;
			
 
				+  int ofs = 0;
			
 
				+
			
 
				+#ifdef DASM_CHECKS
			
 
				+  *szp = 0;
			
 
				+  if (D->status != DASM_S_OK) return D->status;
			
 
				+  {
			
 
				+    int pc;
			
 
				+    for (pc = 0; pc*sizeof(int) < D->pcsize; pc++)
			
 
				+      if (D->pclabels[pc] > 0) return DASM_S_UNDEF_PC|pc;
			
 
				+  }
			
 
				+#endif
			
 
				+
			
 
				+  { /* Handle globals not defined in this translation unit. */
			
 
				+    int idx;
			
 
				+    for (idx = 20; idx*sizeof(int) < D->lgsize; idx++) {
			
 
				+      int n = D->lglabels[idx];
			
 
				+      /* Undefined label: Collapse rel chain and replace with marker (< 0). */
			
 
				+      while (n > 0) { int *pb = DASM_POS2PTR(D, n); n = *pb; *pb = -idx; }
			
 
				+    }
			
 
				+  }
			
 
				+
			
 
				+  /* Combine all code sections. No support for data sections (yet). */
			
 
				+  for (secnum = 0; secnum < D->maxsection; secnum++) {
			
 
				+    dasm_Section *sec = D->sections + secnum;
			
 
				+    int *b = sec->rbuf;
			
 
				+    int pos = DASM_SEC2POS(secnum);
			
 
				+    int lastpos = sec->pos;
			
 
				+
			
 
				+    while (pos != lastpos) {
			
 
				+      dasm_ActList p = D->actionlist + b[pos++];
			
 
				+      while (1) {
			
 
				+	unsigned int ins = *p++;
			
 
				+	unsigned int action = (ins >> 16);
			
 
				+	switch (action) {
			
 
				+	case DASM_STOP: case DASM_SECTION: goto stop;
			
 
				+	case DASM_ESC: p++; break;
			
 
				+	case DASM_REL_EXT: break;
			
 
				+	case DASM_ALIGN: ofs -= (b[pos++] + ofs) & (ins & 255); break;
			
 
				+	case DASM_REL_LG: case DASM_REL_PC: pos++; break;
			
 
				+	case DASM_LABEL_LG: case DASM_LABEL_PC: b[pos++] += ofs; break;
			
 
				+	case DASM_IMM: case DASM_IMM6: case DASM_IMM12: case DASM_IMM13W:
			
 
				+	case DASM_IMML: pos++; break;
			
 
				+	case DASM_IMM13X: pos += 2; break;
			
 
				+	}
			
 
				+      }
			
 
				+      stop: (void)0;
			
 
				+    }
			
 
				+    ofs += sec->ofs;  /* Next section starts right after current section. */
			
 
				+  }
			
 
				+
			
 
				+  D->codesize = ofs;  /* Total size of all code sections */
			
 
				+  *szp = ofs;
			
 
				+  return DASM_S_OK;
			
 
				+}
			
 
				+
			
 
				+#ifdef DASM_CHECKS
			
 
				+#define CK(x, st) \
			
 
				+  do { if (!(x)) return DASM_S_##st|(p-D->actionlist-1); } while (0)
			
 
				+#else
			
 
				+#define CK(x, st)	((void)0)
			
 
				+#endif
			
 
				+
			
 
				+/* Pass 3: Encode sections. */
			
 
				+int dasm_encode(Dst_DECL, void *buffer)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  char *base = (char *)buffer;
			
 
				+  unsigned int *cp = (unsigned int *)buffer;
			
 
				+  int secnum;
			
 
				+
			
 
				+  /* Encode all code sections. No support for data sections (yet). */
			
 
				+  for (secnum = 0; secnum < D->maxsection; secnum++) {
			
 
				+    dasm_Section *sec = D->sections + secnum;
			
 
				+    int *b = sec->buf;
			
 
				+    int *endb = sec->rbuf + sec->pos;
			
 
				+
			
 
				+    while (b != endb) {
			
 
				+      dasm_ActList p = D->actionlist + *b++;
			
 
				+      while (1) {
			
 
				+	unsigned int ins = *p++;
			
 
				+	unsigned int action = (ins >> 16);
			
 
				+	int n = (action >= DASM_ALIGN && action < DASM__MAX) ? *b++ : 0;
			
 
				+	switch (action) {
			
 
				+	case DASM_STOP: case DASM_SECTION: goto stop;
			
 
				+	case DASM_ESC: *cp++ = *p++; break;
			
 
				+	case DASM_REL_EXT:
			
 
				+	  n = DASM_EXTERN(Dst, (unsigned char *)cp, (ins&2047), !(ins&2048));
			
 
				+	  goto patchrel;
			
 
				+	case DASM_ALIGN:
			
 
				+	  ins &= 255; while ((((char *)cp - base) & ins)) *cp++ = 0xe1a00000;
			
 
				+	  break;
			
 
				+	case DASM_REL_LG:
			
 
				+	  CK(n >= 0, UNDEF_LG);
			
 
				+	  /* fallthrough */
			
 
				+	case DASM_REL_PC:
			
 
				+	  CK(n >= 0, UNDEF_PC);
			
 
				+	  n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base) + 4;
			
 
				+	patchrel:
			
 
				+	  if (!(ins & 0xf800)) {  /* B, BL */
			
 
				+	    CK((n & 3) == 0 && ((n+0x08000000) >> 28) == 0, RANGE_REL);
			
 
				+	    cp[-1] |= ((n >> 2) & 0x03ffffff);
			
 
				+	  } else if ((ins & 0x800)) {  /* B.cond, CBZ, CBNZ, LDR* literal */
			
 
				+	    CK((n & 3) == 0 && ((n+0x00100000) >> 21) == 0, RANGE_REL);
			
 
				+	    cp[-1] |= ((n << 3) & 0x00ffffe0);
			
 
				+	  } else if ((ins & 0x3000) == 0x2000) {  /* ADR */
			
 
				+	    CK(((n+0x00100000) >> 21) == 0, RANGE_REL);
			
 
				+	    cp[-1] |= ((n << 3) & 0x00ffffe0) | ((n & 3) << 29);
			
 
				+	  } else if ((ins & 0x3000) == 0x3000) {  /* ADRP */
			
 
				+	    cp[-1] |= ((n >> 9) & 0x00ffffe0) | (((n >> 12) & 3) << 29);
			
 
				+	  } else if ((ins & 0x1000)) {  /* TBZ, TBNZ */
			
 
				+	    CK((n & 3) == 0 && ((n+0x00008000) >> 16) == 0, RANGE_REL);
			
 
				+	    cp[-1] |= ((n << 3) & 0x0007ffe0);
			
 
				+	  }
			
 
				+	  break;
			
 
				+	case DASM_LABEL_LG:
			
 
				+	  ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
			
 
				+	  break;
			
 
				+	case DASM_LABEL_PC: break;
			
 
				+	case DASM_IMM:
			
 
				+	  cp[-1] |= (n & ((1<<((ins>>5)&31))-1)) << (ins&31);
			
 
				+	  break;
			
 
				+	case DASM_IMM6:
			
 
				+	  cp[-1] |= ((n&31) << 19) | ((n&32) << 26);
			
 
				+	  break;
			
 
				+	case DASM_IMM12:
			
 
				+	  cp[-1] |= (dasm_imm12((unsigned int)n) << 10);
			
 
				+	  break;
			
 
				+	case DASM_IMM13W:
			
 
				+	  cp[-1] |= (dasm_imm13(n, n) << 10);
			
 
				+	  break;
			
 
				+	case DASM_IMM13X:
			
 
				+	  cp[-1] |= (dasm_imm13(n, *b++) << 10);
			
 
				+	  break;
			
 
				+	case DASM_IMML: {
			
 
				+	  int scale = (p[-2] >> 30);
			
 
				+	  cp[-1] |= (!(n & ((1<<scale)-1)) && (unsigned int)(n>>scale) < 4096) ?
			
 
				+	    ((n << (10-scale)) | 0x01000000) : ((n & 511) << 12);
			
 
				+	  break;
			
 
				+	  }
			
 
				+	default: *cp++ = ins; break;
			
 
				+	}
			
 
				+      }
			
 
				+      stop: (void)0;
			
 
				+    }
			
 
				+  }
			
 
				+
			
 
				+  if (base + D->codesize != (char *)cp)  /* Check for phase errors. */
			
 
				+    return DASM_S_PHASE;
			
 
				+  return DASM_S_OK;
			
 
				+}
			
 
				+#undef CK
			
 
				+
			
 
				+/* Get PC label offset. */
			
 
				+int dasm_getpclabel(Dst_DECL, unsigned int pc)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  if (pc*sizeof(int) < D->pcsize) {
			
 
				+    int pos = D->pclabels[pc];
			
 
				+    if (pos < 0) return *DASM_POS2PTR(D, -pos);
			
 
				+    if (pos > 0) return -1;  /* Undefined. */
			
 
				+  }
			
 
				+  return -2;  /* Unused or out of range. */
			
 
				+}
			
 
				+
			
 
				+#ifdef DASM_CHECKS
			
 
				+/* Optional sanity checker to call between isolated encoding steps. */
			
 
				+int dasm_checkstep(Dst_DECL, int secmatch)
			
 
				+{
			
 
				+  dasm_State *D = Dst_REF;
			
 
				+  if (D->status == DASM_S_OK) {
			
 
				+    int i;
			
 
				+    for (i = 1; i <= 9; i++) {
			
 
				+      if (D->lglabels[i] > 0) { D->status = DASM_S_UNDEF_LG|i; break; }
			
 
				+      D->lglabels[i] = 0;
			
 
				+    }
			
 
				+  }
			
 
				+  if (D->status == DASM_S_OK && secmatch >= 0 &&
			
 
				+      D->section != &D->sections[secmatch])
			
 
				+    D->status = DASM_S_MATCH_SEC|(D->section-D->sections);
			
 
				+  return D->status;
			
 
				+}
			
 
				+#endif
			
 
				+
			
--- a/luajit.mod/luajit/dynasm/dasm_arm64.lua
+++ b/luajit.mod/luajit/dynasm/dasm_arm64.lua
@@ -0,0 +1,1166 @@
 
				+------------------------------------------------------------------------------
			
 
				+-- DynASM ARM64 module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- See dynasm.lua for full copyright notice.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Module information:
			
 
				+local _info = {
			
 
				+  arch =	"arm",
			
 
				+  description =	"DynASM ARM64 module",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2015-10-18",
			
 
				+  author =	"Mike Pall",
			
 
				+  license =	"MIT",
			
 
				+}
			
 
				+
			
 
				+-- Exported glue functions for the arch-specific module.
			
 
				+local _M = { _info = _info }
			
 
				+
			
 
				+-- Cache library functions.
			
 
				+local type, tonumber, pairs, ipairs = type, tonumber, pairs, ipairs
			
 
				+local assert, setmetatable, rawget = assert, setmetatable, rawget
			
 
				+local _s = string
			
 
				+local sub, format, byte, char = _s.sub, _s.format, _s.byte, _s.char
			
 
				+local match, gmatch, gsub = _s.match, _s.gmatch, _s.gsub
			
 
				+local concat, sort, insert = table.concat, table.sort, table.insert
			
 
				+local bit = bit or require("bit")
			
 
				+local band, shl, shr, sar = bit.band, bit.lshift, bit.rshift, bit.arshift
			
 
				+local ror, tohex = bit.ror, bit.tohex
			
 
				+
			
 
				+-- Inherited tables and callbacks.
			
 
				+local g_opt, g_arch
			
 
				+local wline, werror, wfatal, wwarn
			
 
				+
			
 
				+-- Action name list.
			
 
				+-- CHECK: Keep this in sync with the C code!
			
 
				+local action_names = {
			
 
				+  "STOP", "SECTION", "ESC", "REL_EXT",
			
 
				+  "ALIGN", "REL_LG", "LABEL_LG",
			
 
				+  "REL_PC", "LABEL_PC", "IMM", "IMM6", "IMM12", "IMM13W", "IMM13X", "IMML",
			
 
				+}
			
 
				+
			
 
				+-- Maximum number of section buffer positions for dasm_put().
			
 
				+-- CHECK: Keep this in sync with the C code!
			
 
				+local maxsecpos = 25 -- Keep this low, to avoid excessively long C lines.
			
 
				+
			
 
				+-- Action name -> action number.
			
 
				+local map_action = {}
			
 
				+for n,name in ipairs(action_names) do
			
 
				+  map_action[name] = n-1
			
 
				+end
			
 
				+
			
 
				+-- Action list buffer.
			
 
				+local actlist = {}
			
 
				+
			
 
				+-- Argument list for next dasm_put(). Start with offset 0 into action list.
			
 
				+local actargs = { 0 }
			
 
				+
			
 
				+-- Current number of section buffer positions for dasm_put().
			
 
				+local secpos = 1
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Dump action names and numbers.
			
 
				+local function dumpactions(out)
			
 
				+  out:write("DynASM encoding engine action codes:\n")
			
 
				+  for n,name in ipairs(action_names) do
			
 
				+    local num = map_action[name]
			
 
				+    out:write(format("  %-10s %02X  %d\n", name, num, num))
			
 
				+  end
			
 
				+  out:write("\n")
			
 
				+end
			
 
				+
			
 
				+-- Write action list buffer as a huge static C array.
			
 
				+local function writeactions(out, name)
			
 
				+  local nn = #actlist
			
 
				+  if nn == 0 then nn = 1; actlist[0] = map_action.STOP end
			
 
				+  out:write("static const unsigned int ", name, "[", nn, "] = {\n")
			
 
				+  for i = 1,nn-1 do
			
 
				+    assert(out:write("0x", tohex(actlist[i]), ",\n"))
			
 
				+  end
			
 
				+  assert(out:write("0x", tohex(actlist[nn]), "\n};\n\n"))
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Add word to action list.
			
 
				+local function wputxw(n)
			
 
				+  assert(n >= 0 and n <= 0xffffffff and n % 1 == 0, "word out of range")
			
 
				+  actlist[#actlist+1] = n
			
 
				+end
			
 
				+
			
 
				+-- Add action to list with optional arg. Advance buffer pos, too.
			
 
				+local function waction(action, val, a, num)
			
 
				+  local w = assert(map_action[action], "bad action name `"..action.."'")
			
 
				+  wputxw(w * 0x10000 + (val or 0))
			
 
				+  if a then actargs[#actargs+1] = a end
			
 
				+  if a or num then secpos = secpos + (num or 1) end
			
 
				+end
			
 
				+
			
 
				+-- Flush action list (intervening C code or buffer pos overflow).
			
 
				+local function wflush(term)
			
 
				+  if #actlist == actargs[1] then return end -- Nothing to flush.
			
 
				+  if not term then waction("STOP") end -- Terminate action list.
			
 
				+  wline(format("dasm_put(Dst, %s);", concat(actargs, ", ")), true)
			
 
				+  actargs = { #actlist } -- Actionlist offset is 1st arg to next dasm_put().
			
 
				+  secpos = 1 -- The actionlist offset occupies a buffer position, too.
			
 
				+end
			
 
				+
			
 
				+-- Put escaped word.
			
 
				+local function wputw(n)
			
 
				+  if n <= 0x000fffff then waction("ESC") end
			
 
				+  wputxw(n)
			
 
				+end
			
 
				+
			
 
				+-- Reserve position for word.
			
 
				+local function wpos()
			
 
				+  local pos = #actlist+1
			
 
				+  actlist[pos] = ""
			
 
				+  return pos
			
 
				+end
			
 
				+
			
 
				+-- Store word to reserved position.
			
 
				+local function wputpos(pos, n)
			
 
				+  assert(n >= 0 and n <= 0xffffffff and n % 1 == 0, "word out of range")
			
 
				+  if n <= 0x000fffff then
			
 
				+    insert(actlist, pos+1, n)
			
 
				+    n = map_action.ESC * 0x10000
			
 
				+  end
			
 
				+  actlist[pos] = n
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Global label name -> global label number. With auto assignment on 1st use.
			
 
				+local next_global = 20
			
 
				+local map_global = setmetatable({}, { __index = function(t, name)
			
 
				+  if not match(name, "^[%a_][%w_]*$") then werror("bad global label") end
			
 
				+  local n = next_global
			
 
				+  if n > 2047 then werror("too many global labels") end
			
 
				+  next_global = n + 1
			
 
				+  t[name] = n
			
 
				+  return n
			
 
				+end})
			
 
				+
			
 
				+-- Dump global labels.
			
 
				+local function dumpglobals(out, lvl)
			
 
				+  local t = {}
			
 
				+  for name, n in pairs(map_global) do t[n] = name end
			
 
				+  out:write("Global labels:\n")
			
 
				+  for i=20,next_global-1 do
			
 
				+    out:write(format("  %s\n", t[i]))
			
 
				+  end
			
 
				+  out:write("\n")
			
 
				+end
			
 
				+
			
 
				+-- Write global label enum.
			
 
				+local function writeglobals(out, prefix)
			
 
				+  local t = {}
			
 
				+  for name, n in pairs(map_global) do t[n] = name end
			
 
				+  out:write("enum {\n")
			
 
				+  for i=20,next_global-1 do
			
 
				+    out:write("  ", prefix, t[i], ",\n")
			
 
				+  end
			
 
				+  out:write("  ", prefix, "_MAX\n};\n")
			
 
				+end
			
 
				+
			
 
				+-- Write global label names.
			
 
				+local function writeglobalnames(out, name)
			
 
				+  local t = {}
			
 
				+  for name, n in pairs(map_global) do t[n] = name end
			
 
				+  out:write("static const char *const ", name, "[] = {\n")
			
 
				+  for i=20,next_global-1 do
			
 
				+    out:write("  \"", t[i], "\",\n")
			
 
				+  end
			
 
				+  out:write("  (const char *)0\n};\n")
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Extern label name -> extern label number. With auto assignment on 1st use.
			
 
				+local next_extern = 0
			
 
				+local map_extern_ = {}
			
 
				+local map_extern = setmetatable({}, { __index = function(t, name)
			
 
				+  -- No restrictions on the name for now.
			
 
				+  local n = next_extern
			
 
				+  if n > 2047 then werror("too many extern labels") end
			
 
				+  next_extern = n + 1
			
 
				+  t[name] = n
			
 
				+  map_extern_[n] = name
			
 
				+  return n
			
 
				+end})
			
 
				+
			
 
				+-- Dump extern labels.
			
 
				+local function dumpexterns(out, lvl)
			
 
				+  out:write("Extern labels:\n")
			
 
				+  for i=0,next_extern-1 do
			
 
				+    out:write(format("  %s\n", map_extern_[i]))
			
 
				+  end
			
 
				+  out:write("\n")
			
 
				+end
			
 
				+
			
 
				+-- Write extern label names.
			
 
				+local function writeexternnames(out, name)
			
 
				+  out:write("static const char *const ", name, "[] = {\n")
			
 
				+  for i=0,next_extern-1 do
			
 
				+    out:write("  \"", map_extern_[i], "\",\n")
			
 
				+  end
			
 
				+  out:write("  (const char *)0\n};\n")
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Arch-specific maps.
			
 
				+
			
 
				+-- Ext. register name -> int. name.
			
 
				+local map_archdef = { xzr = "@x31", wzr = "@w31", lr = "x30", }
			
 
				+
			
 
				+-- Int. register name -> ext. name.
			
 
				+local map_reg_rev = { ["@x31"] = "xzr", ["@w31"] = "wzr", x30 = "lr", }
			
 
				+
			
 
				+local map_type = {}		-- Type name -> { ctype, reg }
			
 
				+local ctypenum = 0		-- Type number (for Dt... macros).
			
 
				+
			
 
				+-- Reverse defines for registers.
			
 
				+function _M.revdef(s)
			
 
				+  return map_reg_rev[s] or s
			
 
				+end
			
 
				+
			
 
				+local map_shift = { lsl = 0, lsr = 1, asr = 2, }
			
 
				+
			
 
				+local map_extend = {
			
 
				+  uxtb = 0, uxth = 1, uxtw = 2, uxtx = 3,
			
 
				+  sxtb = 4, sxth = 5, sxtw = 6, sxtx = 7,
			
 
				+}
			
 
				+
			
 
				+local map_cond = {
			
 
				+  eq = 0, ne = 1, cs = 2, cc = 3, mi = 4, pl = 5, vs = 6, vc = 7,
			
 
				+  hi = 8, ls = 9, ge = 10, lt = 11, gt = 12, le = 13, al = 14,
			
 
				+  hs = 2, lo = 3,
			
 
				+}
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local parse_reg_type
			
 
				+
			
 
				+local function parse_reg(expr)
			
 
				+  if not expr then werror("expected register name") end
			
 
				+  local tname, ovreg = match(expr, "^([%w_]+):(@?%l%d+)$")
			
 
				+  local tp = map_type[tname or expr]
			
 
				+  if tp then
			
 
				+    local reg = ovreg or tp.reg
			
 
				+    if not reg then
			
 
				+      werror("type `"..(tname or expr).."' needs a register override")
			
 
				+    end
			
 
				+    expr = reg
			
 
				+  end
			
 
				+  local ok31, rt, r = match(expr, "^(@?)([xwqdshb])([123]?[0-9])$")
			
 
				+  if r then
			
 
				+    r = tonumber(r)
			
 
				+    if r <= 30 or (r == 31 and ok31 ~= "" or (rt ~= "w" and rt ~= "x")) then
			
 
				+      if not parse_reg_type then
			
 
				+	parse_reg_type = rt
			
 
				+      elseif parse_reg_type ~= rt then
			
 
				+	werror("register size mismatch")
			
 
				+      end
			
 
				+      return r, tp
			
 
				+    end
			
 
				+  end
			
 
				+  werror("bad register name `"..expr.."'")
			
 
				+end
			
 
				+
			
 
				+local function parse_reg_base(expr)
			
 
				+  if expr == "sp" then return 0x3e0 end
			
 
				+  local base, tp = parse_reg(expr)
			
 
				+  if parse_reg_type ~= "x" then werror("bad register type") end
			
 
				+  parse_reg_type = false
			
 
				+  return shl(base, 5), tp
			
 
				+end
			
 
				+
			
 
				+local parse_ctx = {}
			
 
				+
			
 
				+local loadenv = setfenv and function(s)
			
 
				+  local code = loadstring(s, "")
			
 
				+  if code then setfenv(code, parse_ctx) end
			
 
				+  return code
			
 
				+end or function(s)
			
 
				+  return load(s, "", nil, parse_ctx)
			
 
				+end
			
 
				+
			
 
				+-- Try to parse simple arithmetic, too, since some basic ops are aliases.
			
 
				+local function parse_number(n)
			
 
				+  local x = tonumber(n)
			
 
				+  if x then return x end
			
 
				+  local code = loadenv("return "..n)
			
 
				+  if code then
			
 
				+    local ok, y = pcall(code)
			
 
				+    if ok then return y end
			
 
				+  end
			
 
				+  return nil
			
 
				+end
			
 
				+
			
 
				+local function parse_imm(imm, bits, shift, scale, signed)
			
 
				+  imm = match(imm, "^#(.*)$")
			
 
				+  if not imm then werror("expected immediate operand") end
			
 
				+  local n = parse_number(imm)
			
 
				+  if n then
			
 
				+    local m = sar(n, scale)
			
 
				+    if shl(m, scale) == n then
			
 
				+      if signed then
			
 
				+	local s = sar(m, bits-1)
			
 
				+	if s == 0 then return shl(m, shift)
			
 
				+	elseif s == -1 then return shl(m + shl(1, bits), shift) end
			
 
				+      else
			
 
				+	if sar(m, bits) == 0 then return shl(m, shift) end
			
 
				+      end
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  else
			
 
				+    waction("IMM", (signed and 32768 or 0)+scale*1024+bits*32+shift, imm)
			
 
				+    return 0
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_imm12(imm)
			
 
				+  imm = match(imm, "^#(.*)$")
			
 
				+  if not imm then werror("expected immediate operand") end
			
 
				+  local n = parse_number(imm)
			
 
				+  if n then
			
 
				+    if shr(n, 12) == 0 then
			
 
				+      return shl(n, 10)
			
 
				+    elseif band(n, 0xff000fff) == 0 then
			
 
				+      return shr(n, 2) + 0x00400000
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  else
			
 
				+    waction("IMM12", 0, imm)
			
 
				+    return 0
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_imm13(imm)
			
 
				+  imm = match(imm, "^#(.*)$")
			
 
				+  if not imm then werror("expected immediate operand") end
			
 
				+  local n = parse_number(imm)
			
 
				+  local r64 = parse_reg_type == "x"
			
 
				+  if n and n % 1 == 0 and n >= 0 and n <= 0xffffffff then
			
 
				+    local inv = false
			
 
				+    if band(n, 1) == 1 then n = bit.bnot(n); inv = true end
			
 
				+    local t = {}
			
 
				+    for i=1,32 do t[i] = band(n, 1); n = shr(n, 1) end
			
 
				+    local b = table.concat(t)
			
 
				+    b = b..(r64 and (inv and "1" or "0"):rep(32) or b)
			
 
				+    local p0, p1, p0a, p1a = b:match("^(0+)(1+)(0*)(1*)")
			
 
				+    if p0 then
			
 
				+      local w = p1a == "" and (r64 and 64 or 32) or #p1+#p0a
			
 
				+      if band(w, w-1) == 0 and b == b:sub(1, w):rep(64/w) then
			
 
				+	local s = band(-2*w, 0x3f) - 1
			
 
				+	if w == 64 then s = s + 0x1000 end
			
 
				+	if inv then
			
 
				+	  return shl(w-#p1-#p0, 16) + shl(s+w-#p1, 10)
			
 
				+	else
			
 
				+	  return shl(w-#p0, 16) + shl(s+#p1, 10)
			
 
				+	end
			
 
				+      end
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  elseif r64 then
			
 
				+    waction("IMM13X", 0, format("(unsigned int)(%s)", imm))
			
 
				+    actargs[#actargs+1] = format("(unsigned int)((unsigned long long)(%s)>>32)", imm)
			
 
				+    return 0
			
 
				+  else
			
 
				+    waction("IMM13W", 0, imm)
			
 
				+    return 0
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_imm6(imm)
			
 
				+  imm = match(imm, "^#(.*)$")
			
 
				+  if not imm then werror("expected immediate operand") end
			
 
				+  local n = parse_number(imm)
			
 
				+  if n then
			
 
				+    if n >= 0 and n <= 63 then
			
 
				+      return shl(band(n, 0x1f), 19) + (n >= 32 and 0x80000000 or 0)
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  else
			
 
				+    waction("IMM6", 0, imm)
			
 
				+    return 0
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_imm_load(imm, scale)
			
 
				+  local n = parse_number(imm)
			
 
				+  if n then
			
 
				+    local m = sar(n, scale)
			
 
				+    if shl(m, scale) == n and m >= 0 and m < 0x1000 then
			
 
				+      return shl(m, 10) + 0x01000000 -- Scaled, unsigned 12 bit offset.
			
 
				+    elseif n >= -256 and n < 256 then
			
 
				+      return shl(band(n, 511), 12) -- Unscaled, signed 9 bit offset.
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  else
			
 
				+    waction("IMML", 0, imm)
			
 
				+    return 0
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_fpimm(imm)
			
 
				+  imm = match(imm, "^#(.*)$")
			
 
				+  if not imm then werror("expected immediate operand") end
			
 
				+  local n = parse_number(imm)
			
 
				+  if n then
			
 
				+    local m, e = math.frexp(n)
			
 
				+    local s, e2 = 0, band(e-2, 7)
			
 
				+    if m < 0 then m = -m; s = 0x00100000 end
			
 
				+    m = m*32-16
			
 
				+    if m % 1 == 0 and m >= 0 and m <= 15 and sar(shl(e2, 29), 29)+2 == e then
			
 
				+      return s + shl(e2, 17) + shl(m, 13)
			
 
				+    end
			
 
				+    werror("out of range immediate `"..imm.."'")
			
 
				+  else
			
 
				+    werror("NYI fpimm action")
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_shift(expr)
			
 
				+  local s, s2 = match(expr, "^(%S+)%s*(.*)$")
			
 
				+  s = map_shift[s]
			
 
				+  if not s then werror("expected shift operand") end
			
 
				+  return parse_imm(s2, 6, 10, 0, false) + shl(s, 22)
			
 
				+end
			
 
				+
			
 
				+local function parse_lslx16(expr)
			
 
				+  local n = match(expr, "^lsl%s*#(%d+)$")
			
 
				+  n = tonumber(n)
			
 
				+  if not n then werror("expected shift operand") end
			
 
				+  if band(n, parse_reg_type == "x" and 0xffffffcf or 0xffffffef) ~= 0 then
			
 
				+    werror("bad shift amount")
			
 
				+  end
			
 
				+  return shl(n, 17)
			
 
				+end
			
 
				+
			
 
				+local function parse_extend(expr)
			
 
				+  local s, s2 = match(expr, "^(%S+)%s*(.*)$")
			
 
				+  if s == "lsl" then
			
 
				+    s = parse_reg_type == "x" and 3 or 2
			
 
				+  else
			
 
				+    s = map_extend[s]
			
 
				+  end
			
 
				+  if not s then werror("expected extend operand") end
			
 
				+  return (s2 == "" and 0 or parse_imm(s2, 3, 10, 0, false)) + shl(s, 13)
			
 
				+end
			
 
				+
			
 
				+local function parse_cond(expr, inv)
			
 
				+  local c = map_cond[expr]
			
 
				+  if not c then werror("expected condition operand") end
			
 
				+  return shl(bit.bxor(c, inv), 12)
			
 
				+end
			
 
				+
			
 
				+local function parse_load(params, nparams, n, op)
			
 
				+  if params[n+2] then werror("too many operands") end
			
 
				+  local pn, p2 = params[n], params[n+1]
			
 
				+  local p1, wb = match(pn, "^%[%s*(.-)%s*%](!?)$")
			
 
				+  if not p1 then
			
 
				+    if not p2 then
			
 
				+      local reg, tailr = match(pn, "^([%w_:]+)%s*(.*)$")
			
 
				+      if reg and tailr ~= "" then
			
 
				+	local base, tp = parse_reg_base(reg)
			
 
				+	if tp then
			
 
				+	  waction("IMML", 0, format(tp.ctypefmt, tailr))
			
 
				+	  return op + base
			
 
				+	end
			
 
				+      end
			
 
				+    end
			
 
				+    werror("expected address operand")
			
 
				+  end
			
 
				+  local scale = shr(op, 30)
			
 
				+  if p2 then
			
 
				+    if wb == "!" then werror("bad use of '!'") end
			
 
				+    op = op + parse_reg_base(p1) + parse_imm(p2, 9, 12, 0, true) + 0x400
			
 
				+  elseif wb == "!" then
			
 
				+    local p1a, p2a = match(p1, "^([^,%s]*)%s*,%s*(.*)$")
			
 
				+    if not p1a then werror("bad use of '!'") end
			
 
				+    op = op + parse_reg_base(p1a) + parse_imm(p2a, 9, 12, 0, true) + 0xc00
			
 
				+  else
			
 
				+    local p1a, p2a = match(p1, "^([^,%s]*)%s*(.*)$")
			
 
				+    op = op + parse_reg_base(p1a)
			
 
				+    if p2a ~= "" then
			
 
				+      local imm = match(p2a, "^,%s*#(.*)$")
			
 
				+      if imm then
			
 
				+	op = op + parse_imm_load(imm, scale)
			
 
				+      else
			
 
				+	local p2b, p3b, p3s = match(p2a, "^,%s*([^,%s]*)%s*,?%s*(%S*)%s*(.*)$")
			
 
				+	op = op + shl(parse_reg(p2b), 16) + 0x00200800
			
 
				+	if parse_reg_type ~= "x" and parse_reg_type ~= "w" then
			
 
				+	  werror("bad index register type")
			
 
				+	end
			
 
				+	if p3b == "" then
			
 
				+	  if parse_reg_type ~= "x" then werror("bad index register type") end
			
 
				+	  op = op + 0x6000
			
 
				+	else
			
 
				+	  if p3s == "" or p3s == "#0" then
			
 
				+	  elseif p3s == "#"..scale then
			
 
				+	    op = op + 0x1000
			
 
				+	  else
			
 
				+	    werror("bad scale")
			
 
				+	  end
			
 
				+	  if parse_reg_type == "x" then
			
 
				+	    if p3b == "lsl" and p3s ~= "" then op = op + 0x6000
			
 
				+	    elseif p3b == "sxtx" then op = op + 0xe000
			
 
				+	    else
			
 
				+	      werror("bad extend/shift specifier")
			
 
				+	    end
			
 
				+	  else
			
 
				+	    if p3b == "uxtw" then op = op + 0x4000
			
 
				+	    elseif p3b == "sxtw" then op = op + 0xc000
			
 
				+	    else
			
 
				+	      werror("bad extend/shift specifier")
			
 
				+	    end
			
 
				+	  end
			
 
				+	end
			
 
				+      end
			
 
				+    else
			
 
				+      if wb == "!" then werror("bad use of '!'") end
			
 
				+      op = op + 0x01000000
			
 
				+    end
			
 
				+  end
			
 
				+  return op
			
 
				+end
			
 
				+
			
 
				+local function parse_load_pair(params, nparams, n, op)
			
 
				+  if params[n+2] then werror("too many operands") end
			
 
				+  local pn, p2 = params[n], params[n+1]
			
 
				+  local scale = shr(op, 30) == 0 and 2 or 3
			
 
				+  local p1, wb = match(pn, "^%[%s*(.-)%s*%](!?)$")
			
 
				+  if not p1 then
			
 
				+    if not p2 then
			
 
				+      local reg, tailr = match(pn, "^([%w_:]+)%s*(.*)$")
			
 
				+      if reg and tailr ~= "" then
			
 
				+	local base, tp = parse_reg_base(reg)
			
 
				+	if tp then
			
 
				+	  waction("IMM", 32768+7*32+15+scale*1024, format(tp.ctypefmt, tailr))
			
 
				+	  return op + base + 0x01000000
			
 
				+	end
			
 
				+      end
			
 
				+    end
			
 
				+    werror("expected address operand")
			
 
				+  end
			
 
				+  if p2 then
			
 
				+    if wb == "!" then werror("bad use of '!'") end
			
 
				+    op = op + 0x00800000
			
 
				+  else
			
 
				+    local p1a, p2a = match(p1, "^([^,%s]*)%s*,%s*(.*)$")
			
 
				+    if p1a then p1, p2 = p1a, p2a else p2 = "#0" end
			
 
				+    op = op + (wb == "!" and 0x01800000 or 0x01000000)
			
 
				+  end
			
 
				+  return op + parse_reg_base(p1) + parse_imm(p2, 7, 15, scale, true)
			
 
				+end
			
 
				+
			
 
				+local function parse_label(label, def)
			
 
				+  local prefix = sub(label, 1, 2)
			
 
				+  -- =>label (pc label reference)
			
 
				+  if prefix == "=>" then
			
 
				+    return "PC", 0, sub(label, 3)
			
 
				+  end
			
 
				+  -- ->name (global label reference)
			
 
				+  if prefix == "->" then
			
 
				+    return "LG", map_global[sub(label, 3)]
			
 
				+  end
			
 
				+  if def then
			
 
				+    -- [1-9] (local label definition)
			
 
				+    if match(label, "^[1-9]$") then
			
 
				+      return "LG", 10+tonumber(label)
			
 
				+    end
			
 
				+  else
			
 
				+    -- [<>][1-9] (local label reference)
			
 
				+    local dir, lnum = match(label, "^([<>])([1-9])$")
			
 
				+    if dir then -- Fwd: 1-9, Bkwd: 11-19.
			
 
				+      return "LG", lnum + (dir == ">" and 0 or 10)
			
 
				+    end
			
 
				+    -- extern label (extern label reference)
			
 
				+    local extname = match(label, "^extern%s+(%S+)$")
			
 
				+    if extname then
			
 
				+      return "EXT", map_extern[extname]
			
 
				+    end
			
 
				+  end
			
 
				+  werror("bad label `"..label.."'")
			
 
				+end
			
 
				+
			
 
				+local function branch_type(op)
			
 
				+  if band(op, 0x7c000000) == 0x14000000 then return 0 -- B, BL
			
 
				+  elseif shr(op, 24) == 0x54 or band(op, 0x7e000000) == 0x34000000 or
			
 
				+	 band(op, 0x3b000000) == 0x18000000 then
			
 
				+    return 0x800 -- B.cond, CBZ, CBNZ, LDR* literal
			
 
				+  elseif band(op, 0x7e000000) == 0x36000000 then return 0x1000 -- TBZ, TBNZ
			
 
				+  elseif band(op, 0x9f000000) == 0x10000000 then return 0x2000 -- ADR
			
 
				+  elseif band(op, 0x9f000000) == band(0x90000000) then return 0x3000 -- ADRP
			
 
				+  else
			
 
				+    assert(false, "unknown branch type")
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local map_op, op_template
			
 
				+
			
 
				+local function op_alias(opname, f)
			
 
				+  return function(params, nparams)
			
 
				+    if not params then return "-> "..opname:sub(1, -3) end
			
 
				+    f(params, nparams)
			
 
				+    op_template(params, map_op[opname], nparams)
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function alias_bfx(p)
			
 
				+  p[4] = "#("..p[3]:sub(2)..")+("..p[4]:sub(2)..")-1"
			
 
				+end
			
 
				+
			
 
				+local function alias_bfiz(p)
			
 
				+  parse_reg(p[1])
			
 
				+  if parse_reg_type == "w" then
			
 
				+    p[3] = "#-("..p[3]:sub(2)..")%32"
			
 
				+    p[4] = "#("..p[4]:sub(2)..")-1"
			
 
				+  else
			
 
				+    p[3] = "#-("..p[3]:sub(2)..")%64"
			
 
				+    p[4] = "#("..p[4]:sub(2)..")-1"
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local alias_lslimm = op_alias("ubfm_4", function(p)
			
 
				+  parse_reg(p[1])
			
 
				+  local sh = p[3]:sub(2)
			
 
				+  if parse_reg_type == "w" then
			
 
				+    p[3] = "#-("..sh..")%32"
			
 
				+    p[4] = "#31-("..sh..")"
			
 
				+  else
			
 
				+    p[3] = "#-("..sh..")%64"
			
 
				+    p[4] = "#63-("..sh..")"
			
 
				+  end
			
 
				+end)
			
 
				+
			
 
				+-- Template strings for ARM instructions.
			
 
				+map_op = {
			
 
				+  -- Basic data processing instructions.
			
 
				+  add_3  = "0b000000DNMg|11000000pDpNIg|8b206000pDpNMx",
			
 
				+  add_4  = "0b000000DNMSg|0b200000DNMXg|8b200000pDpNMXx|8b200000pDpNxMwX",
			
 
				+  adds_3 = "2b000000DNMg|31000000DpNIg|ab206000DpNMx",
			
 
				+  adds_4 = "2b000000DNMSg|2b200000DNMXg|ab200000DpNMXx|ab200000DpNxMwX",
			
 
				+  cmn_2  = "2b00001fNMg|3100001fpNIg|ab20601fpNMx",
			
 
				+  cmn_3  = "2b00001fNMSg|2b20001fNMXg|ab20001fpNMXx|ab20001fpNxMwX",
			
 
				+
			
 
				+  sub_3  = "4b000000DNMg|51000000pDpNIg|cb206000pDpNMx",
			
 
				+  sub_4  = "4b000000DNMSg|4b200000DNMXg|cb200000pDpNMXx|cb200000pDpNxMwX",
			
 
				+  subs_3 = "6b000000DNMg|71000000DpNIg|eb206000DpNMx",
			
 
				+  subs_4 = "6b000000DNMSg|6b200000DNMXg|eb200000DpNMXx|eb200000DpNxMwX",
			
 
				+  cmp_2  = "6b00001fNMg|7100001fpNIg|eb20601fpNMx",
			
 
				+  cmp_3  = "6b00001fNMSg|6b20001fNMXg|eb20001fpNMXx|eb20001fpNxMwX",
			
 
				+
			
 
				+  neg_2  = "4b0003e0DMg",
			
 
				+  neg_3  = "4b0003e0DMSg",
			
 
				+  negs_2 = "6b0003e0DMg",
			
 
				+  negs_3 = "6b0003e0DMSg",
			
 
				+
			
 
				+  adc_3  = "1a000000DNMg",
			
 
				+  adcs_3 = "3a000000DNMg",
			
 
				+  sbc_3  = "5a000000DNMg",
			
 
				+  sbcs_3 = "7a000000DNMg",
			
 
				+  ngc_2  = "5a0003e0DMg",
			
 
				+  ngcs_2 = "7a0003e0DMg",
			
 
				+
			
 
				+  and_3  = "0a000000DNMg|12000000pDNig",
			
 
				+  and_4  = "0a000000DNMSg",
			
 
				+  orr_3  = "2a000000DNMg|32000000pDNig",
			
 
				+  orr_4  = "2a000000DNMSg",
			
 
				+  eor_3  = "4a000000DNMg|52000000pDNig",
			
 
				+  eor_4  = "4a000000DNMSg",
			
 
				+  ands_3 = "6a000000DNMg|72000000DNig",
			
 
				+  ands_4 = "6a000000DNMSg",
			
 
				+  tst_2  = "6a00001fNMg|7200001fNig",
			
 
				+  tst_3  = "6a00001fNMSg",
			
 
				+
			
 
				+  bic_3  = "0a200000DNMg",
			
 
				+  bic_4  = "0a200000DNMSg",
			
 
				+  orn_3  = "2a200000DNMg",
			
 
				+  orn_4  = "2a200000DNMSg",
			
 
				+  eon_3  = "4a200000DNMg",
			
 
				+  eon_4  = "4a200000DNMSg",
			
 
				+  bics_3 = "6a200000DNMg",
			
 
				+  bics_4 = "6a200000DNMSg",
			
 
				+
			
 
				+  movn_2 = "12800000DWg",
			
 
				+  movn_3 = "12800000DWRg",
			
 
				+  movz_2 = "52800000DWg",
			
 
				+  movz_3 = "52800000DWRg",
			
 
				+  movk_2 = "72800000DWg",
			
 
				+  movk_3 = "72800000DWRg",
			
 
				+
			
 
				+  -- TODO: this doesn't cover all valid immediates for mov reg, #imm.
			
 
				+  mov_2  = "2a0003e0DMg|52800000DW|320003e0pDig|11000000pDpNg",
			
 
				+  mov_3  = "2a0003e0DMSg",
			
 
				+  mvn_2  = "2a2003e0DMg",
			
 
				+  mvn_3  = "2a2003e0DMSg",
			
 
				+
			
 
				+  adr_2  = "10000000DBx",
			
 
				+  adrp_2 = "90000000DBx",
			
 
				+
			
 
				+  csel_4  = "1a800000DNMCg",
			
 
				+  csinc_4 = "1a800400DNMCg",
			
 
				+  csinv_4 = "5a800000DNMCg",
			
 
				+  csneg_4 = "5a800400DNMCg",
			
 
				+  cset_2  = "1a9f07e0Dcg",
			
 
				+  csetm_2 = "5a9f03e0Dcg",
			
 
				+  cinc_3  = "1a800400DNmcg",
			
 
				+  cinv_3  = "5a800000DNmcg",
			
 
				+  cneg_3  = "5a800400DNmcg",
			
 
				+
			
 
				+  ccmn_4 = "3a400000NMVCg|3a400800N5VCg",
			
 
				+  ccmp_4 = "7a400000NMVCg|7a400800N5VCg",
			
 
				+
			
 
				+  madd_4 = "1b000000DNMAg",
			
 
				+  msub_4 = "1b008000DNMAg",
			
 
				+  mul_3  = "1b007c00DNMg",
			
 
				+  mneg_3 = "1b00fc00DNMg",
			
 
				+
			
 
				+  smaddl_4 = "9b200000DxNMwAx",
			
 
				+  smsubl_4 = "9b208000DxNMwAx",
			
 
				+  smull_3  = "9b207c00DxNMw",
			
 
				+  smnegl_3 = "9b20fc00DxNMw",
			
 
				+  smulh_3  = "9b407c00DNMx",
			
 
				+  umaddl_4 = "9ba00000DxNMwAx",
			
 
				+  umsubl_4 = "9ba08000DxNMwAx",
			
 
				+  umull_3  = "9ba07c00DxNMw",
			
 
				+  umnegl_3 = "9ba0fc00DxNMw",
			
 
				+  umulh_3  = "9bc07c00DNMx",
			
 
				+
			
 
				+  udiv_3 = "1ac00800DNMg",
			
 
				+  sdiv_3 = "1ac00c00DNMg",
			
 
				+
			
 
				+  -- Bit operations.
			
 
				+  sbfm_4 = "13000000DN12w|93400000DN12x",
			
 
				+  bfm_4  = "33000000DN12w|b3400000DN12x",
			
 
				+  ubfm_4 = "53000000DN12w|d3400000DN12x",
			
 
				+  extr_4 = "13800000DNM2w|93c00000DNM2x",
			
 
				+
			
 
				+  sxtb_2 = "13001c00DNw|93401c00DNx",
			
 
				+  sxth_2 = "13003c00DNw|93403c00DNx",
			
 
				+  sxtw_2 = "93407c00DxNw",
			
 
				+  uxtb_2 = "53001c00DNw",
			
 
				+  uxth_2 = "53003c00DNw",
			
 
				+
			
 
				+  sbfx_4  = op_alias("sbfm_4", alias_bfx),
			
 
				+  bfxil_4 = op_alias("bfm_4", alias_bfx),
			
 
				+  ubfx_4  = op_alias("ubfm_4", alias_bfx),
			
 
				+  sbfiz_4 = op_alias("sbfm_4", alias_bfiz),
			
 
				+  bfi_4   = op_alias("bfm_4", alias_bfiz),
			
 
				+  ubfiz_4 = op_alias("ubfm_4", alias_bfiz),
			
 
				+
			
 
				+  lsl_3  = function(params, nparams)
			
 
				+    if params and params[3]:byte() == 35 then
			
 
				+      return alias_lslimm(params, nparams)
			
 
				+    else
			
 
				+      return op_template(params, "1ac02000DNMg", nparams)
			
 
				+    end
			
 
				+  end,
			
 
				+  lsr_3  = "1ac02400DNMg|53007c00DN1w|d340fc00DN1x",
			
 
				+  asr_3  = "1ac02800DNMg|13007c00DN1w|9340fc00DN1x",
			
 
				+  ror_3  = "1ac02c00DNMg|13800000DNm2w|93c00000DNm2x",
			
 
				+
			
 
				+  clz_2   = "5ac01000DNg",
			
 
				+  cls_2   = "5ac01400DNg",
			
 
				+  rbit_2  = "5ac00000DNg",
			
 
				+  rev_2   = "5ac00800DNw|dac00c00DNx",
			
 
				+  rev16_2 = "5ac00400DNg",
			
 
				+  rev32_2 = "dac00800DNx",
			
 
				+
			
 
				+  -- Loads and stores.
			
 
				+  ["strb_*"]  = "38000000DwL",
			
 
				+  ["ldrb_*"]  = "38400000DwL",
			
 
				+  ["ldrsb_*"] = "38c00000DwL|38800000DxL",
			
 
				+  ["strh_*"]  = "78000000DwL",
			
 
				+  ["ldrh_*"]  = "78400000DwL",
			
 
				+  ["ldrsh_*"] = "78c00000DwL|78800000DxL",
			
 
				+  ["str_*"]   = "b8000000DwL|f8000000DxL|bc000000DsL|fc000000DdL",
			
 
				+  ["ldr_*"]   = "18000000DwB|58000000DxB|1c000000DsB|5c000000DdB|b8400000DwL|f8400000DxL|bc400000DsL|fc400000DdL",
			
 
				+  ["ldrsw_*"] = "98000000DxB|b8800000DxL",
			
 
				+  -- NOTE: ldur etc. are handled by ldr et al.
			
 
				+
			
 
				+  ["stp_*"]   = "28000000DAwP|a8000000DAxP|2c000000DAsP|6c000000DAdP",
			
 
				+  ["ldp_*"]   = "28400000DAwP|a8400000DAxP|2c400000DAsP|6c400000DAdP",
			
 
				+  ["ldpsw_*"] = "68400000DAxP",
			
 
				+
			
 
				+  -- Branches.
			
 
				+  b_1    = "14000000B",
			
 
				+  bl_1   = "94000000B",
			
 
				+  blr_1  = "d63f0000Nx",
			
 
				+  br_1   = "d61f0000Nx",
			
 
				+  ret_0  = "d65f03c0",
			
 
				+  ret_1  = "d65f0000Nx",
			
 
				+  -- b.cond is added below.
			
 
				+  cbz_2  = "34000000DBg",
			
 
				+  cbnz_2 = "35000000DBg",
			
 
				+  tbz_3  = "36000000DTBw|36000000DTBx",
			
 
				+  tbnz_3 = "37000000DTBw|37000000DTBx",
			
 
				+
			
 
				+  -- Miscellaneous instructions.
			
 
				+  -- TODO: hlt, hvc, smc, svc, eret, dcps[123], drps, mrs, msr
			
 
				+  -- TODO: sys, sysl, ic, dc, at, tlbi
			
 
				+  -- TODO: hint, yield, wfe, wfi, sev, sevl
			
 
				+  -- TODO: clrex, dsb, dmb, isb
			
 
				+  nop_0  = "d503201f",
			
 
				+  brk_0  = "d4200000",
			
 
				+  brk_1  = "d4200000W",
			
 
				+
			
 
				+  -- Floating point instructions.
			
 
				+  fmov_2  = "1e204000DNf|1e260000DwNs|1e270000DsNw|9e660000DxNd|9e670000DdNx|1e201000DFf",
			
 
				+  fabs_2  = "1e20c000DNf",
			
 
				+  fneg_2  = "1e214000DNf",
			
 
				+  fsqrt_2 = "1e21c000DNf",
			
 
				+
			
 
				+  fcvt_2  = "1e22c000DdNs|1e624000DsNd",
			
 
				+
			
 
				+  -- TODO: half-precision and fixed-point conversions.
			
 
				+  fcvtas_2 = "1e240000DwNs|9e240000DxNs|1e640000DwNd|9e640000DxNd",
			
 
				+  fcvtau_2 = "1e250000DwNs|9e250000DxNs|1e650000DwNd|9e650000DxNd",
			
 
				+  fcvtms_2 = "1e300000DwNs|9e300000DxNs|1e700000DwNd|9e700000DxNd",
			
 
				+  fcvtmu_2 = "1e310000DwNs|9e310000DxNs|1e710000DwNd|9e710000DxNd",
			
 
				+  fcvtns_2 = "1e200000DwNs|9e200000DxNs|1e600000DwNd|9e600000DxNd",
			
 
				+  fcvtnu_2 = "1e210000DwNs|9e210000DxNs|1e610000DwNd|9e610000DxNd",
			
 
				+  fcvtps_2 = "1e280000DwNs|9e280000DxNs|1e680000DwNd|9e680000DxNd",
			
 
				+  fcvtpu_2 = "1e290000DwNs|9e290000DxNs|1e690000DwNd|9e690000DxNd",
			
 
				+  fcvtzs_2 = "1e380000DwNs|9e380000DxNs|1e780000DwNd|9e780000DxNd",
			
 
				+  fcvtzu_2 = "1e390000DwNs|9e390000DxNs|1e790000DwNd|9e790000DxNd",
			
 
				+
			
 
				+  scvtf_2  = "1e220000DsNw|9e220000DsNx|1e620000DdNw|9e620000DdNx",
			
 
				+  ucvtf_2  = "1e230000DsNw|9e230000DsNx|1e630000DdNw|9e630000DdNx",
			
 
				+
			
 
				+  frintn_2 = "1e244000DNf",
			
 
				+  frintp_2 = "1e24c000DNf",
			
 
				+  frintm_2 = "1e254000DNf",
			
 
				+  frintz_2 = "1e25c000DNf",
			
 
				+  frinta_2 = "1e264000DNf",
			
 
				+  frintx_2 = "1e274000DNf",
			
 
				+  frinti_2 = "1e27c000DNf",
			
 
				+
			
 
				+  fadd_3   = "1e202800DNMf",
			
 
				+  fsub_3   = "1e203800DNMf",
			
 
				+  fmul_3   = "1e200800DNMf",
			
 
				+  fnmul_3  = "1e208800DNMf",
			
 
				+  fdiv_3   = "1e201800DNMf",
			
 
				+
			
 
				+  fmadd_4  = "1f000000DNMAf",
			
 
				+  fmsub_4  = "1f008000DNMAf",
			
 
				+  fnmadd_4 = "1f200000DNMAf",
			
 
				+  fnmsub_4 = "1f208000DNMAf",
			
 
				+
			
 
				+  fmax_3   = "1e204800DNMf",
			
 
				+  fmaxnm_3 = "1e206800DNMf",
			
 
				+  fmin_3   = "1e205800DNMf",
			
 
				+  fminnm_3 = "1e207800DNMf",
			
 
				+
			
 
				+  fcmp_2   = "1e202000NMf|1e202008NZf",
			
 
				+  fcmpe_2  = "1e202010NMf|1e202018NZf",
			
 
				+
			
 
				+  fccmp_4  = "1e200400NMVCf",
			
 
				+  fccmpe_4 = "1e200410NMVCf",
			
 
				+
			
 
				+  fcsel_4  = "1e200c00DNMCf",
			
 
				+
			
 
				+  -- TODO: crc32*, aes*, sha*, pmull
			
 
				+  -- TODO: SIMD instructions.
			
 
				+}
			
 
				+
			
 
				+for cond,c in pairs(map_cond) do
			
 
				+  map_op["b"..cond.."_1"] = tohex(0x54000000+c).."B"
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Handle opcodes defined with template strings.
			
 
				+local function parse_template(params, template, nparams, pos)
			
 
				+  local op = tonumber(sub(template, 1, 8), 16)
			
 
				+  local n = 1
			
 
				+  local rtt = {}
			
 
				+
			
 
				+  parse_reg_type = false
			
 
				+
			
 
				+  -- Process each character.
			
 
				+  for p in gmatch(sub(template, 9), ".") do
			
 
				+    local q = params[n]
			
 
				+    if p == "D" then
			
 
				+      op = op + parse_reg(q); n = n + 1
			
 
				+    elseif p == "N" then
			
 
				+      op = op + shl(parse_reg(q), 5); n = n + 1
			
 
				+    elseif p == "M" then
			
 
				+      op = op + shl(parse_reg(q), 16); n = n + 1
			
 
				+    elseif p == "A" then
			
 
				+      op = op + shl(parse_reg(q), 10); n = n + 1
			
 
				+    elseif p == "m" then
			
 
				+      op = op + shl(parse_reg(params[n-1]), 16)
			
 
				+
			
 
				+    elseif p == "p" then
			
 
				+      if q == "sp" then params[n] = "@x31" end
			
 
				+    elseif p == "g" then
			
 
				+      if parse_reg_type == "x" then
			
 
				+	op = op + 0x80000000
			
 
				+      elseif parse_reg_type ~= "w" then
			
 
				+	werror("bad register type")
			
 
				+      end
			
 
				+      parse_reg_type = false
			
 
				+    elseif p == "f" then
			
 
				+      if parse_reg_type == "d" then
			
 
				+	op = op + 0x00400000
			
 
				+      elseif parse_reg_type ~= "s" then
			
 
				+	werror("bad register type")
			
 
				+      end
			
 
				+      parse_reg_type = false
			
 
				+    elseif p == "x" or p == "w" or p == "d" or p == "s" then
			
 
				+      if parse_reg_type ~= p then
			
 
				+	werror("register size mismatch")
			
 
				+      end
			
 
				+      parse_reg_type = false
			
 
				+
			
 
				+    elseif p == "L" then
			
 
				+      op = parse_load(params, nparams, n, op)
			
 
				+    elseif p == "P" then
			
 
				+      op = parse_load_pair(params, nparams, n, op)
			
 
				+
			
 
				+    elseif p == "B" then
			
 
				+      local mode, v, s = parse_label(q, false); n = n + 1
			
 
				+      local m = branch_type(op)
			
 
				+      waction("REL_"..mode, v+m, s, 1)
			
 
				+
			
 
				+    elseif p == "I" then
			
 
				+      op = op + parse_imm12(q); n = n + 1
			
 
				+    elseif p == "i" then
			
 
				+      op = op + parse_imm13(q); n = n + 1
			
 
				+    elseif p == "W" then
			
 
				+      op = op + parse_imm(q, 16, 5, 0, false); n = n + 1
			
 
				+    elseif p == "T" then
			
 
				+      op = op + parse_imm6(q); n = n + 1
			
 
				+    elseif p == "1" then
			
 
				+      op = op + parse_imm(q, 6, 16, 0, false); n = n + 1
			
 
				+    elseif p == "2" then
			
 
				+      op = op + parse_imm(q, 6, 10, 0, false); n = n + 1
			
 
				+    elseif p == "5" then
			
 
				+      op = op + parse_imm(q, 5, 16, 0, false); n = n + 1
			
 
				+    elseif p == "V" then
			
 
				+      op = op + parse_imm(q, 4, 0, 0, false); n = n + 1
			
 
				+    elseif p == "F" then
			
 
				+      op = op + parse_fpimm(q); n = n + 1
			
 
				+    elseif p == "Z" then
			
 
				+      if q ~= "#0" and q ~= "#0.0" then werror("expected zero immediate") end
			
 
				+      n = n + 1
			
 
				+
			
 
				+    elseif p == "S" then
			
 
				+      op = op + parse_shift(q); n = n + 1
			
 
				+    elseif p == "X" then
			
 
				+      op = op + parse_extend(q); n = n + 1
			
 
				+    elseif p == "R" then
			
 
				+      op = op + parse_lslx16(q); n = n + 1
			
 
				+    elseif p == "C" then
			
 
				+      op = op + parse_cond(q, 0); n = n + 1
			
 
				+    elseif p == "c" then
			
 
				+      op = op + parse_cond(q, 1); n = n + 1
			
 
				+
			
 
				+    else
			
 
				+      assert(false)
			
 
				+    end
			
 
				+  end
			
 
				+  wputpos(pos, op)
			
 
				+end
			
 
				+
			
 
				+function op_template(params, template, nparams)
			
 
				+  if not params then return template:gsub("%x%x%x%x%x%x%x%x", "") end
			
 
				+
			
 
				+  -- Limit number of section buffer positions used by a single dasm_put().
			
 
				+  -- A single opcode needs a maximum of 3 positions.
			
 
				+  if secpos+3 > maxsecpos then wflush() end
			
 
				+  local pos = wpos()
			
 
				+  local lpos, apos, spos = #actlist, #actargs, secpos
			
 
				+
			
 
				+  local ok, err
			
 
				+  for t in gmatch(template, "[^|]+") do
			
 
				+    ok, err = pcall(parse_template, params, t, nparams, pos)
			
 
				+    if ok then return end
			
 
				+    secpos = spos
			
 
				+    actlist[lpos+1] = nil
			
 
				+    actlist[lpos+2] = nil
			
 
				+    actlist[lpos+3] = nil
			
 
				+    actargs[apos+1] = nil
			
 
				+    actargs[apos+2] = nil
			
 
				+    actargs[apos+3] = nil
			
 
				+  end
			
 
				+  error(err, 0)
			
 
				+end
			
 
				+
			
 
				+map_op[".template__"] = op_template
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Pseudo-opcode to mark the position where the action list is to be emitted.
			
 
				+map_op[".actionlist_1"] = function(params)
			
 
				+  if not params then return "cvar" end
			
 
				+  local name = params[1] -- No syntax check. You get to keep the pieces.
			
 
				+  wline(function(out) writeactions(out, name) end)
			
 
				+end
			
 
				+
			
 
				+-- Pseudo-opcode to mark the position where the global enum is to be emitted.
			
 
				+map_op[".globals_1"] = function(params)
			
 
				+  if not params then return "prefix" end
			
 
				+  local prefix = params[1] -- No syntax check. You get to keep the pieces.
			
 
				+  wline(function(out) writeglobals(out, prefix) end)
			
 
				+end
			
 
				+
			
 
				+-- Pseudo-opcode to mark the position where the global names are to be emitted.
			
 
				+map_op[".globalnames_1"] = function(params)
			
 
				+  if not params then return "cvar" end
			
 
				+  local name = params[1] -- No syntax check. You get to keep the pieces.
			
 
				+  wline(function(out) writeglobalnames(out, name) end)
			
 
				+end
			
 
				+
			
 
				+-- Pseudo-opcode to mark the position where the extern names are to be emitted.
			
 
				+map_op[".externnames_1"] = function(params)
			
 
				+  if not params then return "cvar" end
			
 
				+  local name = params[1] -- No syntax check. You get to keep the pieces.
			
 
				+  wline(function(out) writeexternnames(out, name) end)
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Label pseudo-opcode (converted from trailing colon form).
			
 
				+map_op[".label_1"] = function(params)
			
 
				+  if not params then return "[1-9] | ->global | =>pcexpr" end
			
 
				+  if secpos+1 > maxsecpos then wflush() end
			
 
				+  local mode, n, s = parse_label(params[1], true)
			
 
				+  if mode == "EXT" then werror("bad label definition") end
			
 
				+  waction("LABEL_"..mode, n, s, 1)
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Pseudo-opcodes for data storage.
			
 
				+map_op[".long_*"] = function(params)
			
 
				+  if not params then return "imm..." end
			
 
				+  for _,p in ipairs(params) do
			
 
				+    local n = tonumber(p)
			
 
				+    if not n then werror("bad immediate `"..p.."'") end
			
 
				+    if n < 0 then n = n + 2^32 end
			
 
				+    wputw(n)
			
 
				+    if secpos+2 > maxsecpos then wflush() end
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+-- Alignment pseudo-opcode.
			
 
				+map_op[".align_1"] = function(params)
			
 
				+  if not params then return "numpow2" end
			
 
				+  if secpos+1 > maxsecpos then wflush() end
			
 
				+  local align = tonumber(params[1])
			
 
				+  if align then
			
 
				+    local x = align
			
 
				+    -- Must be a power of 2 in the range (2 ... 256).
			
 
				+    for i=1,8 do
			
 
				+      x = x / 2
			
 
				+      if x == 1 then
			
 
				+	waction("ALIGN", align-1, nil, 1) -- Action byte is 2**n-1.
			
 
				+	return
			
 
				+      end
			
 
				+    end
			
 
				+  end
			
 
				+  werror("bad alignment")
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Pseudo-opcode for (primitive) type definitions (map to C types).
			
 
				+map_op[".type_3"] = function(params, nparams)
			
 
				+  if not params then
			
 
				+    return nparams == 2 and "name, ctype" or "name, ctype, reg"
			
 
				+  end
			
 
				+  local name, ctype, reg = params[1], params[2], params[3]
			
 
				+  if not match(name, "^[%a_][%w_]*$") then
			
 
				+    werror("bad type name `"..name.."'")
			
 
				+  end
			
 
				+  local tp = map_type[name]
			
 
				+  if tp then
			
 
				+    werror("duplicate type `"..name.."'")
			
 
				+  end
			
 
				+  -- Add #type to defines. A bit unclean to put it in map_archdef.
			
 
				+  map_archdef["#"..name] = "sizeof("..ctype..")"
			
 
				+  -- Add new type and emit shortcut define.
			
 
				+  local num = ctypenum + 1
			
 
				+  map_type[name] = {
			
 
				+    ctype = ctype,
			
 
				+    ctypefmt = format("Dt%X(%%s)", num),
			
 
				+    reg = reg,
			
 
				+  }
			
 
				+  wline(format("#define Dt%X(_V) (int)(ptrdiff_t)&(((%s *)0)_V)", num, ctype))
			
 
				+  ctypenum = num
			
 
				+end
			
 
				+map_op[".type_2"] = map_op[".type_3"]
			
 
				+
			
 
				+-- Dump type definitions.
			
 
				+local function dumptypes(out, lvl)
			
 
				+  local t = {}
			
 
				+  for name in pairs(map_type) do t[#t+1] = name end
			
 
				+  sort(t)
			
 
				+  out:write("Type definitions:\n")
			
 
				+  for _,name in ipairs(t) do
			
 
				+    local tp = map_type[name]
			
 
				+    local reg = tp.reg or ""
			
 
				+    out:write(format("  %-20s %-20s %s\n", name, tp.ctype, reg))
			
 
				+  end
			
 
				+  out:write("\n")
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Set the current section.
			
 
				+function _M.section(num)
			
 
				+  waction("SECTION", num)
			
 
				+  wflush(true) -- SECTION is a terminal action.
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Dump architecture description.
			
 
				+function _M.dumparch(out)
			
 
				+  out:write(format("DynASM %s version %s, released %s\n\n",
			
 
				+    _info.arch, _info.version, _info.release))
			
 
				+  dumpactions(out)
			
 
				+end
			
 
				+
			
 
				+-- Dump all user defined elements.
			
 
				+function _M.dumpdef(out, lvl)
			
 
				+  dumptypes(out, lvl)
			
 
				+  dumpglobals(out, lvl)
			
 
				+  dumpexterns(out, lvl)
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Pass callbacks from/to the DynASM core.
			
 
				+function _M.passcb(wl, we, wf, ww)
			
 
				+  wline, werror, wfatal, wwarn = wl, we, wf, ww
			
 
				+  return wflush
			
 
				+end
			
 
				+
			
 
				+-- Setup the arch-specific module.
			
 
				+function _M.setup(arch, opt)
			
 
				+  g_arch, g_opt = arch, opt
			
 
				+end
			
 
				+
			
 
				+-- Merge the core maps and the arch-specific maps.
			
 
				+function _M.mergemaps(map_coreop, map_def)
			
 
				+  setmetatable(map_op, { __index = map_coreop })
			
 
				+  setmetatable(map_def, { __index = map_archdef })
			
 
				+  return map_op, map_def
			
 
				+end
			
 
				+
			
 
				+return _M
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
--- a/luajit.mod/luajit/dynasm/dasm_mips.h
+++ b/luajit.mod/luajit/dynasm/dasm_mips.h
@@ -21,7 +21,7 @@ enum {
 
				   /* The following actions need a buffer position. */
			
 
				   DASM_ALIGN, DASM_REL_LG, DASM_LABEL_LG,
			
 
				   /* The following actions also have an argument. */
			
 
				-  DASM_REL_PC, DASM_LABEL_PC, DASM_IMM,
			
 
				+  DASM_REL_PC, DASM_LABEL_PC, DASM_IMM, DASM_IMMS,
			
 
				   DASM__MAX
			
 
				 };
			
 
				 
			
@@ -231,7 +231,7 @@ void dasm_put(Dst_DECL, int start, ...)
 
				 	*pl = -pos;  /* Label exists now. */
			
 
				 	b[pos++] = ofs;  /* Store pass1 offset estimate. */
			
 
				 	break;
			
 
				-      case DASM_IMM:
			
 
				+      case DASM_IMM: case DASM_IMMS:
			
 
				 #ifdef DASM_CHECKS
			
 
				 	CK((n & ((1<<((ins>>10)&31))-1)) == 0, RANGE_I);
			
 
				 #endif
			
@@ -299,7 +299,7 @@ int dasm_link(Dst_DECL, size_t *szp)
 
				 	case DASM_ALIGN: ofs -= (b[pos++] + ofs) & (ins & 255); break;
			
 
				 	case DASM_REL_LG: case DASM_REL_PC: pos++; break;
			
 
				 	case DASM_LABEL_LG: case DASM_LABEL_PC: b[pos++] += ofs; break;
			
 
				-	case DASM_IMM: pos++; break;
			
 
				+	case DASM_IMM: case DASM_IMMS: pos++; break;
			
 
				 	}
			
 
				       }
			
 
				       stop: (void)0;
			
@@ -350,13 +350,14 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	  break;
			
 
				 	case DASM_REL_LG:
			
 
				 	  CK(n >= 0, UNDEF_LG);
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_REL_PC:
			
 
				 	  CK(n >= 0, UNDEF_PC);
			
 
				 	  n = *DASM_POS2PTR(D, n);
			
 
				 	  if (ins & 2048)
			
 
				 	    n = n - (int)((char *)cp - base);
			
 
				 	  else
			
 
				-	    n = (n + (int)base) & 0x0fffffff;
			
 
				+	    n = (n + (int)(size_t)base) & 0x0fffffff;
			
 
				 	patchrel:
			
 
				 	  CK((n & 3) == 0 &&
			
 
				 	     ((n + ((ins & 2048) ? 0x00020000 : 0)) >>
			
@@ -367,6 +368,9 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	  ins &= 2047; if (ins >= 20) D->globals[ins-10] = (void *)(base + n);
			
 
				 	  break;
			
 
				 	case DASM_LABEL_PC: break;
			
 
				+	case DASM_IMMS:
			
 
				+	  cp[-1] |= ((n>>3) & 4); n &= 0x1f;
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_IMM:
			
 
				 	  cp[-1] |= (n & ((1<<((ins>>5)&31))-1)) << (ins&31);
			
 
				 	  break;
			
--- a/luajit.mod/luajit/dynasm/dasm_mips.lua
+++ b/luajit.mod/luajit/dynasm/dasm_mips.lua
@@ -1,17 +1,19 @@
 
				 ------------------------------------------------------------------------------
			
 
				--- DynASM MIPS module.
			
 
				+-- DynASM MIPS32/MIPS64 module.
			
 
				 --
			
 
				 -- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				 -- See dynasm.lua for full copyright notice.
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				+local mips64 = mips64
			
 
				+
			
 
				 -- Module information:
			
 
				 local _info = {
			
 
				-  arch =	"mips",
			
 
				-  description =	"DynASM MIPS module",
			
 
				-  version =	"1.3.0",
			
 
				-  vernum =	 10300,
			
 
				-  release =	"2012-01-23",
			
 
				+  arch =	mips64 and "mips64" or "mips",
			
 
				+  description =	"DynASM MIPS32/MIPS64 module",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2016-05-24",
			
 
				   author =	"Mike Pall",
			
 
				   license =	"MIT",
			
 
				 }
			
@@ -27,7 +29,8 @@ local sub, format, byte, char = _s.sub, _s.format, _s.byte, _s.char
 
				 local match, gmatch = _s.match, _s.gmatch
			
 
				 local concat, sort = table.concat, table.sort
			
 
				 local bit = bit or require("bit")
			
 
				-local band, shl, sar, tohex = bit.band, bit.lshift, bit.arshift, bit.tohex
			
 
				+local band, shl, shr, sar = bit.band, bit.lshift, bit.rshift, bit.arshift
			
 
				+local tohex = bit.tohex
			
 
				 
			
 
				 -- Inherited tables and callbacks.
			
 
				 local g_opt, g_arch
			
@@ -38,7 +41,7 @@ local wline, werror, wfatal, wwarn
 
				 local action_names = {
			
 
				   "STOP", "SECTION", "ESC", "REL_EXT",
			
 
				   "ALIGN", "REL_LG", "LABEL_LG",
			
 
				-  "REL_PC", "LABEL_PC", "IMM",
			
 
				+  "REL_PC", "LABEL_PC", "IMM", "IMMS",
			
 
				 }
			
 
				 
			
 
				 -- Maximum number of section buffer positions for dasm_put().
			
@@ -251,6 +254,10 @@ local map_op = {
 
				   bnel_3 =	"54000000STB",
			
 
				   blezl_2 =	"58000000SB",
			
 
				   bgtzl_2 =	"5c000000SB",
			
 
				+  daddi_3 =	mips64 and "60000000TSI",
			
 
				+  daddiu_3 =	mips64 and "64000000TSI",
			
 
				+  ldl_2 =	mips64 and "68000000TO",
			
 
				+  ldr_2 =	mips64 and "6c000000TO",
			
 
				   lb_2 =	"80000000TO",
			
 
				   lh_2 =	"84000000TO",
			
 
				   lwl_2 =	"88000000TO",
			
@@ -258,23 +265,30 @@ local map_op = {
 
				   lbu_2 =	"90000000TO",
			
 
				   lhu_2 =	"94000000TO",
			
 
				   lwr_2 =	"98000000TO",
			
 
				+  lwu_2 =	mips64 and "9c000000TO",
			
 
				   sb_2 =	"a0000000TO",
			
 
				   sh_2 =	"a4000000TO",
			
 
				   swl_2 =	"a8000000TO",
			
 
				   sw_2 =	"ac000000TO",
			
 
				+  sdl_2 =	mips64 and "b0000000TO",
			
 
				+  sdr_2 =	mips64 and "b1000000TO",
			
 
				   swr_2 =	"b8000000TO",
			
 
				   cache_2 =	"bc000000NO",
			
 
				   ll_2 =	"c0000000TO",
			
 
				   lwc1_2 =	"c4000000HO",
			
 
				   pref_2 =	"cc000000NO",
			
 
				   ldc1_2 =	"d4000000HO",
			
 
				+  ld_2 =	mips64 and "dc000000TO",
			
 
				   sc_2 =	"e0000000TO",
			
 
				   swc1_2 =	"e4000000HO",
			
 
				+  scd_2 =	mips64 and "f0000000TO",
			
 
				   sdc1_2 =	"f4000000HO",
			
 
				+  sd_2 =	mips64 and "fc000000TO",
			
 
				 
			
 
				   -- Opcode SPECIAL.
			
 
				   nop_0 =	"00000000",
			
 
				   sll_3 =	"00000000DTA",
			
 
				+  sextw_2 =	"00000000DT",
			
 
				   movf_2 =	"00000001DS",
			
 
				   movf_3 =	"00000001DSC",
			
 
				   movt_2 =	"00010001DS",
			
@@ -285,6 +299,7 @@ local map_op = {
 
				   sllv_3 =	"00000004DTS",
			
 
				   srlv_3 =	"00000006DTS",
			
 
				   rotrv_3 =	"00000046DTS",
			
 
				+  drotrv_3 =	mips64 and "00000056DTS",
			
 
				   srav_3 =	"00000007DTS",
			
 
				   jr_1 =	"00000008S",
			
 
				   jalr_1 =	"0000f809S",
			
@@ -300,15 +315,22 @@ local map_op = {
 
				   mthi_1 =	"00000011S",
			
 
				   mflo_1 =	"00000012D",
			
 
				   mtlo_1 =	"00000013S",
			
 
				+  dsllv_3 =	mips64 and "00000014DTS",
			
 
				+  dsrlv_3 =	mips64 and "00000016DTS",
			
 
				+  dsrav_3 =	mips64 and "00000017DTS",
			
 
				   mult_2 =	"00000018ST",
			
 
				   multu_2 =	"00000019ST",
			
 
				   div_2 =	"0000001aST",
			
 
				   divu_2 =	"0000001bST",
			
 
				+  dmult_2 =	mips64 and "0000001cST",
			
 
				+  dmultu_2 =	mips64 and "0000001dST",
			
 
				+  ddiv_2 =	mips64 and "0000001eST",
			
 
				+  ddivu_2 =	mips64 and "0000001fST",
			
 
				   add_3 =	"00000020DST",
			
 
				-  move_2 =	"00000021DS",
			
 
				+  move_2 =	mips64 and "00000025DS" or "00000021DS",
			
 
				   addu_3 =	"00000021DST",
			
 
				   sub_3 =	"00000022DST",
			
 
				-  negu_2 =	"00000023DT",
			
 
				+  negu_2 =	mips64 and "0000002fDT" or "00000023DT",
			
 
				   subu_3 =	"00000023DST",
			
 
				   and_3 =	"00000024DST",
			
 
				   or_3 =	"00000025DST",
			
@@ -317,6 +339,10 @@ local map_op = {
 
				   nor_3 =	"00000027DST",
			
 
				   slt_3 =	"0000002aDST",
			
 
				   sltu_3 =	"0000002bDST",
			
 
				+  dadd_3 =	mips64 and "0000002cDST",
			
 
				+  daddu_3 =	mips64 and "0000002dDST",
			
 
				+  dsub_3 =	mips64 and "0000002eDST",
			
 
				+  dsubu_3 =	mips64 and "0000002fDST",
			
 
				   tge_2 =	"00000030ST",
			
 
				   tge_3 =	"00000030STZ",
			
 
				   tgeu_2 =	"00000031ST",
			
@@ -329,6 +355,14 @@ local map_op = {
 
				   teq_3 =	"00000034STZ",
			
 
				   tne_2 =	"00000036ST",
			
 
				   tne_3 =	"00000036STZ",
			
 
				+  dsll_3 =	mips64 and "00000038DTa",
			
 
				+  dsrl_3 =	mips64 and "0000003aDTa",
			
 
				+  drotr_3 =	mips64 and "0020003aDTa",
			
 
				+  dsra_3 =	mips64 and "0000003bDTa",
			
 
				+  dsll32_3 =	mips64 and "0000003cDTA",
			
 
				+  dsrl32_3 =	mips64 and "0000003eDTA",
			
 
				+  drotr32_3 =	mips64 and "0020003eDTA",
			
 
				+  dsra32_3 =	mips64 and "0000003fDTA",
			
 
				 
			
 
				   -- Opcode REGIMM.
			
 
				   bltz_2 =	"04000000SB",
			
@@ -356,13 +390,24 @@ local map_op = {
 
				   msubu_2 =	"70000005ST",
			
 
				   clz_2 =	"70000020DS=",
			
 
				   clo_2 =	"70000021DS=",
			
 
				+  dclz_2 =	mips64 and "70000024DS=",
			
 
				+  dclo_2 =	mips64 and "70000025DS=",
			
 
				   sdbbp_0 =	"7000003f",
			
 
				   sdbbp_1 =	"7000003fY",
			
 
				 
			
 
				   -- Opcode SPECIAL3.
			
 
				   ext_4 =	"7c000000TSAM", -- Note: last arg is msbd = size-1
			
 
				+  dextm_4 =	mips64 and "7c000001TSAM", -- Args: pos    | size-1-32
			
 
				+  dextu_4 =	mips64 and "7c000002TSAM", -- Args: pos-32 | size-1
			
 
				+  dext_4 =	mips64 and "7c000003TSAM", -- Args: pos    | size-1
			
 
				+  zextw_2 =	mips64 and "7c00f803TS",
			
 
				   ins_4 =	"7c000004TSAM", -- Note: last arg is msb = pos+size-1
			
 
				+  dinsm_4 =	mips64 and "7c000005TSAM", -- Args: pos    | pos+size-33
			
 
				+  dinsu_4 =	mips64 and "7c000006TSAM", -- Args: pos-32 | pos+size-33
			
 
				+  dins_4 =	mips64 and "7c000007TSAM", -- Args: pos    | pos+size-1
			
 
				   wsbh_2 =	"7c0000a0DT",
			
 
				+  dsbh_2 =	mips64 and "7c0000a4DT",
			
 
				+  dshd_2 =	mips64 and "7c000164DT",
			
 
				   seb_2 =	"7c000420DT",
			
 
				   seh_2 =	"7c000620DT",
			
 
				   rdhwr_2 =	"7c00003bTD",
			
@@ -370,8 +415,12 @@ local map_op = {
 
				   -- Opcode COP0.
			
 
				   mfc0_2 =	"40000000TD",
			
 
				   mfc0_3 =	"40000000TDW",
			
 
				+  dmfc0_2 =	mips64 and "40200000TD",
			
 
				+  dmfc0_3 =	mips64 and "40200000TDW",
			
 
				   mtc0_2 =	"40800000TD",
			
 
				   mtc0_3 =	"40800000TDW",
			
 
				+  dmtc0_2 =	mips64 and "40a00000TD",
			
 
				+  dmtc0_3 =	mips64 and "40a00000TDW",
			
 
				   rdpgpr_2 =	"41400000DT",
			
 
				   di_0 =	"41606000",
			
 
				   di_1 =	"41606000T",
			
@@ -388,9 +437,11 @@ local map_op = {
 
				 
			
 
				   -- Opcode COP1.
			
 
				   mfc1_2 =	"44000000TG",
			
 
				+  dmfc1_2 =	mips64 and "44200000TG",
			
 
				   cfc1_2 =	"44400000TG",
			
 
				   mfhc1_2 =	"44600000TG",
			
 
				   mtc1_2 =	"44800000TG",
			
 
				+  dmtc1_2 =	mips64 and "44a00000TG",
			
 
				   ctc1_2 =	"44c00000TG",
			
 
				   mthc1_2 =	"44e00000TG",
			
 
				 
			
@@ -633,7 +684,7 @@ local function parse_fpr(expr)
 
				   werror("bad register name `"..expr.."'")
			
 
				 end
			
 
				 
			
 
				-local function parse_imm(imm, bits, shift, scale, signed)
			
 
				+local function parse_imm(imm, bits, shift, scale, signed, action)
			
 
				   local n = tonumber(imm)
			
 
				   if n then
			
 
				     local m = sar(n, scale)
			
@@ -651,7 +702,8 @@ local function parse_imm(imm, bits, shift, scale, signed)
 
				 	 match(imm, "^([%w_]+):([rf][1-3]?[0-9])$") then
			
 
				     werror("expected immediate operand, got register")
			
 
				   else
			
 
				-    waction("IMM", (signed and 32768 or 0)+scale*1024+bits*32+shift, imm)
			
 
				+    waction(action or "IMM",
			
 
				+	    (signed and 32768 or 0)+shl(scale, 10)+shl(bits, 5)+shift, imm)
			
 
				     return 0
			
 
				   end
			
 
				 end
			
@@ -757,12 +809,15 @@ map_op[".template__"] = function(params, template, nparams)
 
				     elseif p == "X" then
			
 
				       op = op + parse_index(params[n]); n = n + 1
			
 
				     elseif p == "B" or p == "J" then
			
 
				-      local mode, n, s = parse_label(params[n], false)
			
 
				-      if p == "B" then n = n + 2048 end
			
 
				-      waction("REL_"..mode, n, s, 1)
			
 
				+      local mode, m, s = parse_label(params[n], false)
			
 
				+      if p == "B" then m = m + 2048 end
			
 
				+      waction("REL_"..mode, m, s, 1)
			
 
				       n = n + 1
			
 
				     elseif p == "A" then
			
 
				       op = op + parse_imm(params[n], 5, 6, 0, false); n = n + 1
			
 
				+    elseif p == "a" then
			
 
				+      local m = parse_imm(params[n], 6, 6, 0, false, "IMMS"); n = n + 1
			
 
				+      op = op + band(m, 0x7c0) + band(shr(m, 9), 4)
			
 
				     elseif p == "M" then
			
 
				       op = op + parse_imm(params[n], 5, 11, 0, false); n = n + 1
			
 
				     elseif p == "N" then
			
--- a/luajit.mod/luajit/dynasm/dasm_mips64.lua
+++ b/luajit.mod/luajit/dynasm/dasm_mips64.lua
@@ -0,0 +1,12 @@
 
				+------------------------------------------------------------------------------
			
 
				+-- DynASM MIPS64 module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- See dynasm.lua for full copyright notice.
			
 
				+------------------------------------------------------------------------------
			
 
				+-- This module just sets 64 bit mode for the combined MIPS/MIPS64 module.
			
 
				+-- All the interesting stuff is there.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+mips64 = true -- Using a global is an ugly, but effective solution.
			
 
				+return require("dasm_mips")
			
--- a/luajit.mod/luajit/dynasm/dasm_ppc.h
+++ b/luajit.mod/luajit/dynasm/dasm_ppc.h
@@ -1,5 +1,5 @@
 
				 /*
			
 
				-** DynASM PPC encoding engine.
			
 
				+** DynASM PPC/PPC64 encoding engine.
			
 
				 ** Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				 ** Released under the MIT license. See dynasm.lua for full copyright notice.
			
 
				 */
			
@@ -21,7 +21,7 @@ enum {
 
				   /* The following actions need a buffer position. */
			
 
				   DASM_ALIGN, DASM_REL_LG, DASM_LABEL_LG,
			
 
				   /* The following actions also have an argument. */
			
 
				-  DASM_REL_PC, DASM_LABEL_PC, DASM_IMM,
			
 
				+  DASM_REL_PC, DASM_LABEL_PC, DASM_IMM, DASM_IMMSH,
			
 
				   DASM__MAX
			
 
				 };
			
 
				 
			
@@ -244,6 +244,10 @@ void dasm_put(Dst_DECL, int start, ...)
 
				 #endif
			
 
				 	b[pos++] = n;
			
 
				 	break;
			
 
				+      case DASM_IMMSH:
			
 
				+	CK((n >> 6) == 0, RANGE_I);
			
 
				+	b[pos++] = n;
			
 
				+	break;
			
 
				       }
			
 
				     }
			
 
				   }
			
@@ -299,7 +303,7 @@ int dasm_link(Dst_DECL, size_t *szp)
 
				 	case DASM_ALIGN: ofs -= (b[pos++] + ofs) & (ins & 255); break;
			
 
				 	case DASM_REL_LG: case DASM_REL_PC: pos++; break;
			
 
				 	case DASM_LABEL_LG: case DASM_LABEL_PC: b[pos++] += ofs; break;
			
 
				-	case DASM_IMM: pos++; break;
			
 
				+	case DASM_IMM: case DASM_IMMSH: pos++; break;
			
 
				 	}
			
 
				       }
			
 
				       stop: (void)0;
			
@@ -350,6 +354,7 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	  break;
			
 
				 	case DASM_REL_LG:
			
 
				 	  CK(n >= 0, UNDEF_LG);
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_REL_PC:
			
 
				 	  CK(n >= 0, UNDEF_PC);
			
 
				 	  n = *DASM_POS2PTR(D, n) - (int)((char *)cp - base);
			
@@ -366,6 +371,9 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	case DASM_IMM:
			
 
				 	  cp[-1] |= (n & ((1<<((ins>>5)&31))-1)) << (ins&31);
			
 
				 	  break;
			
 
				+	case DASM_IMMSH:
			
 
				+	  cp[-1] |= (ins & 1) ? ((n&31)<<11)|((n&32)>>4) : ((n&31)<<6)|(n&32);
			
 
				+	  break;
			
 
				 	default: *cp++ = ins; break;
			
 
				 	}
			
 
				       }
			
--- a/luajit.mod/luajit/dynasm/dasm_ppc.lua
+++ b/luajit.mod/luajit/dynasm/dasm_ppc.lua
@@ -1,17 +1,19 @@
 
				 ------------------------------------------------------------------------------
			
 
				--- DynASM PPC module.
			
 
				+-- DynASM PPC/PPC64 module.
			
 
				 --
			
 
				 -- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				 -- See dynasm.lua for full copyright notice.
			
 
				+--
			
 
				+-- Support for various extensions contributed by Caio Souza Oliveira.
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 -- Module information:
			
 
				 local _info = {
			
 
				   arch =	"ppc",
			
 
				   description =	"DynASM PPC module",
			
 
				-  version =	"1.3.0",
			
 
				-  vernum =	 10300,
			
 
				-  release =	"2011-05-05",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2015-10-18",
			
 
				   author =	"Mike Pall",
			
 
				   license =	"MIT",
			
 
				 }
			
@@ -39,7 +41,7 @@ local wline, werror, wfatal, wwarn
 
				 local action_names = {
			
 
				   "STOP", "SECTION", "ESC", "REL_EXT",
			
 
				   "ALIGN", "REL_LG", "LABEL_LG",
			
 
				-  "REL_PC", "LABEL_PC", "IMM",
			
 
				+  "REL_PC", "LABEL_PC", "IMM", "IMMSH"
			
 
				 }
			
 
				 
			
 
				 -- Maximum number of section buffer positions for dasm_put().
			
@@ -228,8 +230,18 @@ local map_cond = {
 
				 
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				+local map_op, op_template
			
 
				+
			
 
				+local function op_alias(opname, f)
			
 
				+  return function(params, nparams)
			
 
				+    if not params then return "-> "..opname:sub(1, -3) end
			
 
				+    f(params, nparams)
			
 
				+    op_template(params, map_op[opname], nparams)
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				 -- Template strings for PPC instructions.
			
 
				-local map_op = {
			
 
				+map_op = {
			
 
				   tdi_3 =	"08000000ARI",
			
 
				   twi_3 =	"0c000000ARI",
			
 
				   mulli_3 =	"1c000000RRI",
			
@@ -297,6 +309,250 @@ local map_op = {
 
				   std_2 =	"f8000000RD",
			
 
				   stdu_2 =	"f8000001RD",
			
 
				 
			
 
				+  subi_3 =	op_alias("addi_3", function(p) p[3] = "-("..p[3]..")" end),
			
 
				+  subis_3 =	op_alias("addis_3", function(p) p[3] = "-("..p[3]..")" end),
			
 
				+  subic_3 =	op_alias("addic_3", function(p) p[3] = "-("..p[3]..")" end),
			
 
				+  ["subic._3"] = op_alias("addic._3", function(p) p[3] = "-("..p[3]..")" end),
			
 
				+
			
 
				+  rotlwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[4] = "0"; p[5] = "31"
			
 
				+  end),
			
 
				+  rotrwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[3] = "32-("..p[3]..")"; p[4] = "0"; p[5] = "31"
			
 
				+  end),
			
 
				+  rotlw_3 =	op_alias("rlwnm_5", function(p)
			
 
				+    p[4] = "0"; p[5] = "31"
			
 
				+  end),
			
 
				+  slwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[5] = "31-("..p[3]..")"; p[4] = "0"
			
 
				+  end),
			
 
				+  srwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[4] = p[3]; p[3] = "32-("..p[3]..")"; p[5] = "31"
			
 
				+  end),
			
 
				+  clrlwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[4] = p[3]; p[3] = "0"; p[5] = "31"
			
 
				+  end),
			
 
				+  clrrwi_3 =	op_alias("rlwinm_5", function(p)
			
 
				+    p[5] = "31-("..p[3]..")"; p[3] = "0"; p[4] = "0"
			
 
				+  end),
			
 
				+
			
 
				+  -- Primary opcode 4:
			
 
				+  mulhhwu_3 =		"10000010RRR.",
			
 
				+  machhwu_3 =		"10000018RRR.",
			
 
				+  mulhhw_3 =		"10000050RRR.",
			
 
				+  nmachhw_3 =		"1000005cRRR.",
			
 
				+  machhwsu_3 =		"10000098RRR.",
			
 
				+  machhws_3 =		"100000d8RRR.",
			
 
				+  nmachhws_3 =		"100000dcRRR.",
			
 
				+  mulchwu_3 =		"10000110RRR.",
			
 
				+  macchwu_3 =		"10000118RRR.",
			
 
				+  mulchw_3 =		"10000150RRR.",
			
 
				+  macchw_3 =		"10000158RRR.",
			
 
				+  nmacchw_3 =		"1000015cRRR.",
			
 
				+  macchwsu_3 =		"10000198RRR.",
			
 
				+  macchws_3 =		"100001d8RRR.",
			
 
				+  nmacchws_3 =		"100001dcRRR.",
			
 
				+  mullhw_3 =		"10000350RRR.",
			
 
				+  maclhw_3 =		"10000358RRR.",
			
 
				+  nmaclhw_3 =		"1000035cRRR.",
			
 
				+  maclhwsu_3 =		"10000398RRR.",
			
 
				+  maclhws_3 =		"100003d8RRR.",
			
 
				+  nmaclhws_3 =		"100003dcRRR.",
			
 
				+  machhwuo_3 =		"10000418RRR.",
			
 
				+  nmachhwo_3 =		"1000045cRRR.",
			
 
				+  machhwsuo_3 =		"10000498RRR.",
			
 
				+  machhwso_3 =		"100004d8RRR.",
			
 
				+  nmachhwso_3 =		"100004dcRRR.",
			
 
				+  macchwuo_3 =		"10000518RRR.",
			
 
				+  macchwo_3 =		"10000558RRR.",
			
 
				+  nmacchwo_3 =		"1000055cRRR.",
			
 
				+  macchwsuo_3 =		"10000598RRR.",
			
 
				+  macchwso_3 =		"100005d8RRR.",
			
 
				+  nmacchwso_3 =		"100005dcRRR.",
			
 
				+  maclhwo_3 =		"10000758RRR.",
			
 
				+  nmaclhwo_3 =		"1000075cRRR.",
			
 
				+  maclhwsuo_3 =		"10000798RRR.",
			
 
				+  maclhwso_3 =		"100007d8RRR.",
			
 
				+  nmaclhwso_3 =		"100007dcRRR.",
			
 
				+
			
 
				+  vaddubm_3 =		"10000000VVV",
			
 
				+  vmaxub_3 =		"10000002VVV",
			
 
				+  vrlb_3 =		"10000004VVV",
			
 
				+  vcmpequb_3 =		"10000006VVV",
			
 
				+  vmuloub_3 =		"10000008VVV",
			
 
				+  vaddfp_3 =		"1000000aVVV",
			
 
				+  vmrghb_3 =		"1000000cVVV",
			
 
				+  vpkuhum_3 =		"1000000eVVV",
			
 
				+  vmhaddshs_4 =		"10000020VVVV",
			
 
				+  vmhraddshs_4 =	"10000021VVVV",
			
 
				+  vmladduhm_4 =		"10000022VVVV",
			
 
				+  vmsumubm_4 =		"10000024VVVV",
			
 
				+  vmsummbm_4 =		"10000025VVVV",
			
 
				+  vmsumuhm_4 =		"10000026VVVV",
			
 
				+  vmsumuhs_4 =		"10000027VVVV",
			
 
				+  vmsumshm_4 =		"10000028VVVV",
			
 
				+  vmsumshs_4 =		"10000029VVVV",
			
 
				+  vsel_4 =		"1000002aVVVV",
			
 
				+  vperm_4 =		"1000002bVVVV",
			
 
				+  vsldoi_4 =		"1000002cVVVP",
			
 
				+  vpermxor_4 =		"1000002dVVVV",
			
 
				+  vmaddfp_4 =		"1000002eVVVV~",
			
 
				+  vnmsubfp_4 =		"1000002fVVVV~",
			
 
				+  vaddeuqm_4 =		"1000003cVVVV",
			
 
				+  vaddecuq_4 =		"1000003dVVVV",
			
 
				+  vsubeuqm_4 =		"1000003eVVVV",
			
 
				+  vsubecuq_4 =		"1000003fVVVV",
			
 
				+  vadduhm_3 =		"10000040VVV",
			
 
				+  vmaxuh_3 =		"10000042VVV",
			
 
				+  vrlh_3 =		"10000044VVV",
			
 
				+  vcmpequh_3 =		"10000046VVV",
			
 
				+  vmulouh_3 =		"10000048VVV",
			
 
				+  vsubfp_3 =		"1000004aVVV",
			
 
				+  vmrghh_3 =		"1000004cVVV",
			
 
				+  vpkuwum_3 =		"1000004eVVV",
			
 
				+  vadduwm_3 =		"10000080VVV",
			
 
				+  vmaxuw_3 =		"10000082VVV",
			
 
				+  vrlw_3 =		"10000084VVV",
			
 
				+  vcmpequw_3 =		"10000086VVV",
			
 
				+  vmulouw_3 =		"10000088VVV",
			
 
				+  vmuluwm_3 =		"10000089VVV",
			
 
				+  vmrghw_3 =		"1000008cVVV",
			
 
				+  vpkuhus_3 =		"1000008eVVV",
			
 
				+  vaddudm_3 =		"100000c0VVV",
			
 
				+  vmaxud_3 =		"100000c2VVV",
			
 
				+  vrld_3 =		"100000c4VVV",
			
 
				+  vcmpeqfp_3 =		"100000c6VVV",
			
 
				+  vcmpequd_3 =		"100000c7VVV",
			
 
				+  vpkuwus_3 =		"100000ceVVV",
			
 
				+  vadduqm_3 =		"10000100VVV",
			
 
				+  vmaxsb_3 =		"10000102VVV",
			
 
				+  vslb_3 =		"10000104VVV",
			
 
				+  vmulosb_3 =		"10000108VVV",
			
 
				+  vrefp_2 =		"1000010aV-V",
			
 
				+  vmrglb_3 =		"1000010cVVV",
			
 
				+  vpkshus_3 =		"1000010eVVV",
			
 
				+  vaddcuq_3 =		"10000140VVV",
			
 
				+  vmaxsh_3 =		"10000142VVV",
			
 
				+  vslh_3 =		"10000144VVV",
			
 
				+  vmulosh_3 =		"10000148VVV",
			
 
				+  vrsqrtefp_2 =		"1000014aV-V",
			
 
				+  vmrglh_3 =		"1000014cVVV",
			
 
				+  vpkswus_3 =		"1000014eVVV",
			
 
				+  vaddcuw_3 =		"10000180VVV",
			
 
				+  vmaxsw_3 =		"10000182VVV",
			
 
				+  vslw_3 =		"10000184VVV",
			
 
				+  vmulosw_3 =		"10000188VVV",
			
 
				+  vexptefp_2 =		"1000018aV-V",
			
 
				+  vmrglw_3 =		"1000018cVVV",
			
 
				+  vpkshss_3 =		"1000018eVVV",
			
 
				+  vmaxsd_3 =		"100001c2VVV",
			
 
				+  vsl_3 =		"100001c4VVV",
			
 
				+  vcmpgefp_3 =		"100001c6VVV",
			
 
				+  vlogefp_2 =		"100001caV-V",
			
 
				+  vpkswss_3 =		"100001ceVVV",
			
 
				+  vadduhs_3 =		"10000240VVV",
			
 
				+  vminuh_3 =		"10000242VVV",
			
 
				+  vsrh_3 =		"10000244VVV",
			
 
				+  vcmpgtuh_3 =		"10000246VVV",
			
 
				+  vmuleuh_3 =		"10000248VVV",
			
 
				+  vrfiz_2 =		"1000024aV-V",
			
 
				+  vsplth_3 =		"1000024cVV3",
			
 
				+  vupkhsh_2 =		"1000024eV-V",
			
 
				+  vminuw_3 =		"10000282VVV",
			
 
				+  vminud_3 =		"100002c2VVV",
			
 
				+  vcmpgtud_3 =		"100002c7VVV",
			
 
				+  vrfim_2 =		"100002caV-V",
			
 
				+  vcmpgtsb_3 =		"10000306VVV",
			
 
				+  vcfux_3 =		"1000030aVVA~",
			
 
				+  vaddshs_3 =		"10000340VVV",
			
 
				+  vminsh_3 =		"10000342VVV",
			
 
				+  vsrah_3 =		"10000344VVV",
			
 
				+  vcmpgtsh_3 =		"10000346VVV",
			
 
				+  vmulesh_3 =		"10000348VVV",
			
 
				+  vcfsx_3 =		"1000034aVVA~",
			
 
				+  vspltish_2 =		"1000034cVS",
			
 
				+  vupkhpx_2 =		"1000034eV-V",
			
 
				+  vaddsws_3 =		"10000380VVV",
			
 
				+  vminsw_3 =		"10000382VVV",
			
 
				+  vsraw_3 =		"10000384VVV",
			
 
				+  vcmpgtsw_3 =		"10000386VVV",
			
 
				+  vmulesw_3 =		"10000388VVV",
			
 
				+  vctuxs_3 =		"1000038aVVA~",
			
 
				+  vspltisw_2 =		"1000038cVS",
			
 
				+  vminsd_3 =		"100003c2VVV",
			
 
				+  vsrad_3 =		"100003c4VVV",
			
 
				+  vcmpbfp_3 =		"100003c6VVV",
			
 
				+  vcmpgtsd_3 =		"100003c7VVV",
			
 
				+  vctsxs_3 =		"100003caVVA~",
			
 
				+  vupklpx_2 =		"100003ceV-V",
			
 
				+  vsububm_3 =		"10000400VVV",
			
 
				+  ["bcdadd._4"] =	"10000401VVVy.",
			
 
				+  vavgub_3 =		"10000402VVV",
			
 
				+  vand_3 =		"10000404VVV",
			
 
				+  ["vcmpequb._3"] =	"10000406VVV",
			
 
				+  vmaxfp_3 =		"1000040aVVV",
			
 
				+  vsubuhm_3 =		"10000440VVV",
			
 
				+  ["bcdsub._4"] =	"10000441VVVy.",
			
 
				+  vavguh_3 =		"10000442VVV",
			
 
				+  vandc_3 =		"10000444VVV",
			
 
				+  ["vcmpequh._3"] =	"10000446VVV",
			
 
				+  vminfp_3 =		"1000044aVVV",
			
 
				+  vpkudum_3 =		"1000044eVVV",
			
 
				+  vsubuwm_3 =		"10000480VVV",
			
 
				+  vavguw_3 =		"10000482VVV",
			
 
				+  vor_3 =		"10000484VVV",
			
 
				+  ["vcmpequw._3"] =	"10000486VVV",
			
 
				+  vpmsumw_3 =		"10000488VVV",
			
 
				+  ["vcmpeqfp._3"] =	"100004c6VVV",
			
 
				+  ["vcmpequd._3"] =	"100004c7VVV",
			
 
				+  vpkudus_3 =		"100004ceVVV",
			
 
				+  vavgsb_3 =		"10000502VVV",
			
 
				+  vavgsh_3 =		"10000542VVV",
			
 
				+  vorc_3 =		"10000544VVV",
			
 
				+  vbpermq_3 =		"1000054cVVV",
			
 
				+  vpksdus_3 =		"1000054eVVV",
			
 
				+  vavgsw_3 =		"10000582VVV",
			
 
				+  vsld_3 =		"100005c4VVV",
			
 
				+  ["vcmpgefp._3"] =	"100005c6VVV",
			
 
				+  vpksdss_3 =		"100005ceVVV",
			
 
				+  vsububs_3 =		"10000600VVV",
			
 
				+  mfvscr_1 =		"10000604V--",
			
 
				+  vsum4ubs_3 =		"10000608VVV",
			
 
				+  vsubuhs_3 =		"10000640VVV",
			
 
				+  mtvscr_1 =		"10000644--V",
			
 
				+  ["vcmpgtuh._3"] =	"10000646VVV",
			
 
				+  vsum4shs_3 =		"10000648VVV",
			
 
				+  vupkhsw_2 =		"1000064eV-V",
			
 
				+  vsubuws_3 =		"10000680VVV",
			
 
				+  vshasigmaw_4 =	"10000682VVYp",
			
 
				+  veqv_3 =		"10000684VVV",
			
 
				+  vsum2sws_3 =		"10000688VVV",
			
 
				+  vmrgow_3 =		"1000068cVVV",
			
 
				+  vshasigmad_4 =	"100006c2VVYp",
			
 
				+  vsrd_3 =		"100006c4VVV",
			
 
				+  ["vcmpgtud._3"] =	"100006c7VVV",
			
 
				+  vupklsw_2 =		"100006ceV-V",
			
 
				+  vupkslw_2 =		"100006ceV-V",
			
 
				+  vsubsbs_3 =		"10000700VVV",
			
 
				+  vclzb_2 =		"10000702V-V",
			
 
				+  vpopcntb_2 =		"10000703V-V",
			
 
				+  ["vcmpgtsb._3"] =	"10000706VVV",
			
 
				+  vsum4sbs_3 =		"10000708VVV",
			
 
				+  vsubshs_3 =		"10000740VVV",
			
 
				+  vclzh_2 =		"10000742V-V",
			
 
				+  vpopcnth_2 =		"10000743V-V",
			
 
				+  ["vcmpgtsh._3"] =	"10000746VVV",
			
 
				+  vsubsws_3 =		"10000780VVV",
			
 
				+  vclzw_2 =		"10000782V-V",
			
 
				+  vpopcntw_2 =		"10000783V-V",
			
 
				+  ["vcmpgtsw._3"] =	"10000786VVV",
			
 
				+  vsumsws_3 =		"10000788VVV",
			
 
				+  vmrgew_3 =		"1000078cVVV",
			
 
				+  vclzd_2 =		"100007c2V-V",
			
 
				+  vpopcntd_2 =		"100007c3V-V",
			
 
				+  ["vcmpbfp._3"] =	"100007c6VVV",
			
 
				+  ["vcmpgtsd._3"] =	"100007c7VVV",
			
 
				+
			
 
				   -- Primary opcode 19:
			
 
				   mcrf_2 =	"4c000000XX",
			
 
				   isync_0 =	"4c00012c",
			
@@ -316,6 +572,8 @@ local map_op = {
 
				   bclrl_2 =	"4c000021AA",
			
 
				   bcctr_2 =	"4c000420AA",
			
 
				   bcctrl_2 =	"4c000421AA",
			
 
				+  bctar_2 =	"4c000460AA",
			
 
				+  bctarl_2 =	"4c000461AA",
			
 
				   blr_0 =	"4e800020",
			
 
				   blrl_0 =	"4e800021",
			
 
				   bctr_0 =	"4e800420",
			
@@ -327,6 +585,7 @@ local map_op = {
 
				   cmpd_3 =	"7c200000XRR",
			
 
				   cmpd_2 =	"7c200000-RR",
			
 
				   tw_3 =	"7c000008ARR",
			
 
				+  lvsl_3 =	"7c00000cVRR",
			
 
				   subfc_3 =	"7c000010RRR.",
			
 
				   subc_3 =	"7c000010RRR~.",
			
 
				   mulhdu_3 =	"7c000012RRR.",
			
@@ -351,50 +610,68 @@ local map_op = {
 
				   cmplw_2 =	"7c000040-RR",
			
 
				   cmpld_3 =	"7c200040XRR",
			
 
				   cmpld_2 =	"7c200040-RR",
			
 
				+  lvsr_3 =	"7c00004cVRR",
			
 
				   subf_3 =	"7c000050RRR.",
			
 
				   sub_3 =	"7c000050RRR~.",
			
 
				+  lbarx_3 =	"7c000068RR0R",
			
 
				   ldux_3 =	"7c00006aRR0R",
			
 
				   dcbst_2 =	"7c00006c-RR",
			
 
				   lwzux_3 =	"7c00006eRR0R",
			
 
				   cntlzd_2 =	"7c000074RR~",
			
 
				   andc_3 =	"7c000078RR~R.",
			
 
				   td_3 =	"7c000088ARR",
			
 
				+  lvewx_3 =	"7c00008eVRR",
			
 
				   mulhd_3 =	"7c000092RRR.",
			
 
				+  addg6s_3 =	"7c000094RRR",
			
 
				   mulhw_3 =	"7c000096RRR.",
			
 
				+  dlmzb_3 =	"7c00009cRR~R.",
			
 
				   ldarx_3 =	"7c0000a8RR0R",
			
 
				   dcbf_2 =	"7c0000ac-RR",
			
 
				   lbzx_3 =	"7c0000aeRR0R",
			
 
				+  lvx_3 =	"7c0000ceVRR",
			
 
				   neg_2 =	"7c0000d0RR.",
			
 
				+  lharx_3 =	"7c0000e8RR0R",
			
 
				   lbzux_3 =	"7c0000eeRR0R",
			
 
				   popcntb_2 =	"7c0000f4RR~",
			
 
				   not_2 =	"7c0000f8RR~%.",
			
 
				   nor_3 =	"7c0000f8RR~R.",
			
 
				+  stvebx_3 =	"7c00010eVRR",
			
 
				   subfe_3 =	"7c000110RRR.",
			
 
				   sube_3 =	"7c000110RRR~.",
			
 
				   adde_3 =	"7c000114RRR.",
			
 
				   stdx_3 =	"7c00012aRR0R",
			
 
				-  stwcx_3 =	"7c00012cRR0R.",
			
 
				+  ["stwcx._3"] =	"7c00012dRR0R.",
			
 
				   stwx_3 =	"7c00012eRR0R",
			
 
				   prtyw_2 =	"7c000134RR~",
			
 
				+  stvehx_3 =	"7c00014eVRR",
			
 
				   stdux_3 =	"7c00016aRR0R",
			
 
				+  ["stqcx._3"] =	"7c00016dR:R0R.",
			
 
				   stwux_3 =	"7c00016eRR0R",
			
 
				   prtyd_2 =	"7c000174RR~",
			
 
				+  stvewx_3 =	"7c00018eVRR",
			
 
				   subfze_2 =	"7c000190RR.",
			
 
				   addze_2 =	"7c000194RR.",
			
 
				-  stdcx_3 =	"7c0001acRR0R.",
			
 
				+  ["stdcx._3"] =	"7c0001adRR0R.",
			
 
				   stbx_3 =	"7c0001aeRR0R",
			
 
				+  stvx_3 =	"7c0001ceVRR",
			
 
				   subfme_2 =	"7c0001d0RR.",
			
 
				   mulld_3 =	"7c0001d2RRR.",
			
 
				   addme_2 =	"7c0001d4RR.",
			
 
				   mullw_3 =	"7c0001d6RRR.",
			
 
				   dcbtst_2 =	"7c0001ec-RR",
			
 
				   stbux_3 =	"7c0001eeRR0R",
			
 
				+  bpermd_3 =	"7c0001f8RR~R",
			
 
				+  lvepxl_3 =	"7c00020eVRR",
			
 
				   add_3 =	"7c000214RRR.",
			
 
				+  lqarx_3 =	"7c000228R:R0R",
			
 
				   dcbt_2 =	"7c00022c-RR",
			
 
				   lhzx_3 =	"7c00022eRR0R",
			
 
				+  cdtbcd_2 =	"7c000234RR~",
			
 
				   eqv_3 =	"7c000238RR~R.",
			
 
				+  lvepx_3 =	"7c00024eVRR",
			
 
				   eciwx_3 =	"7c00026cRR0R",
			
 
				   lhzux_3 =	"7c00026eRR0R",
			
 
				+  cbcdtd_2 =	"7c000274RR~",
			
 
				   xor_3 =	"7c000278RR~R.",
			
 
				   mfspefscr_1 =	"7c0082a6R",
			
 
				   mfxer_1 =	"7c0102a6R",
			
@@ -404,8 +681,12 @@ local map_op = {
 
				   lhax_3 =	"7c0002aeRR0R",
			
 
				   mftb_1 =	"7c0c42e6R",
			
 
				   mftbu_1 =	"7c0d42e6R",
			
 
				+  lvxl_3 =	"7c0002ceVRR",
			
 
				   lwaux_3 =	"7c0002eaRR0R",
			
 
				   lhaux_3 =	"7c0002eeRR0R",
			
 
				+  popcntw_2 =	"7c0002f4RR~",
			
 
				+  divdeu_3 =	"7c000312RRR.",
			
 
				+  divweu_3 =	"7c000316RRR.",
			
 
				   sthx_3 =	"7c00032eRR0R",
			
 
				   orc_3 =	"7c000338RR~R.",
			
 
				   ecowx_3 =	"7c00036cRR0R",
			
@@ -420,10 +701,14 @@ local map_op = {
 
				   mtctr_1 =	"7c0903a6R",
			
 
				   dcbi_2 =	"7c0003ac-RR",
			
 
				   nand_3 =	"7c0003b8RR~R.",
			
 
				+  dsn_2 =	"7c0003c6-RR",
			
 
				+  stvxl_3 =	"7c0003ceVRR",
			
 
				   divd_3 =	"7c0003d2RRR.",
			
 
				   divw_3 =	"7c0003d6RRR.",
			
 
				+  popcntd_2 =	"7c0003f4RR~",
			
 
				   cmpb_3 =	"7c0003f8RR~R.",
			
 
				   mcrxr_1 =	"7c000400X",
			
 
				+  lbdx_3 =	"7c000406RRR",
			
 
				   subfco_3 =	"7c000410RRR.",
			
 
				   subco_3 =	"7c000410RRR~.",
			
 
				   addco_3 =	"7c000414RRR.",
			
@@ -433,16 +718,20 @@ local map_op = {
 
				   lfsx_3 =	"7c00042eFR0R",
			
 
				   srw_3 =	"7c000430RR~R.",
			
 
				   srd_3 =	"7c000436RR~R.",
			
 
				+  lhdx_3 =	"7c000446RRR",
			
 
				   subfo_3 =	"7c000450RRR.",
			
 
				   subo_3 =	"7c000450RRR~.",
			
 
				   lfsux_3 =	"7c00046eFR0R",
			
 
				+  lwdx_3 =	"7c000486RRR",
			
 
				   lswi_3 =	"7c0004aaRR0A",
			
 
				   sync_0 =	"7c0004ac",
			
 
				   lwsync_0 =	"7c2004ac",
			
 
				   ptesync_0 =	"7c4004ac",
			
 
				   lfdx_3 =	"7c0004aeFR0R",
			
 
				+  lddx_3 =	"7c0004c6RRR",
			
 
				   nego_2 =	"7c0004d0RR.",
			
 
				   lfdux_3 =	"7c0004eeFR0R",
			
 
				+  stbdx_3 =	"7c000506RRR",
			
 
				   subfeo_3 =	"7c000510RRR.",
			
 
				   subeo_3 =	"7c000510RRR~.",
			
 
				   addeo_3 =	"7c000514RRR.",
			
@@ -450,27 +739,42 @@ local map_op = {
 
				   stswx_3 =	"7c00052aRR0R",
			
 
				   stwbrx_3 =	"7c00052cRR0R",
			
 
				   stfsx_3 =	"7c00052eFR0R",
			
 
				+  sthdx_3 =	"7c000546RRR",
			
 
				+  ["stbcx._3"] =	"7c00056dRRR",
			
 
				   stfsux_3 =	"7c00056eFR0R",
			
 
				+  stwdx_3 =	"7c000586RRR",
			
 
				   subfzeo_2 =	"7c000590RR.",
			
 
				   addzeo_2 =	"7c000594RR.",
			
 
				   stswi_3 =	"7c0005aaRR0A",
			
 
				+  ["sthcx._3"] =	"7c0005adRRR",
			
 
				   stfdx_3 =	"7c0005aeFR0R",
			
 
				+  stddx_3 =	"7c0005c6RRR",
			
 
				   subfmeo_2 =	"7c0005d0RR.",
			
 
				   mulldo_3 =	"7c0005d2RRR.",
			
 
				   addmeo_2 =	"7c0005d4RR.",
			
 
				   mullwo_3 =	"7c0005d6RRR.",
			
 
				   dcba_2 =	"7c0005ec-RR",
			
 
				   stfdux_3 =	"7c0005eeFR0R",
			
 
				+  stvepxl_3 =	"7c00060eVRR",
			
 
				   addo_3 =	"7c000614RRR.",
			
 
				   lhbrx_3 =	"7c00062cRR0R",
			
 
				+  lfdpx_3 =	"7c00062eF:RR",
			
 
				   sraw_3 =	"7c000630RR~R.",
			
 
				   srad_3 =	"7c000634RR~R.",
			
 
				+  lfddx_3 =	"7c000646FRR",
			
 
				+  stvepx_3 =	"7c00064eVRR",
			
 
				   srawi_3 =	"7c000670RR~A.",
			
 
				   sradi_3 =	"7c000674RR~H.",
			
 
				   eieio_0 =	"7c0006ac",
			
 
				   lfiwax_3 =	"7c0006aeFR0R",
			
 
				+  divdeuo_3 =	"7c000712RRR.",
			
 
				+  divweuo_3 =	"7c000716RRR.",
			
 
				   sthbrx_3 =	"7c00072cRR0R",
			
 
				+  stfdpx_3 =	"7c00072eF:RR",
			
 
				   extsh_2 =	"7c000734RR~.",
			
 
				+  stfddx_3 =	"7c000746FRR",
			
 
				+  divdeo_3 =	"7c000752RRR.",
			
 
				+  divweo_3 =	"7c000756RRR.",
			
 
				   extsb_2 =	"7c000774RR~.",
			
 
				   divduo_3 =	"7c000792RRR.",
			
 
				   divwou_3 =	"7c000796RRR.",
			
@@ -481,6 +785,40 @@ local map_op = {
 
				   divwo_3 =	"7c0007d6RRR.",
			
 
				   dcbz_2 =	"7c0007ec-RR",
			
 
				 
			
 
				+  ["tbegin._1"] =	"7c00051d1",
			
 
				+  ["tbegin._0"] =	"7c00051d",
			
 
				+  ["tend._1"] =		"7c00055dY",
			
 
				+  ["tend._0"] =		"7c00055d",
			
 
				+  ["tendall._0"] =	"7e00055d",
			
 
				+  tcheck_1 =		"7c00059cX",
			
 
				+  ["tsr._1"] =		"7c0005dd1",
			
 
				+  ["tsuspend._0"] =	"7c0005dd",
			
 
				+  ["tresume._0"] =	"7c2005dd",
			
 
				+  ["tabortwc._3"] =	"7c00061dARR",
			
 
				+  ["tabortdc._3"] =	"7c00065dARR",
			
 
				+  ["tabortwci._3"] =	"7c00069dARS",
			
 
				+  ["tabortdci._3"] =	"7c0006ddARS",
			
 
				+  ["tabort._1"] =	"7c00071d-R-",
			
 
				+  ["treclaim._1"] =	"7c00075d-R",
			
 
				+  ["trechkpt._0"] =	"7c0007dd",
			
 
				+
			
 
				+  lxsiwzx_3 =	"7c000018QRR",
			
 
				+  lxsiwax_3 =	"7c000098QRR",
			
 
				+  mfvsrd_2 =	"7c000066-Rq",
			
 
				+  mfvsrwz_2 =	"7c0000e6-Rq",
			
 
				+  stxsiwx_3 =	"7c000118QRR",
			
 
				+  mtvsrd_2 =	"7c000166QR",
			
 
				+  mtvsrwa_2 =	"7c0001a6QR",
			
 
				+  lxvdsx_3 =	"7c000298QRR",
			
 
				+  lxsspx_3 =	"7c000418QRR",
			
 
				+  lxsdx_3 =	"7c000498QRR",
			
 
				+  stxsspx_3 =	"7c000518QRR",
			
 
				+  stxsdx_3 =	"7c000598QRR",
			
 
				+  lxvw4x_3 =	"7c000618QRR",
			
 
				+  lxvd2x_3 =	"7c000698QRR",
			
 
				+  stxvw4x_3 =	"7c000718QRR",
			
 
				+  stxvd2x_3 =	"7c000798QRR",
			
 
				+
			
 
				   -- Primary opcode 30:
			
 
				   rldicl_4 =	"78000000RR~HM.",
			
 
				   rldicr_4 =	"78000004RR~HM.",
			
@@ -489,6 +827,34 @@ local map_op = {
 
				   rldcl_4 =	"78000010RR~RM.",
			
 
				   rldcr_4 =	"78000012RR~RM.",
			
 
				 
			
 
				+  rotldi_3 =	op_alias("rldicl_4", function(p)
			
 
				+    p[4] = "0"
			
 
				+  end),
			
 
				+  rotrdi_3 =	op_alias("rldicl_4", function(p)
			
 
				+    p[3] = "64-("..p[3]..")"; p[4] = "0"
			
 
				+  end),
			
 
				+  rotld_3 =	op_alias("rldcl_4", function(p)
			
 
				+    p[4] = "0"
			
 
				+  end),
			
 
				+  sldi_3 =	op_alias("rldicr_4", function(p)
			
 
				+    p[4] = "63-("..p[3]..")"
			
 
				+  end),
			
 
				+  srdi_3 =	op_alias("rldicl_4", function(p)
			
 
				+    p[4] = p[3]; p[3] = "64-("..p[3]..")"
			
 
				+  end),
			
 
				+  clrldi_3 =	op_alias("rldicl_4", function(p)
			
 
				+    p[4] = p[3]; p[3] = "0"
			
 
				+  end),
			
 
				+  clrrdi_3 =	op_alias("rldicr_4", function(p)
			
 
				+    p[4] = "63-("..p[3]..")"; p[3] = "0"
			
 
				+  end),
			
 
				+
			
 
				+  -- Primary opcode 56:
			
 
				+  lq_2 =	"e0000000R:D", -- NYI: displacement must be divisible by 8.
			
 
				+
			
 
				+  -- Primary opcode 57:
			
 
				+  lfdp_2 =	"e4000000F:D", -- NYI: displacement must be divisible by 4.
			
 
				+
			
 
				   -- Primary opcode 59:
			
 
				   fdivs_3 =	"ec000024FFF.",
			
 
				   fsubs_3 =	"ec000028FFF.",
			
@@ -501,6 +867,200 @@ local map_op = {
 
				   fmadds_4 =	"ec00003aFFFF~.",
			
 
				   fnmsubs_4 =	"ec00003cFFFF~.",
			
 
				   fnmadds_4 =	"ec00003eFFFF~.",
			
 
				+  fcfids_2 =	"ec00069cF-F.",
			
 
				+  fcfidus_2 =	"ec00079cF-F.",
			
 
				+
			
 
				+  dadd_3 =	"ec000004FFF.",
			
 
				+  dqua_4 =	"ec000006FFFZ.",
			
 
				+  dmul_3 =	"ec000044FFF.",
			
 
				+  drrnd_4 =	"ec000046FFFZ.",
			
 
				+  dscli_3 =	"ec000084FF6.",
			
 
				+  dquai_4 =	"ec000086SF~FZ.",
			
 
				+  dscri_3 =	"ec0000c4FF6.",
			
 
				+  drintx_4 =	"ec0000c61F~FZ.",
			
 
				+  dcmpo_3 =	"ec000104XFF",
			
 
				+  dtstex_3 =	"ec000144XFF",
			
 
				+  dtstdc_3 =	"ec000184XF6",
			
 
				+  dtstdg_3 =	"ec0001c4XF6",
			
 
				+  drintn_4 =	"ec0001c61F~FZ.",
			
 
				+  dctdp_2 =	"ec000204F-F.",
			
 
				+  dctfix_2 =	"ec000244F-F.",
			
 
				+  ddedpd_3 =	"ec000284ZF~F.",
			
 
				+  dxex_2 =	"ec0002c4F-F.",
			
 
				+  dsub_3 =	"ec000404FFF.",
			
 
				+  ddiv_3 =	"ec000444FFF.",
			
 
				+  dcmpu_3 =	"ec000504XFF",
			
 
				+  dtstsf_3 =	"ec000544XFF",
			
 
				+  drsp_2 =	"ec000604F-F.",
			
 
				+  dcffix_2 =	"ec000644F-F.",
			
 
				+  denbcd_3 =	"ec000684YF~F.",
			
 
				+  diex_3 =	"ec0006c4FFF.",
			
 
				+
			
 
				+  -- Primary opcode 60:
			
 
				+  xsaddsp_3 =		"f0000000QQQ",
			
 
				+  xsmaddasp_3 =		"f0000008QQQ",
			
 
				+  xxsldwi_4 =		"f0000010QQQz",
			
 
				+  xsrsqrtesp_2 =	"f0000028Q-Q",
			
 
				+  xssqrtsp_2 =		"f000002cQ-Q",
			
 
				+  xxsel_4 =		"f0000030QQQQ",
			
 
				+  xssubsp_3 =		"f0000040QQQ",
			
 
				+  xsmaddmsp_3 =		"f0000048QQQ",
			
 
				+  xxpermdi_4 =		"f0000050QQQz",
			
 
				+  xsresp_2 =		"f0000068Q-Q",
			
 
				+  xsmulsp_3 =		"f0000080QQQ",
			
 
				+  xsmsubasp_3 =		"f0000088QQQ",
			
 
				+  xxmrghw_3 =		"f0000090QQQ",
			
 
				+  xsdivsp_3 =		"f00000c0QQQ",
			
 
				+  xsmsubmsp_3 =		"f00000c8QQQ",
			
 
				+  xsadddp_3 =		"f0000100QQQ",
			
 
				+  xsmaddadp_3 =		"f0000108QQQ",
			
 
				+  xscmpudp_3 =		"f0000118XQQ",
			
 
				+  xscvdpuxws_2 =	"f0000120Q-Q",
			
 
				+  xsrdpi_2 =		"f0000124Q-Q",
			
 
				+  xsrsqrtedp_2 =	"f0000128Q-Q",
			
 
				+  xssqrtdp_2 =		"f000012cQ-Q",
			
 
				+  xssubdp_3 =		"f0000140QQQ",
			
 
				+  xsmaddmdp_3 =		"f0000148QQQ",
			
 
				+  xscmpodp_3 =		"f0000158XQQ",
			
 
				+  xscvdpsxws_2 =	"f0000160Q-Q",
			
 
				+  xsrdpiz_2 =		"f0000164Q-Q",
			
 
				+  xsredp_2 =		"f0000168Q-Q",
			
 
				+  xsmuldp_3 =		"f0000180QQQ",
			
 
				+  xsmsubadp_3 =		"f0000188QQQ",
			
 
				+  xxmrglw_3 =		"f0000190QQQ",
			
 
				+  xsrdpip_2 =		"f00001a4Q-Q",
			
 
				+  xstsqrtdp_2 =		"f00001a8X-Q",
			
 
				+  xsrdpic_2 =		"f00001acQ-Q",
			
 
				+  xsdivdp_3 =		"f00001c0QQQ",
			
 
				+  xsmsubmdp_3 =		"f00001c8QQQ",
			
 
				+  xsrdpim_2 =		"f00001e4Q-Q",
			
 
				+  xstdivdp_3 =		"f00001e8XQQ",
			
 
				+  xvaddsp_3 =		"f0000200QQQ",
			
 
				+  xvmaddasp_3 =		"f0000208QQQ",
			
 
				+  xvcmpeqsp_3 =		"f0000218QQQ",
			
 
				+  xvcvspuxws_2 =	"f0000220Q-Q",
			
 
				+  xvrspi_2 =		"f0000224Q-Q",
			
 
				+  xvrsqrtesp_2 =	"f0000228Q-Q",
			
 
				+  xvsqrtsp_2 =		"f000022cQ-Q",
			
 
				+  xvsubsp_3 =		"f0000240QQQ",
			
 
				+  xvmaddmsp_3 =		"f0000248QQQ",
			
 
				+  xvcmpgtsp_3 =		"f0000258QQQ",
			
 
				+  xvcvspsxws_2 =	"f0000260Q-Q",
			
 
				+  xvrspiz_2 =		"f0000264Q-Q",
			
 
				+  xvresp_2 =		"f0000268Q-Q",
			
 
				+  xvmulsp_3 =		"f0000280QQQ",
			
 
				+  xvmsubasp_3 =		"f0000288QQQ",
			
 
				+  xxspltw_3 =		"f0000290QQg~",
			
 
				+  xvcmpgesp_3 =		"f0000298QQQ",
			
 
				+  xvcvuxwsp_2 =		"f00002a0Q-Q",
			
 
				+  xvrspip_2 =		"f00002a4Q-Q",
			
 
				+  xvtsqrtsp_2 =		"f00002a8X-Q",
			
 
				+  xvrspic_2 =		"f00002acQ-Q",
			
 
				+  xvdivsp_3 =		"f00002c0QQQ",
			
 
				+  xvmsubmsp_3 =		"f00002c8QQQ",
			
 
				+  xvcvsxwsp_2 =		"f00002e0Q-Q",
			
 
				+  xvrspim_2 =		"f00002e4Q-Q",
			
 
				+  xvtdivsp_3 =		"f00002e8XQQ",
			
 
				+  xvadddp_3 =		"f0000300QQQ",
			
 
				+  xvmaddadp_3 =		"f0000308QQQ",
			
 
				+  xvcmpeqdp_3 =		"f0000318QQQ",
			
 
				+  xvcvdpuxws_2 =	"f0000320Q-Q",
			
 
				+  xvrdpi_2 =		"f0000324Q-Q",
			
 
				+  xvrsqrtedp_2 =	"f0000328Q-Q",
			
 
				+  xvsqrtdp_2 =		"f000032cQ-Q",
			
 
				+  xvsubdp_3 =		"f0000340QQQ",
			
 
				+  xvmaddmdp_3 =		"f0000348QQQ",
			
 
				+  xvcmpgtdp_3 =		"f0000358QQQ",
			
 
				+  xvcvdpsxws_2 =	"f0000360Q-Q",
			
 
				+  xvrdpiz_2 =		"f0000364Q-Q",
			
 
				+  xvredp_2 =		"f0000368Q-Q",
			
 
				+  xvmuldp_3 =		"f0000380QQQ",
			
 
				+  xvmsubadp_3 =		"f0000388QQQ",
			
 
				+  xvcmpgedp_3 =		"f0000398QQQ",
			
 
				+  xvcvuxwdp_2 =		"f00003a0Q-Q",
			
 
				+  xvrdpip_2 =		"f00003a4Q-Q",
			
 
				+  xvtsqrtdp_2 =		"f00003a8X-Q",
			
 
				+  xvrdpic_2 =		"f00003acQ-Q",
			
 
				+  xvdivdp_3 =		"f00003c0QQQ",
			
 
				+  xvmsubmdp_3 =		"f00003c8QQQ",
			
 
				+  xvcvsxwdp_2 =		"f00003e0Q-Q",
			
 
				+  xvrdpim_2 =		"f00003e4Q-Q",
			
 
				+  xvtdivdp_3 =		"f00003e8XQQ",
			
 
				+  xsnmaddasp_3 =	"f0000408QQQ",
			
 
				+  xxland_3 =		"f0000410QQQ",
			
 
				+  xscvdpsp_2 =		"f0000424Q-Q",
			
 
				+  xscvdpspn_2 =		"f000042cQ-Q",
			
 
				+  xsnmaddmsp_3 =	"f0000448QQQ",
			
 
				+  xxlandc_3 =		"f0000450QQQ",
			
 
				+  xsrsp_2 =		"f0000464Q-Q",
			
 
				+  xsnmsubasp_3 =	"f0000488QQQ",
			
 
				+  xxlor_3 =		"f0000490QQQ",
			
 
				+  xscvuxdsp_2 =		"f00004a0Q-Q",
			
 
				+  xsnmsubmsp_3 =	"f00004c8QQQ",
			
 
				+  xxlxor_3 =		"f00004d0QQQ",
			
 
				+  xscvsxdsp_2 =		"f00004e0Q-Q",
			
 
				+  xsmaxdp_3 =		"f0000500QQQ",
			
 
				+  xsnmaddadp_3 =	"f0000508QQQ",
			
 
				+  xxlnor_3 =		"f0000510QQQ",
			
 
				+  xscvdpuxds_2 =	"f0000520Q-Q",
			
 
				+  xscvspdp_2 =		"f0000524Q-Q",
			
 
				+  xscvspdpn_2 =		"f000052cQ-Q",
			
 
				+  xsmindp_3 =		"f0000540QQQ",
			
 
				+  xsnmaddmdp_3 =	"f0000548QQQ",
			
 
				+  xxlorc_3 =		"f0000550QQQ",
			
 
				+  xscvdpsxds_2 =	"f0000560Q-Q",
			
 
				+  xsabsdp_2 =		"f0000564Q-Q",
			
 
				+  xscpsgndp_3 =		"f0000580QQQ",
			
 
				+  xsnmsubadp_3 =	"f0000588QQQ",
			
 
				+  xxlnand_3 =		"f0000590QQQ",
			
 
				+  xscvuxddp_2 =		"f00005a0Q-Q",
			
 
				+  xsnabsdp_2 =		"f00005a4Q-Q",
			
 
				+  xsnmsubmdp_3 =	"f00005c8QQQ",
			
 
				+  xxleqv_3 =		"f00005d0QQQ",
			
 
				+  xscvsxddp_2 =		"f00005e0Q-Q",
			
 
				+  xsnegdp_2 =		"f00005e4Q-Q",
			
 
				+  xvmaxsp_3 =		"f0000600QQQ",
			
 
				+  xvnmaddasp_3 =	"f0000608QQQ",
			
 
				+  ["xvcmpeqsp._3"] =	"f0000618QQQ",
			
 
				+  xvcvspuxds_2 =	"f0000620Q-Q",
			
 
				+  xvcvdpsp_2 =		"f0000624Q-Q",
			
 
				+  xvminsp_3 =		"f0000640QQQ",
			
 
				+  xvnmaddmsp_3 =	"f0000648QQQ",
			
 
				+  ["xvcmpgtsp._3"] =	"f0000658QQQ",
			
 
				+  xvcvspsxds_2 =	"f0000660Q-Q",
			
 
				+  xvabssp_2 =		"f0000664Q-Q",
			
 
				+  xvcpsgnsp_3 =		"f0000680QQQ",
			
 
				+  xvnmsubasp_3 =	"f0000688QQQ",
			
 
				+  ["xvcmpgesp._3"] =	"f0000698QQQ",
			
 
				+  xvcvuxdsp_2 =		"f00006a0Q-Q",
			
 
				+  xvnabssp_2 =		"f00006a4Q-Q",
			
 
				+  xvnmsubmsp_3 =	"f00006c8QQQ",
			
 
				+  xvcvsxdsp_2 =		"f00006e0Q-Q",
			
 
				+  xvnegsp_2 =		"f00006e4Q-Q",
			
 
				+  xvmaxdp_3 =		"f0000700QQQ",
			
 
				+  xvnmaddadp_3 =	"f0000708QQQ",
			
 
				+  ["xvcmpeqdp._3"] =	"f0000718QQQ",
			
 
				+  xvcvdpuxds_2 =	"f0000720Q-Q",
			
 
				+  xvcvspdp_2 =		"f0000724Q-Q",
			
 
				+  xvmindp_3 =		"f0000740QQQ",
			
 
				+  xvnmaddmdp_3 =	"f0000748QQQ",
			
 
				+  ["xvcmpgtdp._3"] =	"f0000758QQQ",
			
 
				+  xvcvdpsxds_2 =	"f0000760Q-Q",
			
 
				+  xvabsdp_2 =		"f0000764Q-Q",
			
 
				+  xvcpsgndp_3 =		"f0000780QQQ",
			
 
				+  xvnmsubadp_3 =	"f0000788QQQ",
			
 
				+  ["xvcmpgedp._3"] =	"f0000798QQQ",
			
 
				+  xvcvuxddp_2 =		"f00007a0Q-Q",
			
 
				+  xvnabsdp_2 =		"f00007a4Q-Q",
			
 
				+  xvnmsubmdp_3 =	"f00007c8QQQ",
			
 
				+  xvcvsxddp_2 =		"f00007e0Q-Q",
			
 
				+  xvnegdp_2 =		"f00007e4Q-Q",
			
 
				+
			
 
				+  -- Primary opcode 61:
			
 
				+  stfdp_2 =	"f4000000F:D", -- NYI: displacement must be divisible by 4.
			
 
				+
			
 
				+  -- Primary opcode 62:
			
 
				+  stq_2 =	"f8000002R:D", -- NYI: displacement must be divisible by 8.
			
 
				 
			
 
				   -- Primary opcode 63:
			
 
				   fdiv_3 =	"fc000024FFF.",
			
@@ -526,8 +1086,12 @@ local map_op = {
 
				   frsp_2 =	"fc000018F-F.",
			
 
				   fctiw_2 =	"fc00001cF-F.",
			
 
				   fctiwz_2 =	"fc00001eF-F.",
			
 
				+  ftdiv_2 =	"fc000100X-F.",
			
 
				+  fctiwu_2 =	"fc00011cF-F.",
			
 
				+  fctiwuz_2 =	"fc00011eF-F.",
			
 
				   mtfsfi_2 =	"fc00010cAA", -- NYI: upshift.
			
 
				   fnabs_2 =	"fc000110F-F.",
			
 
				+  ftsqrt_2 =	"fc000140X-F.",
			
 
				   fabs_2 =	"fc000210F-F.",
			
 
				   frin_2 =	"fc000310F-F.",
			
 
				   friz_2 =	"fc000350F-F.",
			
@@ -537,7 +1101,38 @@ local map_op = {
 
				   -- NYI: mtfsf, mtfsb0, mtfsb1.
			
 
				   fctid_2 =	"fc00065cF-F.",
			
 
				   fctidz_2 =	"fc00065eF-F.",
			
 
				+  fmrgow_3 =	"fc00068cFFF",
			
 
				   fcfid_2 =	"fc00069cF-F.",
			
 
				+  fctidu_2 =	"fc00075cF-F.",
			
 
				+  fctiduz_2 =	"fc00075eF-F.",
			
 
				+  fmrgew_3 =	"fc00078cFFF",
			
 
				+  fcfidu_2 =	"fc00079cF-F.",
			
 
				+
			
 
				+  daddq_3 =	"fc000004F:F:F:.",
			
 
				+  dquaq_4 =	"fc000006F:F:F:Z.",
			
 
				+  dmulq_3 =	"fc000044F:F:F:.",
			
 
				+  drrndq_4 =	"fc000046F:F:F:Z.",
			
 
				+  dscliq_3 =	"fc000084F:F:6.",
			
 
				+  dquaiq_4 =	"fc000086SF:~F:Z.",
			
 
				+  dscriq_3 =	"fc0000c4F:F:6.",
			
 
				+  drintxq_4 =	"fc0000c61F:~F:Z.",
			
 
				+  dcmpoq_3 =	"fc000104XF:F:",
			
 
				+  dtstexq_3 =	"fc000144XF:F:",
			
 
				+  dtstdcq_3 =	"fc000184XF:6",
			
 
				+  dtstdgq_3 =	"fc0001c4XF:6",
			
 
				+  drintnq_4 =	"fc0001c61F:~F:Z.",
			
 
				+  dctqpq_2 =	"fc000204F:-F:.",
			
 
				+  dctfixq_2 =	"fc000244F:-F:.",
			
 
				+  ddedpdq_3 =	"fc000284ZF:~F:.",
			
 
				+  dxexq_2 =	"fc0002c4F:-F:.",
			
 
				+  dsubq_3 =	"fc000404F:F:F:.",
			
 
				+  ddivq_3 =	"fc000444F:F:F:.",
			
 
				+  dcmpuq_3 =	"fc000504XF:F:",
			
 
				+  dtstsfq_3 =	"fc000544XF:F:",
			
 
				+  drdpq_2 =	"fc000604F:-F:.",
			
 
				+  dcffixq_2 =	"fc000644F:-F:.",
			
 
				+  denbcdq_3 =	"fc000684YF:~F:.",
			
 
				+  diexq_3 =	"fc0006c4F:FF:.",
			
 
				 
			
 
				   -- Primary opcode 4, SPE APU extension:
			
 
				   evaddw_3 =		"10000200RRR",
			
@@ -822,7 +1417,7 @@ local map_op = {
 
				 do
			
 
				   local t = {}
			
 
				   for k,v in pairs(map_op) do
			
 
				-    if sub(v, -1) == "." then
			
 
				+    if type(v) == "string" and sub(v, -1) == "." then
			
 
				       local v2 = sub(v, 1, 7)..char(byte(v, 8)+1)..sub(v, 9, -2)
			
 
				       t[sub(k, 1, -3).."."..sub(k, -2)] = v2
			
 
				     end
			
@@ -884,6 +1479,24 @@ local function parse_fpr(expr)
 
				   werror("bad register name `"..expr.."'")
			
 
				 end
			
 
				 
			
 
				+local function parse_vr(expr)
			
 
				+  local r = match(expr, "^v([1-3]?[0-9])$")
			
 
				+  if r then
			
 
				+    r = tonumber(r)
			
 
				+    if r <= 31 then return r end
			
 
				+  end
			
 
				+  werror("bad register name `"..expr.."'")
			
 
				+end
			
 
				+
			
 
				+local function parse_vs(expr)
			
 
				+  local r = match(expr, "^vs([1-6]?[0-9])$")
			
 
				+  if r then
			
 
				+    r = tonumber(r)
			
 
				+    if r <= 63 then return r end
			
 
				+  end
			
 
				+  werror("bad register name `"..expr.."'")
			
 
				+end
			
 
				+
			
 
				 local function parse_cr(expr)
			
 
				   local r = match(expr, "^cr([0-7])$")
			
 
				   if r then return tonumber(r) end
			
@@ -900,8 +1513,30 @@ local function parse_cond(expr)
 
				   werror("bad condition bit name `"..expr.."'")
			
 
				 end
			
 
				 
			
 
				+local parse_ctx = {}
			
 
				+
			
 
				+local loadenv = setfenv and function(s)
			
 
				+  local code = loadstring(s, "")
			
 
				+  if code then setfenv(code, parse_ctx) end
			
 
				+  return code
			
 
				+end or function(s)
			
 
				+  return load(s, "", nil, parse_ctx)
			
 
				+end
			
 
				+
			
 
				+-- Try to parse simple arithmetic, too, since some basic ops are aliases.
			
 
				+local function parse_number(n)
			
 
				+  local x = tonumber(n)
			
 
				+  if x then return x end
			
 
				+  local code = loadenv("return "..n)
			
 
				+  if code then
			
 
				+    local ok, y = pcall(code)
			
 
				+    if ok then return y end
			
 
				+  end
			
 
				+  return nil
			
 
				+end
			
 
				+
			
 
				 local function parse_imm(imm, bits, shift, scale, signed)
			
 
				-  local n = tonumber(imm)
			
 
				+  local n = parse_number(imm)
			
 
				   if n then
			
 
				     local m = sar(n, scale)
			
 
				     if shl(m, scale) == n then
			
@@ -914,7 +1549,8 @@ local function parse_imm(imm, bits, shift, scale, signed)
 
				       end
			
 
				     end
			
 
				     werror("out of range immediate `"..imm.."'")
			
 
				-  elseif match(imm, "^r([1-3]?[0-9])$") or
			
 
				+  elseif match(imm, "^[rfv]([1-3]?[0-9])$") or
			
 
				+	 match(imm, "^vs([1-6]?[0-9])$") or
			
 
				 	 match(imm, "^([%w_]+):(r[1-3]?[0-9])$") then
			
 
				     werror("expected immediate operand, got register")
			
 
				   else
			
@@ -924,11 +1560,11 @@ local function parse_imm(imm, bits, shift, scale, signed)
 
				 end
			
 
				 
			
 
				 local function parse_shiftmask(imm, isshift)
			
 
				-  local n = tonumber(imm)
			
 
				+  local n = parse_number(imm)
			
 
				   if n then
			
 
				     if shr(n, 6) == 0 then
			
 
				-      local lsb = band(imm, 31)
			
 
				-      local msb = imm - lsb
			
 
				+      local lsb = band(n, 31)
			
 
				+      local msb = n - lsb
			
 
				       return isshift and (shl(lsb, 11)+shr(msb, 4)) or (shl(lsb, 6)+msb)
			
 
				     end
			
 
				     werror("out of range immediate `"..imm.."'")
			
@@ -936,7 +1572,8 @@ local function parse_shiftmask(imm, isshift)
 
				 	 match(imm, "^([%w_]+):(r[1-3]?[0-9])$") then
			
 
				     werror("expected immediate operand, got register")
			
 
				   else
			
 
				-    werror("NYI: parameterized 64 bit shift/mask")
			
 
				+    waction("IMMSH", isshift and 1 or 0, imm)
			
 
				+    return 0;
			
 
				   end
			
 
				 end
			
 
				 
			
@@ -1011,7 +1648,7 @@ end
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 -- Handle opcodes defined with template strings.
			
 
				-map_op[".template__"] = function(params, template, nparams)
			
 
				+op_template = function(params, template, nparams)
			
 
				   if not params then return sub(template, 9) end
			
 
				   local op = tonumber(sub(template, 1, 8), 16)
			
 
				   local n, rs = 1, 26
			
@@ -1027,6 +1664,15 @@ map_op[".template__"] = function(params, template, nparams)
 
				       rs = rs - 5; op = op + shl(parse_gpr(params[n]), rs); n = n + 1
			
 
				     elseif p == "F" then
			
 
				       rs = rs - 5; op = op + shl(parse_fpr(params[n]), rs); n = n + 1
			
 
				+    elseif p == "V" then
			
 
				+      rs = rs - 5; op = op + shl(parse_vr(params[n]), rs); n = n + 1
			
 
				+    elseif p == "Q" then
			
 
				+      local vs = parse_vs(params[n]); n = n + 1; rs = rs - 5
			
 
				+      local sh = rs == 6 and 2 or 3 + band(shr(rs, 1), 3)
			
 
				+      op = op + shl(band(vs, 31), rs) + shr(band(vs, 32), sh)
			
 
				+    elseif p == "q" then
			
 
				+      local vs = parse_vs(params[n]); n = n + 1
			
 
				+      op = op + shl(band(vs, 31), 21) + shr(band(vs, 32), 5)
			
 
				     elseif p == "A" then
			
 
				       rs = rs - 5; op = op + parse_imm(params[n], 5, rs, 0, false); n = n + 1
			
 
				     elseif p == "S" then
			
@@ -1047,6 +1693,26 @@ map_op[".template__"] = function(params, template, nparams)
 
				       rs = rs - 5; op = op + shl(parse_cond(params[n]), rs); n = n + 1
			
 
				     elseif p == "X" then
			
 
				       rs = rs - 5; op = op + shl(parse_cr(params[n]), rs+2); n = n + 1
			
 
				+    elseif p == "1" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 1, rs, 0, false); n = n + 1
			
 
				+    elseif p == "g" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 2, rs, 0, false); n = n + 1
			
 
				+    elseif p == "3" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 3, rs, 0, false); n = n + 1
			
 
				+    elseif p == "P" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 4, rs, 0, false); n = n + 1
			
 
				+    elseif p == "p" then
			
 
				+      op = op + parse_imm(params[n], 4, rs, 0, false); n = n + 1
			
 
				+    elseif p == "6" then
			
 
				+      rs = rs - 6; op = op + parse_imm(params[n], 6, rs, 0, false); n = n + 1
			
 
				+    elseif p == "Y" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 1, rs+4, 0, false); n = n + 1
			
 
				+    elseif p == "y" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 1, rs+3, 0, false); n = n + 1
			
 
				+    elseif p == "Z" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 2, rs+3, 0, false); n = n + 1
			
 
				+    elseif p == "z" then
			
 
				+      rs = rs - 5; op = op + parse_imm(params[n], 2, rs+2, 0, false); n = n + 1
			
 
				     elseif p == "W" then
			
 
				       op = op + parse_cr(params[n]); n = n + 1
			
 
				     elseif p == "G" then
			
@@ -1056,9 +1722,9 @@ map_op[".template__"] = function(params, template, nparams)
 
				     elseif p == "M" then
			
 
				       op = op + parse_shiftmask(params[n], false); n = n + 1
			
 
				     elseif p == "J" or p == "K" then
			
 
				-      local mode, n, s = parse_label(params[n], false)
			
 
				-      if p == "K" then n = n + 2048 end
			
 
				-      waction("REL_"..mode, n, s, 1)
			
 
				+      local mode, m, s = parse_label(params[n], false)
			
 
				+      if p == "K" then m = m + 2048 end
			
 
				+      waction("REL_"..mode, m, s, 1)
			
 
				       n = n + 1
			
 
				     elseif p == "0" then
			
 
				       if band(shr(op, rs), 31) == 0 then werror("cannot use r0") end
			
@@ -1071,6 +1737,8 @@ map_op[".template__"] = function(params, template, nparams)
 
				       local lo = band(op, mm)
			
 
				       local hi = band(op, shl(mm, 5))
			
 
				       op = op - lo - hi + shl(lo, 5) + shr(hi, 5)
			
 
				+    elseif p == ":" then
			
 
				+      if band(shr(op, rs), 1) ~= 0 then werror("register pair expected") end
			
 
				     elseif p == "-" then
			
 
				       rs = rs - 5
			
 
				     elseif p == "." then
			
@@ -1082,6 +1750,8 @@ map_op[".template__"] = function(params, template, nparams)
 
				   wputpos(pos, op)
			
 
				 end
			
 
				 
			
 
				+map_op[".template__"] = op_template
			
 
				+
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 -- Pseudo-opcode to mark the position where the action list is to be emitted.
			
--- a/luajit.mod/luajit/dynasm/dasm_proto.h
+++ b/luajit.mod/luajit/dynasm/dasm_proto.h
@@ -10,8 +10,8 @@
 
				 #include <stddef.h>
			
 
				 #include <stdarg.h>
			
 
				 
			
 
				-#define DASM_IDENT	"DynASM 1.3.0"
			
 
				-#define DASM_VERSION	10300	/* 1.3.0 */
			
 
				+#define DASM_IDENT	"DynASM 1.4.0"
			
 
				+#define DASM_VERSION	10400	/* 1.4.0 */
			
 
				 
			
 
				 #ifndef Dst_DECL
			
 
				 #define Dst_DECL	dasm_State **Dst
			
--- a/luajit.mod/luajit/dynasm/dasm_x86.h
+++ b/luajit.mod/luajit/dynasm/dasm_x86.h
@@ -170,7 +170,7 @@ void dasm_put(Dst_DECL, int start, ...)
 
				   dasm_State *D = Dst_REF;
			
 
				   dasm_ActList p = D->actionlist + start;
			
 
				   dasm_Section *sec = D->section;
			
 
				-  int pos = sec->pos, ofs = sec->ofs, mrm = 4;
			
 
				+  int pos = sec->pos, ofs = sec->ofs, mrm = -1;
			
 
				   int *b;
			
 
				 
			
 
				   if (pos >= sec->epos) {
			
@@ -193,21 +193,28 @@ void dasm_put(Dst_DECL, int start, ...)
 
				       b[pos++] = n;
			
 
				       switch (action) {
			
 
				       case DASM_DISP:
			
 
				-	if (n == 0) { if ((mrm&7) == 4) mrm = p[-2]; if ((mrm&7) != 5) break; }
			
 
				-      case DASM_IMM_DB: if (((n+128)&-256) == 0) goto ob;
			
 
				+	if (n == 0) { if (mrm < 0) mrm = p[-2]; if ((mrm&7) != 5) break; }
			
 
				+	/* fallthrough */
			
 
				+      case DASM_IMM_DB: if (((n+128)&-256) == 0) goto ob; /* fallthrough */
			
 
				       case DASM_REL_A: /* Assumes ptrdiff_t is int. !x64 */
			
 
				       case DASM_IMM_D: ofs += 4; break;
			
 
				       case DASM_IMM_S: CK(((n+128)&-256) == 0, RANGE_I); goto ob;
			
 
				       case DASM_IMM_B: CK((n&-256) == 0, RANGE_I); ob: ofs++; break;
			
 
				-      case DASM_IMM_WB: if (((n+128)&-256) == 0) goto ob;
			
 
				+      case DASM_IMM_WB: if (((n+128)&-256) == 0) goto ob; /* fallthrough */
			
 
				       case DASM_IMM_W: CK((n&-65536) == 0, RANGE_I); ofs += 2; break;
			
 
				       case DASM_SPACE: p++; ofs += n; break;
			
 
				       case DASM_SETLABEL: b[pos-2] = -0x40000000; break;  /* Neg. label ofs. */
			
 
				-      case DASM_VREG: CK((n&-8) == 0 && (n != 4 || (*p&1) == 0), RANGE_VREG);
			
 
				-	if (*p++ == 1 && *p == DASM_DISP) mrm = n;
			
 
				+      case DASM_VREG: CK((n&-16) == 0 && (n != 4 || (*p>>5) != 2), RANGE_VREG);
			
 
				+	if (*p < 0x40 && p[1] == DASM_DISP) mrm = n;
			
 
				+	if (*p < 0x20 && (n&7) == 4) ofs++;
			
 
				+	switch ((*p++ >> 3) & 3) {
			
 
				+	case 3: n |= b[pos-3]; /* fallthrough */
			
 
				+	case 2: n |= b[pos-2]; /* fallthrough */
			
 
				+	case 1: if (n <= 7) { b[pos-1] |= 0x10; ofs--; }
			
 
				+	}
			
 
				 	continue;
			
 
				       }
			
 
				-      mrm = 4;
			
 
				+      mrm = -1;
			
 
				     } else {
			
 
				       int *pl, n;
			
 
				       switch (action) {
			
@@ -323,11 +330,14 @@ int dasm_link(Dst_DECL, size_t *szp)
 
				 	  pos += 2;
			
 
				 	  break;
			
 
				 	}
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_SPACE: case DASM_IMM_LG: case DASM_VREG: p++;
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_DISP: case DASM_IMM_S: case DASM_IMM_B: case DASM_IMM_W:
			
 
				 	case DASM_IMM_D: case DASM_IMM_WB: case DASM_IMM_DB:
			
 
				 	case DASM_SETLABEL: case DASM_REL_A: case DASM_IMM_PC: pos++; break;
			
 
				 	case DASM_LABEL_LG: p++;
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_LABEL_PC: b[pos++] += ofs; break; /* Fix label offset. */
			
 
				 	case DASM_ALIGN: ofs -= (b[pos++]+ofs)&*p++; break; /* Adjust ofs. */
			
 
				 	case DASM_EXTERN: p += 2; break;
			
@@ -385,17 +395,42 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	    if (mrm != 5) { mm[-1] -= 0x80; break; } }
			
 
				 	  if (((n+128) & -256) != 0) goto wd; else mm[-1] -= 0x40;
			
 
				 	}
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_IMM_S: case DASM_IMM_B: wb: dasmb(n); break;
			
 
				 	case DASM_IMM_DB: if (((n+128)&-256) == 0) {
			
 
				 	    db: if (!mark) mark = cp; mark[-2] += 2; mark = NULL; goto wb;
			
 
				 	  } else mark = NULL;
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_IMM_D: wd: dasmd(n); break;
			
 
				 	case DASM_IMM_WB: if (((n+128)&-256) == 0) goto db; else mark = NULL;
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_IMM_W: dasmw(n); break;
			
 
				-	case DASM_VREG: { int t = *p++; if (t >= 2) n<<=3; cp[-1] |= n; break; }
			
 
				+	case DASM_VREG: {
			
 
				+	  int t = *p++;
			
 
				+	  unsigned char *ex = cp - (t&7);
			
 
				+	  if ((n & 8) && t < 0xa0) {
			
 
				+	    if (*ex & 0x80) ex[1] ^= 0x20 << (t>>6); else *ex ^= 1 << (t>>6);
			
 
				+	    n &= 7;
			
 
				+	  } else if (n & 0x10) {
			
 
				+	    if (*ex & 0x80) {
			
 
				+	      *ex = 0xc5; ex[1] = (ex[1] & 0x80) | ex[2]; ex += 2;
			
 
				+	    }
			
 
				+	    while (++ex < cp) ex[-1] = *ex;
			
 
				+	    if (mark) mark--;
			
 
				+	    cp--;
			
 
				+	    n &= 7;
			
 
				+	  }
			
 
				+	  if (t >= 0xc0) n <<= 4;
			
 
				+	  else if (t >= 0x40) n <<= 3;
			
 
				+	  else if (n == 4 && t < 0x20) { cp[-1] ^= n; *cp++ = 0x20; }
			
 
				+	  cp[-1] ^= n;
			
 
				+	  break;
			
 
				+	}
			
 
				 	case DASM_REL_LG: p++; if (n >= 0) goto rel_pc;
			
 
				 	  b++; n = (int)(ptrdiff_t)D->globals[-n];
			
 
				-	case DASM_REL_A: rel_a: n -= (int)(ptrdiff_t)(cp+4); goto wd; /* !x64 */
			
 
				+	  /* fallthrough */
			
 
				+	case DASM_REL_A: rel_a:
			
 
				+	  n -= (unsigned int)(ptrdiff_t)(cp+4); goto wd; /* !x64 */
			
 
				 	case DASM_REL_PC: rel_pc: {
			
 
				 	  int shrink = *b++;
			
 
				 	  int *pb = DASM_POS2PTR(D, n); if (*pb < 0) { n = pb[1]; goto rel_a; }
			
@@ -406,6 +441,7 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	}
			
 
				 	case DASM_IMM_LG:
			
 
				 	  p++; if (n < 0) { n = (int)(ptrdiff_t)D->globals[-n]; goto wd; }
			
 
				+	  /* fallthrough */
			
 
				 	case DASM_IMM_PC: {
			
 
				 	  int *pb = DASM_POS2PTR(D, n);
			
 
				 	  n = *pb < 0 ? pb[1] : (*pb + (int)(ptrdiff_t)base);
			
@@ -426,6 +462,7 @@ int dasm_encode(Dst_DECL, void *buffer)
 
				 	case DASM_EXTERN: n = DASM_EXTERN(Dst, cp, p[1], *p); p += 2; goto wd;
			
 
				 	case DASM_MARK: mark = cp; break;
			
 
				 	case DASM_ESC: action = *p++;
			
 
				+	  /* fallthrough */
			
 
				 	default: *cp++ = action; break;
			
 
				 	case DASM_SECTION: case DASM_STOP: goto stop;
			
 
				 	}
			
--- a/luajit.mod/luajit/dynasm/dasm_x86.lua
+++ b/luajit.mod/luajit/dynasm/dasm_x86.lua
@@ -11,9 +11,9 @@ local x64 = x64
 
				 local _info = {
			
 
				   arch =	x64 and "x64" or "x86",
			
 
				   description =	"DynASM x86/x64 module",
			
 
				-  version =	"1.3.0",
			
 
				-  vernum =	 10300,
			
 
				-  release =	"2011-05-05",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2015-10-18",
			
 
				   author =	"Mike Pall",
			
 
				   license =	"MIT",
			
 
				 }
			
@@ -27,9 +27,9 @@ local assert, unpack, setmetatable = assert, unpack or table.unpack, setmetatabl
 
				 local _s = string
			
 
				 local sub, format, byte, char = _s.sub, _s.format, _s.byte, _s.char
			
 
				 local find, match, gmatch, gsub = _s.find, _s.match, _s.gmatch, _s.gsub
			
 
				-local concat, sort = table.concat, table.sort
			
 
				+local concat, sort, remove = table.concat, table.sort, table.remove
			
 
				 local bit = bit or require("bit")
			
 
				-local band, shl, shr = bit.band, bit.lshift, bit.rshift
			
 
				+local band, bxor, shl, shr = bit.band, bit.bxor, bit.lshift, bit.rshift
			
 
				 
			
 
				 -- Inherited tables and callbacks.
			
 
				 local g_opt, g_arch
			
@@ -41,7 +41,7 @@ local action_names = {
 
				   -- int arg, 1 buffer pos:
			
 
				   "DISP",  "IMM_S", "IMM_B", "IMM_W", "IMM_D",  "IMM_WB", "IMM_DB",
			
 
				   -- action arg (1 byte), int arg, 1 buffer pos (reg/num):
			
 
				-  "VREG", "SPACE", -- !x64: VREG support NYI.
			
 
				+  "VREG", "SPACE",
			
 
				   -- ptrdiff_t arg, 1 buffer pos (address): !x64
			
 
				   "SETLABEL", "REL_A",
			
 
				   -- action arg (1 byte) or int arg, 2 buffer pos (link, offset):
			
@@ -83,6 +83,21 @@ local actargs = { 0 }
 
				 -- Current number of section buffer positions for dasm_put().
			
 
				 local secpos = 1
			
 
				 
			
 
				+-- VREG kind encodings, pre-shifted by 5 bits.
			
 
				+local map_vreg = {
			
 
				+  ["modrm.rm.m"] = 0x00,
			
 
				+  ["modrm.rm.r"] = 0x20,
			
 
				+  ["opcode"] =     0x20,
			
 
				+  ["sib.base"] =   0x20,
			
 
				+  ["sib.index"] =  0x40,
			
 
				+  ["modrm.reg"] =  0x80,
			
 
				+  ["vex.v"] =      0xa0,
			
 
				+  ["imm.hi"] =     0xc0,
			
 
				+}
			
 
				+
			
 
				+-- Current number of VREG actions contributing to REX/VEX shrinkage.
			
 
				+local vreg_shrink_count = 0
			
 
				+
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 -- Compute action numbers for action names.
			
@@ -134,6 +149,21 @@ local function waction(action, a, num)
 
				   if a or num then secpos = secpos + (num or 1) end
			
 
				 end
			
 
				 
			
 
				+-- Optionally add a VREG action.
			
 
				+local function wvreg(kind, vreg, psz, sk, defer)
			
 
				+  if not vreg then return end
			
 
				+  waction("VREG", vreg)
			
 
				+  local b = assert(map_vreg[kind], "bad vreg kind `"..vreg.."'")
			
 
				+  if b < (sk or 0) then
			
 
				+    vreg_shrink_count = vreg_shrink_count + 1
			
 
				+  end
			
 
				+  if not defer then
			
 
				+    b = b + vreg_shrink_count * 8
			
 
				+    vreg_shrink_count = 0
			
 
				+  end
			
 
				+  wputxb(b + (psz or 0))
			
 
				+end
			
 
				+
			
 
				 -- Add call to embedded DynASM C code.
			
 
				 local function wcall(func, args)
			
 
				   wline(format("dasm_%s(Dst, %s);", func, concat(args, ", ")), true)
			
@@ -299,7 +329,7 @@ local function mkrmap(sz, cl, names)
 
				     local iname = format("@%s%x%s", sz, i, needrex and "R" or "")
			
 
				     if needrex then map_reg_needrex[iname] = true end
			
 
				     local name
			
 
				-    if sz == "o" then name = format("xmm%d", i)
			
 
				+    if sz == "o" or sz == "y" then name = format("%s%d", cl, i)
			
 
				     elseif sz == "f" then name = format("st%d", i)
			
 
				     else name = format("r%d%s", i, sz == addrsize and "" or sz) end
			
 
				     map_archdef[name] = iname
			
@@ -326,6 +356,7 @@ mkrmap("w", "Rw", {"ax", "cx", "dx", "bx", "sp", "bp", "si", "di"})
 
				 mkrmap("b", "Rb", {"al", "cl", "dl", "bl", "ah", "ch", "dh", "bh"})
			
 
				 map_reg_valid_index[map_archdef.esp] = false
			
 
				 if x64 then map_reg_valid_index[map_archdef.rsp] = false end
			
 
				+if x64 then map_reg_needrex[map_archdef.Rb] = true end
			
 
				 map_archdef["Ra"] = "@"..addrsize
			
 
				 
			
 
				 -- FP registers (internally tword sized, but use "f" as operand size).
			
@@ -334,21 +365,24 @@ mkrmap("f", "Rf")
 
				 -- SSE registers (oword sized, but qword and dword accessible).
			
 
				 mkrmap("o", "xmm")
			
 
				 
			
 
				+-- AVX registers (yword sized, but oword, qword and dword accessible).
			
 
				+mkrmap("y", "ymm")
			
 
				+
			
 
				 -- Operand size prefixes to codes.
			
 
				 local map_opsize = {
			
 
				-  byte = "b", word = "w", dword = "d", qword = "q", oword = "o", tword = "t",
			
 
				-  aword = addrsize,
			
 
				+  byte = "b", word = "w", dword = "d", qword = "q", oword = "o", yword = "y",
			
 
				+  tword = "t", aword = addrsize,
			
 
				 }
			
 
				 
			
 
				 -- Operand size code to number.
			
 
				 local map_opsizenum = {
			
 
				-  b = 1, w = 2, d = 4, q = 8, o = 16, t = 10,
			
 
				+  b = 1, w = 2, d = 4, q = 8, o = 16, y = 32, t = 10,
			
 
				 }
			
 
				 
			
 
				 -- Operand size code to name.
			
 
				 local map_opsizename = {
			
 
				-  b = "byte", w = "word", d = "dword", q = "qword", o = "oword", t = "tword",
			
 
				-  f = "fpword",
			
 
				+  b = "byte", w = "word", d = "dword", q = "qword", o = "oword", y = "yword",
			
 
				+  t = "tword", f = "fpword",
			
 
				 }
			
 
				 
			
 
				 -- Valid index register scale factors.
			
@@ -460,9 +494,45 @@ local function wputszarg(sz, n)
 
				 end
			
 
				 
			
 
				 -- Put multi-byte opcode with operand-size dependent modifications.
			
 
				-local function wputop(sz, op, rex)
			
 
				+local function wputop(sz, op, rex, vex, vregr, vregxb)
			
 
				+  local psz, sk = 0, nil
			
 
				+  if vex then
			
 
				+    local tail
			
 
				+    if vex.m == 1 and band(rex, 11) == 0 then
			
 
				+      if x64 and vregxb then
			
 
				+	sk = map_vreg["modrm.reg"]
			
 
				+      else
			
 
				+	wputb(0xc5)
			
 
				+      tail = shl(bxor(band(rex, 4), 4), 5)
			
 
				+      psz = 3
			
 
				+      end
			
 
				+    end
			
 
				+    if not tail then
			
 
				+      wputb(0xc4)
			
 
				+      wputb(shl(bxor(band(rex, 7), 7), 5) + vex.m)
			
 
				+      tail = shl(band(rex, 8), 4)
			
 
				+      psz = 4
			
 
				+    end
			
 
				+    local reg, vreg = 0, nil
			
 
				+    if vex.v then
			
 
				+      reg = vex.v.reg
			
 
				+      if not reg then werror("bad vex operand") end
			
 
				+      if reg < 0 then reg = 0; vreg = vex.v.vreg end
			
 
				+    end
			
 
				+    if sz == "y" or vex.l then tail = tail + 4 end
			
 
				+    wputb(tail + shl(bxor(reg, 15), 3) + vex.p)
			
 
				+    wvreg("vex.v", vreg)
			
 
				+    rex = 0
			
 
				+    if op >= 256 then werror("bad vex opcode") end
			
 
				+  else
			
 
				+    if rex ~= 0 then
			
 
				+      if not x64 then werror("bad operand size") end
			
 
				+    elseif (vregr or vregxb) and x64 then
			
 
				+      rex = 0x10
			
 
				+      sk = map_vreg["vex.v"]
			
 
				+    end
			
 
				+  end
			
 
				   local r
			
 
				-  if rex ~= 0 and not x64 then werror("bad operand size") end
			
 
				   if sz == "w" then wputb(102) end
			
 
				   -- Needs >32 bit numbers, but only for crc32 eax, word [ebx]
			
 
				   if op >= 4294967296 then r = op%4294967296 wputb((op-r)/4294967296) op = r end
			
@@ -471,20 +541,20 @@ local function wputop(sz, op, rex)
 
				     if rex ~= 0 then
			
 
				       local opc3 = band(op, 0xffff00)
			
 
				       if opc3 == 0x0f3a00 or opc3 == 0x0f3800 then
			
 
				-	wputb(64 + band(rex, 15)); rex = 0
			
 
				+	wputb(64 + band(rex, 15)); rex = 0; psz = 2
			
 
				       end
			
 
				     end
			
 
				-    wputb(shr(op, 16)); op = band(op, 0xffff)
			
 
				+    wputb(shr(op, 16)); op = band(op, 0xffff); psz = psz + 1
			
 
				   end
			
 
				   if op >= 256 then
			
 
				     local b = shr(op, 8)
			
 
				-    if b == 15 and rex ~= 0 then wputb(64 + band(rex, 15)); rex = 0 end
			
 
				-    wputb(b)
			
 
				-    op = band(op, 255)
			
 
				+    if b == 15 and rex ~= 0 then wputb(64 + band(rex, 15)); rex = 0; psz = 2 end
			
 
				+    wputb(b); op = band(op, 255); psz = psz + 1
			
 
				   end
			
 
				-  if rex ~= 0 then wputb(64 + band(rex, 15)) end
			
 
				+  if rex ~= 0 then wputb(64 + band(rex, 15)); psz = 2 end
			
 
				   if sz == "b" then op = op - 1 end
			
 
				   wputb(op)
			
 
				+  return psz, sk
			
 
				 end
			
 
				 
			
 
				 -- Put ModRM or SIB formatted byte.
			
@@ -494,7 +564,7 @@ local function wputmodrm(m, s, rm, vs, vrm)
 
				 end
			
 
				 
			
 
				 -- Put ModRM/SIB plus optional displacement.
			
 
				-local function wputmrmsib(t, imark, s, vsreg)
			
 
				+local function wputmrmsib(t, imark, s, vsreg, psz, sk)
			
 
				   local vreg, vxreg
			
 
				   local reg, xreg = t.reg, t.xreg
			
 
				   if reg and reg < 0 then reg = 0; vreg = t.vreg end
			
@@ -504,8 +574,8 @@ local function wputmrmsib(t, imark, s, vsreg)
 
				   -- Register mode.
			
 
				   if sub(t.mode, 1, 1) == "r" then
			
 
				     wputmodrm(3, s, reg)
			
 
				-    if vsreg then waction("VREG", vsreg); wputxb(2) end
			
 
				-    if vreg then waction("VREG", vreg); wputxb(0) end
			
 
				+    wvreg("modrm.reg", vsreg, psz+1, sk, vreg)
			
 
				+    wvreg("modrm.rm.r", vreg, psz+1, sk)
			
 
				     return
			
 
				   end
			
 
				 
			
@@ -519,21 +589,22 @@ local function wputmrmsib(t, imark, s, vsreg)
 
				       -- [xreg*xsc+disp] -> (0, s, esp) (xsc, xreg, ebp)
			
 
				       wputmodrm(0, s, 4)
			
 
				       if imark == "I" then waction("MARK") end
			
 
				-      if vsreg then waction("VREG", vsreg); wputxb(2) end
			
 
				+      wvreg("modrm.reg", vsreg, psz+1, sk, vxreg)
			
 
				       wputmodrm(t.xsc, xreg, 5)
			
 
				-      if vxreg then waction("VREG", vxreg); wputxb(3) end
			
 
				+      wvreg("sib.index", vxreg, psz+2, sk)
			
 
				     else
			
 
				       -- Pure 32 bit displacement.
			
 
				       if x64 and tdisp ~= "table" then
			
 
				 	wputmodrm(0, s, 4) -- [disp] -> (0, s, esp) (0, esp, ebp)
			
 
				+	wvreg("modrm.reg", vsreg, psz+1, sk)
			
 
				 	if imark == "I" then waction("MARK") end
			
 
				 	wputmodrm(0, 4, 5)
			
 
				       else
			
 
				 	riprel = x64
			
 
				 	wputmodrm(0, s, 5) -- [disp|rip-label] -> (0, s, ebp)
			
 
				+	wvreg("modrm.reg", vsreg, psz+1, sk)
			
 
				 	if imark == "I" then waction("MARK") end
			
 
				       end
			
 
				-      if vsreg then waction("VREG", vsreg); wputxb(2) end
			
 
				     end
			
 
				     if riprel then -- Emit rip-relative displacement.
			
 
				       if match("UWSiI", imark) then
			
@@ -561,16 +632,16 @@ local function wputmrmsib(t, imark, s, vsreg)
 
				   if xreg or band(reg, 7) == 4 then
			
 
				     wputmodrm(m or 2, s, 4) -- ModRM.
			
 
				     if m == nil or imark == "I" then waction("MARK") end
			
 
				-    if vsreg then waction("VREG", vsreg); wputxb(2) end
			
 
				+    wvreg("modrm.reg", vsreg, psz+1, sk, vxreg or vreg)
			
 
				     wputmodrm(t.xsc or 0, xreg or 4, reg) -- SIB.
			
 
				-    if vxreg then waction("VREG", vxreg); wputxb(3) end
			
 
				-    if vreg then waction("VREG", vreg); wputxb(1) end
			
 
				+    wvreg("sib.index", vxreg, psz+2, sk, vreg)
			
 
				+    wvreg("sib.base", vreg, psz+2, sk)
			
 
				   else
			
 
				     wputmodrm(m or 2, s, reg) -- ModRM.
			
 
				     if (imark == "I" and (m == 1 or m == 2)) or
			
 
				        (m == nil and (vsreg or vreg)) then waction("MARK") end
			
 
				-    if vsreg then waction("VREG", vsreg); wputxb(2) end
			
 
				-    if vreg then waction("VREG", vreg); wputxb(1) end
			
 
				+    wvreg("modrm.reg", vsreg, psz+1, sk, vreg)
			
 
				+    wvreg("modrm.rm.m", vreg, psz+1, sk)
			
 
				   end
			
 
				 
			
 
				   -- Put displacement.
			
@@ -881,9 +952,16 @@ end
 
				 --   "m"/"M"   generates ModRM/SIB from the 1st/2nd operand.
			
 
				 --             The spare 3 bits are either filled with the last hex digit or
			
 
				 --             the result from a previous "r"/"R". The opcode is restored.
			
 
				+--   "u"       Use VEX encoding, vvvv unused.
			
 
				+--   "v"/"V"   Use VEX encoding, vvvv from 1st/2nd operand (the operand is
			
 
				+--             removed from the list used by future characters).
			
 
				+--   "w"       Use VEX encoding, vvvv from 3rd operand.
			
 
				+--   "L"       Force VEX.L
			
 
				 --
			
 
				 -- All of the following characters force a flush of the opcode:
			
 
				 --   "o"/"O"   stores a pure 32 bit disp (offset) from the 1st/2nd operand.
			
 
				+--   "s"       stores a 4 bit immediate from the last register operand,
			
 
				+--             followed by 4 zero bits.
			
 
				 --   "S"       stores a signed 8 bit immediate from the last operand.
			
 
				 --   "U"       stores an unsigned 8 bit immediate from the last operand.
			
 
				 --   "W"       stores an unsigned 16 bit immediate from the last operand.
			
@@ -1226,46 +1304,14 @@ local map_op = {
 
				   movups_2 =	"rmo:0F10rM|mro:0F11Rm",
			
 
				   orpd_2 =	"rmo:660F56rM",
			
 
				   orps_2 =	"rmo:0F56rM",
			
 
				-  packssdw_2 =	"rmo:660F6BrM",
			
 
				-  packsswb_2 =	"rmo:660F63rM",
			
 
				-  packuswb_2 =	"rmo:660F67rM",
			
 
				-  paddb_2 =	"rmo:660FFCrM",
			
 
				-  paddd_2 =	"rmo:660FFErM",
			
 
				-  paddq_2 =	"rmo:660FD4rM",
			
 
				-  paddsb_2 =	"rmo:660FECrM",
			
 
				-  paddsw_2 =	"rmo:660FEDrM",
			
 
				-  paddusb_2 =	"rmo:660FDCrM",
			
 
				-  paddusw_2 =	"rmo:660FDDrM",
			
 
				-  paddw_2 =	"rmo:660FFDrM",
			
 
				-  pand_2 =	"rmo:660FDBrM",
			
 
				-  pandn_2 =	"rmo:660FDFrM",
			
 
				   pause_0 =	"F390",
			
 
				-  pavgb_2 =	"rmo:660FE0rM",
			
 
				-  pavgw_2 =	"rmo:660FE3rM",
			
 
				-  pcmpeqb_2 =	"rmo:660F74rM",
			
 
				-  pcmpeqd_2 =	"rmo:660F76rM",
			
 
				-  pcmpeqw_2 =	"rmo:660F75rM",
			
 
				-  pcmpgtb_2 =	"rmo:660F64rM",
			
 
				-  pcmpgtd_2 =	"rmo:660F66rM",
			
 
				-  pcmpgtw_2 =	"rmo:660F65rM",
			
 
				   pextrw_3 =	"rri/do:660FC5rMU|xri/wo:660F3A15nRmU", -- Mem op: SSE4.1 only.
			
 
				   pinsrw_3 =	"rri/od:660FC4rMU|rxi/ow:",
			
 
				-  pmaddwd_2 =	"rmo:660FF5rM",
			
 
				-  pmaxsw_2 =	"rmo:660FEErM",
			
 
				-  pmaxub_2 =	"rmo:660FDErM",
			
 
				-  pminsw_2 =	"rmo:660FEArM",
			
 
				-  pminub_2 =	"rmo:660FDArM",
			
 
				   pmovmskb_2 =	"rr/do:660FD7rM",
			
 
				-  pmulhuw_2 =	"rmo:660FE4rM",
			
 
				-  pmulhw_2 =	"rmo:660FE5rM",
			
 
				-  pmullw_2 =	"rmo:660FD5rM",
			
 
				-  pmuludq_2 =	"rmo:660FF4rM",
			
 
				-  por_2 =	"rmo:660FEBrM",
			
 
				   prefetchnta_1 = "xb:n0F180m",
			
 
				   prefetcht0_1 = "xb:n0F181m",
			
 
				   prefetcht1_1 = "xb:n0F182m",
			
 
				   prefetcht2_1 = "xb:n0F183m",
			
 
				-  psadbw_2 =	"rmo:660FF6rM",
			
 
				   pshufd_3 =	"rmio:660F70rMU",
			
 
				   pshufhw_3 =	"rmio:F30F70rMU",
			
 
				   pshuflw_3 =	"rmio:F20F70rMU",
			
@@ -1279,23 +1325,6 @@ local map_op = {
 
				   psrldq_2 =	"rio:660F733mU",
			
 
				   psrlq_2 =	"rmo:660FD3rM|rio:660F732mU",
			
 
				   psrlw_2 =	"rmo:660FD1rM|rio:660F712mU",
			
 
				-  psubb_2 =	"rmo:660FF8rM",
			
 
				-  psubd_2 =	"rmo:660FFArM",
			
 
				-  psubq_2 =	"rmo:660FFBrM",
			
 
				-  psubsb_2 =	"rmo:660FE8rM",
			
 
				-  psubsw_2 =	"rmo:660FE9rM",
			
 
				-  psubusb_2 =	"rmo:660FD8rM",
			
 
				-  psubusw_2 =	"rmo:660FD9rM",
			
 
				-  psubw_2 =	"rmo:660FF9rM",
			
 
				-  punpckhbw_2 =	"rmo:660F68rM",
			
 
				-  punpckhdq_2 =	"rmo:660F6ArM",
			
 
				-  punpckhqdq_2 = "rmo:660F6DrM",
			
 
				-  punpckhwd_2 =	"rmo:660F69rM",
			
 
				-  punpcklbw_2 =	"rmo:660F60rM",
			
 
				-  punpckldq_2 =	"rmo:660F62rM",
			
 
				-  punpcklqdq_2 = "rmo:660F6CrM",
			
 
				-  punpcklwd_2 =	"rmo:660F61rM",
			
 
				-  pxor_2 =	"rmo:660FEFrM",
			
 
				   rcpps_2 =	"rmo:0F53rM",
			
 
				   rcpss_2 =	"rro:F30F53rM|rx/od:",
			
 
				   rsqrtps_2 =	"rmo:0F52rM",
			
@@ -1413,6 +1442,327 @@ local map_op = {
 
				   movntsd_2 =	"xr/qo:nF20F2BRm",
			
 
				   movntss_2 =	"xr/do:F30F2BRm",
			
 
				   -- popcnt is also in SSE4.2
			
 
				+
			
 
				+  -- AES-NI
			
 
				+  aesdec_2 =	"rmo:660F38DErM",
			
 
				+  aesdeclast_2 = "rmo:660F38DFrM",
			
 
				+  aesenc_2 =	"rmo:660F38DCrM",
			
 
				+  aesenclast_2 = "rmo:660F38DDrM",
			
 
				+  aesimc_2 =	"rmo:660F38DBrM",
			
 
				+  aeskeygenassist_3 = "rmio:660F3ADFrMU",
			
 
				+  pclmulqdq_3 =	"rmio:660F3A44rMU",
			
 
				+
			
 
				+   -- AVX FP ops
			
 
				+  vaddsubpd_3 =	"rrmoy:660FVD0rM",
			
 
				+  vaddsubps_3 =	"rrmoy:F20FVD0rM",
			
 
				+  vandpd_3 =	"rrmoy:660FV54rM",
			
 
				+  vandps_3 =	"rrmoy:0FV54rM",
			
 
				+  vandnpd_3 =	"rrmoy:660FV55rM",
			
 
				+  vandnps_3 =	"rrmoy:0FV55rM",
			
 
				+  vblendpd_4 =	"rrmioy:660F3AV0DrMU",
			
 
				+  vblendps_4 =	"rrmioy:660F3AV0CrMU",
			
 
				+  vblendvpd_4 =	"rrmroy:660F3AV4BrMs",
			
 
				+  vblendvps_4 =	"rrmroy:660F3AV4ArMs",
			
 
				+  vbroadcastf128_2 = "rx/yo:660F38u1ArM",
			
 
				+  vcmppd_4 =	"rrmioy:660FVC2rMU",
			
 
				+  vcmpps_4 =	"rrmioy:0FVC2rMU",
			
 
				+  vcmpsd_4 =	"rrrio:F20FVC2rMU|rrxi/ooq:",
			
 
				+  vcmpss_4 =	"rrrio:F30FVC2rMU|rrxi/ood:",
			
 
				+  vcomisd_2 =	"rro:660Fu2FrM|rx/oq:",
			
 
				+  vcomiss_2 =	"rro:0Fu2FrM|rx/od:",
			
 
				+  vcvtdq2pd_2 =	"rro:F30FuE6rM|rx/oq:|rm/yo:",
			
 
				+  vcvtdq2ps_2 =	"rmoy:0Fu5BrM",
			
 
				+  vcvtpd2dq_2 =	"rmoy:F20FuE6rM",
			
 
				+  vcvtpd2ps_2 =	"rmoy:660Fu5ArM",
			
 
				+  vcvtps2dq_2 =	"rmoy:660Fu5BrM",
			
 
				+  vcvtps2pd_2 =	"rro:0Fu5ArM|rx/oq:|rm/yo:",
			
 
				+  vcvtsd2si_2 =	"rr/do:F20Fu2DrM|rx/dq:|rr/qo:|rxq:",
			
 
				+  vcvtsd2ss_3 =	"rrro:F20FV5ArM|rrx/ooq:",
			
 
				+  vcvtsi2sd_3 =	"rrm/ood:F20FV2ArM|rrm/ooq:F20FVX2ArM",
			
 
				+  vcvtsi2ss_3 =	"rrm/ood:F30FV2ArM|rrm/ooq:F30FVX2ArM",
			
 
				+  vcvtss2sd_3 =	"rrro:F30FV5ArM|rrx/ood:",
			
 
				+  vcvtss2si_2 =	"rr/do:F30Fu2DrM|rxd:|rr/qo:|rx/qd:",
			
 
				+  vcvttpd2dq_2 = "rmo:660FuE6rM|rm/oy:660FuLE6rM",
			
 
				+  vcvttps2dq_2 = "rmoy:F30Fu5BrM",
			
 
				+  vcvttsd2si_2 = "rr/do:F20Fu2CrM|rx/dq:|rr/qo:|rxq:",
			
 
				+  vcvttss2si_2 = "rr/do:F30Fu2CrM|rxd:|rr/qo:|rx/qd:",
			
 
				+  vdppd_4 =	"rrmio:660F3AV41rMU",
			
 
				+  vdpps_4 =	"rrmioy:660F3AV40rMU",
			
 
				+  vextractf128_3 = "mri/oy:660F3AuL19RmU",
			
 
				+  vextractps_3 = "mri/do:660F3Au17RmU",
			
 
				+  vhaddpd_3 =	"rrmoy:660FV7CrM",
			
 
				+  vhaddps_3 =	"rrmoy:F20FV7CrM",
			
 
				+  vhsubpd_3 =	"rrmoy:660FV7DrM",
			
 
				+  vhsubps_3 =	"rrmoy:F20FV7DrM",
			
 
				+  vinsertf128_4 = "rrmi/yyo:660F3AV18rMU",
			
 
				+  vinsertps_4 =	"rrrio:660F3AV21rMU|rrxi/ood:",
			
 
				+  vldmxcsr_1 =	"xd:0FuAE2m",
			
 
				+  vmaskmovps_3 = "rrxoy:660F38V2CrM|xrroy:660F38V2ERm",
			
 
				+  vmaskmovpd_3 = "rrxoy:660F38V2DrM|xrroy:660F38V2FRm",
			
 
				+  vmovapd_2 =	"rmoy:660Fu28rM|mroy:660Fu29Rm",
			
 
				+  vmovaps_2 =	"rmoy:0Fu28rM|mroy:0Fu29Rm",
			
 
				+  vmovd_2 =	"rm/od:660Fu6ErM|rm/oq:660FuX6ErM|mr/do:660Fu7ERm|mr/qo:",
			
 
				+  vmovq_2 =	"rro:F30Fu7ErM|rx/oq:|xr/qo:660FuD6Rm",
			
 
				+  vmovddup_2 =	"rmy:F20Fu12rM|rro:|rx/oq:",
			
 
				+  vmovhlps_3 =	"rrro:0FV12rM",
			
 
				+  vmovhpd_2 =	"xr/qo:660Fu17Rm",
			
 
				+  vmovhpd_3 =	"rrx/ooq:660FV16rM",
			
 
				+  vmovhps_2 =	"xr/qo:0Fu17Rm",
			
 
				+  vmovhps_3 =	"rrx/ooq:0FV16rM",
			
 
				+  vmovlhps_3 =	"rrro:0FV16rM",
			
 
				+  vmovlpd_2 =	"xr/qo:660Fu13Rm",
			
 
				+  vmovlpd_3 =	"rrx/ooq:660FV12rM",
			
 
				+  vmovlps_2 =	"xr/qo:0Fu13Rm",
			
 
				+  vmovlps_3 =	"rrx/ooq:0FV12rM",
			
 
				+  vmovmskpd_2 =	"rr/do:660Fu50rM|rr/dy:660FuL50rM",
			
 
				+  vmovmskps_2 =	"rr/do:0Fu50rM|rr/dy:0FuL50rM",
			
 
				+  vmovntpd_2 =	"xroy:660Fu2BRm",
			
 
				+  vmovntps_2 =	"xroy:0Fu2BRm",
			
 
				+  vmovsd_2 =	"rx/oq:F20Fu10rM|xr/qo:F20Fu11Rm",
			
 
				+  vmovsd_3 =	"rrro:F20FV10rM",
			
 
				+  vmovshdup_2 =	"rmoy:F30Fu16rM",
			
 
				+  vmovsldup_2 =	"rmoy:F30Fu12rM",
			
 
				+  vmovss_2 =	"rx/od:F30Fu10rM|xr/do:F30Fu11Rm",
			
 
				+  vmovss_3 =	"rrro:F30FV10rM",
			
 
				+  vmovupd_2 =	"rmoy:660Fu10rM|mroy:660Fu11Rm",
			
 
				+  vmovups_2 =	"rmoy:0Fu10rM|mroy:0Fu11Rm",
			
 
				+  vorpd_3 =	"rrmoy:660FV56rM",
			
 
				+  vorps_3 =	"rrmoy:0FV56rM",
			
 
				+  vpermilpd_3 =	"rrmoy:660F38V0DrM|rmioy:660F3Au05rMU",
			
 
				+  vpermilps_3 =	"rrmoy:660F38V0CrM|rmioy:660F3Au04rMU",
			
 
				+  vperm2f128_4 = "rrmiy:660F3AV06rMU",
			
 
				+  vptestpd_2 =	"rmoy:660F38u0FrM",
			
 
				+  vptestps_2 =	"rmoy:660F38u0ErM",
			
 
				+  vrcpps_2 =	"rmoy:0Fu53rM",
			
 
				+  vrcpss_3 =	"rrro:F30FV53rM|rrx/ood:",
			
 
				+  vrsqrtps_2 =	"rmoy:0Fu52rM",
			
 
				+  vrsqrtss_3 =	"rrro:F30FV52rM|rrx/ood:",
			
 
				+  vroundpd_3 =	"rmioy:660F3Au09rMU",
			
 
				+  vroundps_3 =	"rmioy:660F3Au08rMU",
			
 
				+  vroundsd_4 =	"rrrio:660F3AV0BrMU|rrxi/ooq:",
			
 
				+  vroundss_4 =	"rrrio:660F3AV0ArMU|rrxi/ood:",
			
 
				+  vshufpd_4 =	"rrmioy:660FVC6rMU",
			
 
				+  vshufps_4 =	"rrmioy:0FVC6rMU",
			
 
				+  vsqrtps_2 =	"rmoy:0Fu51rM",
			
 
				+  vsqrtss_2 =	"rro:F30Fu51rM|rx/od:",
			
 
				+  vsqrtpd_2 =	"rmoy:660Fu51rM",
			
 
				+  vsqrtsd_2 =	"rro:F20Fu51rM|rx/oq:",
			
 
				+  vstmxcsr_1 =	"xd:0FuAE3m",
			
 
				+  vucomisd_2 =	"rro:660Fu2ErM|rx/oq:",
			
 
				+  vucomiss_2 =	"rro:0Fu2ErM|rx/od:",
			
 
				+  vunpckhpd_3 =	"rrmoy:660FV15rM",
			
 
				+  vunpckhps_3 =	"rrmoy:0FV15rM",
			
 
				+  vunpcklpd_3 =	"rrmoy:660FV14rM",
			
 
				+  vunpcklps_3 =	"rrmoy:0FV14rM",
			
 
				+  vxorpd_3 =	"rrmoy:660FV57rM",
			
 
				+  vxorps_3 =	"rrmoy:0FV57rM",
			
 
				+  vzeroall_0 =	"0FuL77",
			
 
				+  vzeroupper_0 = "0Fu77",
			
 
				+
			
 
				+  -- AVX2 FP ops
			
 
				+  vbroadcastss_2 = "rx/od:660F38u18rM|rx/yd:|rro:|rr/yo:",
			
 
				+  vbroadcastsd_2 = "rx/yq:660F38u19rM|rr/yo:",
			
 
				+  -- *vgather* (!vsib)
			
 
				+  vpermpd_3 =	"rmiy:660F3AuX01rMU",
			
 
				+  vpermps_3 =	"rrmy:660F38V16rM",
			
 
				+
			
 
				+  -- AVX, AVX2 integer ops
			
 
				+  -- In general, xmm requires AVX, ymm requires AVX2.
			
 
				+  vaesdec_3 =  "rrmo:660F38VDErM",
			
 
				+  vaesdeclast_3 = "rrmo:660F38VDFrM",
			
 
				+  vaesenc_3 =  "rrmo:660F38VDCrM",
			
 
				+  vaesenclast_3 = "rrmo:660F38VDDrM",
			
 
				+  vaesimc_2 =  "rmo:660F38uDBrM",
			
 
				+  vaeskeygenassist_3 = "rmio:660F3AuDFrMU",
			
 
				+  vlddqu_2 =	"rxoy:F20FuF0rM",
			
 
				+  vmaskmovdqu_2 = "rro:660FuF7rM",
			
 
				+  vmovdqa_2 =	"rmoy:660Fu6FrM|mroy:660Fu7FRm",
			
 
				+  vmovdqu_2 =	"rmoy:F30Fu6FrM|mroy:F30Fu7FRm",
			
 
				+  vmovntdq_2 =	"xroy:660FuE7Rm",
			
 
				+  vmovntdqa_2 =	"rxoy:660F38u2ArM",
			
 
				+  vmpsadbw_4 =	"rrmioy:660F3AV42rMU",
			
 
				+  vpabsb_2 =	"rmoy:660F38u1CrM",
			
 
				+  vpabsd_2 =	"rmoy:660F38u1ErM",
			
 
				+  vpabsw_2 =	"rmoy:660F38u1DrM",
			
 
				+  vpackusdw_3 =	"rrmoy:660F38V2BrM",
			
 
				+  vpalignr_4 =	"rrmioy:660F3AV0FrMU",
			
 
				+  vpblendvb_4 =	"rrmroy:660F3AV4CrMs",
			
 
				+  vpblendw_4 =	"rrmioy:660F3AV0ErMU",
			
 
				+  vpclmulqdq_4 = "rrmio:660F3AV44rMU",
			
 
				+  vpcmpeqq_3 =	"rrmoy:660F38V29rM",
			
 
				+  vpcmpestri_3 = "rmio:660F3Au61rMU",
			
 
				+  vpcmpestrm_3 = "rmio:660F3Au60rMU",
			
 
				+  vpcmpgtq_3 =	"rrmoy:660F38V37rM",
			
 
				+  vpcmpistri_3 = "rmio:660F3Au63rMU",
			
 
				+  vpcmpistrm_3 = "rmio:660F3Au62rMU",
			
 
				+  vpextrb_3 =	"rri/do:660F3Au14nRmU|rri/qo:|xri/bo:",
			
 
				+  vpextrw_3 =	"rri/do:660FuC5rMU|xri/wo:660F3Au15nRmU",
			
 
				+  vpextrd_3 =	"mri/do:660F3Au16RmU",
			
 
				+  vpextrq_3 =	"mri/qo:660F3Au16RmU",
			
 
				+  vphaddw_3 =	"rrmoy:660F38V01rM",
			
 
				+  vphaddd_3 =	"rrmoy:660F38V02rM",
			
 
				+  vphaddsw_3 =	"rrmoy:660F38V03rM",
			
 
				+  vphminposuw_2 = "rmo:660F38u41rM",
			
 
				+  vphsubw_3 =	"rrmoy:660F38V05rM",
			
 
				+  vphsubd_3 =	"rrmoy:660F38V06rM",
			
 
				+  vphsubsw_3 =	"rrmoy:660F38V07rM",
			
 
				+  vpinsrb_4 =	"rrri/ood:660F3AV20rMU|rrxi/oob:",
			
 
				+  vpinsrw_4 =	"rrri/ood:660FVC4rMU|rrxi/oow:",
			
 
				+  vpinsrd_4 =	"rrmi/ood:660F3AV22rMU",
			
 
				+  vpinsrq_4 =	"rrmi/ooq:660F3AVX22rMU",
			
 
				+  vpmaddubsw_3 = "rrmoy:660F38V04rM",
			
 
				+  vpmaxsb_3 =	"rrmoy:660F38V3CrM",
			
 
				+  vpmaxsd_3 =	"rrmoy:660F38V3DrM",
			
 
				+  vpmaxuw_3 =	"rrmoy:660F38V3ErM",
			
 
				+  vpmaxud_3 =	"rrmoy:660F38V3FrM",
			
 
				+  vpminsb_3 =	"rrmoy:660F38V38rM",
			
 
				+  vpminsd_3 =	"rrmoy:660F38V39rM",
			
 
				+  vpminuw_3 =	"rrmoy:660F38V3ArM",
			
 
				+  vpminud_3 =	"rrmoy:660F38V3BrM",
			
 
				+  vpmovmskb_2 =	"rr/do:660FuD7rM|rr/dy:660FuLD7rM",
			
 
				+  vpmovsxbw_2 =	"rroy:660F38u20rM|rx/oq:|rx/yo:",
			
 
				+  vpmovsxbd_2 =	"rroy:660F38u21rM|rx/od:|rx/yq:",
			
 
				+  vpmovsxbq_2 =	"rroy:660F38u22rM|rx/ow:|rx/yd:",
			
 
				+  vpmovsxwd_2 =	"rroy:660F38u23rM|rx/oq:|rx/yo:",
			
 
				+  vpmovsxwq_2 =	"rroy:660F38u24rM|rx/od:|rx/yq:",
			
 
				+  vpmovsxdq_2 =	"rroy:660F38u25rM|rx/oq:|rx/yo:",
			
 
				+  vpmovzxbw_2 =	"rroy:660F38u30rM|rx/oq:|rx/yo:",
			
 
				+  vpmovzxbd_2 =	"rroy:660F38u31rM|rx/od:|rx/yq:",
			
 
				+  vpmovzxbq_2 =	"rroy:660F38u32rM|rx/ow:|rx/yd:",
			
 
				+  vpmovzxwd_2 =	"rroy:660F38u33rM|rx/oq:|rx/yo:",
			
 
				+  vpmovzxwq_2 =	"rroy:660F38u34rM|rx/od:|rx/yq:",
			
 
				+  vpmovzxdq_2 =	"rroy:660F38u35rM|rx/oq:|rx/yo:",
			
 
				+  vpmuldq_3 =	"rrmoy:660F38V28rM",
			
 
				+  vpmulhrsw_3 =	"rrmoy:660F38V0BrM",
			
 
				+  vpmulld_3 =	"rrmoy:660F38V40rM",
			
 
				+  vpshufb_3 =	"rrmoy:660F38V00rM",
			
 
				+  vpshufd_3 =	"rmioy:660Fu70rMU",
			
 
				+  vpshufhw_3 =	"rmioy:F30Fu70rMU",
			
 
				+  vpshuflw_3 =	"rmioy:F20Fu70rMU",
			
 
				+  vpsignb_3 =	"rrmoy:660F38V08rM",
			
 
				+  vpsignw_3 =	"rrmoy:660F38V09rM",
			
 
				+  vpsignd_3 =	"rrmoy:660F38V0ArM",
			
 
				+  vpslldq_3 =	"rrioy:660Fv737mU",
			
 
				+  vpsllw_3 =	"rrmoy:660FVF1rM|rrioy:660Fv716mU",
			
 
				+  vpslld_3 =	"rrmoy:660FVF2rM|rrioy:660Fv726mU",
			
 
				+  vpsllq_3 =	"rrmoy:660FVF3rM|rrioy:660Fv736mU",
			
 
				+  vpsraw_3 =	"rrmoy:660FVE1rM|rrioy:660Fv714mU",
			
 
				+  vpsrad_3 =	"rrmoy:660FVE2rM|rrioy:660Fv724mU",
			
 
				+  vpsrldq_3 =	"rrioy:660Fv733mU",
			
 
				+  vpsrlw_3 =	"rrmoy:660FVD1rM|rrioy:660Fv712mU",
			
 
				+  vpsrld_3 =	"rrmoy:660FVD2rM|rrioy:660Fv722mU",
			
 
				+  vpsrlq_3 =	"rrmoy:660FVD3rM|rrioy:660Fv732mU",
			
 
				+  vptest_2 =	"rmoy:660F38u17rM",
			
 
				+
			
 
				+  -- AVX2 integer ops
			
 
				+  vbroadcasti128_2 = "rx/yo:660F38u5ArM",
			
 
				+  vinserti128_4 = "rrmi/yyo:660F3AV38rMU",
			
 
				+  vextracti128_3 = "mri/oy:660F3AuL39RmU",
			
 
				+  vpblendd_4 =	"rrmioy:660F3AV02rMU",
			
 
				+  vpbroadcastb_2 = "rro:660F38u78rM|rx/ob:|rr/yo:|rx/yb:",
			
 
				+  vpbroadcastw_2 = "rro:660F38u79rM|rx/ow:|rr/yo:|rx/yw:",
			
 
				+  vpbroadcastd_2 = "rro:660F38u58rM|rx/od:|rr/yo:|rx/yd:",
			
 
				+  vpbroadcastq_2 = "rro:660F38u59rM|rx/oq:|rr/yo:|rx/yq:",
			
 
				+  vpermd_3 =	"rrmy:660F38V36rM",
			
 
				+  vpermq_3 =	"rmiy:660F3AuX00rMU",
			
 
				+  -- *vpgather* (!vsib)
			
 
				+  vperm2i128_4 = "rrmiy:660F3AV46rMU",
			
 
				+  vpmaskmovd_3 = "rrxoy:660F38V8CrM|xrroy:660F38V8ERm",
			
 
				+  vpmaskmovq_3 = "rrxoy:660F38VX8CrM|xrroy:660F38VX8ERm",
			
 
				+  vpsllvd_3 =	"rrmoy:660F38V47rM",
			
 
				+  vpsllvq_3 =	"rrmoy:660F38VX47rM",
			
 
				+  vpsravd_3 =	"rrmoy:660F38V46rM",
			
 
				+  vpsrlvd_3 =	"rrmoy:660F38V45rM",
			
 
				+  vpsrlvq_3 =	"rrmoy:660F38VX45rM",
			
 
				+
			
 
				+  -- Intel ADX
			
 
				+  adcx_2 =	"rmqd:660F38F6rM",
			
 
				+  adox_2 =	"rmqd:F30F38F6rM",
			
 
				+
			
 
				+  -- BMI1
			
 
				+  andn_3 =	"rrmqd:0F38VF2rM",
			
 
				+  bextr_3 =	"rmrqd:0F38wF7rM",
			
 
				+  blsi_2 =	"rmqd:0F38vF33m",
			
 
				+  blsmsk_2 =	"rmqd:0F38vF32m",
			
 
				+  blsr_2 =	"rmqd:0F38vF31m",
			
 
				+  tzcnt_2 =	"rmqdw:F30FBCrM",
			
 
				+
			
 
				+  -- BMI2
			
 
				+  bzhi_3 =	"rmrqd:0F38wF5rM",
			
 
				+  mulx_3 =	"rrmqd:F20F38VF6rM",
			
 
				+  pdep_3 =	"rrmqd:F20F38VF5rM",
			
 
				+  pext_3 =	"rrmqd:F30F38VF5rM",
			
 
				+  rorx_3 =	"rmSqd:F20F3AuF0rMS",
			
 
				+  sarx_3 =	"rmrqd:F30F38wF7rM",
			
 
				+  shrx_3 =	"rmrqd:F20F38wF7rM",
			
 
				+  shlx_3 =	"rmrqd:660F38wF7rM",
			
 
				+
			
 
				+  -- FMA3
			
 
				+  vfmaddsub132pd_3 = "rrmoy:660F38VX96rM",
			
 
				+  vfmaddsub132ps_3 = "rrmoy:660F38V96rM",
			
 
				+  vfmaddsub213pd_3 = "rrmoy:660F38VXA6rM",
			
 
				+  vfmaddsub213ps_3 = "rrmoy:660F38VA6rM",
			
 
				+  vfmaddsub231pd_3 = "rrmoy:660F38VXB6rM",
			
 
				+  vfmaddsub231ps_3 = "rrmoy:660F38VB6rM",
			
 
				+
			
 
				+  vfmsubadd132pd_3 = "rrmoy:660F38VX97rM",
			
 
				+  vfmsubadd132ps_3 = "rrmoy:660F38V97rM",
			
 
				+  vfmsubadd213pd_3 = "rrmoy:660F38VXA7rM",
			
 
				+  vfmsubadd213ps_3 = "rrmoy:660F38VA7rM",
			
 
				+  vfmsubadd231pd_3 = "rrmoy:660F38VXB7rM",
			
 
				+  vfmsubadd231ps_3 = "rrmoy:660F38VB7rM",
			
 
				+
			
 
				+  vfmadd132pd_3 = "rrmoy:660F38VX98rM",
			
 
				+  vfmadd132ps_3 = "rrmoy:660F38V98rM",
			
 
				+  vfmadd132sd_3 = "rrro:660F38VX99rM|rrx/ooq:",
			
 
				+  vfmadd132ss_3 = "rrro:660F38V99rM|rrx/ood:",
			
 
				+  vfmadd213pd_3 = "rrmoy:660F38VXA8rM",
			
 
				+  vfmadd213ps_3 = "rrmoy:660F38VA8rM",
			
 
				+  vfmadd213sd_3 = "rrro:660F38VXA9rM|rrx/ooq:",
			
 
				+  vfmadd213ss_3 = "rrro:660F38VA9rM|rrx/ood:",
			
 
				+  vfmadd231pd_3 = "rrmoy:660F38VXB8rM",
			
 
				+  vfmadd231ps_3 = "rrmoy:660F38VB8rM",
			
 
				+  vfmadd231sd_3 = "rrro:660F38VXB9rM|rrx/ooq:",
			
 
				+  vfmadd231ss_3 = "rrro:660F38VB9rM|rrx/ood:",
			
 
				+
			
 
				+  vfmsub132pd_3 = "rrmoy:660F38VX9ArM",
			
 
				+  vfmsub132ps_3 = "rrmoy:660F38V9ArM",
			
 
				+  vfmsub132sd_3 = "rrro:660F38VX9BrM|rrx/ooq:",
			
 
				+  vfmsub132ss_3 = "rrro:660F38V9BrM|rrx/ood:",
			
 
				+  vfmsub213pd_3 = "rrmoy:660F38VXAArM",
			
 
				+  vfmsub213ps_3 = "rrmoy:660F38VAArM",
			
 
				+  vfmsub213sd_3 = "rrro:660F38VXABrM|rrx/ooq:",
			
 
				+  vfmsub213ss_3 = "rrro:660F38VABrM|rrx/ood:",
			
 
				+  vfmsub231pd_3 = "rrmoy:660F38VXBArM",
			
 
				+  vfmsub231ps_3 = "rrmoy:660F38VBArM",
			
 
				+  vfmsub231sd_3 = "rrro:660F38VXBBrM|rrx/ooq:",
			
 
				+  vfmsub231ss_3 = "rrro:660F38VBBrM|rrx/ood:",
			
 
				+
			
 
				+  vfnmadd132pd_3 = "rrmoy:660F38VX9CrM",
			
 
				+  vfnmadd132ps_3 = "rrmoy:660F38V9CrM",
			
 
				+  vfnmadd132sd_3 = "rrro:660F38VX9DrM|rrx/ooq:",
			
 
				+  vfnmadd132ss_3 = "rrro:660F38V9DrM|rrx/ood:",
			
 
				+  vfnmadd213pd_3 = "rrmoy:660F38VXACrM",
			
 
				+  vfnmadd213ps_3 = "rrmoy:660F38VACrM",
			
 
				+  vfnmadd213sd_3 = "rrro:660F38VXADrM|rrx/ooq:",
			
 
				+  vfnmadd213ss_3 = "rrro:660F38VADrM|rrx/ood:",
			
 
				+  vfnmadd231pd_3 = "rrmoy:660F38VXBCrM",
			
 
				+  vfnmadd231ps_3 = "rrmoy:660F38VBCrM",
			
 
				+  vfnmadd231sd_3 = "rrro:660F38VXBDrM|rrx/ooq:",
			
 
				+  vfnmadd231ss_3 = "rrro:660F38VBDrM|rrx/ood:",
			
 
				+
			
 
				+  vfnmsub132pd_3 = "rrmoy:660F38VX9ErM",
			
 
				+  vfnmsub132ps_3 = "rrmoy:660F38V9ErM",
			
 
				+  vfnmsub132sd_3 = "rrro:660F38VX9FrM|rrx/ooq:",
			
 
				+  vfnmsub132ss_3 = "rrro:660F38V9FrM|rrx/ood:",
			
 
				+  vfnmsub213pd_3 = "rrmoy:660F38VXAErM",
			
 
				+  vfnmsub213ps_3 = "rrmoy:660F38VAErM",
			
 
				+  vfnmsub213sd_3 = "rrro:660F38VXAFrM|rrx/ooq:",
			
 
				+  vfnmsub213ss_3 = "rrro:660F38VAFrM|rrx/ood:",
			
 
				+  vfnmsub231pd_3 = "rrmoy:660F38VXBErM",
			
 
				+  vfnmsub231ps_3 = "rrmoy:660F38VBErM",
			
 
				+  vfnmsub231sd_3 = "rrro:660F38VXBFrM|rrx/ooq:",
			
 
				+  vfnmsub231ss_3 = "rrro:660F38VBFrM|rrx/ood:",
			
 
				 }
			
 
				 
			
 
				 ------------------------------------------------------------------------------
			
@@ -1463,28 +1813,58 @@ for cc,n in pairs{ b=0, e=1, be=2, u=3, nb=4, ne=5, nbe=6, nu=7 } do
 
				   map_op["fcmov"..cc.."_2"] = format("Fff:%04XR", nc) -- P6+
			
 
				 end
			
 
				 
			
 
				--- SSE FP arithmetic ops.
			
 
				+-- SSE / AVX FP arithmetic ops.
			
 
				 for name,n in pairs{ sqrt = 1, add = 8, mul = 9,
			
 
				 		     sub = 12, min = 13, div = 14, max = 15 } do
			
 
				   map_op[name.."ps_2"] = format("rmo:0F5%XrM", n)
			
 
				   map_op[name.."ss_2"] = format("rro:F30F5%XrM|rx/od:", n)
			
 
				   map_op[name.."pd_2"] = format("rmo:660F5%XrM", n)
			
 
				   map_op[name.."sd_2"] = format("rro:F20F5%XrM|rx/oq:", n)
			
 
				+  if n ~= 1 then
			
 
				+    map_op["v"..name.."ps_3"] = format("rrmoy:0FV5%XrM", n)
			
 
				+    map_op["v"..name.."ss_3"] = format("rrro:F30FV5%XrM|rrx/ood:", n)
			
 
				+    map_op["v"..name.."pd_3"] = format("rrmoy:660FV5%XrM", n)
			
 
				+    map_op["v"..name.."sd_3"] = format("rrro:F20FV5%XrM|rrx/ooq:", n)
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+-- SSE2 / AVX / AVX2 integer arithmetic ops (66 0F leaf).
			
 
				+for name,n in pairs{
			
 
				+  paddb = 0xFC, paddw = 0xFD, paddd = 0xFE, paddq = 0xD4,
			
 
				+  paddsb = 0xEC, paddsw = 0xED, packssdw = 0x6B,
			
 
				+  packsswb = 0x63, packuswb = 0x67, paddusb = 0xDC,
			
 
				+  paddusw = 0xDD, pand = 0xDB, pandn = 0xDF, pavgb = 0xE0,
			
 
				+  pavgw = 0xE3, pcmpeqb = 0x74, pcmpeqd = 0x76,
			
 
				+  pcmpeqw = 0x75, pcmpgtb = 0x64, pcmpgtd = 0x66,
			
 
				+  pcmpgtw = 0x65, pmaddwd = 0xF5, pmaxsw = 0xEE,
			
 
				+  pmaxub = 0xDE, pminsw = 0xEA, pminub = 0xDA,
			
 
				+  pmulhuw = 0xE4, pmulhw = 0xE5, pmullw = 0xD5,
			
 
				+  pmuludq = 0xF4, por = 0xEB, psadbw = 0xF6, psubb = 0xF8,
			
 
				+  psubw = 0xF9, psubd = 0xFA, psubq = 0xFB, psubsb = 0xE8,
			
 
				+  psubsw = 0xE9, psubusb = 0xD8, psubusw = 0xD9,
			
 
				+  punpckhbw = 0x68, punpckhwd = 0x69, punpckhdq = 0x6A,
			
 
				+  punpckhqdq = 0x6D, punpcklbw = 0x60, punpcklwd = 0x61,
			
 
				+  punpckldq = 0x62, punpcklqdq = 0x6C, pxor = 0xEF
			
 
				+} do
			
 
				+  map_op[name.."_2"] = format("rmo:660F%02XrM", n)
			
 
				+  map_op["v"..name.."_3"] = format("rrmoy:660FV%02XrM", n)
			
 
				 end
			
 
				 
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				+local map_vexarg = { u = false, v = 1, V = 2 }
			
 
				+
			
 
				 -- Process pattern string.
			
 
				 local function dopattern(pat, args, sz, op, needrex)
			
 
				-  local digit, addin
			
 
				+  local digit, addin, vex
			
 
				   local opcode = 0
			
 
				   local szov = sz
			
 
				   local narg = 1
			
 
				   local rex = 0
			
 
				 
			
 
				   -- Limit number of section buffer positions used by a single dasm_put().
			
 
				-  -- A single opcode needs a maximum of 5 positions.
			
 
				-  if secpos+5 > maxsecpos then wflush() end
			
 
				+  -- A single opcode needs a maximum of 6 positions.
			
 
				+  if secpos+6 > maxsecpos then wflush() end
			
 
				 
			
 
				   -- Process each character.
			
 
				   for c in gmatch(pat.."|", ".") do
			
@@ -1498,6 +1878,8 @@ local function dopattern(pat, args, sz, op, needrex)
 
				       szov = nil
			
 
				     elseif c == "X" then	-- Force REX.W.
			
 
				       rex = 8
			
 
				+    elseif c == "L" then	-- Force VEX.L.
			
 
				+      vex.l = true
			
 
				     elseif c == "r" then	-- Merge 1st operand regno. into opcode.
			
 
				       addin = args[1]; opcode = opcode + (addin.reg % 8)
			
 
				       if narg < 2 then narg = 2 end
			
@@ -1521,21 +1903,42 @@ local function dopattern(pat, args, sz, op, needrex)
 
				       if t.xreg and t.xreg > 7 then rex = rex + 2 end
			
 
				       if s > 7 then rex = rex + 4 end
			
 
				       if needrex then rex = rex + 16 end
			
 
				-      wputop(szov, opcode, rex); opcode = nil
			
 
				+      local psz, sk = wputop(szov, opcode, rex, vex, s < 0, t.vreg or t.vxreg)
			
 
				+      opcode = nil
			
 
				       local imark = sub(pat, -1) -- Force a mark (ugly).
			
 
				       -- Put ModRM/SIB with regno/last digit as spare.
			
 
				-      wputmrmsib(t, imark, s, addin and addin.vreg)
			
 
				+      wputmrmsib(t, imark, s, addin and addin.vreg, psz, sk)
			
 
				       addin = nil
			
 
				+    elseif map_vexarg[c] ~= nil then -- Encode using VEX prefix
			
 
				+      local b = band(opcode, 255); opcode = shr(opcode, 8)
			
 
				+      local m = 1
			
 
				+      if b == 0x38 then m = 2
			
 
				+      elseif b == 0x3a then m = 3 end
			
 
				+      if m ~= 1 then b = band(opcode, 255); opcode = shr(opcode, 8) end
			
 
				+      if b ~= 0x0f then
			
 
				+	werror("expected `0F', `0F38', or `0F3A' to precede `"..c..
			
 
				+	  "' in pattern `"..pat.."' for `"..op.."'")
			
 
				+      end
			
 
				+      local v = map_vexarg[c]
			
 
				+      if v then v = remove(args, v) end
			
 
				+      b = band(opcode, 255)
			
 
				+      local p = 0
			
 
				+      if b == 0x66 then p = 1
			
 
				+      elseif b == 0xf3 then p = 2
			
 
				+      elseif b == 0xf2 then p = 3 end
			
 
				+      if p ~= 0 then opcode = shr(opcode, 8) end
			
 
				+      if opcode ~= 0 then wputop(nil, opcode, 0); opcode = 0 end
			
 
				+      vex = { m = m, p = p, v = v }
			
 
				     else
			
 
				       if opcode then -- Flush opcode.
			
 
				 	if szov == "q" and rex == 0 then rex = rex + 8 end
			
 
				 	if needrex then rex = rex + 16 end
			
 
				 	if addin and addin.reg == -1 then
			
 
				-	  wputop(szov, opcode - 7, rex)
			
 
				-	  waction("VREG", addin.vreg); wputxb(0)
			
 
				+	  local psz, sk = wputop(szov, opcode - 7, rex, vex, true)
			
 
				+	  wvreg("opcode", addin.vreg, psz, sk)
			
 
				 	else
			
 
				 	  if addin and addin.reg > 7 then rex = rex + 1 end
			
 
				-	  wputop(szov, opcode, rex)
			
 
				+	  wputop(szov, opcode, rex, vex)
			
 
				 	end
			
 
				 	opcode = nil
			
 
				       end
			
@@ -1572,6 +1975,14 @@ local function dopattern(pat, args, sz, op, needrex)
 
				 	  else
			
 
				 	    wputlabel("REL_", imm, 2)
			
 
				 	  end
			
 
				+	elseif c == "s" then
			
 
				+	  local reg = a.reg
			
 
				+	  if reg < 0 then
			
 
				+	    wputb(0)
			
 
				+	    wvreg("imm.hi", a.vreg)
			
 
				+	  else
			
 
				+	    wputb(shl(reg, 4))
			
 
				+	  end
			
 
				 	else
			
 
				 	  werror("bad char `"..c.."' in pattern `"..pat.."' for `"..op.."'")
			
 
				 	end
			
@@ -1648,11 +2059,14 @@ map_op[".template__"] = function(params, template, nparams)
 
				     if pat == "" then pat = lastpat else lastpat = pat end
			
 
				     if matchtm(tm, args) then
			
 
				       local prefix = sub(szm, 1, 1)
			
 
				-      if prefix == "/" then -- Match both operand sizes.
			
 
				-	if args[1].opsize == sub(szm, 2, 2) and
			
 
				-	   args[2].opsize == sub(szm, 3, 3) then
			
 
				-	  dopattern(pat, args, sz, params.op, needrex) -- Process pattern.
			
 
				-	  return
			
 
				+      if prefix == "/" then -- Exactly match leading operand sizes.
			
 
				+	for i = #szm,1,-1 do
			
 
				+	  if i == 1 then
			
 
				+	    dopattern(pat, args, sz, params.op, needrex) -- Process pattern.
			
 
				+	    return
			
 
				+	  elseif args[i-1].opsize ~= sub(szm, i, i) then
			
 
				+	    break
			
 
				+	  end
			
 
				 	end
			
 
				       else -- Match common operand size.
			
 
				 	local szp = sz
			
@@ -1717,8 +2131,8 @@ if x64 then
 
				 	rex = a.reg > 7 and 9 or 8
			
 
				       end
			
 
				     end
			
 
				-    wputop(sz, opcode, rex)
			
 
				-    if vreg then waction("VREG", vreg); wputxb(0) end
			
 
				+    local psz, sk = wputop(sz, opcode, rex, nil, vreg)
			
 
				+    wvreg("opcode", vreg, psz, sk)
			
 
				     waction("IMM_D", format("(unsigned int)(%s)", op64))
			
 
				     waction("IMM_D", format("(unsigned int)((%s)>>32)", op64))
			
 
				   end
			
--- a/luajit.mod/luajit/dynasm/dynasm.lua
+++ b/luajit.mod/luajit/dynasm/dynasm.lua
@@ -10,9 +10,9 @@
 
				 local _info = {
			
 
				   name =	"DynASM",
			
 
				   description =	"A dynamic assembler for code generation engines",
			
 
				-  version =	"1.3.0",
			
 
				-  vernum =	 10300,
			
 
				-  release =	"2011-05-05",
			
 
				+  version =	"1.4.0",
			
 
				+  vernum =	 10400,
			
 
				+  release =	"2015-10-18",
			
 
				   author =	"Mike Pall",
			
 
				   url =		"http://luajit.org/dynasm.html",
			
 
				   license =	"MIT",
			
--- a/luajit.mod/luajit/etc/luajit.pc
+++ b/luajit.mod/luajit/etc/luajit.pc
@@ -1,8 +1,8 @@
 
				 # Package information for LuaJIT to be used by pkg-config.
			
 
				 majver=2
			
 
				-minver=0
			
 
				-relver=5
			
 
				-version=${majver}.${minver}.${relver}
			
 
				+minver=1
			
 
				+relver=0
			
 
				+version=${majver}.${minver}.${relver}-beta3
			
 
				 abiver=5.1
			
 
				 
			
 
				 prefix=/usr/local
			
--- a/luajit.mod/luajit/src/.gitignore
+++ b/luajit.mod/luajit/src/.gitignore
@@ -0,0 +1,7 @@
 
				+luajit
			
 
				+lj_bcdef.h
			
 
				+lj_ffdef.h
			
 
				+lj_libdef.h
			
 
				+lj_recdef.h
			
 
				+lj_folddef.h
			
 
				+lj_vm.[sS]
			
--- a/luajit.mod/luajit/src/Makefile
+++ b/luajit.mod/luajit/src/Makefile
@@ -11,8 +11,8 @@
 
				 ##############################################################################
			
 
				 
			
 
				 MAJVER=  2
			
 
				-MINVER=  0
			
 
				-RELVER=  5
			
 
				+MINVER=  1
			
 
				+RELVER=  0
			
 
				 ABIVER=  5.1
			
 
				 NODOTABIVER= 51
			
 
				 
			
@@ -44,17 +44,14 @@ CCOPT= -O2 -fomit-frame-pointer
 
				 #
			
 
				 # Target-specific compiler options:
			
 
				 #
			
 
				-# x86 only: it's recommended to compile at least for i686. Better yet,
			
 
				-# compile for an architecture that has SSE2, too (-msse -msse2).
			
 
				-#
			
 
				 # x86/x64 only: For GCC 4.2 or higher and if you don't intend to distribute
			
 
				 # the binaries to a different machine you could also use: -march=native
			
 
				 #
			
 
				-CCOPT_x86= -march=i686
			
 
				+CCOPT_x86= -march=i686 -msse -msse2 -mfpmath=sse
			
 
				 CCOPT_x64=
			
 
				 CCOPT_arm=
			
 
				+CCOPT_arm64=
			
 
				 CCOPT_ppc=
			
 
				-CCOPT_ppcspe=
			
 
				 CCOPT_mips=
			
 
				 #
			
 
				 CCDEBUG=
			
@@ -113,6 +110,9 @@ XCFLAGS=
 
				 #XCFLAGS+= -DLUAJIT_NUMMODE=1
			
 
				 #XCFLAGS+= -DLUAJIT_NUMMODE=2
			
 
				 #
			
 
				+# Enable GC64 mode for x64.
			
 
				+#XCFLAGS+= -DLUAJIT_ENABLE_GC64
			
 
				+#
			
 
				 ##############################################################################
			
 
				 
			
 
				 ##############################################################################
			
@@ -124,8 +124,8 @@ XCFLAGS=
 
				 #
			
 
				 # Use the system provided memory allocator (realloc) instead of the
			
 
				 # bundled memory allocator. This is slower, but sometimes helpful for
			
 
				-# debugging. This option cannot be enabled on x64, since realloc usually
			
 
				-# doesn't return addresses in the right address range.
			
 
				+# debugging. This option cannot be enabled on x64 without GC64, since
			
 
				+# realloc usually doesn't return addresses in the right address range.
			
 
				 # OTOH this option is mandatory for Valgrind's memcheck tool on x64 and
			
 
				 # the only way to get useful results from it for all other architectures.
			
 
				 #XCFLAGS+= -DLUAJIT_USE_SYSMALLOC
			
@@ -165,6 +165,10 @@ else
 
				     HOST_SYS= Windows
			
 
				     HOST_MSYS= mingw
			
 
				   endif
			
 
				+  ifneq (,$(findstring MSYS,$(HOST_SYS)))
			
 
				+    HOST_SYS= Windows
			
 
				+    HOST_MSYS= mingw
			
 
				+  endif
			
 
				   ifneq (,$(findstring CYGWIN,$(HOST_SYS)))
			
 
				     HOST_SYS= Windows
			
 
				     HOST_MSYS= cygwin
			
@@ -186,11 +190,12 @@ endif
 
				 #   make HOST_CC="gcc -m32" CROSS=i586-mingw32msvc- TARGET_SYS=Windows
			
 
				 #   make HOST_CC="gcc -m32" CROSS=powerpc-linux-gnu-
			
 
				 
			
 
				-CCOPTIONS= $(CCDEBUG) $(CCOPT) $(CCWARN) $(XCFLAGS) $(CFLAGS)
			
 
				+ASOPTIONS= $(CCOPT) $(CCWARN) $(XCFLAGS) $(CFLAGS)
			
 
				+CCOPTIONS= $(CCDEBUG) $(ASOPTIONS)
			
 
				 LDOPTIONS= $(CCDEBUG) $(LDFLAGS)
			
 
				 
			
 
				 HOST_CC= $(CC)
			
 
				-HOST_RM= rm -f
			
 
				+HOST_RM?= rm -f
			
 
				 # If left blank, minilua is built and used. You can supply an installed
			
 
				 # copy of (plain) Lua 5.1 or 5.2, plus Lua BitOp. E.g. with: HOST_LUA=lua
			
 
				 HOST_LUA=
			
@@ -208,7 +213,7 @@ TARGET_CC= $(STATIC_CC)
 
				 TARGET_STCC= $(STATIC_CC)
			
 
				 TARGET_DYNCC= $(DYNAMIC_CC)
			
 
				 TARGET_LD= $(CROSS)$(CC)
			
 
				-TARGET_AR= $(CROSS)ar rcus 2>/dev/null
			
 
				+TARGET_AR= $(CROSS)ar rcus
			
 
				 TARGET_STRIP= $(CROSS)strip
			
 
				 
			
 
				 TARGET_LIBPATH= $(or $(PREFIX),/usr/local)/$(or $(MULTILIB),lib)
			
@@ -225,6 +230,7 @@ TARGET_XLDFLAGS=
 
				 TARGET_XLIBS= -lm
			
 
				 TARGET_TCFLAGS= $(CCOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
			
 
				 TARGET_ACFLAGS= $(CCOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
			
 
				+TARGET_ASFLAGS= $(ASOPTIONS) $(TARGET_XCFLAGS) $(TARGET_FLAGS) $(TARGET_CFLAGS)
			
 
				 TARGET_ALDFLAGS= $(LDOPTIONS) $(TARGET_XLDFLAGS) $(TARGET_FLAGS) $(TARGET_LDFLAGS)
			
 
				 TARGET_ASHLDFLAGS= $(LDOPTIONS) $(TARGET_XSHLDFLAGS) $(TARGET_FLAGS) $(TARGET_SHLDFLAGS)
			
 
				 TARGET_ALIBS= $(TARGET_XLIBS) $(LIBS) $(TARGET_LIBS)
			
@@ -239,17 +245,29 @@ else
 
				 ifneq (,$(findstring LJ_TARGET_ARM ,$(TARGET_TESTARCH)))
			
 
				   TARGET_LJARCH= arm
			
 
				 else
			
 
				+ifneq (,$(findstring LJ_TARGET_ARM64 ,$(TARGET_TESTARCH)))
			
 
				+  ifneq (,$(findstring __AARCH64EB__ ,$(TARGET_TESTARCH)))
			
 
				+    TARGET_ARCH= -D__AARCH64EB__=1
			
 
				+  endif
			
 
				+  TARGET_LJARCH= arm64
			
 
				+else
			
 
				 ifneq (,$(findstring LJ_TARGET_PPC ,$(TARGET_TESTARCH)))
			
 
				+  ifneq (,$(findstring LJ_LE 1,$(TARGET_TESTARCH)))
			
 
				+    TARGET_ARCH= -DLJ_ARCH_ENDIAN=LUAJIT_LE
			
 
				+  else
			
 
				+    TARGET_ARCH= -DLJ_ARCH_ENDIAN=LUAJIT_BE
			
 
				+  endif
			
 
				   TARGET_LJARCH= ppc
			
 
				 else
			
 
				-ifneq (,$(findstring LJ_TARGET_PPCSPE ,$(TARGET_TESTARCH)))
			
 
				-  TARGET_LJARCH= ppcspe
			
 
				-else
			
 
				 ifneq (,$(findstring LJ_TARGET_MIPS ,$(TARGET_TESTARCH)))
			
 
				   ifneq (,$(findstring MIPSEL ,$(TARGET_TESTARCH)))
			
 
				     TARGET_ARCH= -D__MIPSEL__=1
			
 
				   endif
			
 
				-  TARGET_LJARCH= mips
			
 
				+  ifneq (,$(findstring LJ_TARGET_MIPS64 ,$(TARGET_TESTARCH)))
			
 
				+    TARGET_LJARCH= mips64
			
 
				+  else
			
 
				+    TARGET_LJARCH= mips
			
 
				+  endif
			
 
				 else
			
 
				   $(error Unsupported target architecture)
			
 
				 endif
			
@@ -263,6 +281,7 @@ ifneq (,$(findstring LJ_TARGET_PS3 1,$(TARGET_TESTARCH)))
 
				   TARGET_SYS= PS3
			
 
				   TARGET_ARCH+= -D__CELLOS_LV2__
			
 
				   TARGET_XCFLAGS+= -DLUAJIT_USE_SYSMALLOC
			
 
				+  TARGET_XLIBS+= -lpthread
			
 
				 endif
			
 
				 
			
 
				 TARGET_XCFLAGS+= $(CCOPT_$(TARGET_LJARCH))
			
@@ -293,6 +312,7 @@ ifeq (Windows,$(TARGET_SYS))
 
				   TARGET_XSHLDFLAGS= -shared
			
 
				   TARGET_DYNXLDOPTS=
			
 
				 else
			
 
				+  TARGET_AR+= 2>/dev/null
			
 
				 ifeq (,$(shell $(TARGET_CC) -o /dev/null -c -x c /dev/null -fno-stack-protector 2>/dev/null || echo 1))
			
 
				   TARGET_XCFLAGS+= -fno-stack-protector
			
 
				 endif
			
@@ -314,6 +334,9 @@ ifeq (iOS,$(TARGET_SYS))
 
				   TARGET_XSHLDFLAGS= -dynamiclib -single_module -undefined dynamic_lookup -fPIC
			
 
				   TARGET_DYNXLDOPTS=
			
 
				   TARGET_XSHLDFLAGS+= -install_name $(TARGET_DYLIBPATH) -compatibility_version $(MAJVER).$(MINVER) -current_version $(MAJVER).$(MINVER).$(RELVER)
			
 
				+  ifeq (arm64,$(TARGET_LJARCH))
			
 
				+    TARGET_XCFLAGS+= -fno-omit-frame-pointer
			
 
				+  endif
			
 
				 else
			
 
				   ifneq (SunOS,$(TARGET_SYS))
			
 
				     ifneq (PS3,$(TARGET_SYS))
			
@@ -374,6 +397,11 @@ DASM_XFLAGS=
 
				 DASM_AFLAGS=
			
 
				 DASM_ARCH= $(TARGET_LJARCH)
			
 
				 
			
 
				+ifneq (,$(findstring LJ_LE 1,$(TARGET_TESTARCH)))
			
 
				+  DASM_AFLAGS+= -D ENDIAN_LE
			
 
				+else
			
 
				+  DASM_AFLAGS+= -D ENDIAN_BE
			
 
				+endif
			
 
				 ifneq (,$(findstring LJ_ARCH_BITS 64,$(TARGET_TESTARCH)))
			
 
				   DASM_AFLAGS+= -D P64
			
 
				 endif
			
@@ -406,13 +434,10 @@ DASM_AFLAGS+= -D VER=$(subst LJ_ARCH_VERSION_,,$(filter LJ_ARCH_VERSION_%,$(subs
 
				 ifeq (Windows,$(TARGET_SYS))
			
 
				   DASM_AFLAGS+= -D WIN
			
 
				 endif
			
 
				-ifeq (x86,$(TARGET_LJARCH))
			
 
				-  ifneq (,$(findstring __SSE2__ 1,$(TARGET_TESTARCH)))
			
 
				-    DASM_AFLAGS+= -D SSE
			
 
				-  endif
			
 
				-else
			
 
				 ifeq (x64,$(TARGET_LJARCH))
			
 
				-  DASM_ARCH= x86
			
 
				+  ifeq (,$(findstring LJ_FR2 1,$(TARGET_TESTARCH)))
			
 
				+    DASM_ARCH= x86
			
 
				+  endif
			
 
				 else
			
 
				 ifeq (arm,$(TARGET_LJARCH))
			
 
				   ifeq (iOS,$(TARGET_SYS))
			
@@ -426,13 +451,15 @@ ifeq (ppc,$(TARGET_LJARCH))
 
				   ifneq (,$(findstring LJ_ARCH_ROUND 1,$(TARGET_TESTARCH)))
			
 
				     DASM_AFLAGS+= -D ROUND
			
 
				   endif
			
 
				-  ifneq (,$(findstring LJ_ARCH_PPC64 1,$(TARGET_TESTARCH)))
			
 
				+  ifneq (,$(findstring LJ_ARCH_PPC32ON64 1,$(TARGET_TESTARCH)))
			
 
				     DASM_AFLAGS+= -D GPR64
			
 
				   endif
			
 
				   ifeq (PS3,$(TARGET_SYS))
			
 
				     DASM_AFLAGS+= -D PPE -D TOC
			
 
				   endif
			
 
				-endif
			
 
				+  ifneq (,$(findstring LJ_ARCH_PPC64 ,$(TARGET_TESTARCH)))
			
 
				+    DASM_ARCH= ppc64
			
 
				+  endif
			
 
				 endif
			
 
				 endif
			
 
				 endif
			
@@ -448,7 +475,7 @@ BUILDVM_X= $(BUILDVM_T)
 
				 HOST_O= $(MINILUA_O) $(BUILDVM_O)
			
 
				 HOST_T= $(MINILUA_T) $(BUILDVM_T)
			
 
				 
			
 
				-LJVM_S= lj_vm.s
			
 
				+LJVM_S= lj_vm.S
			
 
				 LJVM_O= lj_vm.o
			
 
				 LJVM_BOUT= $(LJVM_S)
			
 
				 LJVM_MODE= elfasm
			
@@ -457,10 +484,11 @@ LJLIB_O= lib_base.o lib_math.o lib_bit.o lib_string.o lib_table.o \
 
				 	 lib_io.o lib_os.o lib_package.o lib_debug.o lib_jit.o lib_ffi.o
			
 
				 LJLIB_C= $(LJLIB_O:.o=.c)
			
 
				 
			
 
				-LJCORE_O= lj_gc.o lj_err.o lj_char.o lj_bc.o lj_obj.o \
			
 
				+LJCORE_O= lj_gc.o lj_err.o lj_char.o lj_bc.o lj_obj.o lj_buf.o \
			
 
				 	  lj_str.o lj_tab.o lj_func.o lj_udata.o lj_meta.o lj_debug.o \
			
 
				 	  lj_state.o lj_dispatch.o lj_vmevent.o lj_vmmath.o lj_strscan.o \
			
 
				-	  lj_api.o lj_lex.o lj_parse.o lj_bcread.o lj_bcwrite.o lj_load.o \
			
 
				+	  lj_strfmt.o lj_strfmt_num.o lj_api.o lj_profile.o \
			
 
				+	  lj_lex.o lj_parse.o lj_bcread.o lj_bcwrite.o lj_load.o \
			
 
				 	  lj_ir.o lj_opt_mem.o lj_opt_fold.o lj_opt_narrow.o \
			
 
				 	  lj_opt_dce.o lj_opt_loop.o lj_opt_split.o lj_opt_sink.o \
			
 
				 	  lj_mcode.o lj_snap.o lj_record.o lj_crecord.o lj_ffrecord.o \
			
@@ -580,6 +608,10 @@ amalg:
 
				 clean:
			
 
				 	$(HOST_RM) $(ALL_RM)
			
 
				 
			
 
				+libbc:
			
 
				+	./$(LUAJIT_T) host/genlibbc.lua -o host/buildvm_libbc.h $(LJLIB_C)
			
 
				+	$(MAKE) all
			
 
				+
			
 
				 depend:
			
 
				 	@for file in $(ALL_HDRGEN); do \
			
 
				 	  test -f $$file || touch $$file; \
			
@@ -594,7 +626,7 @@ depend:
 
				 	  test -s $$file || $(HOST_RM) $$file; \
			
 
				 	  done
			
 
				 
			
 
				-.PHONY: default all amalg clean depend
			
 
				+.PHONY: default all amalg clean libbc depend
			
 
				 
			
 
				 ##############################################################################
			
 
				 # Rules for generated files.
			
@@ -604,7 +636,7 @@ $(MINILUA_T): $(MINILUA_O)
 
				 	$(E) "HOSTLINK  $@"
			
 
				 	$(Q)$(HOST_CC) $(HOST_ALDFLAGS) -o $@ $(MINILUA_O) $(MINILUA_LIBS) $(HOST_ALIBS)
			
 
				 
			
 
				-host/buildvm_arch.h: $(DASM_DASC) $(DASM_DEP)
			
 
				+host/buildvm_arch.h: $(DASM_DASC) $(DASM_DEP) $(DASM_DIR)/*.lua
			
 
				 	$(E) "DYNASM    $@"
			
 
				 	$(Q)$(DASM) $(DASM_FLAGS) -o $@ $(DASM_DASC)
			
 
				 
			
@@ -651,10 +683,10 @@ lj_folddef.h: $(BUILDVM_T) lj_opt_fold.c
 
				 	$(Q)$(TARGET_DYNCC) $(TARGET_ACFLAGS) -c -o $(@:.o=_dyn.o) $<
			
 
				 	$(Q)$(TARGET_CC) $(TARGET_ACFLAGS) -c -o $@ $<
			
 
				 
			
 
				-%.o: %.s
			
 
				+%.o: %.S
			
 
				 	$(E) "ASM       $@"
			
 
				-	$(Q)$(TARGET_DYNCC) $(TARGET_ACFLAGS) -c -o $(@:.o=_dyn.o) $<
			
 
				-	$(Q)$(TARGET_CC) $(TARGET_ACFLAGS) -c -o $@ $<
			
 
				+	$(Q)$(TARGET_DYNCC) $(TARGET_ASFLAGS) -c -o $(@:.o=_dyn.o) $<
			
 
				+	$(Q)$(TARGET_CC) $(TARGET_ASFLAGS) -c -o $@ $<
			
 
				 
			
 
				 $(LUAJIT_O):
			
 
				 	$(E) "CC        $@"
			
--- a/luajit.mod/luajit/src/Makefile.dep
+++ b/luajit.mod/luajit/src/Makefile.dep
@@ -3,45 +3,49 @@ lib_aux.o: lib_aux.c lua.h luaconf.h lauxlib.h lj_obj.h lj_def.h \
 
				  lj_dispatch.h lj_bc.h lj_traceerr.h lj_lib.h lj_alloc.h
			
 
				 lib_base.o: lib_base.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				  lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h \
			
 
				- lj_tab.h lj_meta.h lj_state.h lj_ctype.h lj_cconv.h lj_bc.h lj_ff.h \
			
 
				- lj_ffdef.h lj_dispatch.h lj_jit.h lj_ir.h lj_char.h lj_strscan.h \
			
 
				- lj_lib.h lj_libdef.h
			
 
				+ lj_tab.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cconv.h \
			
 
				+ lj_ff.h lj_ffdef.h lj_dispatch.h lj_jit.h lj_ir.h lj_char.h lj_strscan.h \
			
 
				+ lj_strfmt.h lj_lib.h lj_libdef.h
			
 
				 lib_bit.o: lib_bit.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
			
 
				- lj_arch.h lj_err.h lj_errmsg.h lj_str.h lj_lib.h lj_libdef.h
			
 
				+ lj_arch.h lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_strscan.h \
			
 
				+ lj_strfmt.h lj_ctype.h lj_cdata.h lj_cconv.h lj_carith.h lj_ff.h \
			
 
				+ lj_ffdef.h lj_lib.h lj_libdef.h
			
 
				 lib_debug.o: lib_debug.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				  lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_lib.h \
			
 
				  lj_libdef.h
			
 
				 lib_ffi.o: lib_ffi.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
			
 
				  lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_meta.h \
			
 
				  lj_ctype.h lj_cparse.h lj_cdata.h lj_cconv.h lj_carith.h lj_ccall.h \
			
 
				- lj_ccallback.h lj_clib.h lj_ff.h lj_ffdef.h lj_lib.h lj_libdef.h
			
 
				+ lj_ccallback.h lj_clib.h lj_strfmt.h lj_ff.h lj_ffdef.h lj_lib.h \
			
 
				+ lj_libdef.h
			
 
				 lib_init.o: lib_init.c lua.h luaconf.h lauxlib.h lualib.h lj_arch.h
			
 
				 lib_io.o: lib_io.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
			
 
				- lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_state.h lj_ff.h \
			
 
				- lj_ffdef.h lj_lib.h lj_libdef.h
			
 
				-lib_jit.o: lib_jit.c lua.h luaconf.h lauxlib.h lualib.h lj_arch.h \
			
 
				- lj_obj.h lj_def.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h \
			
 
				- lj_bc.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_target.h \
			
 
				- lj_target_*.h lj_dispatch.h lj_vm.h lj_vmevent.h lj_lib.h luajit.h \
			
 
				- lj_libdef.h
			
 
				+ lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_state.h \
			
 
				+ lj_strfmt.h lj_ff.h lj_ffdef.h lj_lib.h lj_libdef.h
			
 
				+lib_jit.o: lib_jit.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
			
 
				+ lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h \
			
 
				+ lj_state.h lj_bc.h lj_ctype.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h \
			
 
				+ lj_target.h lj_target_*.h lj_trace.h lj_dispatch.h lj_traceerr.h \
			
 
				+ lj_vm.h lj_vmevent.h lj_lib.h luajit.h lj_libdef.h
			
 
				 lib_math.o: lib_math.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				  lj_def.h lj_arch.h lj_lib.h lj_vm.h lj_libdef.h
			
 
				 lib_os.o: lib_os.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h lj_def.h \
			
 
				- lj_arch.h lj_err.h lj_errmsg.h lj_lib.h lj_libdef.h
			
 
				+ lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_lib.h \
			
 
				+ lj_libdef.h
			
 
				 lib_package.o: lib_package.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				  lj_def.h lj_arch.h lj_err.h lj_errmsg.h lj_lib.h
			
 
				 lib_string.o: lib_string.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				- lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h \
			
 
				- lj_meta.h lj_state.h lj_ff.h lj_ffdef.h lj_bcdump.h lj_lex.h lj_char.h \
			
 
				- lj_lib.h lj_libdef.h
			
 
				+ lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h \
			
 
				+ lj_tab.h lj_meta.h lj_state.h lj_ff.h lj_ffdef.h lj_bcdump.h lj_lex.h \
			
 
				+ lj_char.h lj_strfmt.h lj_lib.h lj_libdef.h
			
 
				 lib_table.o: lib_table.c lua.h luaconf.h lauxlib.h lualib.h lj_obj.h \
			
 
				- lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_lib.h \
			
 
				- lj_libdef.h
			
 
				+ lj_def.h lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h \
			
 
				+ lj_tab.h lj_ff.h lj_ffdef.h lj_lib.h lj_libdef.h
			
 
				 lj_alloc.o: lj_alloc.c lj_def.h lua.h luaconf.h lj_arch.h lj_alloc.h
			
 
				 lj_api.o: lj_api.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				  lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h lj_func.h lj_udata.h \
			
 
				  lj_meta.h lj_state.h lj_bc.h lj_frame.h lj_trace.h lj_jit.h lj_ir.h \
			
 
				- lj_dispatch.h lj_traceerr.h lj_vm.h lj_strscan.h
			
 
				+ lj_dispatch.h lj_traceerr.h lj_vm.h lj_strscan.h lj_strfmt.h
			
 
				 lj_asm.o: lj_asm.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				  lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ctype.h lj_ir.h lj_jit.h \
			
 
				  lj_ircall.h lj_iropt.h lj_mcode.h lj_trace.h lj_dispatch.h lj_traceerr.h \
			
@@ -50,17 +54,20 @@ lj_asm.o: lj_asm.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
 
				 lj_bc.o: lj_bc.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_bc.h \
			
 
				  lj_bcdef.h
			
 
				 lj_bcread.o: lj_bcread.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_bc.h lj_ctype.h \
			
 
				- lj_cdata.h lualib.h lj_lex.h lj_bcdump.h lj_state.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_bc.h \
			
 
				+ lj_ctype.h lj_cdata.h lualib.h lj_lex.h lj_bcdump.h lj_state.h \
			
 
				+ lj_strfmt.h
			
 
				 lj_bcwrite.o: lj_bcwrite.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_str.h lj_bc.h lj_ctype.h lj_dispatch.h lj_jit.h lj_ir.h \
			
 
				- lj_bcdump.h lj_lex.h lj_err.h lj_errmsg.h lj_vm.h
			
 
				+ lj_gc.h lj_buf.h lj_str.h lj_bc.h lj_ctype.h lj_dispatch.h lj_jit.h \
			
 
				+ lj_ir.h lj_strfmt.h lj_bcdump.h lj_lex.h lj_err.h lj_errmsg.h lj_vm.h
			
 
				+lj_buf.o: lj_buf.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_strfmt.h
			
 
				 lj_carith.o: lj_carith.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_meta.h lj_ctype.h lj_cconv.h \
			
 
				- lj_cdata.h lj_carith.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_meta.h lj_ir.h lj_ctype.h \
			
 
				+ lj_cconv.h lj_cdata.h lj_carith.h lj_strscan.h
			
 
				 lj_ccall.o: lj_ccall.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_ctype.h lj_cconv.h \
			
 
				- lj_cdata.h lj_ccall.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_bc.h \
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_ctype.h lj_cconv.h lj_cdata.h \
			
 
				+ lj_ccall.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_bc.h \
			
 
				  lj_traceerr.h
			
 
				 lj_ccallback.o: lj_ccallback.c lj_obj.h lua.h luaconf.h lj_def.h \
			
 
				  lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_state.h lj_frame.h \
			
@@ -71,107 +78,118 @@ lj_cconv.o: lj_cconv.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
 
				  lj_err.h lj_errmsg.h lj_tab.h lj_ctype.h lj_gc.h lj_cdata.h lj_cconv.h \
			
 
				  lj_ccallback.h
			
 
				 lj_cdata.o: lj_cdata.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_ctype.h lj_cconv.h \
			
 
				- lj_cdata.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_tab.h lj_ctype.h lj_cconv.h lj_cdata.h
			
 
				 lj_char.o: lj_char.c lj_char.h lj_def.h lua.h luaconf.h
			
 
				 lj_clib.o: lj_clib.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				  lj_err.h lj_errmsg.h lj_tab.h lj_str.h lj_udata.h lj_ctype.h lj_cconv.h \
			
 
				- lj_cdata.h lj_clib.h
			
 
				+ lj_cdata.h lj_clib.h lj_strfmt.h
			
 
				 lj_cparse.o: lj_cparse.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_ctype.h lj_cparse.h lj_frame.h \
			
 
				- lj_bc.h lj_vm.h lj_char.h lj_strscan.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_ctype.h lj_cparse.h \
			
 
				+ lj_frame.h lj_bc.h lj_vm.h lj_char.h lj_strscan.h lj_strfmt.h
			
 
				 lj_crecord.o: lj_crecord.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ctype.h \
			
 
				- lj_gc.h lj_cdata.h lj_cparse.h lj_cconv.h lj_clib.h lj_ccall.h lj_ff.h \
			
 
				- lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
			
 
				+ lj_err.h lj_errmsg.h lj_tab.h lj_frame.h lj_bc.h lj_ctype.h lj_gc.h \
			
 
				+ lj_cdata.h lj_cparse.h lj_cconv.h lj_carith.h lj_clib.h lj_ccall.h \
			
 
				+ lj_ff.h lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
			
 
				  lj_dispatch.h lj_traceerr.h lj_record.h lj_ffrecord.h lj_snap.h \
			
 
				- lj_crecord.h
			
 
				+ lj_crecord.h lj_strfmt.h
			
 
				 lj_ctype.o: lj_ctype.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_ctype.h lj_ccallback.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_strfmt.h lj_ctype.h \
			
 
				+ lj_ccallback.h lj_buf.h
			
 
				 lj_debug.o: lj_debug.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h lj_state.h lj_frame.h \
			
 
				- lj_bc.h lj_vm.h lj_jit.h lj_ir.h
			
 
				+ lj_err.h lj_errmsg.h lj_debug.h lj_buf.h lj_gc.h lj_str.h lj_tab.h \
			
 
				+ lj_state.h lj_frame.h lj_bc.h lj_strfmt.h lj_jit.h lj_ir.h
			
 
				 lj_dispatch.o: lj_dispatch.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_err.h lj_errmsg.h lj_func.h lj_str.h lj_tab.h lj_meta.h lj_debug.h \
			
 
				- lj_state.h lj_frame.h lj_bc.h lj_ff.h lj_ffdef.h lj_jit.h lj_ir.h \
			
 
				- lj_ccallback.h lj_ctype.h lj_gc.h lj_trace.h lj_dispatch.h lj_traceerr.h \
			
 
				- lj_vm.h luajit.h
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_func.h lj_tab.h \
			
 
				+ lj_meta.h lj_debug.h lj_state.h lj_frame.h lj_bc.h lj_ff.h lj_ffdef.h \
			
 
				+ lj_strfmt.h lj_jit.h lj_ir.h lj_ccallback.h lj_ctype.h lj_trace.h \
			
 
				+ lj_dispatch.h lj_traceerr.h lj_profile.h lj_vm.h luajit.h
			
 
				 lj_err.o: lj_err.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_err.h \
			
 
				  lj_errmsg.h lj_debug.h lj_str.h lj_func.h lj_state.h lj_frame.h lj_bc.h \
			
 
				  lj_ff.h lj_ffdef.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h \
			
 
				- lj_traceerr.h lj_vm.h
			
 
				+ lj_traceerr.h lj_vm.h lj_strfmt.h
			
 
				 lj_ffrecord.o: lj_ffrecord.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_frame.h lj_bc.h lj_ff.h \
			
 
				  lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
			
 
				  lj_dispatch.h lj_traceerr.h lj_record.h lj_ffrecord.h lj_crecord.h \
			
 
				- lj_vm.h lj_strscan.h lj_recdef.h
			
 
				+ lj_vm.h lj_strscan.h lj_strfmt.h lj_recdef.h
			
 
				 lj_func.o: lj_func.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				  lj_func.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_bc.h \
			
 
				  lj_traceerr.h lj_vm.h
			
 
				 lj_gc.o: lj_gc.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_func.h lj_udata.h lj_meta.h \
			
 
				- lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cdata.h lj_trace.h lj_jit.h \
			
 
				- lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_func.h lj_udata.h \
			
 
				+ lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cdata.h lj_trace.h \
			
 
				+ lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h
			
 
				 lj_gdbjit.o: lj_gdbjit.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_frame.h lj_bc.h lj_jit.h \
			
 
				- lj_ir.h lj_dispatch.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_frame.h lj_bc.h lj_buf.h \
			
 
				+ lj_str.h lj_strfmt.h lj_jit.h lj_ir.h lj_dispatch.h
			
 
				 lj_ir.o: lj_ir.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				- lj_str.h lj_tab.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h lj_trace.h \
			
 
				- lj_dispatch.h lj_bc.h lj_traceerr.h lj_ctype.h lj_cdata.h lj_carith.h \
			
 
				- lj_vm.h lj_strscan.h lj_lib.h
			
 
				+ lj_buf.h lj_str.h lj_tab.h lj_ir.h lj_jit.h lj_ircall.h lj_iropt.h \
			
 
				+ lj_trace.h lj_dispatch.h lj_bc.h lj_traceerr.h lj_ctype.h lj_cdata.h \
			
 
				+ lj_carith.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lib.h
			
 
				 lj_lex.o: lj_lex.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_ctype.h lj_cdata.h lualib.h \
			
 
				- lj_state.h lj_lex.h lj_parse.h lj_char.h lj_strscan.h
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_ctype.h lj_cdata.h \
			
 
				+ lualib.h lj_state.h lj_lex.h lj_parse.h lj_char.h lj_strscan.h \
			
 
				+ lj_strfmt.h
			
 
				 lj_lib.o: lj_lib.c lauxlib.h lua.h luaconf.h lj_obj.h lj_def.h lj_arch.h \
			
 
				  lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_func.h lj_bc.h \
			
 
				- lj_dispatch.h lj_jit.h lj_ir.h lj_vm.h lj_strscan.h lj_lib.h
			
 
				+ lj_dispatch.h lj_jit.h lj_ir.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lex.h \
			
 
				+ lj_bcdump.h lj_lib.h
			
 
				 lj_load.o: lj_load.c lua.h luaconf.h lauxlib.h lj_obj.h lj_def.h \
			
 
				- lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_func.h lj_frame.h \
			
 
				- lj_bc.h lj_vm.h lj_lex.h lj_bcdump.h lj_parse.h
			
 
				+ lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_func.h \
			
 
				+ lj_frame.h lj_bc.h lj_vm.h lj_lex.h lj_bcdump.h lj_parse.h
			
 
				 lj_mcode.o: lj_mcode.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_gc.h lj_err.h lj_errmsg.h lj_jit.h lj_ir.h lj_mcode.h lj_trace.h \
			
 
				  lj_dispatch.h lj_bc.h lj_traceerr.h lj_vm.h
			
 
				 lj_meta.o: lj_meta.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_meta.h lj_frame.h lj_bc.h \
			
 
				- lj_vm.h lj_strscan.h
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_meta.h lj_frame.h \
			
 
				+ lj_bc.h lj_vm.h lj_strscan.h lj_strfmt.h lj_lib.h
			
 
				 lj_obj.o: lj_obj.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h
			
 
				 lj_opt_dce.o: lj_opt_dce.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_ir.h lj_jit.h lj_iropt.h
			
 
				 lj_opt_fold.o: lj_opt_fold.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_str.h lj_tab.h lj_ir.h lj_jit.h lj_iropt.h lj_trace.h lj_dispatch.h \
			
 
				- lj_bc.h lj_traceerr.h lj_ctype.h lj_gc.h lj_carith.h lj_vm.h \
			
 
				- lj_strscan.h lj_folddef.h
			
 
				+ lj_buf.h lj_gc.h lj_str.h lj_tab.h lj_ir.h lj_jit.h lj_ircall.h \
			
 
				+ lj_iropt.h lj_trace.h lj_dispatch.h lj_bc.h lj_traceerr.h lj_ctype.h \
			
 
				+ lj_carith.h lj_vm.h lj_strscan.h lj_strfmt.h lj_folddef.h
			
 
				 lj_opt_loop.o: lj_opt_loop.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_ir.h lj_jit.h lj_iropt.h lj_trace.h \
			
 
				- lj_dispatch.h lj_bc.h lj_traceerr.h lj_snap.h lj_vm.h
			
 
				+ lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_ir.h lj_jit.h \
			
 
				+ lj_iropt.h lj_trace.h lj_dispatch.h lj_bc.h lj_traceerr.h lj_snap.h \
			
 
				+ lj_vm.h
			
 
				 lj_opt_mem.o: lj_opt_mem.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_tab.h lj_ir.h lj_jit.h lj_iropt.h
			
 
				+ lj_tab.h lj_ir.h lj_jit.h lj_iropt.h lj_ircall.h
			
 
				 lj_opt_narrow.o: lj_opt_narrow.c lj_obj.h lua.h luaconf.h lj_def.h \
			
 
				  lj_arch.h lj_bc.h lj_ir.h lj_jit.h lj_iropt.h lj_trace.h lj_dispatch.h \
			
 
				  lj_traceerr.h lj_vm.h lj_strscan.h
			
 
				 lj_opt_sink.o: lj_opt_sink.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_ir.h lj_jit.h lj_iropt.h lj_target.h lj_target_*.h
			
 
				 lj_opt_split.o: lj_opt_split.c lj_obj.h lua.h luaconf.h lj_def.h \
			
 
				- lj_arch.h lj_err.h lj_errmsg.h lj_str.h lj_ir.h lj_jit.h lj_ircall.h \
			
 
				- lj_iropt.h lj_vm.h
			
 
				+ lj_arch.h lj_err.h lj_errmsg.h lj_buf.h lj_gc.h lj_str.h lj_ir.h \
			
 
				+ lj_jit.h lj_ircall.h lj_iropt.h lj_dispatch.h lj_bc.h lj_vm.h
			
 
				 lj_parse.o: lj_parse.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_str.h lj_tab.h lj_func.h \
			
 
				- lj_state.h lj_bc.h lj_ctype.h lj_lex.h lj_parse.h lj_vm.h lj_vmevent.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_debug.h lj_buf.h lj_str.h lj_tab.h \
			
 
				+ lj_func.h lj_state.h lj_bc.h lj_ctype.h lj_strfmt.h lj_lex.h lj_parse.h \
			
 
				+ lj_vm.h lj_vmevent.h
			
 
				+lj_profile.o: lj_profile.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				+ lj_buf.h lj_gc.h lj_str.h lj_frame.h lj_bc.h lj_debug.h lj_dispatch.h \
			
 
				+ lj_jit.h lj_ir.h lj_trace.h lj_traceerr.h lj_profile.h luajit.h
			
 
				 lj_record.o: lj_record.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_meta.h lj_frame.h lj_bc.h \
			
 
				- lj_ctype.h lj_gc.h lj_ff.h lj_ffdef.h lj_ir.h lj_jit.h lj_ircall.h \
			
 
				- lj_iropt.h lj_trace.h lj_dispatch.h lj_traceerr.h lj_record.h \
			
 
				- lj_ffrecord.h lj_snap.h lj_vm.h
			
 
				+ lj_ctype.h lj_gc.h lj_ff.h lj_ffdef.h lj_debug.h lj_ir.h lj_jit.h \
			
 
				+ lj_ircall.h lj_iropt.h lj_trace.h lj_dispatch.h lj_traceerr.h \
			
 
				+ lj_record.h lj_ffrecord.h lj_snap.h lj_vm.h
			
 
				 lj_snap.o: lj_snap.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				  lj_tab.h lj_state.h lj_frame.h lj_bc.h lj_ir.h lj_jit.h lj_iropt.h \
			
 
				  lj_trace.h lj_dispatch.h lj_traceerr.h lj_snap.h lj_target.h \
			
 
				  lj_target_*.h lj_ctype.h lj_cdata.h
			
 
				 lj_state.o: lj_state.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				- lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_func.h lj_meta.h \
			
 
				- lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_trace.h lj_jit.h lj_ir.h \
			
 
				- lj_dispatch.h lj_traceerr.h lj_vm.h lj_lex.h lj_alloc.h
			
 
				+ lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h lj_func.h \
			
 
				+ lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_trace.h lj_jit.h \
			
 
				+ lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h lj_lex.h lj_alloc.h luajit.h
			
 
				 lj_str.o: lj_str.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
 
				- lj_err.h lj_errmsg.h lj_str.h lj_state.h lj_char.h
			
 
				+ lj_err.h lj_errmsg.h lj_str.h lj_char.h
			
 
				+lj_strfmt.o: lj_strfmt.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				+ lj_buf.h lj_gc.h lj_str.h lj_state.h lj_char.h lj_strfmt.h
			
 
				+lj_strfmt_num.o: lj_strfmt_num.c lj_obj.h lua.h luaconf.h lj_def.h \
			
 
				+ lj_arch.h lj_buf.h lj_gc.h lj_str.h lj_strfmt.h
			
 
				 lj_strscan.o: lj_strscan.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_char.h lj_strscan.h
			
 
				 lj_tab.o: lj_tab.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h lj_gc.h \
			
@@ -189,26 +207,27 @@ lj_vmevent.o: lj_vmevent.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
 
				 lj_vmmath.o: lj_vmmath.c lj_obj.h lua.h luaconf.h lj_def.h lj_arch.h \
			
 
				  lj_ir.h lj_vm.h
			
 
				 ljamalg.o: ljamalg.c lua.h luaconf.h lauxlib.h lj_gc.c lj_obj.h lj_def.h \
			
 
				- lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_str.h lj_tab.h lj_func.h \
			
 
				- lj_udata.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h lj_cdata.h \
			
 
				- lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h lj_vm.h lj_err.c \
			
 
				- lj_debug.h lj_ff.h lj_ffdef.h lj_char.c lj_char.h lj_bc.c lj_bcdef.h \
			
 
				- lj_obj.c lj_str.c lj_tab.c lj_func.c lj_udata.c lj_meta.c lj_strscan.h \
			
 
				- lj_debug.c lj_state.c lj_lex.h lj_alloc.h lj_dispatch.c lj_ccallback.h \
			
 
				- luajit.h lj_vmevent.c lj_vmevent.h lj_vmmath.c lj_strscan.c lj_api.c \
			
 
				- lj_lex.c lualib.h lj_parse.h lj_parse.c lj_bcread.c lj_bcdump.h \
			
 
				- lj_bcwrite.c lj_load.c lj_ctype.c lj_cdata.c lj_cconv.h lj_cconv.c \
			
 
				- lj_ccall.c lj_ccall.h lj_ccallback.c lj_target.h lj_target_*.h \
			
 
				- lj_mcode.h lj_carith.c lj_carith.h lj_clib.c lj_clib.h lj_cparse.c \
			
 
				- lj_cparse.h lj_lib.c lj_lib.h lj_ir.c lj_ircall.h lj_iropt.h \
			
 
				- lj_opt_mem.c lj_opt_fold.c lj_folddef.h lj_opt_narrow.c lj_opt_dce.c \
			
 
				- lj_opt_loop.c lj_snap.h lj_opt_split.c lj_opt_sink.c lj_mcode.c \
			
 
				- lj_snap.c lj_record.c lj_record.h lj_ffrecord.h lj_crecord.c \
			
 
				- lj_crecord.h lj_ffrecord.c lj_recdef.h lj_asm.c lj_asm.h lj_emit_*.h \
			
 
				- lj_asm_*.h lj_trace.c lj_gdbjit.h lj_gdbjit.c lj_alloc.c lib_aux.c \
			
 
				- lib_base.c lj_libdef.h lib_math.c lib_string.c lib_table.c lib_io.c \
			
 
				- lib_os.c lib_package.c lib_debug.c lib_bit.c lib_jit.c lib_ffi.c \
			
 
				- lib_init.c
			
 
				+ lj_arch.h lj_gc.h lj_err.h lj_errmsg.h lj_buf.h lj_str.h lj_tab.h \
			
 
				+ lj_func.h lj_udata.h lj_meta.h lj_state.h lj_frame.h lj_bc.h lj_ctype.h \
			
 
				+ lj_cdata.h lj_trace.h lj_jit.h lj_ir.h lj_dispatch.h lj_traceerr.h \
			
 
				+ lj_vm.h lj_err.c lj_debug.h lj_ff.h lj_ffdef.h lj_strfmt.h lj_char.c \
			
 
				+ lj_char.h lj_bc.c lj_bcdef.h lj_obj.c lj_buf.c lj_str.c lj_tab.c \
			
 
				+ lj_func.c lj_udata.c lj_meta.c lj_strscan.h lj_lib.h lj_debug.c \
			
 
				+ lj_state.c lj_lex.h lj_alloc.h luajit.h lj_dispatch.c lj_ccallback.h \
			
 
				+ lj_profile.h lj_vmevent.c lj_vmevent.h lj_vmmath.c lj_strscan.c \
			
 
				+ lj_strfmt.c lj_strfmt_num.c lj_api.c lj_profile.c lj_lex.c lualib.h \
			
 
				+ lj_parse.h lj_parse.c lj_bcread.c lj_bcdump.h lj_bcwrite.c lj_load.c \
			
 
				+ lj_ctype.c lj_cdata.c lj_cconv.h lj_cconv.c lj_ccall.c lj_ccall.h \
			
 
				+ lj_ccallback.c lj_target.h lj_target_*.h lj_mcode.h lj_carith.c \
			
 
				+ lj_carith.h lj_clib.c lj_clib.h lj_cparse.c lj_cparse.h lj_lib.c lj_ir.c \
			
 
				+ lj_ircall.h lj_iropt.h lj_opt_mem.c lj_opt_fold.c lj_folddef.h \
			
 
				+ lj_opt_narrow.c lj_opt_dce.c lj_opt_loop.c lj_snap.h lj_opt_split.c \
			
 
				+ lj_opt_sink.c lj_mcode.c lj_snap.c lj_record.c lj_record.h lj_ffrecord.h \
			
 
				+ lj_crecord.c lj_crecord.h lj_ffrecord.c lj_recdef.h lj_asm.c lj_asm.h \
			
 
				+ lj_emit_*.h lj_asm_*.h lj_trace.c lj_gdbjit.h lj_gdbjit.c lj_alloc.c \
			
 
				+ lib_aux.c lib_base.c lj_libdef.h lib_math.c lib_string.c lib_table.c \
			
 
				+ lib_io.c lib_os.c lib_package.c lib_debug.c lib_bit.c lib_jit.c \
			
 
				+ lib_ffi.c lib_init.c
			
 
				 luajit.o: luajit.c lua.h luaconf.h lauxlib.h lualib.h luajit.h lj_arch.h
			
 
				 host/buildvm.o: host/buildvm.c host/buildvm.h lj_def.h lua.h luaconf.h \
			
 
				  lj_arch.h lj_obj.h lj_def.h lj_arch.h lj_gc.h lj_obj.h lj_bc.h lj_ir.h \
			
@@ -220,7 +239,8 @@ host/buildvm_asm.o: host/buildvm_asm.c host/buildvm.h lj_def.h lua.h luaconf.h \
 
				 host/buildvm_fold.o: host/buildvm_fold.c host/buildvm.h lj_def.h lua.h \
			
 
				  luaconf.h lj_arch.h lj_obj.h lj_def.h lj_arch.h lj_ir.h lj_obj.h
			
 
				 host/buildvm_lib.o: host/buildvm_lib.c host/buildvm.h lj_def.h lua.h luaconf.h \
			
 
				- lj_arch.h lj_obj.h lj_def.h lj_arch.h lj_lib.h lj_obj.h
			
 
				+ lj_arch.h lj_obj.h lj_def.h lj_arch.h lj_bc.h lj_lib.h lj_obj.h \
			
 
				+ host/buildvm_libbc.h
			
 
				 host/buildvm_peobj.o: host/buildvm_peobj.c host/buildvm.h lj_def.h lua.h \
			
 
				  luaconf.h lj_arch.h lj_bc.h lj_def.h lj_arch.h
			
 
				 host/minilua.o: host/minilua.c
			
--- a/luajit.mod/luajit/src/host/.gitignore
+++ b/luajit.mod/luajit/src/host/.gitignore
@@ -0,0 +1,3 @@
 
				+minilua
			
 
				+buildvm
			
 
				+buildvm_arch.h
			
--- a/luajit.mod/luajit/src/host/buildvm.c
+++ b/luajit.mod/luajit/src/host/buildvm.c
@@ -59,10 +59,10 @@ static int collect_reloc(BuildCtx *ctx, uint8_t *addr, int idx, int type);
 
				 #include "../dynasm/dasm_x86.h"
			
 
				 #elif LJ_TARGET_ARM
			
 
				 #include "../dynasm/dasm_arm.h"
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+#include "../dynasm/dasm_arm64.h"
			
 
				 #elif LJ_TARGET_PPC
			
 
				 #include "../dynasm/dasm_ppc.h"
			
 
				-#elif LJ_TARGET_PPCSPE
			
 
				-#include "../dynasm/dasm_ppc.h"
			
 
				 #elif LJ_TARGET_MIPS
			
 
				 #include "../dynasm/dasm_mips.h"
			
 
				 #else
			
@@ -110,11 +110,11 @@ static const char *sym_decorate(BuildCtx *ctx,
 
				   if (p) {
			
 
				 #if LJ_TARGET_X86ORX64
			
 
				     if (!LJ_64 && (ctx->mode == BUILD_coffasm || ctx->mode == BUILD_peobj))
			
 
				-      name[0] = '@';
			
 
				+      name[0] = name[1] == 'R' ? '_' : '@';  /* Just for _RtlUnwind@16. */
			
 
				     else
			
 
				       *p = '\0';
			
 
				-#elif (LJ_TARGET_PPC  || LJ_TARGET_PPCSPE) && !LJ_TARGET_CONSOLE
			
 
				-    /* Keep @plt. */
			
 
				+#elif LJ_TARGET_PPC && !LJ_TARGET_CONSOLE
			
 
				+    /* Keep @plt etc. */
			
 
				 #else
			
 
				     *p = '\0';
			
 
				 #endif
			
@@ -179,6 +179,7 @@ static int build_code(BuildCtx *ctx)
 
				   ctx->nreloc = 0;
			
 
				 
			
 
				   ctx->globnames = globnames;
			
 
				+  ctx->extnames = extnames;
			
 
				   ctx->relocsym = (const char **)malloc(NRELOCSYM*sizeof(const char *));
			
 
				   ctx->nrelocsym = 0;
			
 
				   for (i = 0; i < (int)NRELOCSYM; i++) relocmap[i] = -1;
			
@@ -320,20 +321,20 @@ static void emit_vmdef(BuildCtx *ctx)
 
				   char buf[80];
			
 
				   int i;
			
 
				   fprintf(ctx->fp, "-- This is a generated file. DO NOT EDIT!\n\n");
			
 
				-  fprintf(ctx->fp, "module(...)\n\n");
			
 
				+  fprintf(ctx->fp, "return {\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "bcnames = \"");
			
 
				   for (i = 0; bc_names[i]; i++) fprintf(ctx->fp, "%-6s", bc_names[i]);
			
 
				-  fprintf(ctx->fp, "\"\n\n");
			
 
				+  fprintf(ctx->fp, "\",\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "irnames = \"");
			
 
				   for (i = 0; ir_names[i]; i++) fprintf(ctx->fp, "%-6s", ir_names[i]);
			
 
				-  fprintf(ctx->fp, "\"\n\n");
			
 
				+  fprintf(ctx->fp, "\",\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "irfpm = { [0]=");
			
 
				   for (i = 0; irfpm_names[i]; i++)
			
 
				     fprintf(ctx->fp, "\"%s\", ", lower(buf, irfpm_names[i]));
			
 
				-  fprintf(ctx->fp, "}\n\n");
			
 
				+  fprintf(ctx->fp, "},\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "irfield = { [0]=");
			
 
				   for (i = 0; irfield_names[i]; i++) {
			
@@ -343,17 +344,17 @@ static void emit_vmdef(BuildCtx *ctx)
 
				     if (p) *p = '.';
			
 
				     fprintf(ctx->fp, "\"%s\", ", buf);
			
 
				   }
			
 
				-  fprintf(ctx->fp, "}\n\n");
			
 
				+  fprintf(ctx->fp, "},\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "ircall = {\n[0]=");
			
 
				   for (i = 0; ircall_names[i]; i++)
			
 
				     fprintf(ctx->fp, "\"%s\",\n", ircall_names[i]);
			
 
				-  fprintf(ctx->fp, "}\n\n");
			
 
				+  fprintf(ctx->fp, "},\n\n");
			
 
				 
			
 
				   fprintf(ctx->fp, "traceerr = {\n[0]=");
			
 
				   for (i = 0; trace_errors[i]; i++)
			
 
				     fprintf(ctx->fp, "\"%s\",\n", trace_errors[i]);
			
 
				-  fprintf(ctx->fp, "}\n\n");
			
 
				+  fprintf(ctx->fp, "},\n\n");
			
 
				 }
			
 
				 
			
 
				 /* -- Argument parsing ---------------------------------------------------- */
			
@@ -490,6 +491,7 @@ int main(int argc, char **argv)
 
				   case BUILD_vmdef:
			
 
				     emit_vmdef(ctx);
			
 
				     emit_lib(ctx);
			
 
				+    fprintf(ctx->fp, "}\n\n");
			
 
				     break;
			
 
				   case BUILD_ffdef:
			
 
				   case BUILD_libdef:
			
--- a/luajit.mod/luajit/src/host/buildvm.h
+++ b/luajit.mod/luajit/src/host/buildvm.h
@@ -82,6 +82,7 @@ typedef struct BuildCtx {
 
				   const char *beginsym;
			
 
				   /* Strings generated by DynASM. */
			
 
				   const char *const *globnames;
			
 
				+  const char *const *extnames;
			
 
				   const char *dasm_ident;
			
 
				   const char *dasm_arch;
			
 
				   /* Relocations. */
			
--- a/luajit.mod/luajit/src/host/buildvm_asm.c
+++ b/luajit.mod/luajit/src/host/buildvm_asm.c
@@ -51,8 +51,8 @@ static const char *const jccnames[] = {
 
				   "js", "jns", "jpe", "jpo", "jl", "jge", "jle", "jg"
			
 
				 };
			
 
				 
			
 
				-/* Emit relocation for the incredibly stupid OSX assembler. */
			
 
				-static void emit_asm_reloc_mach(BuildCtx *ctx, uint8_t *cp, int n,
			
 
				+/* Emit x86/x64 text relocations. */
			
 
				+static void emit_asm_reloc_text(BuildCtx *ctx, uint8_t *cp, int n,
			
 
				 				const char *sym)
			
 
				 {
			
 
				   const char *opname = NULL;
			
@@ -71,6 +71,20 @@ err:
 
				     exit(1);
			
 
				   }
			
 
				   emit_asm_bytes(ctx, cp, n);
			
 
				+  if (strncmp(sym+(*sym == '_'), LABEL_PREFIX, sizeof(LABEL_PREFIX)-1)) {
			
 
				+    /* Various fixups for external symbols outside of our binary. */
			
 
				+    if (ctx->mode == BUILD_elfasm) {
			
 
				+      if (LJ_32)
			
 
				+	fprintf(ctx->fp, "#if __PIC__\n\t%s lj_wrap_%s\n#else\n", opname, sym);
			
 
				+      fprintf(ctx->fp, "\t%s %s@PLT\n", opname, sym);
			
 
				+      if (LJ_32)
			
 
				+	fprintf(ctx->fp, "#endif\n");
			
 
				+      return;
			
 
				+    } else if (LJ_32 && ctx->mode == BUILD_machasm) {
			
 
				+      fprintf(ctx->fp, "\t%s L%s$stub\n", opname, sym);
			
 
				+      return;
			
 
				+    }
			
 
				+  }
			
 
				   fprintf(ctx->fp, "\t%s %s\n", opname, sym);
			
 
				 }
			
 
				 #else
			
@@ -79,10 +93,14 @@ static void emit_asm_words(BuildCtx *ctx, uint8_t *p, int n)
 
				 {
			
 
				   int i;
			
 
				   for (i = 0; i < n; i += 4) {
			
 
				+    uint32_t ins = *(uint32_t *)(p+i);
			
 
				+#if LJ_TARGET_ARM64 && LJ_BE
			
 
				+    ins = lj_bswap(ins);  /* ARM64 instructions are always little-endian. */
			
 
				+#endif
			
 
				     if ((i & 15) == 0)
			
 
				-      fprintf(ctx->fp, "\t.long 0x%08x", *(uint32_t *)(p+i));
			
 
				+      fprintf(ctx->fp, "\t.long 0x%08x", ins);
			
 
				     else
			
 
				-      fprintf(ctx->fp, ",0x%08x", *(uint32_t *)(p+i));
			
 
				+      fprintf(ctx->fp, ",0x%08x", ins);
			
 
				     if ((i & 15) == 12) putc('\n', ctx->fp);
			
 
				   }
			
 
				   if ((n & 15) != 0) putc('\n', ctx->fp);
			
@@ -107,7 +125,16 @@ static void emit_asm_wordreloc(BuildCtx *ctx, uint8_t *p, int n,
 
				 	    ins, sym);
			
 
				     exit(1);
			
 
				   }
			
 
				-#elif LJ_TARGET_PPC || LJ_TARGET_PPCSPE
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+  if ((ins >> 26) == 0x25u) {
			
 
				+    fprintf(ctx->fp, "\tbl %s\n", sym);
			
 
				+  } else {
			
 
				+    fprintf(stderr,
			
 
				+	    "Error: unsupported opcode %08x for %s symbol relocation.\n",
			
 
				+	    ins, sym);
			
 
				+    exit(1);
			
 
				+  }
			
 
				+#elif LJ_TARGET_PPC
			
 
				 #if LJ_TARGET_PS3
			
 
				 #define TOCPREFIX "."
			
 
				 #else
			
@@ -117,6 +144,14 @@ static void emit_asm_wordreloc(BuildCtx *ctx, uint8_t *p, int n,
 
				     fprintf(ctx->fp, "\t%s %d, %d, " TOCPREFIX "%s\n",
			
 
				 	    (ins & 1) ? "bcl" : "bc", (ins >> 21) & 31, (ins >> 16) & 31, sym);
			
 
				   } else if ((ins >> 26) == 18) {
			
 
				+#if LJ_ARCH_PPC64
			
 
				+    const char *suffix = strchr(sym, '@');
			
 
				+    if (suffix && suffix[1] == 'h') {
			
 
				+      fprintf(ctx->fp, "\taddis 11, 2, %s\n", sym);
			
 
				+    } else if (suffix && suffix[1] == 'l') {
			
 
				+      fprintf(ctx->fp, "\tld 12, %s\n", sym);
			
 
				+    } else
			
 
				+#endif
			
 
				     fprintf(ctx->fp, "\t%s " TOCPREFIX "%s\n", (ins & 1) ? "bl" : "b", sym);
			
 
				   } else {
			
 
				     fprintf(stderr,
			
@@ -215,6 +250,9 @@ void emit_asm(BuildCtx *ctx)
 
				   int i, rel;
			
 
				 
			
 
				   fprintf(ctx->fp, "\t.file \"buildvm_%s.dasc\"\n", ctx->dasm_arch);
			
 
				+#if LJ_ARCH_PPC64
			
 
				+  fprintf(ctx->fp, "\t.abiversion 2\n");
			
 
				+#endif
			
 
				   fprintf(ctx->fp, "\t.text\n");
			
 
				   emit_asm_align(ctx, 4);
			
 
				 
			
@@ -228,11 +266,20 @@ void emit_asm(BuildCtx *ctx)
 
				 
			
 
				 #if LJ_TARGET_ARM && defined(__GNUC__) && !LJ_NO_UNWIND
			
 
				   /* This should really be moved into buildvm_arm.dasc. */
			
 
				+#if LJ_ARCH_HASFPU
			
 
				+  fprintf(ctx->fp,
			
 
				+	  ".fnstart\n"
			
 
				+	  ".save {r5, r6, r7, r8, r9, r10, r11, lr}\n"
			
 
				+	  ".vsave {d8-d15}\n"
			
 
				+	  ".save {r4}\n"
			
 
				+	  ".pad #28\n");
			
 
				+#else
			
 
				   fprintf(ctx->fp,
			
 
				 	  ".fnstart\n"
			
 
				 	  ".save {r4, r5, r6, r7, r8, r9, r10, r11, lr}\n"
			
 
				 	  ".pad #28\n");
			
 
				 #endif
			
 
				+#endif
			
 
				 #if LJ_TARGET_MIPS
			
 
				   fprintf(ctx->fp, ".set nomips16\n.abicalls\n.set noreorder\n.set nomacro\n");
			
 
				 #endif
			
@@ -255,8 +302,9 @@ void emit_asm(BuildCtx *ctx)
 
				       BuildReloc *r = &ctx->reloc[rel];
			
 
				       int n = r->ofs - ofs;
			
 
				 #if LJ_TARGET_X86ORX64
			
 
				-      if (ctx->mode == BUILD_machasm && r->type != 0) {
			
 
				-	emit_asm_reloc_mach(ctx, ctx->code+ofs, n, ctx->relocsym[r->sym]);
			
 
				+      if (r->type != 0 &&
			
 
				+	  (ctx->mode == BUILD_elfasm || ctx->mode == BUILD_machasm)) {
			
 
				+	emit_asm_reloc_text(ctx, ctx->code+ofs, n, ctx->relocsym[r->sym]);
			
 
				       } else {
			
 
				 	emit_asm_bytes(ctx, ctx->code+ofs, n);
			
 
				 	emit_asm_reloc(ctx, r->type, ctx->relocsym[r->sym]);
			
@@ -290,10 +338,7 @@ void emit_asm(BuildCtx *ctx)
 
				 #if !(LJ_TARGET_PS3 || LJ_TARGET_PSVITA)
			
 
				     fprintf(ctx->fp, "\t.section .note.GNU-stack,\"\"," ELFASM_PX "progbits\n");
			
 
				 #endif
			
 
				-#if LJ_TARGET_PPCSPE
			
 
				-    /* Soft-float ABI + SPE. */
			
 
				-    fprintf(ctx->fp, "\t.gnu_attribute 4, 2\n\t.gnu_attribute 8, 3\n");
			
 
				-#elif LJ_TARGET_PPC && !LJ_TARGET_PS3
			
 
				+#if LJ_TARGET_PPC && !LJ_TARGET_PS3 && !LJ_ABI_SOFTFP
			
 
				     /* Hard-float ABI. */
			
 
				     fprintf(ctx->fp, "\t.gnu_attribute 4, 1\n");
			
 
				 #endif
			
--- a/luajit.mod/luajit/src/host/buildvm_lib.c
+++ b/luajit.mod/luajit/src/host/buildvm_lib.c
@@ -5,7 +5,9 @@
 
				 
			
 
				 #include "buildvm.h"
			
 
				 #include "lj_obj.h"
			
 
				+#include "lj_bc.h"
			
 
				 #include "lj_lib.h"
			
 
				+#include "buildvm_libbc.h"
			
 
				 
			
 
				 /* Context for library definitions. */
			
 
				 static uint8_t obuf[8192];
			
@@ -151,6 +153,62 @@ static void libdef_func(BuildCtx *ctx, char *p, int arg)
 
				   regfunc = REGFUNC_OK;
			
 
				 }
			
 
				 
			
 
				+static uint8_t *libdef_uleb128(uint8_t *p, uint32_t *vv)
			
 
				+{
			
 
				+  uint32_t v = *p++;
			
 
				+  if (v >= 0x80) {
			
 
				+    int sh = 0; v &= 0x7f;
			
 
				+    do { v |= ((*p & 0x7f) << (sh += 7)); } while (*p++ >= 0x80);
			
 
				+  }
			
 
				+  *vv = v;
			
 
				+  return p;
			
 
				+}
			
 
				+
			
 
				+static void libdef_fixupbc(uint8_t *p)
			
 
				+{
			
 
				+  uint32_t i, sizebc;
			
 
				+  p += 4;
			
 
				+  p = libdef_uleb128(p, &sizebc);
			
 
				+  p = libdef_uleb128(p, &sizebc);
			
 
				+  p = libdef_uleb128(p, &sizebc);
			
 
				+  for (i = 0; i < sizebc; i++, p += 4) {
			
 
				+    uint8_t op = p[libbc_endian ? 3 : 0];
			
 
				+    uint8_t ra = p[libbc_endian ? 2 : 1];
			
 
				+    uint8_t rc = p[libbc_endian ? 1 : 2];
			
 
				+    uint8_t rb = p[libbc_endian ? 0 : 3];
			
 
				+    if (!LJ_DUALNUM && op == BC_ISTYPE && rc == ~LJ_TNUMX+1) {
			
 
				+      op = BC_ISNUM; rc++;
			
 
				+    }
			
 
				+    p[LJ_ENDIAN_SELECT(0, 3)] = op;
			
 
				+    p[LJ_ENDIAN_SELECT(1, 2)] = ra;
			
 
				+    p[LJ_ENDIAN_SELECT(2, 1)] = rc;
			
 
				+    p[LJ_ENDIAN_SELECT(3, 0)] = rb;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void libdef_lua(BuildCtx *ctx, char *p, int arg)
			
 
				+{
			
 
				+  UNUSED(arg);
			
 
				+  if (ctx->mode == BUILD_libdef) {
			
 
				+    int i;
			
 
				+    for (i = 0; libbc_map[i].name != NULL; i++) {
			
 
				+      if (!strcmp(libbc_map[i].name, p)) {
			
 
				+	int ofs = libbc_map[i].ofs;
			
 
				+	int len = libbc_map[i+1].ofs - ofs;
			
 
				+	obuf[2]++;  /* Bump hash table size. */
			
 
				+	*optr++ = LIBINIT_LUA;
			
 
				+	libdef_name(p, 0);
			
 
				+	memcpy(optr, libbc_code + ofs, len);
			
 
				+	libdef_fixupbc(optr);
			
 
				+	optr += len;
			
 
				+	return;
			
 
				+      }
			
 
				+    }
			
 
				+    fprintf(stderr, "Error: missing libbc definition for %s\n", p);
			
 
				+    exit(1);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				 static uint32_t find_rec(char *name)
			
 
				 {
			
 
				   char *p = (char *)obuf;
			
@@ -277,6 +335,7 @@ static const LibDefHandler libdef_handlers[] = {
 
				   { "CF(",	")",		libdef_func,		LIBINIT_CF },
			
 
				   { "ASM(",	")",		libdef_func,		LIBINIT_ASM },
			
 
				   { "ASM_(",	")",		libdef_func,		LIBINIT_ASM_ },
			
 
				+  { "LUA(",	")",		libdef_lua,		0 },
			
 
				   { "REC(",	")",		libdef_rec,		0 },
			
 
				   { "PUSH(",	")",		libdef_push,		0 },
			
 
				   { "SET(",	")",		libdef_set,		0 },
			
@@ -373,7 +432,7 @@ void emit_lib(BuildCtx *ctx)
 
				       "#ifndef FF_NUM_ASMFUNC\n#define FF_NUM_ASMFUNC %d\n#endif\n\n",
			
 
				       ffasmfunc);
			
 
				   } else if (ctx->mode == BUILD_vmdef) {
			
 
				-    fprintf(ctx->fp, "}\n\n");
			
 
				+    fprintf(ctx->fp, "},\n\n");
			
 
				   } else if (ctx->mode == BUILD_bcdef) {
			
 
				     int i;
			
 
				     fprintf(ctx->fp, "\n};\n\n");
			
--- a/luajit.mod/luajit/src/host/buildvm_libbc.h
+++ b/luajit.mod/luajit/src/host/buildvm_libbc.h
@@ -0,0 +1,56 @@
 
				+/* This is a generated file. DO NOT EDIT! */
			
 
				+
			
 
				+static const int libbc_endian = 0;
			
 
				+
			
 
				+static const uint8_t libbc_code[] = {
			
 
				+#if LJ_FR2
			
 
				+0,1,2,0,0,1,2,24,1,0,0,76,1,2,0,241,135,158,166,3,220,203,178,130,4,0,1,2,0,
			
 
				+0,1,2,24,1,0,0,76,1,2,0,243,244,148,165,20,198,190,199,252,3,0,1,2,0,0,0,3,
			
 
				+16,0,5,0,21,1,0,0,76,1,2,0,0,2,10,0,0,0,15,16,0,12,0,16,1,9,0,41,2,1,0,21,3,
			
 
				+0,0,41,4,1,0,77,2,8,128,18,6,1,0,18,8,5,0,59,9,5,0,66,6,3,2,10,6,0,0,88,7,1,
			
 
				+128,76,6,2,0,79,2,248,127,75,0,1,0,0,2,11,0,0,0,16,16,0,12,0,16,1,9,0,43,2,
			
 
				+0,0,18,3,0,0,41,4,0,0,88,5,7,128,18,7,1,0,18,9,5,0,18,10,6,0,66,7,3,2,10,7,
			
 
				+0,0,88,8,1,128,76,7,2,0,70,5,3,3,82,5,247,127,75,0,1,0,0,1,2,0,0,0,3,16,0,12,
			
 
				+0,21,1,0,0,76,1,2,0,0,2,10,0,0,2,30,16,0,12,0,21,2,0,0,11,1,0,0,88,3,7,128,
			
 
				+8,2,0,0,88,3,23,128,59,3,2,0,43,4,0,0,64,4,2,0,76,3,2,0,88,3,18,128,16,1,14,
			
 
				+0,41,3,1,0,3,3,1,0,88,3,14,128,3,1,2,0,88,3,12,128,59,3,1,0,22,4,1,1,18,5,2,
			
 
				+0,41,6,1,0,77,4,4,128,23,8,1,7,59,9,7,0,64,9,8,0,79,4,252,127,43,4,0,0,64,4,
			
 
				+2,0,76,3,2,0,75,0,1,0,0,2,0,5,12,0,0,0,35,16,0,12,0,16,1,14,0,16,2,14,0,16,
			
 
				+3,14,0,11,4,0,0,88,5,1,128,18,4,0,0,16,4,12,0,3,1,2,0,88,5,24,128,33,5,1,3,
			
 
				+0,2,3,0,88,6,4,128,2,3,1,0,88,6,2,128,4,4,0,0,88,6,9,128,18,6,1,0,18,7,2,0,
			
 
				+41,8,1,0,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,6,252,127,88,6,8,128,
			
 
				+18,6,2,0,18,7,1,0,41,8,255,255,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,
			
 
				+6,252,127,76,4,2,0,0
			
 
				+#else
			
 
				+0,1,2,0,0,1,2,24,1,0,0,76,1,2,0,241,135,158,166,3,220,203,178,130,4,0,1,2,0,
			
 
				+0,1,2,24,1,0,0,76,1,2,0,243,244,148,165,20,198,190,199,252,3,0,1,2,0,0,0,3,
			
 
				+16,0,5,0,21,1,0,0,76,1,2,0,0,2,9,0,0,0,15,16,0,12,0,16,1,9,0,41,2,1,0,21,3,
			
 
				+0,0,41,4,1,0,77,2,8,128,18,6,1,0,18,7,5,0,59,8,5,0,66,6,3,2,10,6,0,0,88,7,1,
			
 
				+128,76,6,2,0,79,2,248,127,75,0,1,0,0,2,10,0,0,0,16,16,0,12,0,16,1,9,0,43,2,
			
 
				+0,0,18,3,0,0,41,4,0,0,88,5,7,128,18,7,1,0,18,8,5,0,18,9,6,0,66,7,3,2,10,7,0,
			
 
				+0,88,8,1,128,76,7,2,0,70,5,3,3,82,5,247,127,75,0,1,0,0,1,2,0,0,0,3,16,0,12,
			
 
				+0,21,1,0,0,76,1,2,0,0,2,10,0,0,2,30,16,0,12,0,21,2,0,0,11,1,0,0,88,3,7,128,
			
 
				+8,2,0,0,88,3,23,128,59,3,2,0,43,4,0,0,64,4,2,0,76,3,2,0,88,3,18,128,16,1,14,
			
 
				+0,41,3,1,0,3,3,1,0,88,3,14,128,3,1,2,0,88,3,12,128,59,3,1,0,22,4,1,1,18,5,2,
			
 
				+0,41,6,1,0,77,4,4,128,23,8,1,7,59,9,7,0,64,9,8,0,79,4,252,127,43,4,0,0,64,4,
			
 
				+2,0,76,3,2,0,75,0,1,0,0,2,0,5,12,0,0,0,35,16,0,12,0,16,1,14,0,16,2,14,0,16,
			
 
				+3,14,0,11,4,0,0,88,5,1,128,18,4,0,0,16,4,12,0,3,1,2,0,88,5,24,128,33,5,1,3,
			
 
				+0,2,3,0,88,6,4,128,2,3,1,0,88,6,2,128,4,4,0,0,88,6,9,128,18,6,1,0,18,7,2,0,
			
 
				+41,8,1,0,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,6,252,127,88,6,8,128,
			
 
				+18,6,2,0,18,7,1,0,41,8,255,255,77,6,4,128,32,10,5,9,59,11,9,0,64,11,10,4,79,
			
 
				+6,252,127,76,4,2,0,0
			
 
				+#endif
			
 
				+};
			
 
				+
			
 
				+static const struct { const char *name; int ofs; } libbc_map[] = {
			
 
				+{"math_deg",0},
			
 
				+{"math_rad",25},
			
 
				+{"string_len",50},
			
 
				+{"table_foreachi",69},
			
 
				+{"table_foreach",136},
			
 
				+{"table_getn",207},
			
 
				+{"table_remove",226},
			
 
				+{"table_move",355},
			
 
				+{NULL,502}
			
 
				+};
			
 
				+
			
--- a/luajit.mod/luajit/src/host/buildvm_peobj.c
+++ b/luajit.mod/luajit/src/host/buildvm_peobj.c
@@ -109,6 +109,8 @@ enum {
 
				 #if LJ_TARGET_X64
			
 
				   PEOBJ_SECT_PDATA,
			
 
				   PEOBJ_SECT_XDATA,
			
 
				+#elif LJ_TARGET_X86
			
 
				+  PEOBJ_SECT_SXDATA,
			
 
				 #endif
			
 
				   PEOBJ_SECT_RDATA_Z,
			
 
				   PEOBJ_NSECTIONS
			
@@ -208,6 +210,13 @@ void emit_peobj(BuildCtx *ctx)
 
				   sofs += (pesect[PEOBJ_SECT_XDATA].nreloc = 1) * PEOBJ_RELOC_SIZE;
			
 
				   /* Flags: 40 = read, 30 = align4, 40 = initialized data. */
			
 
				   pesect[PEOBJ_SECT_XDATA].flags = 0x40300040;
			
 
				+#elif LJ_TARGET_X86
			
 
				+  memcpy(pesect[PEOBJ_SECT_SXDATA].name, ".sxdata", sizeof(".sxdata")-1);
			
 
				+  pesect[PEOBJ_SECT_SXDATA].ofs = sofs;
			
 
				+  sofs += (pesect[PEOBJ_SECT_SXDATA].size = 4);
			
 
				+  pesect[PEOBJ_SECT_SXDATA].relocofs = sofs;
			
 
				+  /* Flags: 40 = read, 30 = align4, 02 = lnk_info, 40 = initialized data. */
			
 
				+  pesect[PEOBJ_SECT_SXDATA].flags = 0x40300240;
			
 
				 #endif
			
 
				 
			
 
				   memcpy(pesect[PEOBJ_SECT_RDATA_Z].name, ".rdata$Z", sizeof(".rdata$Z")-1);
			
@@ -232,7 +241,7 @@ void emit_peobj(BuildCtx *ctx)
 
				   nrsym = ctx->nrelocsym;
			
 
				   pehdr.nsyms = 1+PEOBJ_NSECTIONS*2 + 1+ctx->nsym + nrsym;
			
 
				 #if LJ_TARGET_X64
			
 
				-  pehdr.nsyms += 1;  /* Symbol for lj_err_unwind_win64. */
			
 
				+  pehdr.nsyms += 1;  /* Symbol for lj_err_unwind_win. */
			
 
				 #endif
			
 
				 
			
 
				   /* Write PE object header and all sections. */
			
@@ -312,6 +321,19 @@ void emit_peobj(BuildCtx *ctx)
 
				     reloc.type = PEOBJ_RELOC_ADDR32NB;
			
 
				     owrite(ctx, &reloc, PEOBJ_RELOC_SIZE);
			
 
				   }
			
 
				+#elif LJ_TARGET_X86
			
 
				+  /* Write .sxdata section. */
			
 
				+  for (i = 0; i < nrsym; i++) {
			
 
				+    if (!strcmp(ctx->relocsym[i], "_lj_err_unwind_win")) {
			
 
				+      uint32_t symidx = 1+2+i;
			
 
				+      owrite(ctx, &symidx, 4);
			
 
				+      break;
			
 
				+    }
			
 
				+  }
			
 
				+  if (i == nrsym) {
			
 
				+    fprintf(stderr, "Error: extern lj_err_unwind_win not used\n");
			
 
				+    exit(1);
			
 
				+  }
			
 
				 #endif
			
 
				 
			
 
				   /* Write .rdata$Z section. */
			
@@ -333,8 +355,10 @@ void emit_peobj(BuildCtx *ctx)
 
				 #if LJ_TARGET_X64
			
 
				     emit_peobj_sym_sect(ctx, pesect, PEOBJ_SECT_PDATA);
			
 
				     emit_peobj_sym_sect(ctx, pesect, PEOBJ_SECT_XDATA);
			
 
				-    emit_peobj_sym(ctx, "lj_err_unwind_win64", 0,
			
 
				+    emit_peobj_sym(ctx, "lj_err_unwind_win", 0,
			
 
				 		   PEOBJ_SECT_UNDEF, PEOBJ_TYPE_FUNC, PEOBJ_SCL_EXTERN);
			
 
				+#elif LJ_TARGET_X86
			
 
				+    emit_peobj_sym_sect(ctx, pesect, PEOBJ_SECT_SXDATA);
			
 
				 #endif
			
 
				 
			
 
				     emit_peobj_sym(ctx, ctx->beginsym, 0,
			
--- a/luajit.mod/luajit/src/host/genlibbc.lua
+++ b/luajit.mod/luajit/src/host/genlibbc.lua
@@ -0,0 +1,197 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- Lua script to dump the bytecode of the library functions written in Lua.
			
 
				+-- The resulting 'buildvm_libbc.h' is used for the build process of LuaJIT.
			
 
				+----------------------------------------------------------------------------
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+
			
 
				+local ffi = require("ffi")
			
 
				+local bit = require("bit")
			
 
				+local vmdef = require("jit.vmdef")
			
 
				+local bcnames = vmdef.bcnames
			
 
				+
			
 
				+local format = string.format
			
 
				+
			
 
				+local isbe = (string.byte(string.dump(function() end), 5) % 2 == 1)
			
 
				+
			
 
				+local function usage(arg)
			
 
				+  io.stderr:write("Usage: ", arg and arg[0] or "genlibbc",
			
 
				+		  " [-o buildvm_libbc.h] lib_*.c\n")
			
 
				+  os.exit(1)
			
 
				+end
			
 
				+
			
 
				+local function parse_arg(arg)
			
 
				+  local outfile = "-"
			
 
				+  if not (arg and arg[1]) then
			
 
				+    usage(arg)
			
 
				+  end
			
 
				+  if arg[1] == "-o" then
			
 
				+    outfile = arg[2]
			
 
				+    if not outfile then usage(arg) end
			
 
				+    table.remove(arg, 1)
			
 
				+    table.remove(arg, 1)
			
 
				+  end
			
 
				+  return outfile
			
 
				+end
			
 
				+
			
 
				+local function read_files(names)
			
 
				+  local src = ""
			
 
				+  for _,name in ipairs(names) do
			
 
				+    local fp = assert(io.open(name))
			
 
				+    src = src .. fp:read("*a")
			
 
				+    fp:close()
			
 
				+  end
			
 
				+  return src
			
 
				+end
			
 
				+
			
 
				+local function transform_lua(code)
			
 
				+  local fixup = {}
			
 
				+  local n = -30000
			
 
				+  code = string.gsub(code, "CHECK_(%w*)%((.-)%)", function(tp, var)
			
 
				+    n = n + 1
			
 
				+    fixup[n] = { "CHECK", tp }
			
 
				+    return format("%s=%d", var, n)
			
 
				+  end)
			
 
				+  code = string.gsub(code, "PAIRS%((.-)%)", function(var)
			
 
				+    fixup.PAIRS = true
			
 
				+    return format("nil, %s, 0", var)
			
 
				+  end)
			
 
				+  return "return "..code, fixup
			
 
				+end
			
 
				+
			
 
				+local function read_uleb128(p)
			
 
				+  local v = p[0]; p = p + 1
			
 
				+  if v >= 128 then
			
 
				+    local sh = 7; v = v - 128
			
 
				+    repeat
			
 
				+      local r = p[0]
			
 
				+      v = v + bit.lshift(bit.band(r, 127), sh)
			
 
				+      sh = sh + 7
			
 
				+      p = p + 1
			
 
				+    until r < 128
			
 
				+  end
			
 
				+  return p, v
			
 
				+end
			
 
				+
			
 
				+-- ORDER LJ_T
			
 
				+local name2itype = {
			
 
				+  str = 5, func = 9, tab = 12, int = 14, num = 15
			
 
				+}
			
 
				+
			
 
				+local BC = {}
			
 
				+for i=0,#bcnames/6-1 do
			
 
				+  BC[string.gsub(string.sub(bcnames, i*6+1, i*6+6), " ", "")] = i
			
 
				+end
			
 
				+local xop, xra = isbe and 3 or 0, isbe and 2 or 1
			
 
				+local xrc, xrb = isbe and 1 or 2, isbe and 0 or 3
			
 
				+
			
 
				+local function fixup_dump(dump, fixup)
			
 
				+  local buf = ffi.new("uint8_t[?]", #dump+1, dump)
			
 
				+  local p = buf+5
			
 
				+  local n, sizebc
			
 
				+  p, n = read_uleb128(p)
			
 
				+  local start = p
			
 
				+  p = p + 4
			
 
				+  p = read_uleb128(p)
			
 
				+  p = read_uleb128(p)
			
 
				+  p, sizebc = read_uleb128(p)
			
 
				+  local rawtab = {}
			
 
				+  for i=0,sizebc-1 do
			
 
				+    local op = p[xop]
			
 
				+    if op == BC.KSHORT then
			
 
				+      local rd = p[xrc] + 256*p[xrb]
			
 
				+      rd = bit.arshift(bit.lshift(rd, 16), 16)
			
 
				+      local f = fixup[rd]
			
 
				+      if f then
			
 
				+	if f[1] == "CHECK" then
			
 
				+	  local tp = f[2]
			
 
				+	  if tp == "tab" then rawtab[p[xra]] = true end
			
 
				+	  p[xop] = tp == "num" and BC.ISNUM or BC.ISTYPE
			
 
				+	  p[xrb] = 0
			
 
				+	  p[xrc] = name2itype[tp]
			
 
				+	else
			
 
				+	  error("unhandled fixup type: "..f[1])
			
 
				+	end
			
 
				+      end
			
 
				+    elseif op == BC.TGETV then
			
 
				+      if rawtab[p[xrb]] then
			
 
				+	p[xop] = BC.TGETR
			
 
				+      end
			
 
				+    elseif op == BC.TSETV then
			
 
				+      if rawtab[p[xrb]] then
			
 
				+	p[xop] = BC.TSETR
			
 
				+      end
			
 
				+    elseif op == BC.ITERC then
			
 
				+      if fixup.PAIRS then
			
 
				+	p[xop] = BC.ITERN
			
 
				+      end
			
 
				+    end
			
 
				+    p = p + 4
			
 
				+  end
			
 
				+  return ffi.string(start, n)
			
 
				+end
			
 
				+
			
 
				+local function find_defs(src)
			
 
				+  local defs = {}
			
 
				+  for name, code in string.gmatch(src, "LJLIB_LUA%(([^)]*)%)%s*/%*(.-)%*/") do
			
 
				+    local env = {}
			
 
				+    local tcode, fixup = transform_lua(code)
			
 
				+    local func = assert(load(tcode, "", nil, env))()
			
 
				+    defs[name] = fixup_dump(string.dump(func, true), fixup)
			
 
				+    defs[#defs+1] = name
			
 
				+  end
			
 
				+  return defs
			
 
				+end
			
 
				+
			
 
				+local function gen_header(defs)
			
 
				+  local t = {}
			
 
				+  local function w(x) t[#t+1] = x end
			
 
				+  w("/* This is a generated file. DO NOT EDIT! */\n\n")
			
 
				+  w("static const int libbc_endian = ") w(isbe and 1 or 0) w(";\n\n")
			
 
				+  local s = ""
			
 
				+  for _,name in ipairs(defs) do
			
 
				+    s = s .. defs[name]
			
 
				+  end
			
 
				+  w("static const uint8_t libbc_code[] = {\n")
			
 
				+  local n = 0
			
 
				+  for i=1,#s do
			
 
				+    local x = string.byte(s, i)
			
 
				+    w(x); w(",")
			
 
				+    n = n + (x < 10 and 2 or (x < 100 and 3 or 4))
			
 
				+    if n >= 75 then n = 0; w("\n") end
			
 
				+  end
			
 
				+  w("0\n};\n\n")
			
 
				+  w("static const struct { const char *name; int ofs; } libbc_map[] = {\n")
			
 
				+  local m = 0
			
 
				+  for _,name in ipairs(defs) do
			
 
				+    w('{"'); w(name); w('",'); w(m) w('},\n')
			
 
				+    m = m + #defs[name]
			
 
				+  end
			
 
				+  w("{NULL,"); w(m); w("}\n};\n\n")
			
 
				+  return table.concat(t)
			
 
				+end
			
 
				+
			
 
				+local function write_file(name, data)
			
 
				+  if name == "-" then
			
 
				+    assert(io.write(data))
			
 
				+    assert(io.flush())
			
 
				+  else
			
 
				+    local fp = io.open(name)
			
 
				+    if fp then
			
 
				+      local old = fp:read("*a")
			
 
				+      fp:close()
			
 
				+      if data == old then return end
			
 
				+    end
			
 
				+    fp = assert(io.open(name, "w"))
			
 
				+    assert(fp:write(data))
			
 
				+    assert(fp:close())
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local outfile = parse_arg(arg)
			
 
				+local src = read_files(arg)
			
 
				+local defs = find_defs(src)
			
 
				+local hdr = gen_header(defs)
			
 
				+write_file(outfile, hdr)
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/.gitignore
+++ b/luajit.mod/luajit/src/jit/.gitignore
@@ -0,0 +1 @@
 
				+vmdef.lua
			
--- a/luajit.mod/luajit/src/jit/bc.lua
+++ b/luajit.mod/luajit/src/jit/bc.lua
@@ -41,7 +41,7 @@
 
				 
			
 
				 -- Cache some library functions and objects.
			
 
				 local jit = require("jit")
			
 
				-assert(jit.version_num == 20005, "LuaJIT core/library version mismatch")
			
 
				+assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
			
 
				 local jutil = require("jit.util")
			
 
				 local vmdef = require("jit.vmdef")
			
 
				 local bit = require("bit")
			
@@ -179,13 +179,12 @@ local function bcliston(outfile)
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-line = bcline
			
 
				-dump = bcdump
			
 
				-targets = bctargets
			
 
				-
			
 
				-on = bcliston
			
 
				-off = bclistoff
			
 
				-start = bcliston -- For -j command line option.
			
 
				+return {
			
 
				+  line = bcline,
			
 
				+  dump = bcdump,
			
 
				+  targets = bctargets,
			
 
				+  on = bcliston,
			
 
				+  off = bclistoff,
			
 
				+  start = bcliston -- For -j command line option.
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/bcsave.lua
+++ b/luajit.mod/luajit/src/jit/bcsave.lua
@@ -11,7 +11,7 @@
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 local jit = require("jit")
			
 
				-assert(jit.version_num == 20005, "LuaJIT core/library version mismatch")
			
 
				+assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
			
 
				 local bit = require("bit")
			
 
				 
			
 
				 -- Symbol name prefix for LuaJIT bytecode.
			
@@ -63,8 +63,8 @@ local map_type = {
 
				 }
			
 
				 
			
 
				 local map_arch = {
			
 
				-  x86 = true, x64 = true, arm = true, ppc = true, ppcspe = true,
			
 
				-  mips = true, mipsel = true,
			
 
				+  x86 = true, x64 = true, arm = true, arm64 = true, arm64be = true,
			
 
				+  ppc = true, mips = true, mipsel = true,
			
 
				 }
			
 
				 
			
 
				 local map_os = {
			
@@ -125,12 +125,12 @@ extern "C"
 
				 #ifdef _WIN32
			
 
				 __declspec(dllexport)
			
 
				 #endif
			
 
				-const char %s%s[] = {
			
 
				+const unsigned char %s%s[] = {
			
 
				 ]], LJBC_PREFIX, ctx.modname))
			
 
				   else
			
 
				     fp:write(string.format([[
			
 
				 #define %s%s_SIZE %d
			
 
				-static const char %s%s[] = {
			
 
				+static const unsigned char %s%s[] = {
			
 
				 ]], LJBC_PREFIX, ctx.modname, #s, LJBC_PREFIX, ctx.modname))
			
 
				   end
			
 
				   local t, n, m = {}, 0, 0
			
@@ -200,9 +200,9 @@ typedef struct {
 
				 ]]
			
 
				   local symname = LJBC_PREFIX..ctx.modname
			
 
				   local is64, isbe = false, false
			
 
				-  if ctx.arch == "x64" then
			
 
				+  if ctx.arch == "x64" or ctx.arch == "arm64" or ctx.arch == "arm64be" then
			
 
				     is64 = true
			
 
				-  elseif ctx.arch == "ppc" or ctx.arch == "ppcspe" or ctx.arch == "mips" then
			
 
				+  elseif ctx.arch == "ppc" or ctx.arch == "mips" then
			
 
				     isbe = true
			
 
				   end
			
 
				 
			
@@ -237,7 +237,7 @@ typedef struct {
 
				   hdr.eendian = isbe and 2 or 1
			
 
				   hdr.eversion = 1
			
 
				   hdr.type = f16(1)
			
 
				-  hdr.machine = f16(({ x86=3, x64=62, arm=40, ppc=20, ppcspe=20, mips=8, mipsel=8 })[ctx.arch])
			
 
				+  hdr.machine = f16(({ x86=3, x64=62, arm=40, arm64=183, arm64be=183, ppc=20, mips=8, mipsel=8 })[ctx.arch])
			
 
				   if ctx.arch == "mips" or ctx.arch == "mipsel" then
			
 
				     hdr.flags = f32(0x50001006)
			
 
				   end
			
@@ -275,7 +275,7 @@ typedef struct {
 
				   o.sect[2].size = fofs(ofs)
			
 
				   o.sect[3].type = f32(3) -- .strtab
			
 
				   o.sect[3].ofs = fofs(sofs + ofs)
			
 
				-  o.sect[3].size = fofs(#symname+1)
			
 
				+  o.sect[3].size = fofs(#symname+2)
			
 
				   ffi.copy(o.space+ofs+1, symname)
			
 
				   ofs = ofs + #symname + 2
			
 
				   o.sect[4].type = f32(1) -- .rodata
			
@@ -477,13 +477,13 @@ typedef struct {
 
				 } mach_obj_64;
			
 
				 typedef struct {
			
 
				   mach_fat_header fat;
			
 
				-  mach_fat_arch fat_arch[4];
			
 
				+  mach_fat_arch fat_arch[2];
			
 
				   struct {
			
 
				     mach_header hdr;
			
 
				     mach_segment_command seg;
			
 
				     mach_section sec;
			
 
				     mach_symtab_command sym;
			
 
				-  } arch[4];
			
 
				+  } arch[2];
			
 
				   mach_nlist sym_entry;
			
 
				   uint8_t space[4096];
			
 
				 } mach_fat_obj;
			
@@ -494,6 +494,8 @@ typedef struct {
 
				     is64, align, mobj = true, 8, "mach_obj_64"
			
 
				   elseif ctx.arch == "arm" then
			
 
				     isfat, mobj = true, "mach_fat_obj"
			
 
				+  elseif ctx.arch == "arm64" then
			
 
				+    is64, align, isfat, mobj = true, 8, true, "mach_fat_obj"
			
 
				   else
			
 
				     check(ctx.arch == "x86", "unsupported architecture for OSX")
			
 
				   end
			
@@ -503,8 +505,8 @@ typedef struct {
 
				   -- Create Mach-O object and fill in header.
			
 
				   local o = ffi.new(mobj)
			
 
				   local mach_size = aligned(ffi.offsetof(o, "space")+#symname+2, align)
			
 
				-  local cputype = ({ x86={7}, x64={0x01000007}, arm={7,12,12,12} })[ctx.arch]
			
 
				-  local cpusubtype = ({ x86={3}, x64={3}, arm={3,6,9,11} })[ctx.arch]
			
 
				+  local cputype = ({ x86={7}, x64={0x01000007}, arm={7,12}, arm64={0x01000007,0x0100000c} })[ctx.arch]
			
 
				+  local cpusubtype = ({ x86={3}, x64={3}, arm={3,9}, arm64={3,0} })[ctx.arch]
			
 
				   if isfat then
			
 
				     o.fat.magic = be32(0xcafebabe)
			
 
				     o.fat.nfat_arch = be32(#cpusubtype)
			
@@ -653,7 +655,7 @@ end
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-start = docmd -- Process -b command line option.
			
 
				+return {
			
 
				+  start = docmd -- Process -b command line option.
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_arm.lua
+++ b/luajit.mod/luajit/src/jit/dis_arm.lua
@@ -658,7 +658,7 @@ local function disass_block(ctx, ofs, len)
 
				 end
			
 
				 
			
 
				 -- Extended API: create a disassembler context. Then call ctx:disass(ofs, len).
			
 
				-local function create_(code, addr, out)
			
 
				+local function create(code, addr, out)
			
 
				   local ctx = {}
			
 
				   ctx.code = code
			
 
				   ctx.addr = addr or 0
			
@@ -670,20 +670,20 @@ local function create_(code, addr, out)
 
				 end
			
 
				 
			
 
				 -- Simple API: disassemble code (a string) at address and output via out.
			
 
				-local function disass_(code, addr, out)
			
 
				-  create_(code, addr, out):disass()
			
 
				+local function disass(code, addr, out)
			
 
				+  create(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				 -- Return register name for RID.
			
 
				-local function regname_(r)
			
 
				+local function regname(r)
			
 
				   if r < 16 then return map_gpr[r] end
			
 
				   return "d"..(r-16)
			
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-create = create_
			
 
				-disass = disass_
			
 
				-regname = regname_
			
 
				+return {
			
 
				+  create = create,
			
 
				+  disass = disass,
			
 
				+  regname = regname
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_arm64.lua
+++ b/luajit.mod/luajit/src/jit/dis_arm64.lua
@@ -0,0 +1,1216 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT ARM64 disassembler module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+--
			
 
				+-- Contributed by Djordje Kovacevic and Stefan Pejic from RT-RK.com.
			
 
				+-- Sponsored by Cisco Systems, Inc.
			
 
				+----------------------------------------------------------------------------
			
 
				+-- This is a helper module used by the LuaJIT machine code dumper module.
			
 
				+--
			
 
				+-- It disassembles most user-mode AArch64 instructions.
			
 
				+-- NYI: Advanced SIMD and VFP instructions.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local type = type
			
 
				+local sub, byte, format = string.sub, string.byte, string.format
			
 
				+local match, gmatch, gsub = string.match, string.gmatch, string.gsub
			
 
				+local concat = table.concat
			
 
				+local bit = require("bit")
			
 
				+local band, bor, bxor, tohex = bit.band, bit.bor, bit.bxor, bit.tohex
			
 
				+local lshift, rshift, arshift = bit.lshift, bit.rshift, bit.arshift
			
 
				+local ror = bit.ror
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+-- Opcode maps
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local map_adr = { -- PC-relative addressing.
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = "adrDBx", "adrpDBx"
			
 
				+}
			
 
				+
			
 
				+local map_addsubi = { -- Add/subtract immediate.
			
 
				+  shift = 29, mask = 3,
			
 
				+  [0] = "add|movDNIg", "adds|cmnD0NIg", "subDNIg", "subs|cmpD0NIg",
			
 
				+}
			
 
				+
			
 
				+local map_logi = { -- Logical immediate.
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 22, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "andDNig", "orr|movDN0ig", "eorDNig", "ands|tstD0Nig"
			
 
				+    },
			
 
				+    false -- unallocated
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = "andDNig", "orr|movDN0ig", "eorDNig", "ands|tstD0Nig"
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_movwi = { -- Move wide immediate.
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 22, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "movnDWRg", false, "movz|movDYRg", "movkDWRg"
			
 
				+    }, false -- unallocated
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = "movnDWRg", false, "movz|movDYRg", "movkDWRg"
			
 
				+  },
			
 
				+}
			
 
				+
			
 
				+local map_bitf = { -- Bitfield.
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 22, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "sbfm|sbfiz|sbfx|asr|sxtw|sxth|sxtbDN12w",
			
 
				+      "bfm|bfi|bfxilDN13w",
			
 
				+      "ubfm|ubfiz|ubfx|lsr|lsl|uxth|uxtbDN12w"
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 22, mask = 1,
			
 
				+    {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "sbfm|sbfiz|sbfx|asr|sxtw|sxth|sxtbDN12x",
			
 
				+      "bfm|bfi|bfxilDN13x",
			
 
				+      "ubfm|ubfiz|ubfx|lsr|lsl|uxth|uxtbDN12x"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_datai = { -- Data processing - immediate.
			
 
				+  shift = 23, mask = 7,
			
 
				+  [0] = map_adr, map_adr, map_addsubi, false,
			
 
				+  map_logi, map_movwi, map_bitf,
			
 
				+  {
			
 
				+    shift = 15, mask = 0x1c0c1,
			
 
				+    [0] = "extr|rorDNM4w", [0x10080] = "extr|rorDNM4x",
			
 
				+    [0x10081] = "extr|rorDNM4x"
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_logsr = { -- Logical, shifted register.
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 15, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = {
			
 
				+	shift = 21, mask = 7,
			
 
				+	[0] = "andDNMSg", "bicDNMSg", "andDNMSg", "bicDNMSg",
			
 
				+	"andDNMSg", "bicDNMSg", "andDNMg", "bicDNMg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 21, mask = 7,
			
 
				+	[0] ="orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0MSg", "orn|mvnDN0MSg",
			
 
				+	     "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0Mg", "orn|mvnDN0Mg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 21, mask = 7,
			
 
				+	[0] = "eorDNMSg", "eonDNMSg", "eorDNMSg", "eonDNMSg",
			
 
				+	"eorDNMSg", "eonDNMSg", "eorDNMg", "eonDNMg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 21, mask = 7,
			
 
				+	[0] = "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMSg", "bicsDNMSg",
			
 
				+	"ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMg", "bicsDNMg"
			
 
				+      }
			
 
				+    },
			
 
				+    false -- unallocated
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "andDNMSg", "bicDNMSg", "andDNMSg", "bicDNMSg",
			
 
				+      "andDNMSg", "bicDNMSg", "andDNMg", "bicDNMg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0MSg", "orn|mvnDN0MSg",
			
 
				+      "orr|movDN0MSg", "orn|mvnDN0MSg", "orr|movDN0Mg", "orn|mvnDN0Mg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "eorDNMSg", "eonDNMSg", "eorDNMSg", "eonDNMSg",
			
 
				+      "eorDNMSg", "eonDNMSg", "eorDNMg", "eonDNMg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMSg", "bicsDNMSg",
			
 
				+      "ands|tstD0NMSg", "bicsDNMSg", "ands|tstD0NMg", "bicsDNMg"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_assh = {
			
 
				+  shift = 31, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 15, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "addDNMSg", "addDNMSg", "addDNMSg", "addDNMg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "adds|cmnD0NMSg", "adds|cmnD0NMSg",
			
 
				+	      "adds|cmnD0NMSg", "adds|cmnD0NMg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "sub|negDN0MSg", "sub|negDN0MSg", "sub|negDN0MSg", "sub|negDN0Mg"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "subs|cmp|negsD0N0MzSg", "subs|cmp|negsD0N0MzSg",
			
 
				+	      "subs|cmp|negsD0N0MzSg", "subs|cmp|negsD0N0Mzg"
			
 
				+      },
			
 
				+    },
			
 
				+    false -- unallocated
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "addDNMSg", "addDNMSg", "addDNMSg", "addDNMg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "adds|cmnD0NMSg", "adds|cmnD0NMSg", "adds|cmnD0NMSg",
			
 
				+	    "adds|cmnD0NMg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "sub|negDN0MSg", "sub|negDN0MSg", "sub|negDN0MSg", "sub|negDN0Mg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "subs|cmp|negsD0N0MzSg", "subs|cmp|negsD0N0MzSg",
			
 
				+	    "subs|cmp|negsD0N0MzSg", "subs|cmp|negsD0N0Mzg"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_addsubsh = { -- Add/subtract, shifted register.
			
 
				+  shift = 22, mask = 3,
			
 
				+  [0] = map_assh, map_assh, map_assh
			
 
				+}
			
 
				+
			
 
				+local map_addsubex = { -- Add/subtract, extended register.
			
 
				+  shift = 22, mask = 3,
			
 
				+  [0] = {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = "addDNMXg", "adds|cmnD0NMXg", "subDNMXg", "subs|cmpD0NMzXg",
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_addsubc = { -- Add/subtract, with carry.
			
 
				+  shift = 10, mask = 63,
			
 
				+  [0] = {
			
 
				+    shift = 29, mask = 3,
			
 
				+    [0] = "adcDNMg", "adcsDNMg", "sbc|ngcDN0Mg", "sbcs|ngcsDN0Mg",
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_ccomp = {
			
 
				+  shift = 4, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 10, mask = 3,
			
 
				+    [0] = { -- Conditional compare register.
			
 
				+      shift = 29, mask = 3,
			
 
				+      "ccmnNMVCg", false, "ccmpNMVCg",
			
 
				+    },
			
 
				+    [2] = {  -- Conditional compare immediate.
			
 
				+      shift = 29, mask = 3,
			
 
				+      "ccmnN5VCg", false, "ccmpN5VCg",
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_csel = { -- Conditional select.
			
 
				+  shift = 11, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 10, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "cselDNMzCg", false, "csinv|cinv|csetmDNMcg", false,
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 29, mask = 3,
			
 
				+      [0] = "csinc|cinc|csetDNMcg", false, "csneg|cnegDNMcg", false,
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_data1s = { -- Data processing, 1 source.
			
 
				+  shift = 29, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 31, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 10, mask = 0x7ff,
			
 
				+      [0] = "rbitDNg", "rev16DNg", "revDNw", false, "clzDNg", "clsDNg"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 10, mask = 0x7ff,
			
 
				+      [0] = "rbitDNg", "rev16DNg", "rev32DNx", "revDNx", "clzDNg", "clsDNg"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_data2s = { -- Data processing, 2 sources.
			
 
				+  shift = 29, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 10, mask = 63,
			
 
				+    false, "udivDNMg", "sdivDNMg", false, false, false, false, "lslDNMg",
			
 
				+    "lsrDNMg", "asrDNMg", "rorDNMg"
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_data3s = { -- Data processing, 3 sources.
			
 
				+  shift = 29, mask = 7,
			
 
				+  [0] = {
			
 
				+    shift = 21, mask = 7,
			
 
				+    [0] = {
			
 
				+      shift = 15, mask = 1,
			
 
				+      [0] = "madd|mulDNMA0g", "msub|mnegDNMA0g"
			
 
				+    }
			
 
				+  }, false, false, false,
			
 
				+  {
			
 
				+    shift = 15, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "madd|mulDNMA0g", "smaddl|smullDxNMwA0x", "smulhDNMx", false,
			
 
				+      false, "umaddl|umullDxNMwA0x", "umulhDNMx"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 21, mask = 7,
			
 
				+      [0] = "msub|mnegDNMA0g", "smsubl|smneglDxNMwA0x", false, false,
			
 
				+      false, "umsubl|umneglDxNMwA0x"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_datar = { -- Data processing, register.
			
 
				+  shift = 28, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 24, mask = 1,
			
 
				+    [0] = map_logsr,
			
 
				+    {
			
 
				+      shift = 21, mask = 1,
			
 
				+      [0] = map_addsubsh, map_addsubex
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 21, mask = 15,
			
 
				+    [0] = map_addsubc, false, map_ccomp, false, map_csel, false,
			
 
				+    {
			
 
				+      shift = 30, mask = 1,
			
 
				+      [0] = map_data2s, map_data1s
			
 
				+    },
			
 
				+    false, map_data3s, map_data3s, map_data3s, map_data3s, map_data3s,
			
 
				+    map_data3s, map_data3s, map_data3s
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_lrl = { -- Load register, literal.
			
 
				+  shift = 26, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 30, mask = 3,
			
 
				+    [0] = "ldrDwB", "ldrDxB", "ldrswDxB"
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 30, mask = 3,
			
 
				+    [0] = "ldrDsB", "ldrDdB"
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_lsriind = { -- Load/store register, immediate pre/post-indexed.
			
 
				+  shift = 30, mask = 3,
			
 
				+  [0] = {
			
 
				+    shift = 26, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strbDwzL", "ldrbDwzL", "ldrsbDxzL", "ldrsbDwzL"
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 26, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strhDwzL", "ldrhDwzL", "ldrshDxzL", "ldrshDwzL"
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 26, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strDwzL", "ldrDwzL", "ldrswDxzL"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strDszL", "ldrDszL"
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 26, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strDxzL", "ldrDxzL"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 22, mask = 3,
			
 
				+      [0] = "strDdzL", "ldrDdzL"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_lsriro = {
			
 
				+  shift = 21, mask = 1,
			
 
				+  [0] = {  -- Load/store register immediate.
			
 
				+    shift = 10, mask = 3,
			
 
				+    [0] = { -- Unscaled immediate.
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = {
			
 
				+	shift = 30, mask = 3,
			
 
				+	[0] = {
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "sturbDwK", "ldurbDwK"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "sturhDwK", "ldurhDwK"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "sturDwK", "ldurDwK"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "sturDxK", "ldurDxK"
			
 
				+	}
			
 
				+      }
			
 
				+    }, map_lsriind, false, map_lsriind
			
 
				+  },
			
 
				+  {  -- Load/store register, register offset.
			
 
				+    shift = 10, mask = 3,
			
 
				+    [2] = {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = {
			
 
				+	shift = 30, mask = 3,
			
 
				+	[0] = {
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strbDwO", "ldrbDwO", "ldrsbDxO", "ldrsbDwO"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strhDwO", "ldrhDwO", "ldrshDxO", "ldrshDwO"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strDwO", "ldrDwO", "ldrswDxO"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strDxO", "ldrDxO"
			
 
				+	}
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 30, mask = 3,
			
 
				+	[2] = {
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strDsO", "ldrDsO"
			
 
				+	},
			
 
				+	[3] = {
			
 
				+	  shift = 22, mask = 3,
			
 
				+	  [0] = "strDdO", "ldrDdO"
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_lsp = { -- Load/store register pair, offset.
			
 
				+  shift = 22, mask = 1,
			
 
				+  [0] = {
			
 
				+    shift = 30, mask = 3,
			
 
				+    [0] = {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = "stpDzAzwP", "stpDzAzsP",
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 26, mask = 1,
			
 
				+      "stpDzAzdP"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = "stpDzAzxP"
			
 
				+    }
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 30, mask = 3,
			
 
				+    [0] = {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = "ldpDzAzwP", "ldpDzAzsP",
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = "ldpswDAxP", "ldpDzAzdP"
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 26, mask = 1,
			
 
				+      [0] = "ldpDzAzxP"
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_ls = { -- Loads and stores.
			
 
				+  shift = 24, mask = 0x31,
			
 
				+  [0x10] = map_lrl, [0x30] = map_lsriro,
			
 
				+  [0x20] = {
			
 
				+    shift = 23, mask = 3,
			
 
				+    map_lsp, map_lsp, map_lsp
			
 
				+  },
			
 
				+  [0x21] = {
			
 
				+    shift = 23, mask = 3,
			
 
				+    map_lsp, map_lsp, map_lsp
			
 
				+  },
			
 
				+  [0x31] = {
			
 
				+    shift = 26, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 30, mask = 3,
			
 
				+      [0] = {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strbDwzU", "ldrbDwzU"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strhDwzU", "ldrhDwzU"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strDwzU", "ldrDwzU"
			
 
				+      },
			
 
				+      {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strDxzU", "ldrDxzU"
			
 
				+      }
			
 
				+    },
			
 
				+    {
			
 
				+      shift = 30, mask = 3,
			
 
				+      [2] = {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strDszU", "ldrDszU"
			
 
				+      },
			
 
				+      [3] = {
			
 
				+	shift = 22, mask = 3,
			
 
				+	[0] = "strDdzU", "ldrDdzU"
			
 
				+      }
			
 
				+    }
			
 
				+  },
			
 
				+}
			
 
				+
			
 
				+local map_datafp = { -- Data processing, SIMD and FP.
			
 
				+  shift = 28, mask = 7,
			
 
				+  { -- 001
			
 
				+    shift = 24, mask = 1,
			
 
				+    [0] = {
			
 
				+      shift = 21, mask = 1,
			
 
				+      {
			
 
				+	shift = 10, mask = 3,
			
 
				+	[0] = {
			
 
				+	  shift = 12, mask = 1,
			
 
				+	  [0] = {
			
 
				+	    shift = 13, mask = 1,
			
 
				+	    [0] = {
			
 
				+	      shift = 14, mask = 1,
			
 
				+	      [0] = {
			
 
				+		shift = 15, mask = 1,
			
 
				+		[0] = { -- FP/int conversion.
			
 
				+		  shift = 31, mask = 1,
			
 
				+		  [0] = {
			
 
				+		    shift = 16, mask = 0xff,
			
 
				+		    [0x20] = "fcvtnsDwNs", [0x21] = "fcvtnuDwNs",
			
 
				+		    [0x22] = "scvtfDsNw", [0x23] = "ucvtfDsNw",
			
 
				+		    [0x24] = "fcvtasDwNs", [0x25] = "fcvtauDwNs",
			
 
				+		    [0x26] = "fmovDwNs", [0x27] = "fmovDsNw",
			
 
				+		    [0x28] = "fcvtpsDwNs", [0x29] = "fcvtpuDwNs",
			
 
				+		    [0x30] = "fcvtmsDwNs", [0x31] = "fcvtmuDwNs",
			
 
				+		    [0x38] = "fcvtzsDwNs", [0x39] = "fcvtzuDwNs",
			
 
				+		    [0x60] = "fcvtnsDwNd", [0x61] = "fcvtnuDwNd",
			
 
				+		    [0x62] = "scvtfDdNw", [0x63] = "ucvtfDdNw",
			
 
				+		    [0x64] = "fcvtasDwNd", [0x65] = "fcvtauDwNd",
			
 
				+		    [0x68] = "fcvtpsDwNd", [0x69] = "fcvtpuDwNd",
			
 
				+		    [0x70] = "fcvtmsDwNd", [0x71] = "fcvtmuDwNd",
			
 
				+		    [0x78] = "fcvtzsDwNd", [0x79] = "fcvtzuDwNd"
			
 
				+		  },
			
 
				+		  {
			
 
				+		    shift = 16, mask = 0xff,
			
 
				+		    [0x20] = "fcvtnsDxNs", [0x21] = "fcvtnuDxNs",
			
 
				+		    [0x22] = "scvtfDsNx", [0x23] = "ucvtfDsNx",
			
 
				+		    [0x24] = "fcvtasDxNs", [0x25] = "fcvtauDxNs",
			
 
				+		    [0x28] = "fcvtpsDxNs", [0x29] = "fcvtpuDxNs",
			
 
				+		    [0x30] = "fcvtmsDxNs", [0x31] = "fcvtmuDxNs",
			
 
				+		    [0x38] = "fcvtzsDxNs", [0x39] = "fcvtzuDxNs",
			
 
				+		    [0x60] = "fcvtnsDxNd", [0x61] = "fcvtnuDxNd",
			
 
				+		    [0x62] = "scvtfDdNx", [0x63] = "ucvtfDdNx",
			
 
				+		    [0x64] = "fcvtasDxNd", [0x65] = "fcvtauDxNd",
			
 
				+		    [0x66] = "fmovDxNd", [0x67] = "fmovDdNx",
			
 
				+		    [0x68] = "fcvtpsDxNd", [0x69] = "fcvtpuDxNd",
			
 
				+		    [0x70] = "fcvtmsDxNd", [0x71] = "fcvtmuDxNd",
			
 
				+		    [0x78] = "fcvtzsDxNd", [0x79] = "fcvtzuDxNd"
			
 
				+		  }
			
 
				+		}
			
 
				+	      },
			
 
				+	      { -- FP data-processing, 1 source.
			
 
				+		shift = 31, mask = 1,
			
 
				+		[0] = {
			
 
				+		  shift = 22, mask = 3,
			
 
				+		  [0] = {
			
 
				+		    shift = 15, mask = 63,
			
 
				+		    [0] = "fmovDNf", "fabsDNf", "fnegDNf",
			
 
				+		    "fsqrtDNf", false, "fcvtDdNs", false, false,
			
 
				+		    "frintnDNf", "frintpDNf", "frintmDNf", "frintzDNf",
			
 
				+		    "frintaDNf", false, "frintxDNf", "frintiDNf",
			
 
				+		  },
			
 
				+		  {
			
 
				+		    shift = 15, mask = 63,
			
 
				+		    [0] = "fmovDNf", "fabsDNf", "fnegDNf",
			
 
				+		    "fsqrtDNf", "fcvtDsNd", false, false, false,
			
 
				+		    "frintnDNf", "frintpDNf", "frintmDNf", "frintzDNf",
			
 
				+		    "frintaDNf", false, "frintxDNf", "frintiDNf",
			
 
				+		  }
			
 
				+		}
			
 
				+	      }
			
 
				+	    },
			
 
				+	    { -- FP compare.
			
 
				+	      shift = 31, mask = 1,
			
 
				+	      [0] = {
			
 
				+		shift = 14, mask = 3,
			
 
				+		[0] = {
			
 
				+		  shift = 23, mask = 1,
			
 
				+		  [0] = {
			
 
				+		    shift = 0, mask = 31,
			
 
				+		    [0] = "fcmpNMf", [8] = "fcmpNZf",
			
 
				+		    [16] = "fcmpeNMf", [24] = "fcmpeNZf",
			
 
				+		  }
			
 
				+		}
			
 
				+	      }
			
 
				+	    }
			
 
				+	  },
			
 
				+	  { -- FP immediate.
			
 
				+	    shift = 31, mask = 1,
			
 
				+	    [0] = {
			
 
				+	      shift = 5, mask = 31,
			
 
				+	      [0] = {
			
 
				+		shift = 23, mask = 1,
			
 
				+		[0] = "fmovDFf"
			
 
				+	      }
			
 
				+	    }
			
 
				+	  }
			
 
				+	},
			
 
				+	{ -- FP conditional compare.
			
 
				+	  shift = 31, mask = 1,
			
 
				+	  [0] = {
			
 
				+	    shift = 23, mask = 1,
			
 
				+	    [0] = {
			
 
				+	      shift = 4, mask = 1,
			
 
				+	      [0] = "fccmpNMVCf", "fccmpeNMVCf"
			
 
				+	    }
			
 
				+	  }
			
 
				+	},
			
 
				+	{ -- FP data-processing, 2 sources.
			
 
				+	  shift = 31, mask = 1,
			
 
				+	  [0] = {
			
 
				+	    shift = 23, mask = 1,
			
 
				+	    [0] = {
			
 
				+	      shift = 12, mask = 15,
			
 
				+	      [0] = "fmulDNMf", "fdivDNMf", "faddDNMf", "fsubDNMf",
			
 
				+	      "fmaxDNMf", "fminDNMf", "fmaxnmDNMf", "fminnmDNMf",
			
 
				+	      "fnmulDNMf"
			
 
				+	    }
			
 
				+	  }
			
 
				+	},
			
 
				+	{ -- FP conditional select.
			
 
				+	  shift = 31, mask = 1,
			
 
				+	  [0] = {
			
 
				+	    shift = 23, mask = 1,
			
 
				+	    [0] = "fcselDNMCf"
			
 
				+	  }
			
 
				+	}
			
 
				+      }
			
 
				+    },
			
 
				+    { -- FP data-processing, 3 sources.
			
 
				+      shift = 31, mask = 1,
			
 
				+      [0] = {
			
 
				+	shift = 15, mask = 1,
			
 
				+	[0] = {
			
 
				+	  shift = 21, mask = 5,
			
 
				+	  [0] = "fmaddDNMAf", "fnmaddDNMAf"
			
 
				+	},
			
 
				+	{
			
 
				+	  shift = 21, mask = 5,
			
 
				+	  [0] = "fmsubDNMAf", "fnmsubDNMAf"
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_br = { -- Branches, exception generating and system instructions.
			
 
				+  shift = 29, mask = 7,
			
 
				+  [0] = "bB",
			
 
				+  { -- Compare & branch, immediate.
			
 
				+    shift = 24, mask = 3,
			
 
				+    [0] = "cbzDBg", "cbnzDBg", "tbzDTBw", "tbnzDTBw"
			
 
				+  },
			
 
				+  { -- Conditional branch, immediate.
			
 
				+    shift = 24, mask = 3,
			
 
				+    [0] = {
			
 
				+      shift = 4, mask = 1,
			
 
				+      [0] = {
			
 
				+	shift = 0, mask = 15,
			
 
				+	[0] = "beqB", "bneB", "bhsB", "bloB", "bmiB", "bplB", "bvsB", "bvcB",
			
 
				+	"bhiB", "blsB", "bgeB", "bltB", "bgtB", "bleB", "balB"
			
 
				+      }
			
 
				+    }
			
 
				+  }, false, "blB",
			
 
				+  { -- Compare & branch, immediate.
			
 
				+    shift = 24, mask = 3,
			
 
				+    [0] = "cbzDBg", "cbnzDBg", "tbzDTBx", "tbnzDTBx"
			
 
				+  },
			
 
				+  {
			
 
				+    shift = 24, mask = 3,
			
 
				+    [0] = { -- Exception generation.
			
 
				+      shift = 0, mask = 0xe0001f,
			
 
				+      [0x200000] = "brkW"
			
 
				+    },
			
 
				+    { -- System instructions.
			
 
				+      shift = 0, mask = 0x3fffff,
			
 
				+      [0x03201f] = "nop"
			
 
				+    },
			
 
				+    { -- Unconditional branch, register.
			
 
				+      shift = 0, mask = 0xfffc1f,
			
 
				+      [0x1f0000] = "brNx", [0x3f0000] = "blrNx",
			
 
				+      [0x5f0000] = "retNx"
			
 
				+    },
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+local map_init = {
			
 
				+  shift = 25, mask = 15,
			
 
				+  [0] = false, false, false, false, map_ls, map_datar, map_ls, map_datafp,
			
 
				+  map_datai, map_datai, map_br, map_br, map_ls, map_datar, map_ls, map_datafp
			
 
				+}
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local map_regs = { x = {}, w = {}, d = {}, s = {} }
			
 
				+
			
 
				+for i=0,30 do
			
 
				+  map_regs.x[i] = "x"..i
			
 
				+  map_regs.w[i] = "w"..i
			
 
				+  map_regs.d[i] = "d"..i
			
 
				+  map_regs.s[i] = "s"..i
			
 
				+end
			
 
				+map_regs.x[31] = "sp"
			
 
				+map_regs.w[31] = "wsp"
			
 
				+map_regs.d[31] = "d31"
			
 
				+map_regs.s[31] = "s31"
			
 
				+
			
 
				+local map_cond = {
			
 
				+  [0] = "eq", "ne", "cs", "cc", "mi", "pl", "vs", "vc",
			
 
				+  "hi", "ls", "ge", "lt", "gt", "le", "al",
			
 
				+}
			
 
				+
			
 
				+local map_shift = { [0] = "lsl", "lsr", "asr", }
			
 
				+
			
 
				+local map_extend = {
			
 
				+  [0] = "uxtb", "uxth", "uxtw", "uxtx", "sxtb", "sxth", "sxtw", "sxtx",
			
 
				+}
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Output a nicely formatted line with an opcode and operands.
			
 
				+local function putop(ctx, text, operands)
			
 
				+  local pos = ctx.pos
			
 
				+  local extra = ""
			
 
				+  if ctx.rel then
			
 
				+    local sym = ctx.symtab[ctx.rel]
			
 
				+    if sym then
			
 
				+      extra = "\t->"..sym
			
 
				+    end
			
 
				+  end
			
 
				+  if ctx.hexdump > 0 then
			
 
				+    ctx.out(format("%08x  %s  %-5s %s%s\n",
			
 
				+      ctx.addr+pos, tohex(ctx.op), text, concat(operands, ", "), extra))
			
 
				+  else
			
 
				+    ctx.out(format("%08x  %-5s %s%s\n",
			
 
				+      ctx.addr+pos, text, concat(operands, ", "), extra))
			
 
				+  end
			
 
				+  ctx.pos = pos + 4
			
 
				+end
			
 
				+
			
 
				+-- Fallback for unknown opcodes.
			
 
				+local function unknown(ctx)
			
 
				+  return putop(ctx, ".long", { "0x"..tohex(ctx.op) })
			
 
				+end
			
 
				+
			
 
				+local function match_reg(p, pat, regnum)
			
 
				+  return map_regs[match(pat, p.."%w-([xwds])")][regnum]
			
 
				+end
			
 
				+
			
 
				+local function fmt_hex32(x)
			
 
				+  if x < 0 then
			
 
				+    return tohex(x)
			
 
				+  else
			
 
				+    return format("%x", x)
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local imm13_rep = { 0x55555555, 0x11111111, 0x01010101, 0x00010001, 0x00000001 }
			
 
				+
			
 
				+local function decode_imm13(op)
			
 
				+  local imms = band(rshift(op, 10), 63)
			
 
				+  local immr = band(rshift(op, 16), 63)
			
 
				+  if band(op, 0x00400000) == 0 then
			
 
				+    local len = 5
			
 
				+    if imms >= 56 then
			
 
				+      if imms >= 60 then len = 1 else len = 2 end
			
 
				+    elseif imms >= 48 then len = 3 elseif imms >= 32 then len = 4 end
			
 
				+    local l = lshift(1, len)-1
			
 
				+    local s = band(imms, l)
			
 
				+    local r = band(immr, l)
			
 
				+    local imm = ror(rshift(-1, 31-s), r)
			
 
				+    if len ~= 5 then imm = band(imm, lshift(1, l)-1) + rshift(imm, 31-l) end
			
 
				+    imm = imm * imm13_rep[len]
			
 
				+    local ix = fmt_hex32(imm)
			
 
				+    if rshift(op, 31) ~= 0 then
			
 
				+      return ix..tohex(imm)
			
 
				+    else
			
 
				+      return ix
			
 
				+    end
			
 
				+  else
			
 
				+    local lo, hi = -1, 0
			
 
				+    if imms < 32 then lo = rshift(-1, 31-imms) else hi = rshift(-1, 63-imms) end
			
 
				+    if immr ~= 0 then
			
 
				+      lo, hi = ror(lo, immr), ror(hi, immr)
			
 
				+      local x = immr == 32 and 0 or band(bxor(lo, hi), lshift(-1, 32-immr))
			
 
				+      lo, hi = bxor(lo, x), bxor(hi, x)
			
 
				+      if immr >= 32 then lo, hi = hi, lo end
			
 
				+    end
			
 
				+    if hi ~= 0 then
			
 
				+      return fmt_hex32(hi)..tohex(lo)
			
 
				+    else
			
 
				+      return fmt_hex32(lo)
			
 
				+    end
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_immpc(op, name)
			
 
				+  if name == "b" or name == "bl" then
			
 
				+    return arshift(lshift(op, 6), 4)
			
 
				+  elseif name == "adr" or name == "adrp" then
			
 
				+    local immlo = band(rshift(op, 29), 3)
			
 
				+    local immhi = lshift(arshift(lshift(op, 8), 13), 2)
			
 
				+    return bor(immhi, immlo)
			
 
				+  elseif name == "tbz" or name == "tbnz" then
			
 
				+    return lshift(arshift(lshift(op, 13), 18), 2)
			
 
				+  else
			
 
				+    return lshift(arshift(lshift(op, 8), 13), 2)
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+local function parse_fpimm8(op)
			
 
				+  local sign = band(op, 0x100000) == 0 and 1 or -1
			
 
				+  local exp = bxor(rshift(arshift(lshift(op, 12), 5), 24), 0x80) - 131
			
 
				+  local frac = 16+band(rshift(op, 13), 15)
			
 
				+  return sign * frac * 2^exp
			
 
				+end
			
 
				+
			
 
				+local function prefer_bfx(sf, uns, imms, immr)
			
 
				+  if imms < immr or imms == 31 or imms == 63 then
			
 
				+    return false
			
 
				+  end
			
 
				+  if immr == 0 then
			
 
				+    if sf == 0 and (imms == 7 or imms == 15) then
			
 
				+      return false
			
 
				+    end
			
 
				+    if sf ~= 0 and uns == 0 and (imms == 7 or imms == 15 or imms == 31) then
			
 
				+      return false
			
 
				+    end
			
 
				+  end
			
 
				+  return true
			
 
				+end
			
 
				+
			
 
				+-- Disassemble a single instruction.
			
 
				+local function disass_ins(ctx)
			
 
				+  local pos = ctx.pos
			
 
				+  local b0, b1, b2, b3 = byte(ctx.code, pos+1, pos+4)
			
 
				+  local op = bor(lshift(b3, 24), lshift(b2, 16), lshift(b1, 8), b0)
			
 
				+  local operands = {}
			
 
				+  local suffix = ""
			
 
				+  local last, name, pat
			
 
				+  local map_reg
			
 
				+  ctx.op = op
			
 
				+  ctx.rel = nil
			
 
				+  last = nil
			
 
				+  local opat
			
 
				+  opat = map_init[band(rshift(op, 25), 15)]
			
 
				+  while type(opat) ~= "string" do
			
 
				+    if not opat then return unknown(ctx) end
			
 
				+    opat = opat[band(rshift(op, opat.shift), opat.mask)] or opat._
			
 
				+  end
			
 
				+  name, pat = match(opat, "^([a-z0-9]*)(.*)")
			
 
				+  local altname, pat2 = match(pat, "|([a-z0-9_.|]*)(.*)")
			
 
				+  if altname then pat = pat2 end
			
 
				+  if sub(pat, 1, 1) == "." then
			
 
				+    local s2, p2 = match(pat, "^([a-z0-9.]*)(.*)")
			
 
				+    suffix = suffix..s2
			
 
				+    pat = p2
			
 
				+  end
			
 
				+
			
 
				+  local rt = match(pat, "[gf]")
			
 
				+  if rt then
			
 
				+    if rt == "g" then
			
 
				+      map_reg = band(op, 0x80000000) ~= 0 and map_regs.x or map_regs.w
			
 
				+    else
			
 
				+      map_reg = band(op, 0x400000) ~= 0 and map_regs.d or map_regs.s
			
 
				+    end
			
 
				+  end
			
 
				+
			
 
				+  local second0, immr
			
 
				+
			
 
				+  for p in gmatch(pat, ".") do
			
 
				+    local x = nil
			
 
				+    if p == "D" then
			
 
				+      local regnum = band(op, 31)
			
 
				+      x = rt and map_reg[regnum] or match_reg(p, pat, regnum)
			
 
				+    elseif p == "N" then
			
 
				+      local regnum = band(rshift(op, 5), 31)
			
 
				+      x = rt and map_reg[regnum] or match_reg(p, pat, regnum)
			
 
				+    elseif p == "M" then
			
 
				+      local regnum = band(rshift(op, 16), 31)
			
 
				+      x = rt and map_reg[regnum] or match_reg(p, pat, regnum)
			
 
				+    elseif p == "A" then
			
 
				+      local regnum = band(rshift(op, 10), 31)
			
 
				+      x = rt and map_reg[regnum] or match_reg(p, pat, regnum)
			
 
				+    elseif p == "B" then
			
 
				+      local addr = ctx.addr + pos + parse_immpc(op, name)
			
 
				+      ctx.rel = addr
			
 
				+      x = "0x"..tohex(addr)
			
 
				+    elseif p == "T" then
			
 
				+      x = bor(band(rshift(op, 26), 32), band(rshift(op, 19), 31))
			
 
				+    elseif p == "V" then
			
 
				+      x = band(op, 15)
			
 
				+    elseif p == "C" then
			
 
				+      x = map_cond[band(rshift(op, 12), 15)]
			
 
				+    elseif p == "c" then
			
 
				+      local rn = band(rshift(op, 5), 31)
			
 
				+      local rm = band(rshift(op, 16), 31)
			
 
				+      local cond = band(rshift(op, 12), 15)
			
 
				+      local invc = bxor(cond, 1)
			
 
				+      x = map_cond[cond]
			
 
				+      if altname and cond ~= 14 and cond ~= 15 then
			
 
				+	local a1, a2 = match(altname, "([^|]*)|(.*)")
			
 
				+	if rn == rm then
			
 
				+	  local n = #operands
			
 
				+	  operands[n] = nil
			
 
				+	  x = map_cond[invc]
			
 
				+	  if rn ~= 31 then
			
 
				+	    if a1 then name = a1 else name = altname end
			
 
				+	  else
			
 
				+	    operands[n-1] = nil
			
 
				+	    name = a2
			
 
				+	  end
			
 
				+	end
			
 
				+      end
			
 
				+    elseif p == "W" then
			
 
				+      x = band(rshift(op, 5), 0xffff)
			
 
				+    elseif p == "Y" then
			
 
				+      x = band(rshift(op, 5), 0xffff)
			
 
				+      local hw = band(rshift(op, 21), 3)
			
 
				+      if altname and (hw == 0 or x ~= 0) then
			
 
				+	name = altname
			
 
				+      end
			
 
				+    elseif p == "L" then
			
 
				+      local rn = map_regs.x[band(rshift(op, 5), 31)]
			
 
				+      local imm9 = arshift(lshift(op, 11), 23)
			
 
				+      if band(op, 0x800) ~= 0 then
			
 
				+	x = "["..rn..", #"..imm9.."]!"
			
 
				+      else
			
 
				+	x = "["..rn.."], #"..imm9
			
 
				+      end
			
 
				+    elseif p == "U" then
			
 
				+      local rn = map_regs.x[band(rshift(op, 5), 31)]
			
 
				+      local sz = band(rshift(op, 30), 3)
			
 
				+      local imm12 = lshift(arshift(lshift(op, 10), 20), sz)
			
 
				+      if imm12 ~= 0 then
			
 
				+	x = "["..rn..", #"..imm12.."]"
			
 
				+      else
			
 
				+	x = "["..rn.."]"
			
 
				+      end
			
 
				+    elseif p == "K" then
			
 
				+      local rn = map_regs.x[band(rshift(op, 5), 31)]
			
 
				+      local imm9 = arshift(lshift(op, 11), 23)
			
 
				+      if imm9 ~= 0 then
			
 
				+	x = "["..rn..", #"..imm9.."]"
			
 
				+      else
			
 
				+	x = "["..rn.."]"
			
 
				+      end
			
 
				+    elseif p == "O" then
			
 
				+      local rn, rm = map_regs.x[band(rshift(op, 5), 31)]
			
 
				+      local m = band(rshift(op, 13), 1)
			
 
				+      if m == 0 then
			
 
				+	rm = map_regs.w[band(rshift(op, 16), 31)]
			
 
				+      else
			
 
				+	rm = map_regs.x[band(rshift(op, 16), 31)]
			
 
				+      end
			
 
				+      x = "["..rn..", "..rm
			
 
				+      local opt = band(rshift(op, 13), 7)
			
 
				+      local s = band(rshift(op, 12), 1)
			
 
				+      local sz = band(rshift(op, 30), 3)
			
 
				+      -- extension to be applied
			
 
				+      if opt == 3 then
			
 
				+       if s == 0 then x = x.."]"
			
 
				+       else x = x..", lsl #"..sz.."]" end
			
 
				+      elseif opt == 2 or opt == 6 or opt == 7 then
			
 
				+	if s == 0 then x = x..", "..map_extend[opt].."]"
			
 
				+	else x = x..", "..map_extend[opt].." #"..sz.."]" end
			
 
				+      else
			
 
				+	x = x.."]"
			
 
				+      end
			
 
				+    elseif p == "P" then
			
 
				+      local opcv, sh = rshift(op, 26), 2
			
 
				+      if opcv >= 0x2a then sh = 4 elseif opcv >= 0x1b then sh = 3 end
			
 
				+      local imm7 = lshift(arshift(lshift(op, 10), 25), sh)
			
 
				+      local rn = map_regs.x[band(rshift(op, 5), 31)]
			
 
				+      local ind = band(rshift(op, 23), 3)
			
 
				+      if ind == 1 then
			
 
				+	x = "["..rn.."], #"..imm7
			
 
				+      elseif ind == 2 then
			
 
				+	if imm7 == 0 then
			
 
				+	  x = "["..rn.."]"
			
 
				+	else
			
 
				+	  x = "["..rn..", #"..imm7.."]"
			
 
				+	end
			
 
				+      elseif ind == 3 then
			
 
				+	x = "["..rn..", #"..imm7.."]!"
			
 
				+      end
			
 
				+    elseif p == "I" then
			
 
				+      local shf = band(rshift(op, 22), 3)
			
 
				+      local imm12 = band(rshift(op, 10), 0x0fff)
			
 
				+      local rn, rd = band(rshift(op, 5), 31), band(op, 31)
			
 
				+      if altname == "mov" and shf == 0 and imm12 == 0 and (rn == 31 or rd == 31) then
			
 
				+	name = altname
			
 
				+	x = nil
			
 
				+      elseif shf == 0 then
			
 
				+	x = imm12
			
 
				+      elseif shf == 1 then
			
 
				+	x = imm12..", lsl #12"
			
 
				+      end
			
 
				+    elseif p == "i" then
			
 
				+      x = "#0x"..decode_imm13(op)
			
 
				+    elseif p == "1" then
			
 
				+      immr = band(rshift(op, 16), 63)
			
 
				+      x = immr
			
 
				+    elseif p == "2" then
			
 
				+      x = band(rshift(op, 10), 63)
			
 
				+      if altname then
			
 
				+	local a1, a2, a3, a4, a5, a6 =
			
 
				+	  match(altname, "([^|]*)|([^|]*)|([^|]*)|([^|]*)|([^|]*)|(.*)")
			
 
				+	local sf = band(rshift(op, 26), 32)
			
 
				+	local uns = band(rshift(op, 30), 1)
			
 
				+	if prefer_bfx(sf, uns, x, immr) then
			
 
				+	  name = a2
			
 
				+	  x = x - immr + 1
			
 
				+	elseif immr == 0 and x == 7 then
			
 
				+	  local n = #operands
			
 
				+	  operands[n] = nil
			
 
				+	  if sf ~= 0 then
			
 
				+	    operands[n-1] = gsub(operands[n-1], "x", "w")
			
 
				+	  end
			
 
				+	  last = operands[n-1]
			
 
				+	  name = a6
			
 
				+	  x = nil
			
 
				+	elseif immr == 0 and x == 15 then
			
 
				+	  local n = #operands
			
 
				+	  operands[n] = nil
			
 
				+	  if sf ~= 0 then
			
 
				+	    operands[n-1] = gsub(operands[n-1], "x", "w")
			
 
				+	  end
			
 
				+	  last = operands[n-1]
			
 
				+	  name = a5
			
 
				+	  x = nil
			
 
				+	elseif x == 31 or x == 63 then
			
 
				+	  if x == 31 and immr == 0 and name == "sbfm" then
			
 
				+	    name = a4
			
 
				+	    local n = #operands
			
 
				+	    operands[n] = nil
			
 
				+	    if sf ~= 0 then
			
 
				+	      operands[n-1] = gsub(operands[n-1], "x", "w")
			
 
				+	    end
			
 
				+	    last = operands[n-1]
			
 
				+	  else
			
 
				+	    name = a3
			
 
				+	  end
			
 
				+	  x = nil
			
 
				+	elseif band(x, 31) ~= 31 and immr == x+1 and name == "ubfm" then
			
 
				+	  name = a4
			
 
				+	  last = "#"..(sf+32 - immr)
			
 
				+	  operands[#operands] = last
			
 
				+	  x = nil
			
 
				+	elseif x < immr then
			
 
				+	  name = a1
			
 
				+	  last = "#"..(sf+32 - immr)
			
 
				+	  operands[#operands] = last
			
 
				+	  x = x + 1
			
 
				+	end
			
 
				+      end
			
 
				+    elseif p == "3" then
			
 
				+      x = band(rshift(op, 10), 63)
			
 
				+      if altname then
			
 
				+	local a1, a2 = match(altname, "([^|]*)|(.*)")
			
 
				+	if x < immr then
			
 
				+	  name = a1
			
 
				+	  local sf = band(rshift(op, 26), 32)
			
 
				+	  last = "#"..(sf+32 - immr)
			
 
				+	  operands[#operands] = last
			
 
				+	  x = x + 1
			
 
				+	elseif x >= immr then
			
 
				+	  name = a2
			
 
				+	  x = x - immr + 1
			
 
				+	end
			
 
				+      end
			
 
				+    elseif p == "4" then
			
 
				+      x = band(rshift(op, 10), 63)
			
 
				+      local rn = band(rshift(op, 5), 31)
			
 
				+      local rm = band(rshift(op, 16), 31)
			
 
				+      if altname and rn == rm then
			
 
				+	local n = #operands
			
 
				+	operands[n] = nil
			
 
				+	last = operands[n-1]
			
 
				+	name = altname
			
 
				+      end
			
 
				+    elseif p == "5" then
			
 
				+      x = band(rshift(op, 16), 31)
			
 
				+    elseif p == "S" then
			
 
				+      x = band(rshift(op, 10), 63)
			
 
				+      if x == 0 then x = nil
			
 
				+      else x = map_shift[band(rshift(op, 22), 3)].." #"..x end
			
 
				+    elseif p == "X" then
			
 
				+      local opt = band(rshift(op, 13), 7)
			
 
				+      -- Width specifier <R>.
			
 
				+      if opt ~= 3 and opt ~= 7 then
			
 
				+	last = map_regs.w[band(rshift(op, 16), 31)]
			
 
				+	operands[#operands] = last
			
 
				+      end
			
 
				+      x = band(rshift(op, 10), 7)
			
 
				+      -- Extension.
			
 
				+      if opt == 2 + band(rshift(op, 31), 1) and
			
 
				+	 band(rshift(op, second0 and 5 or 0), 31) == 31 then
			
 
				+	if x == 0 then x = nil
			
 
				+	else x = "lsl #"..x end
			
 
				+      else
			
 
				+	if x == 0 then x = map_extend[band(rshift(op, 13), 7)]
			
 
				+	else x = map_extend[band(rshift(op, 13), 7)].." #"..x end
			
 
				+      end
			
 
				+    elseif p == "R" then
			
 
				+      x = band(rshift(op,21), 3)
			
 
				+      if x == 0 then x = nil
			
 
				+      else x = "lsl #"..x*16 end
			
 
				+    elseif p == "z" then
			
 
				+      local n = #operands
			
 
				+      if operands[n] == "sp" then operands[n] = "xzr"
			
 
				+      elseif operands[n] == "wsp" then operands[n] = "wzr"
			
 
				+      end
			
 
				+    elseif p == "Z" then
			
 
				+      x = 0
			
 
				+    elseif p == "F" then
			
 
				+      x = parse_fpimm8(op)
			
 
				+    elseif p == "g" or p == "f" or p == "x" or p == "w" or
			
 
				+	   p == "d" or p == "s" then
			
 
				+      -- These are handled in D/N/M/A.
			
 
				+    elseif p == "0" then
			
 
				+      if last == "sp" or last == "wsp" then
			
 
				+	local n = #operands
			
 
				+	operands[n] = nil
			
 
				+	last = operands[n-1]
			
 
				+	if altname then
			
 
				+	  local a1, a2 = match(altname, "([^|]*)|(.*)")
			
 
				+	  if not a1 then
			
 
				+	    name = altname
			
 
				+	  elseif second0 then
			
 
				+	    name, altname = a2, a1
			
 
				+	  else
			
 
				+	    name, altname = a1, a2
			
 
				+	  end
			
 
				+	end
			
 
				+      end
			
 
				+      second0 = true
			
 
				+    else
			
 
				+      assert(false)
			
 
				+    end
			
 
				+    if x then
			
 
				+      last = x
			
 
				+      if type(x) == "number" then x = "#"..x end
			
 
				+      operands[#operands+1] = x
			
 
				+    end
			
 
				+  end
			
 
				+
			
 
				+  return putop(ctx, name..suffix, operands)
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Disassemble a block of code.
			
 
				+local function disass_block(ctx, ofs, len)
			
 
				+  if not ofs then ofs = 0 end
			
 
				+  local stop = len and ofs+len or #ctx.code
			
 
				+  ctx.pos = ofs
			
 
				+  ctx.rel = nil
			
 
				+  while ctx.pos < stop do disass_ins(ctx) end
			
 
				+end
			
 
				+
			
 
				+-- Extended API: create a disassembler context. Then call ctx:disass(ofs, len).
			
 
				+local function create(code, addr, out)
			
 
				+  local ctx = {}
			
 
				+  ctx.code = code
			
 
				+  ctx.addr = addr or 0
			
 
				+  ctx.out = out or io.write
			
 
				+  ctx.symtab = {}
			
 
				+  ctx.disass = disass_block
			
 
				+  ctx.hexdump = 8
			
 
				+  return ctx
			
 
				+end
			
 
				+
			
 
				+-- Simple API: disassemble code (a string) at address and output via out.
			
 
				+local function disass(code, addr, out)
			
 
				+  create(code, addr, out):disass()
			
 
				+end
			
 
				+
			
 
				+-- Return register name for RID.
			
 
				+local function regname(r)
			
 
				+  if r < 32 then return map_regs.x[r] end
			
 
				+  return map_regs.d[r-32]
			
 
				+end
			
 
				+
			
 
				+-- Public module functions.
			
 
				+return {
			
 
				+  create = create,
			
 
				+  disass = disass,
			
 
				+  regname = regname
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/dis_arm64be.lua
+++ b/luajit.mod/luajit/src/jit/dis_arm64be.lua
@@ -0,0 +1,12 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT ARM64BE disassembler wrapper module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+-- ARM64 instructions are always little-endian. So just forward to the
			
 
				+-- common ARM64 disassembler module. All the interesting stuff is there.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+return require((string.match(..., ".*%.") or "").."dis_arm64")
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/dis_mips.lua
+++ b/luajit.mod/luajit/src/jit/dis_mips.lua
@@ -34,15 +34,17 @@ local map_special = {
 
				   "jrS",	"jalrD1S",	"movzDST",	"movnDST",
			
 
				   "syscallY",	"breakY",	false,		"sync",
			
 
				   "mfhiD",	"mthiS",	"mfloD",	"mtloS",
			
 
				-  false,	false,		false,		false,
			
 
				+  "dsllvDST",	false,		"dsrlvDST",	"dsravDST",
			
 
				   "multST",	"multuST",	"divST",	"divuST",
			
 
				-  false,	false,		false,		false,
			
 
				+  "dmultST",	"dmultuST",	"ddivST",	"ddivuST",
			
 
				   "addDST",	"addu|moveDST0", "subDST",	"subu|neguDS0T",
			
 
				-  "andDST",	"orDST",	"xorDST",	"nor|notDST0",
			
 
				+  "andDST",	"or|moveDST0",	"xorDST",	"nor|notDST0",
			
 
				   false,	false,		"sltDST",	"sltuDST",
			
 
				-  false,	false,		false,		false,
			
 
				+  "daddDST",	"dadduDST",	"dsubDST",	"dsubuDST",
			
 
				   "tgeSTZ",	"tgeuSTZ",	"tltSTZ",	"tltuSTZ",
			
 
				-  "teqSTZ",	false,		"tneSTZ",
			
 
				+  "teqSTZ",	false,		"tneSTZ",	false,
			
 
				+  "dsllDTA",	false,		"dsrlDTA",	"dsraDTA",
			
 
				+  "dsll32DTA",	false,		"dsrl32DTA",	"dsra32DTA",
			
 
				 }
			
 
				 
			
 
				 local map_special2 = {
			
@@ -60,11 +62,17 @@ local map_bshfl = {
 
				   [24] = "sehDT",
			
 
				 }
			
 
				 
			
 
				+local map_dbshfl = {
			
 
				+  shift = 6, mask = 31,
			
 
				+  [2] = "dsbhDT",
			
 
				+  [5] = "dshdDT",
			
 
				+}
			
 
				+
			
 
				 local map_special3 = {
			
 
				   shift = 0, mask = 63,
			
 
				-  [0] = "extTSAK", [4] = "insTSAL",
			
 
				-  [32] = map_bshfl,
			
 
				-  [59] = "rdhwrTD",
			
 
				+  [0]  = "extTSAK", [1]  = "dextmTSAP", [3]  = "dextTSAK",
			
 
				+  [4]  = "insTSAL", [6]  = "dinsuTSEQ", [7]  = "dinsTSAL",
			
 
				+  [32] = map_bshfl, [36] = map_dbshfl,  [59] = "rdhwrTD",
			
 
				 }
			
 
				 
			
 
				 local map_regimm = {
			
@@ -178,8 +186,8 @@ local map_cop1bc = {
 
				 
			
 
				 local map_cop1 = {
			
 
				   shift = 21, mask = 31,
			
 
				-  [0] = "mfc1TG", false,	"cfc1TG",	"mfhc1TG",
			
 
				-  "mtc1TG",	false,		"ctc1TG",	"mthc1TG",
			
 
				+  [0] = "mfc1TG", "dmfc1TG",	"cfc1TG",	"mfhc1TG",
			
 
				+  "mtc1TG",	"dmtc1TG",	"ctc1TG",	"mthc1TG",
			
 
				   map_cop1bc,	false,		false,		false,
			
 
				   false,	false,		false,		false,
			
 
				   map_cop1s,	map_cop1d,	false,		false,
			
@@ -213,16 +221,16 @@ local map_pri = {
 
				   "andiTSU",	"ori|liTS0U",	"xoriTSU",	"luiTU",
			
 
				   map_cop0,	map_cop1,	false,		map_cop1x,
			
 
				   "beql|beqzlST0B",	"bnel|bnezlST0B",	"blezlSB",	"bgtzlSB",
			
 
				-  false,	false,		false,		false,
			
 
				-  map_special2,	false,		false,		map_special3,
			
 
				+  "daddiTSI",	"daddiuTSI",	false,		false,
			
 
				+  map_special2,	"jalxJ",	false,		map_special3,
			
 
				   "lbTSO",	"lhTSO",	"lwlTSO",	"lwTSO",
			
 
				   "lbuTSO",	"lhuTSO",	"lwrTSO",	false,
			
 
				   "sbTSO",	"shTSO",	"swlTSO",	"swTSO",
			
 
				   false,	false,		"swrTSO",	"cacheNSO",
			
 
				   "llTSO",	"lwc1HSO",	"lwc2TSO",	"prefNSO",
			
 
				-  false,	"ldc1HSO",	"ldc2TSO",	false,
			
 
				+  false,	"ldc1HSO",	"ldc2TSO",	"ldTSO",
			
 
				   "scTSO",	"swc1HSO",	"swc2TSO",	false,
			
 
				-  false,	"sdc1HSO",	"sdc2TSO",	false,
			
 
				+  false,	"sdc1HSO",	"sdc2TSO",	"sdTSO",
			
 
				 }
			
 
				 
			
 
				 ------------------------------------------------------------------------------
			
@@ -306,6 +314,8 @@ local function disass_ins(ctx)
 
				       x = "f"..band(rshift(op, 21), 31)
			
 
				     elseif p == "A" then
			
 
				       x = band(rshift(op, 6), 31)
			
 
				+    elseif p == "E" then
			
 
				+      x = band(rshift(op, 6), 31) + 32
			
 
				     elseif p == "M" then
			
 
				       x = band(rshift(op, 11), 31)
			
 
				     elseif p == "N" then
			
@@ -315,8 +325,12 @@ local function disass_ins(ctx)
 
				       if x == 0 then x = nil end
			
 
				     elseif p == "K" then
			
 
				       x = band(rshift(op, 11), 31) + 1
			
 
				+    elseif p == "P" then
			
 
				+      x = band(rshift(op, 11), 31) + 33
			
 
				     elseif p == "L" then
			
 
				       x = band(rshift(op, 11), 31) - last + 1
			
 
				+    elseif p == "Q" then
			
 
				+      x = band(rshift(op, 11), 31) - last + 33
			
 
				     elseif p == "I" then
			
 
				       x = arshift(lshift(op, 16), 16)
			
 
				     elseif p == "U" then
			
@@ -330,11 +344,12 @@ local function disass_ins(ctx)
 
				     elseif p == "B" then
			
 
				       x = ctx.addr + ctx.pos + arshift(lshift(op, 16), 16)*4 + 4
			
 
				       ctx.rel = x
			
 
				-      x = "0x"..tohex(x)
			
 
				+      x = format("0x%08x", x)
			
 
				     elseif p == "J" then
			
 
				-      x = band(ctx.addr + ctx.pos, 0xf0000000) + band(op, 0x03ffffff)*4
			
 
				+      local a = ctx.addr + ctx.pos
			
 
				+      x = a - band(a, 0x0fffffff) + band(op, 0x03ffffff)*4
			
 
				       ctx.rel = x
			
 
				-      x = "0x"..tohex(x)
			
 
				+      x = format("0x%08x", x)
			
 
				     elseif p == "V" then
			
 
				       x = band(rshift(op, 8), 7)
			
 
				       if x == 0 then x = nil end
			
@@ -384,7 +399,7 @@ local function disass_block(ctx, ofs, len)
 
				 end
			
 
				 
			
 
				 -- Extended API: create a disassembler context. Then call ctx:disass(ofs, len).
			
 
				-local function create_(code, addr, out)
			
 
				+local function create(code, addr, out)
			
 
				   local ctx = {}
			
 
				   ctx.code = code
			
 
				   ctx.addr = addr or 0
			
@@ -396,33 +411,33 @@ local function create_(code, addr, out)
 
				   return ctx
			
 
				 end
			
 
				 
			
 
				-local function create_el_(code, addr, out)
			
 
				-  local ctx = create_(code, addr, out)
			
 
				+local function create_el(code, addr, out)
			
 
				+  local ctx = create(code, addr, out)
			
 
				   ctx.get = get_le
			
 
				   return ctx
			
 
				 end
			
 
				 
			
 
				 -- Simple API: disassemble code (a string) at address and output via out.
			
 
				-local function disass_(code, addr, out)
			
 
				-  create_(code, addr, out):disass()
			
 
				+local function disass(code, addr, out)
			
 
				+  create(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				-local function disass_el_(code, addr, out)
			
 
				-  create_el_(code, addr, out):disass()
			
 
				+local function disass_el(code, addr, out)
			
 
				+  create_el(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				 -- Return register name for RID.
			
 
				-local function regname_(r)
			
 
				+local function regname(r)
			
 
				   if r < 32 then return map_gpr[r] end
			
 
				   return "f"..(r-32)
			
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-create = create_
			
 
				-create_el = create_el_
			
 
				-disass = disass_
			
 
				-disass_el = disass_el_
			
 
				-regname = regname_
			
 
				+return {
			
 
				+  create = create,
			
 
				+  create_el = create_el,
			
 
				+  disass = disass,
			
 
				+  disass_el = disass_el,
			
 
				+  regname = regname
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_mips64.lua
+++ b/luajit.mod/luajit/src/jit/dis_mips64.lua
@@ -0,0 +1,17 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT MIPS64 disassembler wrapper module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+-- This module just exports the big-endian functions from the
			
 
				+-- MIPS disassembler module. All the interesting stuff is there.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local dis_mips = require((string.match(..., ".*%.") or "").."dis_mips")
			
 
				+return {
			
 
				+  create = dis_mips.create,
			
 
				+  disass = dis_mips.disass,
			
 
				+  regname = dis_mips.regname
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/dis_mips64el.lua
+++ b/luajit.mod/luajit/src/jit/dis_mips64el.lua
@@ -0,0 +1,17 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT MIPS64EL disassembler wrapper module.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+-- This module just exports the little-endian functions from the
			
 
				+-- MIPS disassembler module. All the interesting stuff is there.
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local dis_mips = require((string.match(..., ".*%.") or "").."dis_mips")
			
 
				+return {
			
 
				+  create = dis_mips.create_el,
			
 
				+  disass = dis_mips.disass_el,
			
 
				+  regname = dis_mips.regname
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/dis_mipsel.lua
+++ b/luajit.mod/luajit/src/jit/dis_mipsel.lua
@@ -8,13 +8,10 @@
 
				 -- MIPS disassembler module. All the interesting stuff is there.
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				-local require = require
			
 
				-
			
 
				-module(...)
			
 
				-
			
 
				-local dis_mips = require(_PACKAGE.."dis_mips")
			
 
				-
			
 
				-create = dis_mips.create_el
			
 
				-disass = dis_mips.disass_el
			
 
				-regname = dis_mips.regname
			
 
				+local dis_mips = require((string.match(..., ".*%.") or "").."dis_mips")
			
 
				+return {
			
 
				+  create = dis_mips.create_el,
			
 
				+  disass = dis_mips.disass_el,
			
 
				+  regname = dis_mips.regname
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_ppc.lua
+++ b/luajit.mod/luajit/src/jit/dis_ppc.lua
@@ -560,7 +560,7 @@ local function disass_block(ctx, ofs, len)
 
				 end
			
 
				 
			
 
				 -- Extended API: create a disassembler context. Then call ctx:disass(ofs, len).
			
 
				-local function create_(code, addr, out)
			
 
				+local function create(code, addr, out)
			
 
				   local ctx = {}
			
 
				   ctx.code = code
			
 
				   ctx.addr = addr or 0
			
@@ -572,20 +572,20 @@ local function create_(code, addr, out)
 
				 end
			
 
				 
			
 
				 -- Simple API: disassemble code (a string) at address and output via out.
			
 
				-local function disass_(code, addr, out)
			
 
				-  create_(code, addr, out):disass()
			
 
				+local function disass(code, addr, out)
			
 
				+  create(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				 -- Return register name for RID.
			
 
				-local function regname_(r)
			
 
				+local function regname(r)
			
 
				   if r < 32 then return map_gpr[r] end
			
 
				   return "f"..(r-32)
			
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-create = create_
			
 
				-disass = disass_
			
 
				-regname = regname_
			
 
				+return {
			
 
				+  create = create,
			
 
				+  disass = disass,
			
 
				+  regname = regname
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_x64.lua
+++ b/luajit.mod/luajit/src/jit/dis_x64.lua
@@ -8,13 +8,10 @@
 
				 -- x86/x64 disassembler module. All the interesting stuff is there.
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				-local require = require
			
 
				-
			
 
				-module(...)
			
 
				-
			
 
				-local dis_x86 = require(_PACKAGE.."dis_x86")
			
 
				-
			
 
				-create = dis_x86.create64
			
 
				-disass = dis_x86.disass64
			
 
				-regname = dis_x86.regname64
			
 
				+local dis_x86 = require((string.match(..., ".*%.") or "").."dis_x86")
			
 
				+return {
			
 
				+  create = dis_x86.create64,
			
 
				+  disass = dis_x86.disass64,
			
 
				+  regname = dis_x86.regname64
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dis_x86.lua
+++ b/luajit.mod/luajit/src/jit/dis_x86.lua
@@ -15,19 +15,20 @@
 
				 -- Intel and AMD manuals. The supported instruction set is quite extensive
			
 
				 -- and reflects what a current generation Intel or AMD CPU implements in
			
 
				 -- 32 bit and 64 bit mode. Yes, this includes MMX, SSE, SSE2, SSE3, SSSE3,
			
 
				--- SSE4.1, SSE4.2, SSE4a and even privileged and hypervisor (VMX/SVM)
			
 
				--- instructions.
			
 
				+-- SSE4.1, SSE4.2, SSE4a, AVX, AVX2 and even privileged and hypervisor
			
 
				+-- (VMX/SVM) instructions.
			
 
				 --
			
 
				 -- Notes:
			
 
				 -- * The (useless) a16 prefix, 3DNow and pre-586 opcodes are unsupported.
			
 
				 -- * No attempt at optimization has been made -- it's fast enough for my needs.
			
 
				--- * The public API may change when more architectures are added.
			
 
				 ------------------------------------------------------------------------------
			
 
				 
			
 
				 local type = type
			
 
				 local sub, byte, format = string.sub, string.byte, string.format
			
 
				 local match, gmatch, gsub = string.match, string.gmatch, string.gsub
			
 
				 local lower, rep = string.lower, string.rep
			
 
				+local bit = require("bit")
			
 
				+local tohex = bit.tohex
			
 
				 
			
 
				 -- Map for 1st opcode byte in 32 bit mode. Ugly? Well ... read on.
			
 
				 local map_opc1_32 = {
			
@@ -76,7 +77,7 @@ local map_opc1_32 = {
 
				 "movBRi","movBRi","movBRi","movBRi","movBRi","movBRi","movBRi","movBRi",
			
 
				 "movVRI","movVRI","movVRI","movVRI","movVRI","movVRI","movVRI","movVRI",
			
 
				 --Cx
			
 
				-"shift!Bmu","shift!Vmu","retBw","ret","$lesVrm","$ldsVrm","movBmi","movVmi",
			
 
				+"shift!Bmu","shift!Vmu","retBw","ret","vex*3$lesVrm","vex*2$ldsVrm","movBmi","movVmi",
			
 
				 "enterBwu","leave","retfBw","retf","int3","intBu","into","iretVS",
			
 
				 --Dx
			
 
				 "shift!Bm1","shift!Vm1","shift!Bmc","shift!Vmc","aamBu","aadBu","salc","xlatb",
			
@@ -101,7 +102,7 @@ local map_opc1_64 = setmetatable({
 
				   [0x44]="rex*r",  [0x45]="rex*rb",  [0x46]="rex*rx",  [0x47]="rex*rxb",
			
 
				   [0x48]="rex*w",  [0x49]="rex*wb",  [0x4a]="rex*wx",  [0x4b]="rex*wxb",
			
 
				   [0x4c]="rex*wr", [0x4d]="rex*wrb", [0x4e]="rex*wrx", [0x4f]="rex*wrxb",
			
 
				-  [0x82]=false, [0x9a]=false, [0xc4]=false, [0xc5]=false, [0xce]=false,
			
 
				+  [0x82]=false, [0x9a]=false, [0xc4]="vex*3", [0xc5]="vex*2", [0xce]=false,
			
 
				   [0xd4]=false, [0xd5]=false, [0xd6]=false, [0xea]=false,
			
 
				 }, { __index = map_opc1_32 })
			
 
				 
			
@@ -112,12 +113,12 @@ local map_opc2 = {
 
				 [0]="sldt!Dmp","sgdt!Ump","larVrm","lslVrm",nil,"syscall","clts","sysret",
			
 
				 "invd","wbinvd",nil,"ud1",nil,"$prefetch!Bm","femms","3dnowMrmu",
			
 
				 --1x
			
 
				-"movupsXrm|movssXrm|movupdXrm|movsdXrm",
			
 
				-"movupsXmr|movssXmr|movupdXmr|movsdXmr",
			
 
				+"movupsXrm|movssXrvm|movupdXrm|movsdXrvm",
			
 
				+"movupsXmr|movssXmvr|movupdXmr|movsdXmvr",
			
 
				 "movhlpsXrm$movlpsXrm|movsldupXrm|movlpdXrm|movddupXrm",
			
 
				 "movlpsXmr||movlpdXmr",
			
 
				-"unpcklpsXrm||unpcklpdXrm",
			
 
				-"unpckhpsXrm||unpckhpdXrm",
			
 
				+"unpcklpsXrvm||unpcklpdXrvm",
			
 
				+"unpckhpsXrvm||unpckhpdXrvm",
			
 
				 "movlhpsXrm$movhpsXrm|movshdupXrm|movhpdXrm",
			
 
				 "movhpsXmr||movhpdXmr",
			
 
				 "$prefetcht!Bm","hintnopVm","hintnopVm","hintnopVm",
			
@@ -126,7 +127,7 @@ local map_opc2 = {
 
				 "movUmx$","movUmy$","movUxm$","movUym$","movUmz$",nil,"movUzm$",nil,
			
 
				 "movapsXrm||movapdXrm",
			
 
				 "movapsXmr||movapdXmr",
			
 
				-"cvtpi2psXrMm|cvtsi2ssXrVmt|cvtpi2pdXrMm|cvtsi2sdXrVmt",
			
 
				+"cvtpi2psXrMm|cvtsi2ssXrvVmt|cvtpi2pdXrMm|cvtsi2sdXrvVmt",
			
 
				 "movntpsXmr|movntssXmr|movntpdXmr|movntsdXmr",
			
 
				 "cvttps2piMrXm|cvttss2siVrXm|cvttpd2piMrXm|cvttsd2siVrXm",
			
 
				 "cvtps2piMrXm|cvtss2siVrXm|cvtpd2piMrXm|cvtsd2siVrXm",
			
@@ -142,27 +143,27 @@ local map_opc2 = {
 
				 "cmovlVrm","cmovgeVrm","cmovleVrm","cmovgVrm",
			
 
				 --5x
			
 
				 "movmskpsVrXm$||movmskpdVrXm$","sqrtpsXrm|sqrtssXrm|sqrtpdXrm|sqrtsdXrm",
			
 
				-"rsqrtpsXrm|rsqrtssXrm","rcppsXrm|rcpssXrm",
			
 
				-"andpsXrm||andpdXrm","andnpsXrm||andnpdXrm",
			
 
				-"orpsXrm||orpdXrm","xorpsXrm||xorpdXrm",
			
 
				-"addpsXrm|addssXrm|addpdXrm|addsdXrm","mulpsXrm|mulssXrm|mulpdXrm|mulsdXrm",
			
 
				-"cvtps2pdXrm|cvtss2sdXrm|cvtpd2psXrm|cvtsd2ssXrm",
			
 
				+"rsqrtpsXrm|rsqrtssXrvm","rcppsXrm|rcpssXrvm",
			
 
				+"andpsXrvm||andpdXrvm","andnpsXrvm||andnpdXrvm",
			
 
				+"orpsXrvm||orpdXrvm","xorpsXrvm||xorpdXrvm",
			
 
				+"addpsXrvm|addssXrvm|addpdXrvm|addsdXrvm","mulpsXrvm|mulssXrvm|mulpdXrvm|mulsdXrvm",
			
 
				+"cvtps2pdXrm|cvtss2sdXrvm|cvtpd2psXrm|cvtsd2ssXrvm",
			
 
				 "cvtdq2psXrm|cvttps2dqXrm|cvtps2dqXrm",
			
 
				-"subpsXrm|subssXrm|subpdXrm|subsdXrm","minpsXrm|minssXrm|minpdXrm|minsdXrm",
			
 
				-"divpsXrm|divssXrm|divpdXrm|divsdXrm","maxpsXrm|maxssXrm|maxpdXrm|maxsdXrm",
			
 
				+"subpsXrvm|subssXrvm|subpdXrvm|subsdXrvm","minpsXrvm|minssXrvm|minpdXrvm|minsdXrvm",
			
 
				+"divpsXrvm|divssXrvm|divpdXrvm|divsdXrvm","maxpsXrvm|maxssXrvm|maxpdXrvm|maxsdXrvm",
			
 
				 --6x
			
 
				-"punpcklbwPrm","punpcklwdPrm","punpckldqPrm","packsswbPrm",
			
 
				-"pcmpgtbPrm","pcmpgtwPrm","pcmpgtdPrm","packuswbPrm",
			
 
				-"punpckhbwPrm","punpckhwdPrm","punpckhdqPrm","packssdwPrm",
			
 
				-"||punpcklqdqXrm","||punpckhqdqXrm",
			
 
				+"punpcklbwPrvm","punpcklwdPrvm","punpckldqPrvm","packsswbPrvm",
			
 
				+"pcmpgtbPrvm","pcmpgtwPrvm","pcmpgtdPrvm","packuswbPrvm",
			
 
				+"punpckhbwPrvm","punpckhwdPrvm","punpckhdqPrvm","packssdwPrvm",
			
 
				+"||punpcklqdqXrvm","||punpckhqdqXrvm",
			
 
				 "movPrVSm","movqMrm|movdquXrm|movdqaXrm",
			
 
				 --7x
			
 
				-"pshufwMrmu|pshufhwXrmu|pshufdXrmu|pshuflwXrmu","pshiftw!Pmu",
			
 
				-"pshiftd!Pmu","pshiftq!Mmu||pshiftdq!Xmu",
			
 
				-"pcmpeqbPrm","pcmpeqwPrm","pcmpeqdPrm","emms|",
			
 
				+"pshufwMrmu|pshufhwXrmu|pshufdXrmu|pshuflwXrmu","pshiftw!Pvmu",
			
 
				+"pshiftd!Pvmu","pshiftq!Mvmu||pshiftdq!Xvmu",
			
 
				+"pcmpeqbPrvm","pcmpeqwPrvm","pcmpeqdPrvm","emms*|",
			
 
				 "vmreadUmr||extrqXmuu$|insertqXrmuu$","vmwriteUrm||extrqXrm$|insertqXrm$",
			
 
				 nil,nil,
			
 
				-"||haddpdXrm|haddpsXrm","||hsubpdXrm|hsubpsXrm",
			
 
				+"||haddpdXrvm|haddpsXrvm","||hsubpdXrvm|hsubpsXrvm",
			
 
				 "movVSmMr|movqXrm|movVSmXr","movqMmr|movdquXmr|movdqaXmr",
			
 
				 --8x
			
 
				 "joVj","jnoVj","jbVj","jnbVj","jzVj","jnzVj","jbeVj","jaVj",
			
@@ -180,27 +181,27 @@ nil,nil,
 
				 "bsfVrm","bsrVrm|lzcntVrm|bsrWrm","movsxVrBmt","movsxVrWmt",
			
 
				 --Cx
			
 
				 "xaddBmr","xaddVmr",
			
 
				-"cmppsXrmu|cmpssXrmu|cmppdXrmu|cmpsdXrmu","$movntiVmr|",
			
 
				-"pinsrwPrWmu","pextrwDrPmu",
			
 
				-"shufpsXrmu||shufpdXrmu","$cmpxchg!Qmp",
			
 
				+"cmppsXrvmu|cmpssXrvmu|cmppdXrvmu|cmpsdXrvmu","$movntiVmr|",
			
 
				+"pinsrwPrvWmu","pextrwDrPmu",
			
 
				+"shufpsXrvmu||shufpdXrvmu","$cmpxchg!Qmp",
			
 
				 "bswapVR","bswapVR","bswapVR","bswapVR","bswapVR","bswapVR","bswapVR","bswapVR",
			
 
				 --Dx
			
 
				-"||addsubpdXrm|addsubpsXrm","psrlwPrm","psrldPrm","psrlqPrm",
			
 
				-"paddqPrm","pmullwPrm",
			
 
				+"||addsubpdXrvm|addsubpsXrvm","psrlwPrvm","psrldPrvm","psrlqPrvm",
			
 
				+"paddqPrvm","pmullwPrvm",
			
 
				 "|movq2dqXrMm|movqXmr|movdq2qMrXm$","pmovmskbVrMm||pmovmskbVrXm",
			
 
				-"psubusbPrm","psubuswPrm","pminubPrm","pandPrm",
			
 
				-"paddusbPrm","padduswPrm","pmaxubPrm","pandnPrm",
			
 
				+"psubusbPrvm","psubuswPrvm","pminubPrvm","pandPrvm",
			
 
				+"paddusbPrvm","padduswPrvm","pmaxubPrvm","pandnPrvm",
			
 
				 --Ex
			
 
				-"pavgbPrm","psrawPrm","psradPrm","pavgwPrm",
			
 
				-"pmulhuwPrm","pmulhwPrm",
			
 
				+"pavgbPrvm","psrawPrvm","psradPrvm","pavgwPrvm",
			
 
				+"pmulhuwPrvm","pmulhwPrvm",
			
 
				 "|cvtdq2pdXrm|cvttpd2dqXrm|cvtpd2dqXrm","$movntqMmr||$movntdqXmr",
			
 
				-"psubsbPrm","psubswPrm","pminswPrm","porPrm",
			
 
				-"paddsbPrm","paddswPrm","pmaxswPrm","pxorPrm",
			
 
				+"psubsbPrvm","psubswPrvm","pminswPrvm","porPrvm",
			
 
				+"paddsbPrvm","paddswPrvm","pmaxswPrvm","pxorPrvm",
			
 
				 --Fx
			
 
				-"|||lddquXrm","psllwPrm","pslldPrm","psllqPrm",
			
 
				-"pmuludqPrm","pmaddwdPrm","psadbwPrm","maskmovqMrm||maskmovdquXrm$",
			
 
				-"psubbPrm","psubwPrm","psubdPrm","psubqPrm",
			
 
				-"paddbPrm","paddwPrm","padddPrm","ud",
			
 
				+"|||lddquXrm","psllwPrvm","pslldPrvm","psllqPrvm",
			
 
				+"pmuludqPrvm","pmaddwdPrvm","psadbwPrvm","maskmovqMrm||maskmovdquXrm$",
			
 
				+"psubbPrvm","psubwPrvm","psubdPrvm","psubqPrvm",
			
 
				+"paddbPrvm","paddwPrvm","padddPrvm","ud",
			
 
				 }
			
 
				 assert(map_opc2[255] == "ud")
			
 
				 
			
@@ -208,49 +209,91 @@ assert(map_opc2[255] == "ud")
 
				 local map_opc3 = {
			
 
				 ["38"] = { -- [66] 0f 38 xx
			
 
				 --0x
			
 
				-[0]="pshufbPrm","phaddwPrm","phadddPrm","phaddswPrm",
			
 
				-"pmaddubswPrm","phsubwPrm","phsubdPrm","phsubswPrm",
			
 
				-"psignbPrm","psignwPrm","psigndPrm","pmulhrswPrm",
			
 
				-nil,nil,nil,nil,
			
 
				+[0]="pshufbPrvm","phaddwPrvm","phadddPrvm","phaddswPrvm",
			
 
				+"pmaddubswPrvm","phsubwPrvm","phsubdPrvm","phsubswPrvm",
			
 
				+"psignbPrvm","psignwPrvm","psigndPrvm","pmulhrswPrvm",
			
 
				+"||permilpsXrvm","||permilpdXrvm",nil,nil,
			
 
				 --1x
			
 
				 "||pblendvbXrma",nil,nil,nil,
			
 
				-"||blendvpsXrma","||blendvpdXrma",nil,"||ptestXrm",
			
 
				-nil,nil,nil,nil,
			
 
				+"||blendvpsXrma","||blendvpdXrma","||permpsXrvm","||ptestXrm",
			
 
				+"||broadcastssXrm","||broadcastsdXrm","||broadcastf128XrlXm",nil,
			
 
				 "pabsbPrm","pabswPrm","pabsdPrm",nil,
			
 
				 --2x
			
 
				 "||pmovsxbwXrm","||pmovsxbdXrm","||pmovsxbqXrm","||pmovsxwdXrm",
			
 
				 "||pmovsxwqXrm","||pmovsxdqXrm",nil,nil,
			
 
				-"||pmuldqXrm","||pcmpeqqXrm","||$movntdqaXrm","||packusdwXrm",
			
 
				-nil,nil,nil,nil,
			
 
				+"||pmuldqXrvm","||pcmpeqqXrvm","||$movntdqaXrm","||packusdwXrvm",
			
 
				+"||maskmovpsXrvm","||maskmovpdXrvm","||maskmovpsXmvr","||maskmovpdXmvr",
			
 
				 --3x
			
 
				 "||pmovzxbwXrm","||pmovzxbdXrm","||pmovzxbqXrm","||pmovzxwdXrm",
			
 
				-"||pmovzxwqXrm","||pmovzxdqXrm",nil,"||pcmpgtqXrm",
			
 
				-"||pminsbXrm","||pminsdXrm","||pminuwXrm","||pminudXrm",
			
 
				-"||pmaxsbXrm","||pmaxsdXrm","||pmaxuwXrm","||pmaxudXrm",
			
 
				+"||pmovzxwqXrm","||pmovzxdqXrm","||permdXrvm","||pcmpgtqXrvm",
			
 
				+"||pminsbXrvm","||pminsdXrvm","||pminuwXrvm","||pminudXrvm",
			
 
				+"||pmaxsbXrvm","||pmaxsdXrvm","||pmaxuwXrvm","||pmaxudXrvm",
			
 
				 --4x
			
 
				-"||pmulddXrm","||phminposuwXrm",
			
 
				+"||pmulddXrvm","||phminposuwXrm",nil,nil,
			
 
				+nil,"||psrlvVSXrvm","||psravdXrvm","||psllvVSXrvm",
			
 
				+--5x
			
 
				+[0x58] = "||pbroadcastdXrlXm",[0x59] = "||pbroadcastqXrlXm",
			
 
				+[0x5a] = "||broadcasti128XrlXm",
			
 
				+--7x
			
 
				+[0x78] = "||pbroadcastbXrlXm",[0x79] = "||pbroadcastwXrlXm",
			
 
				+--8x
			
 
				+[0x8c] = "||pmaskmovXrvVSm",
			
 
				+[0x8e] = "||pmaskmovVSmXvr",
			
 
				+--9x
			
 
				+[0x96] = "||fmaddsub132pHXrvm",[0x97] = "||fmsubadd132pHXrvm",
			
 
				+[0x98] = "||fmadd132pHXrvm",[0x99] = "||fmadd132sHXrvm",
			
 
				+[0x9a] = "||fmsub132pHXrvm",[0x9b] = "||fmsub132sHXrvm",
			
 
				+[0x9c] = "||fnmadd132pHXrvm",[0x9d] = "||fnmadd132sHXrvm",
			
 
				+[0x9e] = "||fnmsub132pHXrvm",[0x9f] = "||fnmsub132sHXrvm",
			
 
				+--Ax
			
 
				+[0xa6] = "||fmaddsub213pHXrvm",[0xa7] = "||fmsubadd213pHXrvm",
			
 
				+[0xa8] = "||fmadd213pHXrvm",[0xa9] = "||fmadd213sHXrvm",
			
 
				+[0xaa] = "||fmsub213pHXrvm",[0xab] = "||fmsub213sHXrvm",
			
 
				+[0xac] = "||fnmadd213pHXrvm",[0xad] = "||fnmadd213sHXrvm",
			
 
				+[0xae] = "||fnmsub213pHXrvm",[0xaf] = "||fnmsub213sHXrvm",
			
 
				+--Bx
			
 
				+[0xb6] = "||fmaddsub231pHXrvm",[0xb7] = "||fmsubadd231pHXrvm",
			
 
				+[0xb8] = "||fmadd231pHXrvm",[0xb9] = "||fmadd231sHXrvm",
			
 
				+[0xba] = "||fmsub231pHXrvm",[0xbb] = "||fmsub231sHXrvm",
			
 
				+[0xbc] = "||fnmadd231pHXrvm",[0xbd] = "||fnmadd231sHXrvm",
			
 
				+[0xbe] = "||fnmsub231pHXrvm",[0xbf] = "||fnmsub231sHXrvm",
			
 
				+--Dx
			
 
				+[0xdc] = "||aesencXrvm", [0xdd] = "||aesenclastXrvm",
			
 
				+[0xde] = "||aesdecXrvm", [0xdf] = "||aesdeclastXrvm",
			
 
				 --Fx
			
 
				 [0xf0] = "|||crc32TrBmt",[0xf1] = "|||crc32TrVmt",
			
 
				+[0xf7] = "| sarxVrmv| shlxVrmv| shrxVrmv",
			
 
				 },
			
 
				 
			
 
				 ["3a"] = { -- [66] 0f 3a xx
			
 
				 --0x
			
 
				-[0x00]=nil,nil,nil,nil,nil,nil,nil,nil,
			
 
				-"||roundpsXrmu","||roundpdXrmu","||roundssXrmu","||roundsdXrmu",
			
 
				-"||blendpsXrmu","||blendpdXrmu","||pblendwXrmu","palignrPrmu",
			
 
				+[0x00]="||permqXrmu","||permpdXrmu","||pblenddXrvmu",nil,
			
 
				+"||permilpsXrmu","||permilpdXrmu","||perm2f128Xrvmu",nil,
			
 
				+"||roundpsXrmu","||roundpdXrmu","||roundssXrvmu","||roundsdXrvmu",
			
 
				+"||blendpsXrvmu","||blendpdXrvmu","||pblendwXrvmu","palignrPrvmu",
			
 
				 --1x
			
 
				 nil,nil,nil,nil,
			
 
				 "||pextrbVmXru","||pextrwVmXru","||pextrVmSXru","||extractpsVmXru",
			
 
				-nil,nil,nil,nil,nil,nil,nil,nil,
			
 
				+"||insertf128XrvlXmu","||extractf128XlXmYru",nil,nil,
			
 
				+nil,nil,nil,nil,
			
 
				 --2x
			
 
				-"||pinsrbXrVmu","||insertpsXrmu","||pinsrXrVmuS",nil,
			
 
				+"||pinsrbXrvVmu","||insertpsXrvmu","||pinsrXrvVmuS",nil,
			
 
				+--3x
			
 
				+[0x38] = "||inserti128Xrvmu",[0x39] = "||extracti128XlXmYru",
			
 
				 --4x
			
 
				-[0x40] = "||dppsXrmu",
			
 
				-[0x41] = "||dppdXrmu",
			
 
				-[0x42] = "||mpsadbwXrmu",
			
 
				+[0x40] = "||dppsXrvmu",
			
 
				+[0x41] = "||dppdXrvmu",
			
 
				+[0x42] = "||mpsadbwXrvmu",
			
 
				+[0x44] = "||pclmulqdqXrvmu",
			
 
				+[0x46] = "||perm2i128Xrvmu",
			
 
				+[0x4a] = "||blendvpsXrvmb",[0x4b] = "||blendvpdXrvmb",
			
 
				+[0x4c] = "||pblendvbXrvmb",
			
 
				 --6x
			
 
				 [0x60] = "||pcmpestrmXrmu",[0x61] = "||pcmpestriXrmu",
			
 
				 [0x62] = "||pcmpistrmXrmu",[0x63] = "||pcmpistriXrmu",
			
 
				+[0xdf] = "||aeskeygenassistXrmu",
			
 
				+--Fx
			
 
				+[0xf0] = "||| rorxVrmu",
			
 
				 },
			
 
				 }
			
 
				 
			
@@ -354,17 +397,19 @@ local map_regs = {
 
				 	"mm0", "mm1", "mm2", "mm3", "mm4", "mm5", "mm6", "mm7" }, -- No x64 ext!
			
 
				   X = { "xmm0", "xmm1", "xmm2", "xmm3", "xmm4", "xmm5", "xmm6", "xmm7",
			
 
				 	"xmm8", "xmm9", "xmm10", "xmm11", "xmm12", "xmm13", "xmm14", "xmm15" },
			
 
				+  Y = { "ymm0", "ymm1", "ymm2", "ymm3", "ymm4", "ymm5", "ymm6", "ymm7",
			
 
				+	"ymm8", "ymm9", "ymm10", "ymm11", "ymm12", "ymm13", "ymm14", "ymm15" },
			
 
				 }
			
 
				 local map_segregs = { "es", "cs", "ss", "ds", "fs", "gs", "segr6", "segr7" }
			
 
				 
			
 
				 -- Maps for size names.
			
 
				 local map_sz2n = {
			
 
				-  B = 1, W = 2, D = 4, Q = 8, M = 8, X = 16,
			
 
				+  B = 1, W = 2, D = 4, Q = 8, M = 8, X = 16, Y = 32,
			
 
				 }
			
 
				 local map_sz2prefix = {
			
 
				   B = "byte", W = "word", D = "dword",
			
 
				   Q = "qword",
			
 
				-  M = "qword", X = "xword",
			
 
				+  M = "qword", X = "xword", Y = "yword",
			
 
				   F = "dword", G = "qword", -- No need for sizes/register names for these two.
			
 
				 }
			
 
				 
			
@@ -387,10 +432,13 @@ local function putop(ctx, text, operands)
 
				   if ctx.rep then text = ctx.rep.." "..text; ctx.rep = false end
			
 
				   if ctx.rex then
			
 
				     local t = (ctx.rexw and "w" or "")..(ctx.rexr and "r" or "")..
			
 
				-	      (ctx.rexx and "x" or "")..(ctx.rexb and "b" or "")
			
 
				-    if t ~= "" then text = "rex."..t.." "..text end
			
 
				+	      (ctx.rexx and "x" or "")..(ctx.rexb and "b" or "")..
			
 
				+	      (ctx.vexl and "l" or "")
			
 
				+    if ctx.vexv and ctx.vexv ~= 0 then t = t.."v"..ctx.vexv end
			
 
				+    if t ~= "" then text = ctx.rex.."."..t.." "..gsub(text, "^ ", "")
			
 
				+    elseif ctx.rex == "vex" then text = gsub("v"..text, "^v ", "") end
			
 
				     ctx.rexw = false; ctx.rexr = false; ctx.rexx = false; ctx.rexb = false
			
 
				-    ctx.rex = false
			
 
				+    ctx.rex = false; ctx.vexl = false; ctx.vexv = false
			
 
				   end
			
 
				   if ctx.seg then
			
 
				     local text2, n = gsub(text, "%[", "["..ctx.seg..":")
			
@@ -405,6 +453,7 @@ local function putop(ctx, text, operands)
 
				   end
			
 
				   ctx.out(format("%08x  %s%s\n", ctx.addr+ctx.start, hex, text))
			
 
				   ctx.mrm = false
			
 
				+  ctx.vexv = false
			
 
				   ctx.start = pos
			
 
				   ctx.imm = nil
			
 
				 end
			
@@ -413,7 +462,7 @@ end
 
				 local function clearprefixes(ctx)
			
 
				   ctx.o16 = false; ctx.seg = false; ctx.lock = false; ctx.rep = false
			
 
				   ctx.rexw = false; ctx.rexr = false; ctx.rexx = false; ctx.rexb = false
			
 
				-  ctx.rex = false; ctx.a32 = false
			
 
				+  ctx.rex = false; ctx.a32 = false; ctx.vexl = false
			
 
				 end
			
 
				 
			
 
				 -- Fallback for incomplete opcodes at the end.
			
@@ -450,9 +499,9 @@ end
 
				 -- Process pattern string and generate the operands.
			
 
				 local function putpat(ctx, name, pat)
			
 
				   local operands, regs, sz, mode, sp, rm, sc, rx, sdisp
			
 
				-  local code, pos, stop = ctx.code, ctx.pos, ctx.stop
			
 
				+  local code, pos, stop, vexl = ctx.code, ctx.pos, ctx.stop, ctx.vexl
			
 
				 
			
 
				-  -- Chars used: 1DFGIMPQRSTUVWXacdfgijmoprstuwxyz
			
 
				+  -- Chars used: 1DFGHIMPQRSTUVWXYabcdfgijlmoprstuvwxyz
			
 
				   for p in gmatch(pat, ".") do
			
 
				     local x = nil
			
 
				     if p == "V" or p == "U" then
			
@@ -467,12 +516,17 @@ local function putpat(ctx, name, pat)
 
				     elseif p == "B" then
			
 
				       sz = "B"
			
 
				       regs = ctx.rex and map_regs.B64 or map_regs.B
			
 
				-    elseif match(p, "[WDQMXFG]") then
			
 
				+    elseif match(p, "[WDQMXYFG]") then
			
 
				       sz = p
			
 
				+      if sz == "X" and vexl then sz = "Y"; ctx.vexl = false end
			
 
				       regs = map_regs[sz]
			
 
				     elseif p == "P" then
			
 
				       sz = ctx.o16 and "X" or "M"; ctx.o16 = false
			
 
				+      if sz == "X" and vexl then sz = "Y"; ctx.vexl = false end
			
 
				       regs = map_regs[sz]
			
 
				+    elseif p == "H" then
			
 
				+      name = name..(ctx.rexw and "d" or "s")
			
 
				+      ctx.rexw = false
			
 
				     elseif p == "S" then
			
 
				       name = name..lower(sz)
			
 
				     elseif p == "s" then
			
@@ -484,6 +538,10 @@ local function putpat(ctx, name, pat)
 
				       local imm = getimm(ctx, pos, 1); if not imm then return end
			
 
				       x = format("0x%02x", imm)
			
 
				       pos = pos+1
			
 
				+    elseif p == "b" then
			
 
				+      local imm = getimm(ctx, pos, 1); if not imm then return end
			
 
				+      x = regs[imm/16+1]
			
 
				+      pos = pos+1
			
 
				     elseif p == "w" then
			
 
				       local imm = getimm(ctx, pos, 2); if not imm then return end
			
 
				       x = format("0x%x", imm)
			
@@ -532,7 +590,7 @@ local function putpat(ctx, name, pat)
 
				 	local lo = imm % 0x1000000
			
 
				 	x = format("0x%02x%06x", (imm-lo) / 0x1000000, lo)
			
 
				       else
			
 
				-	x = format("0x%08x", imm)
			
 
				+	x = "0x"..tohex(imm)
			
 
				       end
			
 
				     elseif p == "R" then
			
 
				       local r = byte(code, pos-1, pos-1)%8
			
@@ -616,8 +674,13 @@ local function putpat(ctx, name, pat)
 
				 	else
			
 
				 	  x = "CR"..sp
			
 
				 	end
			
 
				+      elseif p == "v" then
			
 
				+	if ctx.vexv then
			
 
				+	  x = regs[ctx.vexv+1]; ctx.vexv = false
			
 
				+	end
			
 
				       elseif p == "y" then x = "DR"..sp
			
 
				       elseif p == "z" then x = "TR"..sp
			
 
				+      elseif p == "l" then vexl = false
			
 
				       elseif p == "t" then
			
 
				       else
			
 
				 	error("bad pattern `"..pat.."'")
			
@@ -692,7 +755,8 @@ map_act = {
 
				   B = putpat, W = putpat, D = putpat, Q = putpat,
			
 
				   V = putpat, U = putpat, T = putpat,
			
 
				   M = putpat, X = putpat, P = putpat,
			
 
				-  F = putpat, G = putpat,
			
 
				+  F = putpat, G = putpat, Y = putpat,
			
 
				+  H = putpat,
			
 
				 
			
 
				   -- Collect prefixes.
			
 
				   [":"] = function(ctx, name, pat)
			
@@ -753,15 +817,68 @@ map_act = {
 
				 
			
 
				   -- REX prefix.
			
 
				   rex = function(ctx, name, pat)
			
 
				-    if ctx.rex then return unknown(ctx) end -- Only 1 REX prefix allowed.
			
 
				+    if ctx.rex then return unknown(ctx) end -- Only 1 REX or VEX prefix allowed.
			
 
				     for p in gmatch(pat, ".") do ctx["rex"..p] = true end
			
 
				-    ctx.rex = true
			
 
				+    ctx.rex = "rex"
			
 
				+  end,
			
 
				+
			
 
				+  -- VEX prefix.
			
 
				+  vex = function(ctx, name, pat)
			
 
				+    if ctx.rex then return unknown(ctx) end -- Only 1 REX or VEX prefix allowed.
			
 
				+    ctx.rex = "vex"
			
 
				+    local pos = ctx.pos
			
 
				+    if ctx.mrm then
			
 
				+      ctx.mrm = nil
			
 
				+      pos = pos-1
			
 
				+    end
			
 
				+    local b = byte(ctx.code, pos, pos)
			
 
				+    if not b then return incomplete(ctx) end
			
 
				+    pos = pos+1
			
 
				+    if b < 128 then ctx.rexr = true end
			
 
				+    local m = 1
			
 
				+    if pat == "3" then
			
 
				+      m = b%32; b = (b-m)/32
			
 
				+      local nb = b%2; b = (b-nb)/2
			
 
				+      if nb == 0 then ctx.rexb = true end
			
 
				+      local nx = b%2
			
 
				+      if nx == 0 then ctx.rexx = true end
			
 
				+      b = byte(ctx.code, pos, pos)
			
 
				+      if not b then return incomplete(ctx) end
			
 
				+      pos = pos+1
			
 
				+      if b >= 128 then ctx.rexw = true end
			
 
				+    end
			
 
				+    ctx.pos = pos
			
 
				+    local map
			
 
				+    if m == 1 then map = map_opc2
			
 
				+    elseif m == 2 then map = map_opc3["38"]
			
 
				+    elseif m == 3 then map = map_opc3["3a"]
			
 
				+    else return unknown(ctx) end
			
 
				+    local p = b%4; b = (b-p)/4
			
 
				+    if p == 1 then ctx.o16 = "o16"
			
 
				+    elseif p == 2 then ctx.rep = "rep"
			
 
				+    elseif p == 3 then ctx.rep = "repne" end
			
 
				+    local l = b%2; b = (b-l)/2
			
 
				+    if l ~= 0 then ctx.vexl = true end
			
 
				+    ctx.vexv = (-1-b)%16
			
 
				+    return dispatchmap(ctx, map)
			
 
				   end,
			
 
				 
			
 
				   -- Special case for nop with REX prefix.
			
 
				   nop = function(ctx, name, pat)
			
 
				     return dispatch(ctx, ctx.rex and pat or "nop")
			
 
				   end,
			
 
				+
			
 
				+  -- Special case for 0F 77.
			
 
				+  emms = function(ctx, name, pat)
			
 
				+    if ctx.rex ~= "vex" then
			
 
				+      return putop(ctx, "emms")
			
 
				+    elseif ctx.vexl then
			
 
				+      ctx.vexl = false
			
 
				+      return putop(ctx, "zeroall")
			
 
				+    else
			
 
				+      return putop(ctx, "zeroupper")
			
 
				+    end
			
 
				+  end,
			
 
				 }
			
 
				 
			
 
				 ------------------------------------------------------------------------------
			
@@ -782,7 +899,7 @@ local function disass_block(ctx, ofs, len)
 
				 end
			
 
				 
			
 
				 -- Extended API: create a disassembler context. Then call ctx:disass(ofs, len).
			
 
				-local function create_(code, addr, out)
			
 
				+local function create(code, addr, out)
			
 
				   local ctx = {}
			
 
				   ctx.code = code
			
 
				   ctx.addr = (addr or 0) - 1
			
@@ -796,8 +913,8 @@ local function create_(code, addr, out)
 
				   return ctx
			
 
				 end
			
 
				 
			
 
				-local function create64_(code, addr, out)
			
 
				-  local ctx = create_(code, addr, out)
			
 
				+local function create64(code, addr, out)
			
 
				+  local ctx = create(code, addr, out)
			
 
				   ctx.x64 = true
			
 
				   ctx.map1 = map_opc1_64
			
 
				   ctx.aregs = map_regs.Q
			
@@ -805,32 +922,32 @@ local function create64_(code, addr, out)
 
				 end
			
 
				 
			
 
				 -- Simple API: disassemble code (a string) at address and output via out.
			
 
				-local function disass_(code, addr, out)
			
 
				-  create_(code, addr, out):disass()
			
 
				+local function disass(code, addr, out)
			
 
				+  create(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				-local function disass64_(code, addr, out)
			
 
				-  create64_(code, addr, out):disass()
			
 
				+local function disass64(code, addr, out)
			
 
				+  create64(code, addr, out):disass()
			
 
				 end
			
 
				 
			
 
				 -- Return register name for RID.
			
 
				-local function regname_(r)
			
 
				+local function regname(r)
			
 
				   if r < 8 then return map_regs.D[r+1] end
			
 
				   return map_regs.X[r-7]
			
 
				 end
			
 
				 
			
 
				-local function regname64_(r)
			
 
				+local function regname64(r)
			
 
				   if r < 16 then return map_regs.Q[r+1] end
			
 
				   return map_regs.X[r-15]
			
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-create = create_
			
 
				-create64 = create64_
			
 
				-disass = disass_
			
 
				-disass64 = disass64_
			
 
				-regname = regname_
			
 
				-regname64 = regname64_
			
 
				+return {
			
 
				+  create = create,
			
 
				+  create64 = create64,
			
 
				+  disass = disass,
			
 
				+  disass64 = disass64,
			
 
				+  regname = regname,
			
 
				+  regname64 = regname64
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/dump.lua
+++ b/luajit.mod/luajit/src/jit/dump.lua
@@ -55,7 +55,7 @@
 
				 
			
 
				 -- Cache some library functions and objects.
			
 
				 local jit = require("jit")
			
 
				-assert(jit.version_num == 20005, "LuaJIT core/library version mismatch")
			
 
				+assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
			
 
				 local jutil = require("jit.util")
			
 
				 local vmdef = require("jit.vmdef")
			
 
				 local funcinfo, funcbc = jutil.funcinfo, jutil.funcbc
			
@@ -63,7 +63,7 @@ local traceinfo, traceir, tracek = jutil.traceinfo, jutil.traceir, jutil.tracek
 
				 local tracemc, tracesnap = jutil.tracemc, jutil.tracesnap
			
 
				 local traceexitstub, ircalladdr = jutil.traceexitstub, jutil.ircalladdr
			
 
				 local bit = require("bit")
			
 
				-local band, shr = bit.band, bit.rshift
			
 
				+local band, shr, tohex = bit.band, bit.rshift, bit.tohex
			
 
				 local sub, gsub, format = string.sub, string.gsub, string.format
			
 
				 local byte, rep = string.byte, string.rep
			
 
				 local type, tostring = type, tostring
			
@@ -85,12 +85,13 @@ local nexitsym = 0
 
				 local function fillsymtab_tr(tr, nexit)
			
 
				   local t = {}
			
 
				   symtabmt.__index = t
			
 
				-  if jit.arch == "mips" or jit.arch == "mipsel" then
			
 
				+  if jit.arch:sub(1, 4) == "mips" then
			
 
				     t[traceexitstub(tr, 0)] = "exit"
			
 
				     return
			
 
				   end
			
 
				   for i=0,nexit-1 do
			
 
				     local addr = traceexitstub(tr, i)
			
 
				+    if addr < 0 then addr = addr + 2^32 end
			
 
				     t[addr] = tostring(i)
			
 
				   end
			
 
				   local addr = traceexitstub(tr, nexit)
			
@@ -104,7 +105,10 @@ local function fillsymtab(tr, nexit)
 
				     local ircall = vmdef.ircall
			
 
				     for i=0,#ircall do
			
 
				       local addr = ircalladdr(i)
			
 
				-      if addr ~= 0 then t[addr] = ircall[i] end
			
 
				+      if addr ~= 0 then
			
 
				+	if addr < 0 then addr = addr + 2^32 end
			
 
				+	t[addr] = ircall[i]
			
 
				+      end
			
 
				     end
			
 
				   end
			
 
				   if nexitsym == 1000000 then -- Per-trace exit stubs.
			
@@ -118,6 +122,7 @@ local function fillsymtab(tr, nexit)
 
				 	nexit = 1000000
			
 
				 	break
			
 
				       end
			
 
				+      if addr < 0 then addr = addr + 2^32 end
			
 
				       t[addr] = tostring(i)
			
 
				     end
			
 
				     nexitsym = nexit
			
@@ -136,6 +141,7 @@ local function dump_mcode(tr)
 
				   local mcode, addr, loop = tracemc(tr)
			
 
				   if not mcode then return end
			
 
				   if not disass then disass = require("jit.dis_"..jit.arch) end
			
 
				+  if addr < 0 then addr = addr + 2^32 end
			
 
				   out:write("---- TRACE ", tr, " mcode ", #mcode, "\n")
			
 
				   local ctx = disass.create(mcode, addr, dumpwrite)
			
 
				   ctx.hexdump = 0
			
@@ -270,8 +276,7 @@ local litname = {
 
				   ["CONV  "] = setmetatable({}, { __index = function(t, mode)
			
 
				     local s = irtype[band(mode, 31)]
			
 
				     s = irtype[band(shr(mode, 5), 31)].."."..s
			
 
				-    if band(mode, 0x400) ~= 0 then s = s.." trunc"
			
 
				-    elseif band(mode, 0x800) ~= 0 then s = s.." sext" end
			
 
				+    if band(mode, 0x800) ~= 0 then s = s.." sext" end
			
 
				     local c = shr(mode, 14)
			
 
				     if c == 2 then s = s.." index" elseif c == 3 then s = s.." check" end
			
 
				     t[mode] = s
			
@@ -280,6 +285,8 @@ local litname = {
 
				   ["FLOAD "] = vmdef.irfield,
			
 
				   ["FREF  "] = vmdef.irfield,
			
 
				   ["FPMATH"] = vmdef.irfpm,
			
 
				+  ["BUFHDR"] = { [0] = "RESET", "APPEND" },
			
 
				+  ["TOSTR "] = { [0] = "INT", "NUM", "CHAR" },
			
 
				 }
			
 
				 
			
 
				 local function ctlsub(c)
			
@@ -303,15 +310,17 @@ local function fmtfunc(func, pc)
 
				   end
			
 
				 end
			
 
				 
			
 
				-local function formatk(tr, idx)
			
 
				+local function formatk(tr, idx, sn)
			
 
				   local k, t, slot = tracek(tr, idx)
			
 
				   local tn = type(k)
			
 
				   local s
			
 
				   if tn == "number" then
			
 
				-    if k == 2^52+2^51 then
			
 
				+    if band(sn or 0, 0x30000) ~= 0 then
			
 
				+      s = band(sn, 0x20000) ~= 0 and "contpc" or "ftsz"
			
 
				+    elseif k == 2^52+2^51 then
			
 
				       s = "bias"
			
 
				     else
			
 
				-      s = format("%+.14g", k)
			
 
				+      s = format(0 < k and k < 0x1p-1026 and "%+a" or "%+.14g", k)
			
 
				     end
			
 
				   elseif tn == "string" then
			
 
				     s = format(#k > 20 and '"%.20s"~' or '"%s"', gsub(k, "%c", ctlsub))
			
@@ -329,6 +338,8 @@ local function formatk(tr, idx)
 
				   elseif t == 21 then -- int64_t
			
 
				     s = sub(tostring(k), 1, -3)
			
 
				     if sub(s, 1, 1) ~= "-" then s = "+"..s end
			
 
				+  elseif sn == 0x1057fff then -- SNAP(1, SNAP_FRAME | SNAP_NORESTORE, REF_NIL)
			
 
				+    return "----" -- Special case for LJ_FR2 slot 1.
			
 
				   else
			
 
				     s = tostring(k) -- For primitives.
			
 
				   end
			
@@ -347,7 +358,7 @@ local function printsnap(tr, snap)
 
				       n = n + 1
			
 
				       local ref = band(sn, 0xffff) - 0x8000 -- REF_BIAS
			
 
				       if ref < 0 then
			
 
				-	out:write(formatk(tr, ref))
			
 
				+	out:write(formatk(tr, ref, sn))
			
 
				       elseif band(sn, 0x80000) ~= 0 then -- SNAP_SOFTFPNUM
			
 
				 	out:write(colorize(format("%04d/%04d", ref, ref+1), 14))
			
 
				       else
			
@@ -545,7 +556,7 @@ local function dump_trace(what, tr, func, pc, otr, oex)
 
				   if what == "start" then
			
 
				     if dumpmode.H then out:write('<pre class="ljdump">\n') end
			
 
				     out:write("---- TRACE ", tr, " ", what)
			
 
				-    if otr then out:write(" ", otr, "/", oex) end
			
 
				+    if otr then out:write(" ", otr, "/", oex == -1 and "stitch" or oex) end
			
 
				     out:write(" ", fmtfunc(func, pc), "\n")
			
 
				   elseif what == "stop" or what == "abort" then
			
 
				     out:write("---- TRACE ", tr, " ", what)
			
@@ -608,7 +619,7 @@ local function dump_texit(tr, ex, ngpr, nfpr, ...)
 
				       end
			
 
				     else
			
 
				       for i=1,ngpr do
			
 
				-	out:write(format(" %08x", regs[i]))
			
 
				+	out:write(" ", tohex(regs[i]))
			
 
				 	if i % 8 == 0 then out:write("\n") end
			
 
				       end
			
 
				     end
			
@@ -693,9 +704,9 @@ local function dumpon(opt, outfile)
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-on = dumpon
			
 
				-off = dumpoff
			
 
				-start = dumpon -- For -j command line option.
			
 
				+return {
			
 
				+  on = dumpon,
			
 
				+  off = dumpoff,
			
 
				+  start = dumpon -- For -j command line option.
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/p.lua
+++ b/luajit.mod/luajit/src/jit/p.lua
@@ -0,0 +1,311 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT profiler.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+--
			
 
				+-- This module is a simple command line interface to the built-in
			
 
				+-- low-overhead profiler of LuaJIT.
			
 
				+--
			
 
				+-- The lower-level API of the profiler is accessible via the "jit.profile"
			
 
				+-- module or the luaJIT_profile_* C API.
			
 
				+--
			
 
				+-- Example usage:
			
 
				+--
			
 
				+--   luajit -jp myapp.lua
			
 
				+--   luajit -jp=s myapp.lua
			
 
				+--   luajit -jp=-s myapp.lua
			
 
				+--   luajit -jp=vl myapp.lua
			
 
				+--   luajit -jp=G,profile.txt myapp.lua
			
 
				+--
			
 
				+-- The following dump features are available:
			
 
				+--
			
 
				+--   f  Stack dump: function name, otherwise module:line. Default mode.
			
 
				+--   F  Stack dump: ditto, but always prepend module.
			
 
				+--   l  Stack dump: module:line.
			
 
				+--   <number> stack dump depth (callee < caller). Default: 1.
			
 
				+--   -<number> Inverse stack dump depth (caller > callee).
			
 
				+--   s  Split stack dump after first stack level. Implies abs(depth) >= 2.
			
 
				+--   p  Show full path for module names.
			
 
				+--   v  Show VM states. Can be combined with stack dumps, e.g. vf or fv.
			
 
				+--   z  Show zones. Can be combined with stack dumps, e.g. zf or fz.
			
 
				+--   r  Show raw sample counts. Default: show percentages.
			
 
				+--   a  Annotate excerpts from source code files.
			
 
				+--   A  Annotate complete source code files.
			
 
				+--   G  Produce raw output suitable for graphical tools (e.g. flame graphs).
			
 
				+--   m<number> Minimum sample percentage to be shown. Default: 3.
			
 
				+--   i<number> Sampling interval in milliseconds. Default: 10.
			
 
				+--
			
 
				+----------------------------------------------------------------------------
			
 
				+
			
 
				+-- Cache some library functions and objects.
			
 
				+local jit = require("jit")
			
 
				+assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
			
 
				+local profile = require("jit.profile")
			
 
				+local vmdef = require("jit.vmdef")
			
 
				+local math = math
			
 
				+local pairs, ipairs, tonumber, floor = pairs, ipairs, tonumber, math.floor
			
 
				+local sort, format = table.sort, string.format
			
 
				+local stdout = io.stdout
			
 
				+local zone -- Load jit.zone module on demand.
			
 
				+
			
 
				+-- Output file handle.
			
 
				+local out
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local prof_ud
			
 
				+local prof_states, prof_split, prof_min, prof_raw, prof_fmt, prof_depth
			
 
				+local prof_ann, prof_count1, prof_count2, prof_samples
			
 
				+
			
 
				+local map_vmmode = {
			
 
				+  N = "Compiled",
			
 
				+  I = "Interpreted",
			
 
				+  C = "C code",
			
 
				+  G = "Garbage Collector",
			
 
				+  J = "JIT Compiler",
			
 
				+}
			
 
				+
			
 
				+-- Profiler callback.
			
 
				+local function prof_cb(th, samples, vmmode)
			
 
				+  prof_samples = prof_samples + samples
			
 
				+  local key_stack, key_stack2, key_state
			
 
				+  -- Collect keys for sample.
			
 
				+  if prof_states then
			
 
				+    if prof_states == "v" then
			
 
				+      key_state = map_vmmode[vmmode] or vmmode
			
 
				+    else
			
 
				+      key_state = zone:get() or "(none)"
			
 
				+    end
			
 
				+  end
			
 
				+  if prof_fmt then
			
 
				+    key_stack = profile.dumpstack(th, prof_fmt, prof_depth)
			
 
				+    key_stack = key_stack:gsub("%[builtin#(%d+)%]", function(x)
			
 
				+      return vmdef.ffnames[tonumber(x)]
			
 
				+    end)
			
 
				+    if prof_split == 2 then
			
 
				+      local k1, k2 = key_stack:match("(.-) [<>] (.*)")
			
 
				+      if k2 then key_stack, key_stack2 = k1, k2 end
			
 
				+    elseif prof_split == 3 then
			
 
				+      key_stack2 = profile.dumpstack(th, "l", 1)
			
 
				+    end
			
 
				+  end
			
 
				+  -- Order keys.
			
 
				+  local k1, k2
			
 
				+  if prof_split == 1 then
			
 
				+    if key_state then
			
 
				+      k1 = key_state
			
 
				+      if key_stack then k2 = key_stack end
			
 
				+    end
			
 
				+  elseif key_stack then
			
 
				+    k1 = key_stack
			
 
				+    if key_stack2 then k2 = key_stack2 elseif key_state then k2 = key_state end
			
 
				+  end
			
 
				+  -- Coalesce samples in one or two levels.
			
 
				+  if k1 then
			
 
				+    local t1 = prof_count1
			
 
				+    t1[k1] = (t1[k1] or 0) + samples
			
 
				+    if k2 then
			
 
				+      local t2 = prof_count2
			
 
				+      local t3 = t2[k1]
			
 
				+      if not t3 then t3 = {}; t2[k1] = t3 end
			
 
				+      t3[k2] = (t3[k2] or 0) + samples
			
 
				+    end
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Show top N list.
			
 
				+local function prof_top(count1, count2, samples, indent)
			
 
				+  local t, n = {}, 0
			
 
				+  for k in pairs(count1) do
			
 
				+    n = n + 1
			
 
				+    t[n] = k
			
 
				+  end
			
 
				+  sort(t, function(a, b) return count1[a] > count1[b] end)
			
 
				+  for i=1,n do
			
 
				+    local k = t[i]
			
 
				+    local v = count1[k]
			
 
				+    local pct = floor(v*100/samples + 0.5)
			
 
				+    if pct < prof_min then break end
			
 
				+    if not prof_raw then
			
 
				+      out:write(format("%s%2d%%  %s\n", indent, pct, k))
			
 
				+    elseif prof_raw == "r" then
			
 
				+      out:write(format("%s%5d  %s\n", indent, v, k))
			
 
				+    else
			
 
				+      out:write(format("%s %d\n", k, v))
			
 
				+    end
			
 
				+    if count2 then
			
 
				+      local r = count2[k]
			
 
				+      if r then
			
 
				+	prof_top(r, nil, v, (prof_split == 3 or prof_split == 1) and "  -- " or
			
 
				+			    (prof_depth < 0 and "  -> " or "  <- "))
			
 
				+      end
			
 
				+    end
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+-- Annotate source code
			
 
				+local function prof_annotate(count1, samples)
			
 
				+  local files = {}
			
 
				+  local ms = 0
			
 
				+  for k, v in pairs(count1) do
			
 
				+    local pct = floor(v*100/samples + 0.5)
			
 
				+    ms = math.max(ms, v)
			
 
				+    if pct >= prof_min then
			
 
				+      local file, line = k:match("^(.*):(%d+)$")
			
 
				+      if not file then file = k; line = 0 end
			
 
				+      local fl = files[file]
			
 
				+      if not fl then fl = {}; files[file] = fl; files[#files+1] = file end
			
 
				+      line = tonumber(line)
			
 
				+      fl[line] = prof_raw and v or pct
			
 
				+    end
			
 
				+  end
			
 
				+  sort(files)
			
 
				+  local fmtv, fmtn = " %3d%% | %s\n", "      | %s\n"
			
 
				+  if prof_raw then
			
 
				+    local n = math.max(5, math.ceil(math.log10(ms)))
			
 
				+    fmtv = "%"..n.."d | %s\n"
			
 
				+    fmtn = (" "):rep(n).." | %s\n"
			
 
				+  end
			
 
				+  local ann = prof_ann
			
 
				+  for _, file in ipairs(files) do
			
 
				+    local f0 = file:byte()
			
 
				+    if f0 == 40 or f0 == 91 then
			
 
				+      out:write(format("\n====== %s ======\n[Cannot annotate non-file]\n", file))
			
 
				+      break
			
 
				+    end
			
 
				+    local fp, err = io.open(file)
			
 
				+    if not fp then
			
 
				+      out:write(format("====== ERROR: %s: %s\n", file, err))
			
 
				+      break
			
 
				+    end
			
 
				+    out:write(format("\n====== %s ======\n", file))
			
 
				+    local fl = files[file]
			
 
				+    local n, show = 1, false
			
 
				+    if ann ~= 0 then
			
 
				+      for i=1,ann do
			
 
				+	if fl[i] then show = true; out:write("@@ 1 @@\n"); break end
			
 
				+      end
			
 
				+    end
			
 
				+    for line in fp:lines() do
			
 
				+      if line:byte() == 27 then
			
 
				+	out:write("[Cannot annotate bytecode file]\n")
			
 
				+	break
			
 
				+      end
			
 
				+      local v = fl[n]
			
 
				+      if ann ~= 0 then
			
 
				+	local v2 = fl[n+ann]
			
 
				+	if show then
			
 
				+	  if v2 then show = n+ann elseif v then show = n
			
 
				+	  elseif show+ann < n then show = false end
			
 
				+	elseif v2 then
			
 
				+	  show = n+ann
			
 
				+	  out:write(format("@@ %d @@\n", n))
			
 
				+	end
			
 
				+	if not show then goto next end
			
 
				+      end
			
 
				+      if v then
			
 
				+	out:write(format(fmtv, v, line))
			
 
				+      else
			
 
				+	out:write(format(fmtn, line))
			
 
				+      end
			
 
				+    ::next::
			
 
				+      n = n + 1
			
 
				+    end
			
 
				+    fp:close()
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+-- Finish profiling and dump result.
			
 
				+local function prof_finish()
			
 
				+  if prof_ud then
			
 
				+    profile.stop()
			
 
				+    local samples = prof_samples
			
 
				+    if samples == 0 then
			
 
				+      if prof_raw ~= true then out:write("[No samples collected]\n") end
			
 
				+      return
			
 
				+    end
			
 
				+    if prof_ann then
			
 
				+      prof_annotate(prof_count1, samples)
			
 
				+    else
			
 
				+      prof_top(prof_count1, prof_count2, samples, "")
			
 
				+    end
			
 
				+    prof_count1 = nil
			
 
				+    prof_count2 = nil
			
 
				+    prof_ud = nil
			
 
				+  end
			
 
				+end
			
 
				+
			
 
				+-- Start profiling.
			
 
				+local function prof_start(mode)
			
 
				+  local interval = ""
			
 
				+  mode = mode:gsub("i%d*", function(s) interval = s; return "" end)
			
 
				+  prof_min = 3
			
 
				+  mode = mode:gsub("m(%d+)", function(s) prof_min = tonumber(s); return "" end)
			
 
				+  prof_depth = 1
			
 
				+  mode = mode:gsub("%-?%d+", function(s) prof_depth = tonumber(s); return "" end)
			
 
				+  local m = {}
			
 
				+  for c in mode:gmatch(".") do m[c] = c end
			
 
				+  prof_states = m.z or m.v
			
 
				+  if prof_states == "z" then zone = require("jit.zone") end
			
 
				+  local scope = m.l or m.f or m.F or (prof_states and "" or "f")
			
 
				+  local flags = (m.p or "")
			
 
				+  prof_raw = m.r
			
 
				+  if m.s then
			
 
				+    prof_split = 2
			
 
				+    if prof_depth == -1 or m["-"] then prof_depth = -2
			
 
				+    elseif prof_depth == 1 then prof_depth = 2 end
			
 
				+  elseif mode:find("[fF].*l") then
			
 
				+    scope = "l"
			
 
				+    prof_split = 3
			
 
				+  else
			
 
				+    prof_split = (scope == "" or mode:find("[zv].*[lfF]")) and 1 or 0
			
 
				+  end
			
 
				+  prof_ann = m.A and 0 or (m.a and 3)
			
 
				+  if prof_ann then
			
 
				+    scope = "l"
			
 
				+    prof_fmt = "pl"
			
 
				+    prof_split = 0
			
 
				+    prof_depth = 1
			
 
				+  elseif m.G and scope ~= "" then
			
 
				+    prof_fmt = flags..scope.."Z;"
			
 
				+    prof_depth = -100
			
 
				+    prof_raw = true
			
 
				+    prof_min = 0
			
 
				+  elseif scope == "" then
			
 
				+    prof_fmt = false
			
 
				+  else
			
 
				+    local sc = prof_split == 3 and m.f or m.F or scope
			
 
				+    prof_fmt = flags..sc..(prof_depth >= 0 and "Z < " or "Z > ")
			
 
				+  end
			
 
				+  prof_count1 = {}
			
 
				+  prof_count2 = {}
			
 
				+  prof_samples = 0
			
 
				+  profile.start(scope:lower()..interval, prof_cb)
			
 
				+  prof_ud = newproxy(true)
			
 
				+  getmetatable(prof_ud).__gc = prof_finish
			
 
				+end
			
 
				+
			
 
				+------------------------------------------------------------------------------
			
 
				+
			
 
				+local function start(mode, outfile)
			
 
				+  if not outfile then outfile = os.getenv("LUAJIT_PROFILEFILE") end
			
 
				+  if outfile then
			
 
				+    out = outfile == "-" and stdout or assert(io.open(outfile, "w"))
			
 
				+  else
			
 
				+    out = stdout
			
 
				+  end
			
 
				+  prof_start(mode or "f")
			
 
				+end
			
 
				+
			
 
				+-- Public module functions.
			
 
				+return {
			
 
				+  start = start, -- For -j command line option.
			
 
				+  stop = prof_finish
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/jit/v.lua
+++ b/luajit.mod/luajit/src/jit/v.lua
@@ -59,7 +59,7 @@
 
				 
			
 
				 -- Cache some library functions and objects.
			
 
				 local jit = require("jit")
			
 
				-assert(jit.version_num == 20005, "LuaJIT core/library version mismatch")
			
 
				+assert(jit.version_num == 20100, "LuaJIT core/library version mismatch")
			
 
				 local jutil = require("jit.util")
			
 
				 local vmdef = require("jit.vmdef")
			
 
				 local funcinfo, traceinfo = jutil.funcinfo, jutil.traceinfo
			
@@ -99,7 +99,7 @@ end
 
				 local function dump_trace(what, tr, func, pc, otr, oex)
			
 
				   if what == "start" then
			
 
				     startloc = fmtfunc(func, pc)
			
 
				-    startex = otr and "("..otr.."/"..oex..") " or ""
			
 
				+    startex = otr and "("..otr.."/"..(oex == -1 and "stitch" or oex)..") " or ""
			
 
				   else
			
 
				     if what == "abort" then
			
 
				       local loc = fmtfunc(func, pc)
			
@@ -116,6 +116,9 @@ local function dump_trace(what, tr, func, pc, otr, oex)
 
				       if ltype == "interpreter" then
			
 
				 	out:write(format("[TRACE %3s %s%s -- fallback to interpreter]\n",
			
 
				 	  tr, startex, startloc))
			
 
				+      elseif ltype == "stitch" then
			
 
				+	out:write(format("[TRACE %3s %s%s %s %s]\n",
			
 
				+	  tr, startex, startloc, ltype, fmtfunc(func, pc)))
			
 
				       elseif link == tr or link == 0 then
			
 
				 	out:write(format("[TRACE %3s %s%s %s]\n",
			
 
				 	  tr, startex, startloc, ltype))
			
@@ -159,9 +162,9 @@ local function dumpon(outfile)
 
				 end
			
 
				 
			
 
				 -- Public module functions.
			
 
				-module(...)
			
 
				-
			
 
				-on = dumpon
			
 
				-off = dumpoff
			
 
				-start = dumpon -- For -j command line option.
			
 
				+return {
			
 
				+  on = dumpon,
			
 
				+  off = dumpoff,
			
 
				+  start = dumpon -- For -j command line option.
			
 
				+}
			
 
				 
			
--- a/luajit.mod/luajit/src/jit/zone.lua
+++ b/luajit.mod/luajit/src/jit/zone.lua
@@ -0,0 +1,45 @@
 
				+----------------------------------------------------------------------------
			
 
				+-- LuaJIT profiler zones.
			
 
				+--
			
 
				+-- Copyright (C) 2005-2017 Mike Pall. All rights reserved.
			
 
				+-- Released under the MIT license. See Copyright Notice in luajit.h
			
 
				+----------------------------------------------------------------------------
			
 
				+--
			
 
				+-- This module implements a simple hierarchical zone model.
			
 
				+--
			
 
				+-- Example usage:
			
 
				+--
			
 
				+--   local zone = require("jit.zone")
			
 
				+--   zone("AI")
			
 
				+--   ...
			
 
				+--     zone("A*")
			
 
				+--     ...
			
 
				+--     print(zone:get()) --> "A*"
			
 
				+--     ...
			
 
				+--     zone()
			
 
				+--   ...
			
 
				+--   print(zone:get()) --> "AI"
			
 
				+--   ...
			
 
				+--   zone()
			
 
				+--
			
 
				+----------------------------------------------------------------------------
			
 
				+
			
 
				+local remove = table.remove
			
 
				+
			
 
				+return setmetatable({
			
 
				+  flush = function(t)
			
 
				+    for i=#t,1,-1 do t[i] = nil end
			
 
				+  end,
			
 
				+  get = function(t)
			
 
				+    return t[#t]
			
 
				+  end
			
 
				+}, {
			
 
				+  __call = function(t, zone)
			
 
				+    if zone then
			
 
				+      t[#t+1] = zone
			
 
				+    else
			
 
				+      return (assert(remove(t), "empty zone stack"))
			
 
				+    end
			
 
				+  end
			
 
				+})
			
 
				+
			
--- a/luajit.mod/luajit/src/lauxlib.h
+++ b/luajit.mod/luajit/src/lauxlib.h
@@ -15,9 +15,6 @@
 
				 #include "lua.h"
			
 
				 
			
 
				 
			
 
				-#define luaL_getn(L,i)          ((int)lua_objlen(L, i))
			
 
				-#define luaL_setn(L,i,j)        ((void)0)  /* no op! */
			
 
				-
			
 
				 /* extra error code for `luaL_load' */
			
 
				 #define LUA_ERRFILE     (LUA_ERRERR+1)
			
 
				 
			
@@ -58,6 +55,10 @@ LUALIB_API int (luaL_error) (lua_State *L, const char *fmt, ...);
 
				 LUALIB_API int (luaL_checkoption) (lua_State *L, int narg, const char *def,
			
 
				                                    const char *const lst[]);
			
 
				 
			
 
				+/* pre-defined references */
			
 
				+#define LUA_NOREF       (-2)
			
 
				+#define LUA_REFNIL      (-1)
			
 
				+
			
 
				 LUALIB_API int (luaL_ref) (lua_State *L, int t);
			
 
				 LUALIB_API void (luaL_unref) (lua_State *L, int t, int ref);
			
 
				 
			
@@ -84,6 +85,11 @@ LUALIB_API int (luaL_loadbufferx) (lua_State *L, const char *buff, size_t sz,
 
				 				   const char *name, const char *mode);
			
 
				 LUALIB_API void luaL_traceback (lua_State *L, lua_State *L1, const char *msg,
			
 
				 				int level);
			
 
				+LUALIB_API void (luaL_setfuncs) (lua_State *L, const luaL_Reg *l, int nup);
			
 
				+LUALIB_API void (luaL_pushmodule) (lua_State *L, const char *modname,
			
 
				+				   int sizehint);
			
 
				+LUALIB_API void *(luaL_testudata) (lua_State *L, int ud, const char *tname);
			
 
				+LUALIB_API void (luaL_setmetatable) (lua_State *L, const char *tname);
			
 
				 
			
 
				 
			
 
				 /*
			
@@ -113,6 +119,11 @@ LUALIB_API void luaL_traceback (lua_State *L, lua_State *L1, const char *msg,
 
				 
			
 
				 #define luaL_opt(L,f,n,d)	(lua_isnoneornil(L,(n)) ? (d) : f(L,(n)))
			
 
				 
			
 
				+/* From Lua 5.2. */
			
 
				+#define luaL_newlibtable(L, l) \
			
 
				+	lua_createtable(L, 0, sizeof(l)/sizeof((l)[0]) - 1)
			
 
				+#define luaL_newlib(L, l)	(luaL_newlibtable(L, l), luaL_setfuncs(L, l, 0))
			
 
				+
			
 
				 /*
			
 
				 ** {======================================================
			
 
				 ** Generic Buffer manipulation
			
@@ -147,21 +158,4 @@ LUALIB_API void (luaL_pushresult) (luaL_Buffer *B);
 
				 
			
 
				 /* }====================================================== */
			
 
				 
			
 
				-
			
 
				-/* compatibility with ref system */
			
 
				-
			
 
				-/* pre-defined references */
			
 
				-#define LUA_NOREF       (-2)
			
 
				-#define LUA_REFNIL      (-1)
			
 
				-
			
 
				-#define lua_ref(L,lock) ((lock) ? luaL_ref(L, LUA_REGISTRYINDEX) : \
			
 
				-      (lua_pushstring(L, "unlocked references are obsolete"), lua_error(L), 0))
			
 
				-
			
 
				-#define lua_unref(L,ref)        luaL_unref(L, LUA_REGISTRYINDEX, (ref))
			
 
				-
			
 
				-#define lua_getref(L,ref)       lua_rawgeti(L, LUA_REGISTRYINDEX, (ref))
			
 
				-
			
 
				-
			
 
				-#define luaL_reg	luaL_Reg
			
 
				-
			
 
				 #endif
			
--- a/luajit.mod/luajit/src/lib_aux.c
+++ b/luajit.mod/luajit/src/lib_aux.c
@@ -107,38 +107,36 @@ LUALIB_API const char *luaL_findtable(lua_State *L, int idx,
 
				 static int libsize(const luaL_Reg *l)
			
 
				 {
			
 
				   int size = 0;
			
 
				-  for (; l->name; l++) size++;
			
 
				+  for (; l && l->name; l++) size++;
			
 
				   return size;
			
 
				 }
			
 
				 
			
 
				+LUALIB_API void luaL_pushmodule(lua_State *L, const char *modname, int sizehint)
			
 
				+{
			
 
				+  luaL_findtable(L, LUA_REGISTRYINDEX, "_LOADED", 16);
			
 
				+  lua_getfield(L, -1, modname);
			
 
				+  if (!lua_istable(L, -1)) {
			
 
				+    lua_pop(L, 1);
			
 
				+    if (luaL_findtable(L, LUA_GLOBALSINDEX, modname, sizehint) != NULL)
			
 
				+      lj_err_callerv(L, LJ_ERR_BADMODN, modname);
			
 
				+    lua_pushvalue(L, -1);
			
 
				+    lua_setfield(L, -3, modname);  /* _LOADED[modname] = new table. */
			
 
				+  }
			
 
				+  lua_remove(L, -2);  /* Remove _LOADED table. */
			
 
				+}
			
 
				+
			
 
				 LUALIB_API void luaL_openlib(lua_State *L, const char *libname,
			
 
				 			     const luaL_Reg *l, int nup)
			
 
				 {
			
 
				   lj_lib_checkfpu(L);
			
 
				   if (libname) {
			
 
				-    int size = libsize(l);
			
 
				-    /* check whether lib already exists */
			
 
				-    luaL_findtable(L, LUA_REGISTRYINDEX, "_LOADED", 16);
			
 
				-    lua_getfield(L, -1, libname);  /* get _LOADED[libname] */
			
 
				-    if (!lua_istable(L, -1)) {  /* not found? */
			
 
				-      lua_pop(L, 1);  /* remove previous result */
			
 
				-      /* try global variable (and create one if it does not exist) */
			
 
				-      if (luaL_findtable(L, LUA_GLOBALSINDEX, libname, size) != NULL)
			
 
				-	lj_err_callerv(L, LJ_ERR_BADMODN, libname);
			
 
				-      lua_pushvalue(L, -1);
			
 
				-      lua_setfield(L, -3, libname);  /* _LOADED[libname] = new table */
			
 
				-    }
			
 
				-    lua_remove(L, -2);  /* remove _LOADED table */
			
 
				-    lua_insert(L, -(nup+1));  /* move library table to below upvalues */
			
 
				+    luaL_pushmodule(L, libname, libsize(l));
			
 
				+    lua_insert(L, -(nup + 1));  /* Move module table below upvalues. */
			
 
				   }
			
 
				-  for (; l->name; l++) {
			
 
				-    int i;
			
 
				-    for (i = 0; i < nup; i++)  /* copy upvalues to the top */
			
 
				-      lua_pushvalue(L, -nup);
			
 
				-    lua_pushcclosure(L, l->func, nup);
			
 
				-    lua_setfield(L, -(nup+2), l->name);
			
 
				-  }
			
 
				-  lua_pop(L, nup);  /* remove upvalues */
			
 
				+  if (l)
			
 
				+    luaL_setfuncs(L, l, nup);
			
 
				+  else
			
 
				+    lua_pop(L, nup);  /* Remove upvalues. */
			
 
				 }
			
 
				 
			
 
				 LUALIB_API void luaL_register(lua_State *L, const char *libname,
			
@@ -147,6 +145,19 @@ LUALIB_API void luaL_register(lua_State *L, const char *libname,
 
				   luaL_openlib(L, libname, l, 0);
			
 
				 }
			
 
				 
			
 
				+LUALIB_API void luaL_setfuncs(lua_State *L, const luaL_Reg *l, int nup)
			
 
				+{
			
 
				+  luaL_checkstack(L, nup, "too many upvalues");
			
 
				+  for (; l->name; l++) {
			
 
				+    int i;
			
 
				+    for (i = 0; i < nup; i++)  /* Copy upvalues to the top. */
			
 
				+      lua_pushvalue(L, -nup);
			
 
				+    lua_pushcclosure(L, l->func, nup);
			
 
				+    lua_setfield(L, -(nup + 2), l->name);
			
 
				+  }
			
 
				+  lua_pop(L, nup);  /* Remove upvalues. */
			
 
				+}
			
 
				+
			
 
				 LUALIB_API const char *luaL_gsub(lua_State *L, const char *s,
			
 
				 				 const char *p, const char *r)
			
 
				 {
			
@@ -207,8 +218,15 @@ LUALIB_API char *luaL_prepbuffer(luaL_Buffer *B)
 
				 
			
 
				 LUALIB_API void luaL_addlstring(luaL_Buffer *B, const char *s, size_t l)
			
 
				 {
			
 
				-  while (l--)
			
 
				-    luaL_addchar(B, *s++);
			
 
				+  if (l <= bufffree(B)) {
			
 
				+    memcpy(B->p, s, l);
			
 
				+    B->p += l;
			
 
				+  } else {
			
 
				+    emptybuffer(B);
			
 
				+    lua_pushlstring(B->L, s, l);
			
 
				+    B->lvl++;
			
 
				+    adjuststack(B);
			
 
				+  }
			
 
				 }
			
 
				 
			
 
				 LUALIB_API void luaL_addstring(luaL_Buffer *B, const char *s)
			
@@ -302,7 +320,7 @@ static int panic(lua_State *L)
 
				 
			
 
				 #ifdef LUAJIT_USE_SYSMALLOC
			
 
				 
			
 
				-#if LJ_64 && !defined(LUAJIT_USE_VALGRIND)
			
 
				+#if LJ_64 && !LJ_GC64 && !defined(LUAJIT_USE_VALGRIND)
			
 
				 #error "Must use builtin allocator for 64 bit target"
			
 
				 #endif
			
 
				 
			
@@ -334,7 +352,7 @@ LUALIB_API lua_State *luaL_newstate(void)
 
				   lua_State *L;
			
 
				   void *ud = lj_alloc_create();
			
 
				   if (ud == NULL) return NULL;
			
 
				-#if LJ_64
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				   L = lj_state_newstate(lj_alloc_f, ud);
			
 
				 #else
			
 
				   L = lua_newstate(lj_alloc_f, ud);
			
@@ -343,7 +361,7 @@ LUALIB_API lua_State *luaL_newstate(void)
 
				   return L;
			
 
				 }
			
 
				 
			
 
				-#if LJ_64
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				 LUA_API lua_State *lua_newstate(lua_Alloc f, void *ud)
			
 
				 {
			
 
				   UNUSED(f); UNUSED(ud);
			
--- a/luajit.mod/luajit/src/lib_base.c
+++ b/luajit.mod/luajit/src/lib_base.c
@@ -23,6 +23,7 @@
 
				 #include "lj_tab.h"
			
 
				 #include "lj_meta.h"
			
 
				 #include "lj_state.h"
			
 
				+#include "lj_frame.h"
			
 
				 #if LJ_HASFFI
			
 
				 #include "lj_ctype.h"
			
 
				 #include "lj_cconv.h"
			
@@ -32,6 +33,7 @@
 
				 #include "lj_dispatch.h"
			
 
				 #include "lj_char.h"
			
 
				 #include "lj_strscan.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
 
				 /* -- Base library: checks ------------------------------------------------ */
			
@@ -40,13 +42,13 @@
 
				 
			
 
				 LJLIB_ASM(assert)		LJLIB_REC(.)
			
 
				 {
			
 
				-  GCstr *s;
			
 
				   lj_lib_checkany(L, 1);
			
 
				-  s = lj_lib_optstr(L, 2);
			
 
				-  if (s)
			
 
				-    lj_err_callermsg(L, strdata(s));
			
 
				-  else
			
 
				+  if (L->top == L->base+1)
			
 
				     lj_err_caller(L, LJ_ERR_ASSERT);
			
 
				+  else if (tvisstr(L->base+1) || tvisnumber(L->base+1))
			
 
				+    lj_err_callermsg(L, strdata(lj_lib_checkstr(L, 2)));
			
 
				+  else
			
 
				+    lj_err_run(L);
			
 
				   return FFH_UNREACHABLE;
			
 
				 }
			
 
				 
			
@@ -86,10 +88,11 @@ static int ffh_pairs(lua_State *L, MMS mm)
 
				   cTValue *mo = lj_meta_lookup(L, o, mm);
			
 
				   if ((LJ_52 || tviscdata(o)) && !tvisnil(mo)) {
			
 
				     L->top = o+1;  /* Only keep one argument. */
			
 
				-    copyTV(L, L->base-1, mo);  /* Replace callable. */
			
 
				+    copyTV(L, L->base-1-LJ_FR2, mo);  /* Replace callable. */
			
 
				     return FFH_TAILCALL;
			
 
				   } else {
			
 
				     if (!tvistab(o)) lj_err_argt(L, 1, LUA_TTABLE);
			
 
				+    if (LJ_FR2) { copyTV(L, o-1, o); o--; }
			
 
				     setfuncV(L, o-1, funcV(lj_lib_upvalue(L, 1)));
			
 
				     if (mm == MM_pairs) setnilV(o+1); else setintV(o+1, 0);
			
 
				     return FFH_RES(3);
			
@@ -100,7 +103,7 @@ static int ffh_pairs(lua_State *L, MMS mm)
 
				 #endif
			
 
				 
			
 
				 LJLIB_PUSH(lastcl)
			
 
				-LJLIB_ASM(pairs)
			
 
				+LJLIB_ASM(pairs)		LJLIB_REC(xpairs 0)
			
 
				 {
			
 
				   return ffh_pairs(L, MM_pairs);
			
 
				 }
			
@@ -113,7 +116,7 @@ LJLIB_NOREGUV LJLIB_ASM(ipairs_aux)	LJLIB_REC(.)
 
				 }
			
 
				 
			
 
				 LJLIB_PUSH(lastcl)
			
 
				-LJLIB_ASM(ipairs)		LJLIB_REC(.)
			
 
				+LJLIB_ASM(ipairs)		LJLIB_REC(xpairs 1)
			
 
				 {
			
 
				   return ffh_pairs(L, MM_ipairs);
			
 
				 }
			
@@ -131,11 +134,11 @@ LJLIB_ASM(setmetatable)		LJLIB_REC(.)
 
				     lj_err_caller(L, LJ_ERR_PROTMT);
			
 
				   setgcref(t->metatable, obj2gco(mt));
			
 
				   if (mt) { lj_gc_objbarriert(L, t, mt); }
			
 
				-  settabV(L, L->base-1, t);
			
 
				+  settabV(L, L->base-1-LJ_FR2, t);
			
 
				   return FFH_RES(1);
			
 
				 }
			
 
				 
			
 
				-LJLIB_CF(getfenv)
			
 
				+LJLIB_CF(getfenv)		LJLIB_REC(.)
			
 
				 {
			
 
				   GCfunc *fn;
			
 
				   cTValue *o = L->base;
			
@@ -144,6 +147,7 @@ LJLIB_CF(getfenv)
 
				     o = lj_debug_frame(L, level, &level);
			
 
				     if (o == NULL)
			
 
				       lj_err_arg(L, 1, LJ_ERR_INVLVL);
			
 
				+    if (LJ_FR2) o--;
			
 
				   }
			
 
				   fn = &gcval(o)->fn;
			
 
				   settabV(L, L->top++, isluafunc(fn) ? tabref(fn->l.env) : tabref(L->env));
			
@@ -165,6 +169,7 @@ LJLIB_CF(setfenv)
 
				     o = lj_debug_frame(L, level, &level);
			
 
				     if (o == NULL)
			
 
				       lj_err_arg(L, 1, LJ_ERR_INVLVL);
			
 
				+    if (LJ_FR2) o--;
			
 
				   }
			
 
				   fn = &gcval(o)->fn;
			
 
				   if (!isluafunc(fn))
			
@@ -257,7 +262,7 @@ LJLIB_ASM(tonumber)		LJLIB_REC(.)
 
				   if (base == 10) {
			
 
				     TValue *o = lj_lib_checkany(L, 1);
			
 
				     if (lj_strscan_numberobj(o)) {
			
 
				-      copyTV(L, L->base-1, o);
			
 
				+      copyTV(L, L->base-1-LJ_FR2, o);
			
 
				       return FFH_RES(1);
			
 
				     }
			
 
				 #if LJ_HASFFI
			
@@ -270,11 +275,11 @@ LJLIB_ASM(tonumber)		LJLIB_REC(.)
 
				 	    ct->size <= 4 && !(ct->size == 4 && (ct->info & CTF_UNSIGNED))) {
			
 
				 	  int32_t i;
			
 
				 	  lj_cconv_ct_tv(cts, ctype_get(cts, CTID_INT32), (uint8_t *)&i, o, 0);
			
 
				-	  setintV(L->base-1, i);
			
 
				+	  setintV(L->base-1-LJ_FR2, i);
			
 
				 	  return FFH_RES(1);
			
 
				 	}
			
 
				 	lj_cconv_ct_tv(cts, ctype_get(cts, CTID_DOUBLE),
			
 
				-		       (uint8_t *)&(L->base-1)->n, o, 0);
			
 
				+		       (uint8_t *)&(L->base-1-LJ_FR2)->n, o, 0);
			
 
				 	return FFH_RES(1);
			
 
				       }
			
 
				     }
			
@@ -282,53 +287,46 @@ LJLIB_ASM(tonumber)		LJLIB_REC(.)
 
				   } else {
			
 
				     const char *p = strdata(lj_lib_checkstr(L, 1));
			
 
				     char *ep;
			
 
				+    unsigned int neg = 0;
			
 
				     unsigned long ul;
			
 
				     if (base < 2 || base > 36)
			
 
				       lj_err_arg(L, 2, LJ_ERR_BASERNG);
			
 
				-    ul = strtoul(p, &ep, base);
			
 
				-    if (p != ep) {
			
 
				-      while (lj_char_isspace((unsigned char)(*ep))) ep++;
			
 
				-      if (*ep == '\0') {
			
 
				-	if (LJ_DUALNUM && LJ_LIKELY(ul < 0x80000000u))
			
 
				-	  setintV(L->base-1, (int32_t)ul);
			
 
				-	else
			
 
				-	  setnumV(L->base-1, (lua_Number)ul);
			
 
				-	return FFH_RES(1);
			
 
				+    while (lj_char_isspace((unsigned char)(*p))) p++;
			
 
				+    if (*p == '-') { p++; neg = 1; } else if (*p == '+') { p++; }
			
 
				+    if (lj_char_isalnum((unsigned char)(*p))) {
			
 
				+      ul = strtoul(p, &ep, base);
			
 
				+      if (p != ep) {
			
 
				+	while (lj_char_isspace((unsigned char)(*ep))) ep++;
			
 
				+	if (*ep == '\0') {
			
 
				+	  if (LJ_DUALNUM && LJ_LIKELY(ul < 0x80000000u+neg)) {
			
 
				+	    if (neg) ul = -ul;
			
 
				+	    setintV(L->base-1-LJ_FR2, (int32_t)ul);
			
 
				+	  } else {
			
 
				+	    lua_Number n = (lua_Number)ul;
			
 
				+	    if (neg) n = -n;
			
 
				+	    setnumV(L->base-1-LJ_FR2, n);
			
 
				+	  }
			
 
				+	  return FFH_RES(1);
			
 
				+	}
			
 
				       }
			
 
				     }
			
 
				   }
			
 
				-  setnilV(L->base-1);
			
 
				+  setnilV(L->base-1-LJ_FR2);
			
 
				   return FFH_RES(1);
			
 
				 }
			
 
				 
			
 
				-LJLIB_PUSH("nil")
			
 
				-LJLIB_PUSH("false")
			
 
				-LJLIB_PUSH("true")
			
 
				 LJLIB_ASM(tostring)		LJLIB_REC(.)
			
 
				 {
			
 
				   TValue *o = lj_lib_checkany(L, 1);
			
 
				   cTValue *mo;
			
 
				   L->top = o+1;  /* Only keep one argument. */
			
 
				   if (!tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
			
 
				-    copyTV(L, L->base-1, mo);  /* Replace callable. */
			
 
				+    copyTV(L, L->base-1-LJ_FR2, mo);  /* Replace callable. */
			
 
				     return FFH_TAILCALL;
			
 
				-  } else {
			
 
				-    GCstr *s;
			
 
				-    if (tvisnumber(o)) {
			
 
				-      s = lj_str_fromnumber(L, o);
			
 
				-    } else if (tvispri(o)) {
			
 
				-      s = strV(lj_lib_upvalue(L, -(int32_t)itype(o)));
			
 
				-    } else {
			
 
				-      if (tvisfunc(o) && isffunc(funcV(o)))
			
 
				-	lua_pushfstring(L, "function: builtin#%d", funcV(o)->c.ffid);
			
 
				-      else
			
 
				-	lua_pushfstring(L, "%s: %p", lj_typename(o), lua_topointer(L, 1));
			
 
				-      /* Note: lua_pushfstring calls the GC which may invalidate o. */
			
 
				-      s = strV(L->top-1);
			
 
				-    }
			
 
				-    setstrV(L, L->base-1, s);
			
 
				-    return FFH_RES(1);
			
 
				   }
			
 
				+  lj_gc_check(L);
			
 
				+  setstrV(L, L->base-1-LJ_FR2, lj_strfmt_obj(L, L->base));
			
 
				+  return FFH_RES(1);
			
 
				 }
			
 
				 
			
 
				 /* -- Base library: throw and catch errors -------------------------------- */
			
@@ -357,7 +355,7 @@ LJLIB_ASM_(xpcall)		LJLIB_REC(.)
 
				 
			
 
				 static int load_aux(lua_State *L, int status, int envarg)
			
 
				 {
			
 
				-  if (status == 0) {
			
 
				+  if (status == LUA_OK) {
			
 
				     if (tvistab(L->base+envarg-1)) {
			
 
				       GCfunc *fn = funcV(L->top-1);
			
 
				       GCtab *t = tabV(L->base+envarg-1);
			
@@ -430,7 +428,7 @@ LJLIB_CF(dofile)
 
				   GCstr *fname = lj_lib_optstr(L, 1);
			
 
				   setnilV(L->top);
			
 
				   L->top = L->base+1;
			
 
				-  if (luaL_loadfile(L, fname ? strdata(fname) : NULL) != 0)
			
 
				+  if (luaL_loadfile(L, fname ? strdata(fname) : NULL) != LUA_OK)
			
 
				     lua_error(L);
			
 
				   lua_call(L, 0, LUA_MULTRET);
			
 
				   return (int)(L->top - L->base) - 1;
			
@@ -440,20 +438,20 @@ LJLIB_CF(dofile)
 
				 
			
 
				 LJLIB_CF(gcinfo)
			
 
				 {
			
 
				-  setintV(L->top++, (G(L)->gc.total >> 10));
			
 
				+  setintV(L->top++, (int32_t)(G(L)->gc.total >> 10));
			
 
				   return 1;
			
 
				 }
			
 
				 
			
 
				 LJLIB_CF(collectgarbage)
			
 
				 {
			
 
				   int opt = lj_lib_checkopt(L, 1, LUA_GCCOLLECT,  /* ORDER LUA_GC* */
			
 
				-    "\4stop\7restart\7collect\5count\1\377\4step\10setpause\12setstepmul");
			
 
				+    "\4stop\7restart\7collect\5count\1\377\4step\10setpause\12setstepmul\1\377\11isrunning");
			
 
				   int32_t data = lj_lib_optint(L, 2, 0);
			
 
				   if (opt == LUA_GCCOUNT) {
			
 
				     setnumV(L->top, (lua_Number)G(L)->gc.total/1024.0);
			
 
				   } else {
			
 
				     int res = lua_gc(L, opt, data);
			
 
				-    if (opt == LUA_GCSTEP)
			
 
				+    if (opt == LUA_GCSTEP || opt == LUA_GCISRUNNING)
			
 
				       setboolV(L->top, res);
			
 
				     else
			
 
				       setintV(L->top, res);
			
@@ -506,21 +504,12 @@ LJLIB_CF(print)
 
				   }
			
 
				   shortcut = (tvisfunc(tv) && funcV(tv)->c.ffid == FF_tostring);
			
 
				   for (i = 0; i < nargs; i++) {
			
 
				+    cTValue *o = &L->base[i];
			
 
				     const char *str;
			
 
				     size_t size;
			
 
				-    cTValue *o = &L->base[i];
			
 
				-    if (shortcut && tvisstr(o)) {
			
 
				-      str = strVdata(o);
			
 
				-      size = strV(o)->len;
			
 
				-    } else if (shortcut && tvisint(o)) {
			
 
				-      char buf[LJ_STR_INTBUF];
			
 
				-      char *p = lj_str_bufint(buf, intV(o));
			
 
				-      size = (size_t)(buf+LJ_STR_INTBUF-p);
			
 
				-      str = p;
			
 
				-    } else if (shortcut && tvisnum(o)) {
			
 
				-      char buf[LJ_STR_NUMBUF];
			
 
				-      size = lj_str_bufnum(buf, o);
			
 
				-      str = buf;
			
 
				+    MSize len;
			
 
				+    if (shortcut && (str = lj_strfmt_wstrnum(L, o, &len)) != NULL) {
			
 
				+      size = len;
			
 
				     } else {
			
 
				       copyTV(L, L->top+1, o);
			
 
				       copyTV(L, L->top, L->top-1);
			
@@ -557,8 +546,8 @@ LJLIB_CF(coroutine_status)
 
				   co = threadV(L->base);
			
 
				   if (co == L) s = "running";
			
 
				   else if (co->status == LUA_YIELD) s = "suspended";
			
 
				-  else if (co->status != 0) s = "dead";
			
 
				-  else if (co->base > tvref(co->stack)+1) s = "normal";
			
 
				+  else if (co->status != LUA_OK) s = "dead";
			
 
				+  else if (co->base > tvref(co->stack)+1+LJ_FR2) s = "normal";
			
 
				   else if (co->top == co->base) s = "dead";
			
 
				   else s = "suspended";
			
 
				   lua_pushstring(L, s);
			
@@ -578,6 +567,12 @@ LJLIB_CF(coroutine_running)
 
				 #endif
			
 
				 }
			
 
				 
			
 
				+LJLIB_CF(coroutine_isyieldable)
			
 
				+{
			
 
				+  setboolV(L->top++, cframe_canyield(L->cframe));
			
 
				+  return 1;
			
 
				+}
			
 
				+
			
 
				 LJLIB_CF(coroutine_create)
			
 
				 {
			
 
				   lua_State *L1;
			
@@ -597,11 +592,11 @@ LJLIB_ASM(coroutine_yield)
 
				 static int ffh_resume(lua_State *L, lua_State *co, int wrap)
			
 
				 {
			
 
				   if (co->cframe != NULL || co->status > LUA_YIELD ||
			
 
				-      (co->status == 0 && co->top == co->base)) {
			
 
				+      (co->status == LUA_OK && co->top == co->base)) {
			
 
				     ErrMsg em = co->cframe ? LJ_ERR_CORUN : LJ_ERR_CODEAD;
			
 
				     if (wrap) lj_err_caller(L, em);
			
 
				-    setboolV(L->base-1, 0);
			
 
				-    setstrV(L, L->base, lj_err_str(L, em));
			
 
				+    setboolV(L->base-1-LJ_FR2, 0);
			
 
				+    setstrV(L, L->base-LJ_FR2, lj_err_str(L, em));
			
 
				     return FFH_RES(2);
			
 
				   }
			
 
				   lj_state_growstack(co, (MSize)(L->top - L->base));
			
@@ -642,9 +637,10 @@ static void setpc_wrap_aux(lua_State *L, GCfunc *fn);
 
				 
			
 
				 LJLIB_CF(coroutine_wrap)
			
 
				 {
			
 
				+  GCfunc *fn;
			
 
				   lj_cf_coroutine_create(L);
			
 
				-  lj_lib_pushcc(L, lj_ffh_coroutine_wrap_aux, FF_coroutine_wrap_aux, 1);
			
 
				-  setpc_wrap_aux(L, funcV(L->top-1));
			
 
				+  fn = lj_lib_pushcc(L, lj_ffh_coroutine_wrap_aux, FF_coroutine_wrap_aux, 1);
			
 
				+  setpc_wrap_aux(L, fn);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lib_bit.c
+++ b/luajit.mod/luajit/src/lib_bit.c
@@ -12,26 +12,99 @@
 
				 
			
 
				 #include "lj_obj.h"
			
 
				 #include "lj_err.h"
			
 
				-#include "lj_str.h"
			
 
				+#include "lj_buf.h"
			
 
				+#include "lj_strscan.h"
			
 
				+#include "lj_strfmt.h"
			
 
				+#if LJ_HASFFI
			
 
				+#include "lj_ctype.h"
			
 
				+#include "lj_cdata.h"
			
 
				+#include "lj_cconv.h"
			
 
				+#include "lj_carith.h"
			
 
				+#endif
			
 
				+#include "lj_ff.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				 #define LJLIB_MODULE_bit
			
 
				 
			
 
				-LJLIB_ASM(bit_tobit)		LJLIB_REC(bit_unary IR_TOBIT)
			
 
				+#if LJ_HASFFI
			
 
				+static int bit_result64(lua_State *L, CTypeID id, uint64_t x)
			
 
				 {
			
 
				+  GCcdata *cd = lj_cdata_new_(L, id, 8);
			
 
				+  *(uint64_t *)cdataptr(cd) = x;
			
 
				+  setcdataV(L, L->base-1-LJ_FR2, cd);
			
 
				+  return FFH_RES(1);
			
 
				+}
			
 
				+#else
			
 
				+static int32_t bit_checkbit(lua_State *L, int narg)
			
 
				+{
			
 
				+  TValue *o = L->base + narg-1;
			
 
				+  if (!(o < L->top && lj_strscan_numberobj(o)))
			
 
				+    lj_err_argt(L, narg, LUA_TNUMBER);
			
 
				+  if (LJ_LIKELY(tvisint(o))) {
			
 
				+    return intV(o);
			
 
				+  } else {
			
 
				+    int32_t i = lj_num2bit(numV(o));
			
 
				+    if (LJ_DUALNUM) setintV(o, i);
			
 
				+    return i;
			
 
				+  }
			
 
				+}
			
 
				+#endif
			
 
				+
			
 
				+LJLIB_ASM(bit_tobit)		LJLIB_REC(bit_tobit)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0;
			
 
				+  setintV(L->base-1-LJ_FR2, (int32_t)lj_carith_check64(L, 1, &id));
			
 
				+  return FFH_RES(1);
			
 
				+#else
			
 
				+  lj_lib_checknumber(L, 1);
			
 
				+  return FFH_RETRY;
			
 
				+#endif
			
 
				+}
			
 
				+
			
 
				+LJLIB_ASM(bit_bnot)		LJLIB_REC(bit_unary IR_BNOT)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0;
			
 
				+  uint64_t x = lj_carith_check64(L, 1, &id);
			
 
				+  return id ? bit_result64(L, id, ~x) : FFH_RETRY;
			
 
				+#else
			
 
				   lj_lib_checknumber(L, 1);
			
 
				   return FFH_RETRY;
			
 
				+#endif
			
 
				+}
			
 
				+
			
 
				+LJLIB_ASM(bit_bswap)		LJLIB_REC(bit_unary IR_BSWAP)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0;
			
 
				+  uint64_t x = lj_carith_check64(L, 1, &id);
			
 
				+  return id ? bit_result64(L, id, lj_bswap64(x)) : FFH_RETRY;
			
 
				+#else
			
 
				+  lj_lib_checknumber(L, 1);
			
 
				+  return FFH_RETRY;
			
 
				+#endif
			
 
				 }
			
 
				-LJLIB_ASM_(bit_bnot)		LJLIB_REC(bit_unary IR_BNOT)
			
 
				-LJLIB_ASM_(bit_bswap)		LJLIB_REC(bit_unary IR_BSWAP)
			
 
				 
			
 
				 LJLIB_ASM(bit_lshift)		LJLIB_REC(bit_shift IR_BSHL)
			
 
				 {
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0, id2 = 0;
			
 
				+  uint64_t x = lj_carith_check64(L, 1, &id);
			
 
				+  int32_t sh = (int32_t)lj_carith_check64(L, 2, &id2);
			
 
				+  if (id) {
			
 
				+    x = lj_carith_shift64(x, sh, curr_func(L)->c.ffid - (int)FF_bit_lshift);
			
 
				+    return bit_result64(L, id, x);
			
 
				+  }
			
 
				+  if (id2) setintV(L->base+1, sh);
			
 
				+  return FFH_RETRY;
			
 
				+#else
			
 
				   lj_lib_checknumber(L, 1);
			
 
				-  lj_lib_checkbit(L, 2);
			
 
				+  bit_checkbit(L, 2);
			
 
				   return FFH_RETRY;
			
 
				+#endif
			
 
				 }
			
 
				 LJLIB_ASM_(bit_rshift)		LJLIB_REC(bit_shift IR_BSHR)
			
 
				 LJLIB_ASM_(bit_arshift)		LJLIB_REC(bit_shift IR_BSAR)
			
@@ -40,25 +113,58 @@ LJLIB_ASM_(bit_ror)		LJLIB_REC(bit_shift IR_BROR)
 
				 
			
 
				 LJLIB_ASM(bit_band)		LJLIB_REC(bit_nary IR_BAND)
			
 
				 {
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0;
			
 
				+  TValue *o = L->base, *top = L->top;
			
 
				+  int i = 0;
			
 
				+  do { lj_carith_check64(L, ++i, &id); } while (++o < top);
			
 
				+  if (id) {
			
 
				+    CTState *cts = ctype_cts(L);
			
 
				+    CType *ct = ctype_get(cts, id);
			
 
				+    int op = curr_func(L)->c.ffid - (int)FF_bit_bor;
			
 
				+    uint64_t x, y = op >= 0 ? 0 : ~(uint64_t)0;
			
 
				+    o = L->base;
			
 
				+    do {
			
 
				+      lj_cconv_ct_tv(cts, ct, (uint8_t *)&x, o, 0);
			
 
				+      if (op < 0) y &= x; else if (op == 0) y |= x; else y ^= x;
			
 
				+    } while (++o < top);
			
 
				+    return bit_result64(L, id, y);
			
 
				+  }
			
 
				+  return FFH_RETRY;
			
 
				+#else
			
 
				   int i = 0;
			
 
				   do { lj_lib_checknumber(L, ++i); } while (L->base+i < L->top);
			
 
				   return FFH_RETRY;
			
 
				+#endif
			
 
				 }
			
 
				 LJLIB_ASM_(bit_bor)		LJLIB_REC(bit_nary IR_BOR)
			
 
				 LJLIB_ASM_(bit_bxor)		LJLIB_REC(bit_nary IR_BXOR)
			
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				-LJLIB_CF(bit_tohex)
			
 
				+LJLIB_CF(bit_tohex)		LJLIB_REC(.)
			
 
				 {
			
 
				-  uint32_t b = (uint32_t)lj_lib_checkbit(L, 1);
			
 
				-  int32_t i, n = L->base+1 >= L->top ? 8 : lj_lib_checkbit(L, 2);
			
 
				-  const char *hexdigits = "0123456789abcdef";
			
 
				-  char buf[8];
			
 
				-  if (n < 0) { n = -n; hexdigits = "0123456789ABCDEF"; }
			
 
				-  if (n > 8) n = 8;
			
 
				-  for (i = n; --i >= 0; ) { buf[i] = hexdigits[b & 15]; b >>= 4; }
			
 
				-  lua_pushlstring(L, buf, (size_t)n);
			
 
				+#if LJ_HASFFI
			
 
				+  CTypeID id = 0, id2 = 0;
			
 
				+  uint64_t b = lj_carith_check64(L, 1, &id);
			
 
				+  int32_t n = L->base+1>=L->top ? (id ? 16 : 8) :
			
 
				+				  (int32_t)lj_carith_check64(L, 2, &id2);
			
 
				+#else
			
 
				+  uint32_t b = (uint32_t)bit_checkbit(L, 1);
			
 
				+  int32_t n = L->base+1>=L->top ? 8 : bit_checkbit(L, 2);
			
 
				+#endif
			
 
				+  SBuf *sb = lj_buf_tmp_(L);
			
 
				+  SFormat sf = (STRFMT_UINT|STRFMT_T_HEX);
			
 
				+  if (n < 0) { n = -n; sf |= STRFMT_F_UPPER; }
			
 
				+  sf |= ((SFormat)((n+1)&255) << STRFMT_SH_PREC);
			
 
				+#if LJ_HASFFI
			
 
				+  if (n < 16) b &= ((uint64_t)1 << 4*n)-1;
			
 
				+#else
			
 
				+  if (n < 8) b &= (1u << 4*n)-1;
			
 
				+#endif
			
 
				+  sb = lj_strfmt_putfxint(sb, sf, b);
			
 
				+  setstrV(L, L->top-1, lj_buf_str(L, sb));
			
 
				+  lj_gc_check(L);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lib_debug.c
+++ b/luajit.mod/luajit/src/lib_debug.c
@@ -29,7 +29,7 @@ LJLIB_CF(debug_getregistry)
 
				   return 1;
			
 
				 }
			
 
				 
			
 
				-LJLIB_CF(debug_getmetatable)
			
 
				+LJLIB_CF(debug_getmetatable)	LJLIB_REC(.)
			
 
				 {
			
 
				   lj_lib_checkany(L, 1);
			
 
				   if (!lua_getmetatable(L, 1)) {
			
@@ -283,13 +283,13 @@ LJLIB_CF(debug_setuservalue)
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				-static const char KEY_HOOK = 'h';
			
 
				+#define KEY_HOOK	((void *)0x3004)
			
 
				 
			
 
				 static void hookf(lua_State *L, lua_Debug *ar)
			
 
				 {
			
 
				   static const char *const hooknames[] =
			
 
				     {"call", "return", "line", "count", "tail return"};
			
 
				-  lua_pushlightuserdata(L, (void *)&KEY_HOOK);
			
 
				+  lua_pushlightuserdata(L, KEY_HOOK);
			
 
				   lua_rawget(L, LUA_REGISTRYINDEX);
			
 
				   if (lua_isfunction(L, -1)) {
			
 
				     lua_pushstring(L, hooknames[(int)ar->event]);
			
@@ -334,7 +334,7 @@ LJLIB_CF(debug_sethook)
 
				     count = luaL_optint(L, arg+3, 0);
			
 
				     func = hookf; mask = makemask(smask, count);
			
 
				   }
			
 
				-  lua_pushlightuserdata(L, (void *)&KEY_HOOK);
			
 
				+  lua_pushlightuserdata(L, KEY_HOOK);
			
 
				   lua_pushvalue(L, arg+1);
			
 
				   lua_rawset(L, LUA_REGISTRYINDEX);
			
 
				   lua_sethook(L, func, mask, count);
			
@@ -349,7 +349,7 @@ LJLIB_CF(debug_gethook)
 
				   if (hook != NULL && hook != hookf) {  /* external hook? */
			
 
				     lua_pushliteral(L, "external hook");
			
 
				   } else {
			
 
				-    lua_pushlightuserdata(L, (void *)&KEY_HOOK);
			
 
				+    lua_pushlightuserdata(L, KEY_HOOK);
			
 
				     lua_rawget(L, LUA_REGISTRYINDEX);   /* get hook */
			
 
				   }
			
 
				   lua_pushstring(L, unmakemask(mask, buff));
			
--- a/luajit.mod/luajit/src/lib_ffi.c
+++ b/luajit.mod/luajit/src/lib_ffi.c
@@ -29,6 +29,7 @@
 
				 #include "lj_ccall.h"
			
 
				 #include "lj_ccallback.h"
			
 
				 #include "lj_clib.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 #include "lj_ff.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
@@ -137,7 +138,7 @@ static int ffi_index_meta(lua_State *L, CTState *cts, CType *ct, MMS mm)
 
				       }
			
 
				     }
			
 
				     copyTV(L, base, L->top);
			
 
				-    tv = L->top-1;
			
 
				+    tv = L->top-1-LJ_FR2;
			
 
				   }
			
 
				   return lj_meta_tailcall(L, tv);
			
 
				 }
			
@@ -193,7 +194,7 @@ LJLIB_CF(ffi_meta___eq)		LJLIB_REC(cdata_arith MM_eq)
 
				 
			
 
				 LJLIB_CF(ffi_meta___len)	LJLIB_REC(cdata_arith MM_len)
			
 
				 {
			
 
				-  return ffi_arith(L);
			
 
				+  return lj_carith_len(L);
			
 
				 }
			
 
				 
			
 
				 LJLIB_CF(ffi_meta___lt)		LJLIB_REC(cdata_arith MM_lt)
			
@@ -318,7 +319,7 @@ LJLIB_CF(ffi_meta___tostring)
 
				       }
			
 
				     }
			
 
				   }
			
 
				-  lj_str_pushf(L, msg, strdata(lj_ctype_repr(L, id, NULL)), p);
			
 
				+  lj_strfmt_pushf(L, msg, strdata(lj_ctype_repr(L, id, NULL)), p);
			
 
				 checkgc:
			
 
				   lj_gc_check(L);
			
 
				   return 1;
			
@@ -504,10 +505,7 @@ LJLIB_CF(ffi_new)	LJLIB_REC(.)
 
				   }
			
 
				   if (sz == CTSIZE_INVALID)
			
 
				     lj_err_arg(L, 1, LJ_ERR_FFI_INVSIZE);
			
 
				-  if (!(info & CTF_VLA) && ctype_align(info) <= CT_MEMALIGN)
			
 
				-    cd = lj_cdata_new(cts, id, sz);
			
 
				-  else
			
 
				-    cd = lj_cdata_newv(cts, id, sz, ctype_align(info));
			
 
				+  cd = lj_cdata_newx(cts, id, sz, info);
			
 
				   setcdataV(L, o-1, cd);  /* Anchor the uninitialized cdata. */
			
 
				   lj_cconv_ct_init(cts, ct, sz, cdataptr(cd),
			
 
				 		   o, (MSize)(L->top - o));  /* Initialize cdata. */
			
@@ -558,6 +556,31 @@ LJLIB_CF(ffi_typeof)	LJLIB_REC(.)
 
				   return 1;
			
 
				 }
			
 
				 
			
 
				+/* Internal and unsupported API. */
			
 
				+LJLIB_CF(ffi_typeinfo)
			
 
				+{
			
 
				+  CTState *cts = ctype_cts(L);
			
 
				+  CTypeID id = (CTypeID)ffi_checkint(L, 1);
			
 
				+  if (id > 0 && id < cts->top) {
			
 
				+    CType *ct = ctype_get(cts, id);
			
 
				+    GCtab *t;
			
 
				+    lua_createtable(L, 0, 4);  /* Increment hash size if fields are added. */
			
 
				+    t = tabV(L->top-1);
			
 
				+    setintV(lj_tab_setstr(L, t, lj_str_newlit(L, "info")), (int32_t)ct->info);
			
 
				+    if (ct->size != CTSIZE_INVALID)
			
 
				+      setintV(lj_tab_setstr(L, t, lj_str_newlit(L, "size")), (int32_t)ct->size);
			
 
				+    if (ct->sib)
			
 
				+      setintV(lj_tab_setstr(L, t, lj_str_newlit(L, "sib")), (int32_t)ct->sib);
			
 
				+    if (gcref(ct->name)) {
			
 
				+      GCstr *s = gco2str(gcref(ct->name));
			
 
				+      setstrV(L, lj_tab_setstr(L, t, lj_str_newlit(L, "name")), s);
			
 
				+    }
			
 
				+    lj_gc_check(L);
			
 
				+    return 1;
			
 
				+  }
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				 LJLIB_CF(ffi_istype)	LJLIB_REC(.)
			
 
				 {
			
 
				   CTState *cts = ctype_cts(L);
			
@@ -723,8 +746,14 @@ LJLIB_CF(ffi_abi)	LJLIB_REC(.)
 
				 #endif
			
 
				 #if LJ_ABI_WIN
			
 
				   case H_(4ab624a8,4ab624a8): b = 1; break;  /* win */
			
 
				+#endif
			
 
				+#if LJ_TARGET_UWP
			
 
				+  case H_(a40f0bcb,a40f0bcb): b = 1; break;  /* uwp */
			
 
				 #endif
			
 
				   case H_(3af93066,1f001464): b = 1; break;  /* le/be */
			
 
				+#if LJ_GC64
			
 
				+  case H_(9e89d2c9,13c83c92): b = 1; break;  /* gc64 */
			
 
				+#endif
			
 
				   default:
			
 
				     break;
			
 
				   }
			
@@ -768,19 +797,11 @@ LJLIB_CF(ffi_gc)	LJLIB_REC(.)
 
				   GCcdata *cd = ffi_checkcdata(L, 1);
			
 
				   TValue *fin = lj_lib_checkany(L, 2);
			
 
				   CTState *cts = ctype_cts(L);
			
 
				-  GCtab *t = cts->finalizer;
			
 
				   CType *ct = ctype_raw(cts, cd->ctypeid);
			
 
				   if (!(ctype_isptr(ct->info) || ctype_isstruct(ct->info) ||
			
 
				 	ctype_isrefarray(ct->info)))
			
 
				     lj_err_arg(L, 1, LJ_ERR_FFI_INVTYPE);
			
 
				-  if (gcref(t->metatable)) {  /* Update finalizer table, if still enabled. */
			
 
				-    copyTV(L, lj_tab_set(L, t, L->base), fin);
			
 
				-    lj_gc_anybarriert(L, t);
			
 
				-    if (!tvisnil(fin))
			
 
				-      cd->marked |= LJ_GC_CDATA_FIN;
			
 
				-    else
			
 
				-      cd->marked &= ~LJ_GC_CDATA_FIN;
			
 
				-  }
			
 
				+  lj_cdata_setfin(L, cd, gcval(fin), itype(fin));
			
 
				   L->top = L->base+1;  /* Pass through the cdata object. */
			
 
				   return 1;
			
 
				 }
			
--- a/luajit.mod/luajit/src/lib_io.c
+++ b/luajit.mod/luajit/src/lib_io.c
@@ -19,8 +19,10 @@
 
				 #include "lj_obj.h"
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				 #include "lj_str.h"
			
 
				 #include "lj_state.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 #include "lj_ff.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
@@ -84,7 +86,7 @@ static IOFileUD *io_file_open(lua_State *L, const char *mode)
 
				   IOFileUD *iof = io_file_new(L);
			
 
				   iof->fp = fopen(fname, mode);
			
 
				   if (iof->fp == NULL)
			
 
				-    luaL_argerror(L, 1, lj_str_pushf(L, "%s: %s", fname, strerror(errno)));
			
 
				+    luaL_argerror(L, 1, lj_strfmt_pushf(L, "%s: %s", fname, strerror(errno)));
			
 
				   return iof;
			
 
				 }
			
 
				 
			
@@ -97,7 +99,7 @@ static int io_file_close(lua_State *L, IOFileUD *iof)
 
				     int stat = -1;
			
 
				 #if LJ_TARGET_POSIX
			
 
				     stat = pclose(iof->fp);
			
 
				-#elif LJ_TARGET_WINDOWS
			
 
				+#elif LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE && !LJ_TARGET_UWP
			
 
				     stat = _pclose(iof->fp);
			
 
				 #else
			
 
				     lua_assert(0);
			
@@ -145,7 +147,7 @@ static int io_file_readline(lua_State *L, FILE *fp, MSize chop)
 
				   MSize m = LUAL_BUFFERSIZE, n = 0, ok = 0;
			
 
				   char *buf;
			
 
				   for (;;) {
			
 
				-    buf = lj_str_needbuf(L, &G(L)->tmpbuf, m);
			
 
				+    buf = lj_buf_tmp(L, m);
			
 
				     if (fgets(buf+n, m-n, fp) == NULL) break;
			
 
				     n += (MSize)strlen(buf+n);
			
 
				     ok |= n;
			
@@ -161,7 +163,7 @@ static void io_file_readall(lua_State *L, FILE *fp)
 
				 {
			
 
				   MSize m, n;
			
 
				   for (m = LUAL_BUFFERSIZE, n = 0; ; m += m) {
			
 
				-    char *buf = lj_str_needbuf(L, &G(L)->tmpbuf, m);
			
 
				+    char *buf = lj_buf_tmp(L, m);
			
 
				     n += (MSize)fread(buf+n, 1, m-n, fp);
			
 
				     if (n != m) {
			
 
				       setstrV(L, L->top++, lj_str_new(L, buf, (size_t)n));
			
@@ -174,7 +176,7 @@ static void io_file_readall(lua_State *L, FILE *fp)
 
				 static int io_file_readlen(lua_State *L, FILE *fp, MSize m)
			
 
				 {
			
 
				   if (m) {
			
 
				-    char *buf = lj_str_needbuf(L, &G(L)->tmpbuf, m);
			
 
				+    char *buf = lj_buf_tmp(L, m);
			
 
				     MSize n = (MSize)fread(buf, 1, m, fp);
			
 
				     setstrV(L, L->top++, lj_str_new(L, buf, (size_t)n));
			
 
				     lj_gc_check(L);
			
@@ -201,13 +203,12 @@ static int io_file_read(lua_State *L, FILE *fp, int start)
 
				     for (n = start; nargs-- && ok; n++) {
			
 
				       if (tvisstr(L->base+n)) {
			
 
				 	const char *p = strVdata(L->base+n);
			
 
				-	if (p[0] != '*')
			
 
				-	  lj_err_arg(L, n+1, LJ_ERR_INVOPT);
			
 
				-	if (p[1] == 'n')
			
 
				+	if (p[0] == '*') p++;
			
 
				+	if (p[0] == 'n')
			
 
				 	  ok = io_file_readnum(L, fp);
			
 
				-	else if ((p[1] & ~0x20) == 'L')
			
 
				-	  ok = io_file_readline(L, fp, (p[1] == 'l'));
			
 
				-	else if (p[1] == 'a')
			
 
				+	else if ((p[0] & ~0x20) == 'L')
			
 
				+	  ok = io_file_readline(L, fp, (p[0] == 'l'));
			
 
				+	else if (p[0] == 'a')
			
 
				 	  io_file_readall(L, fp);
			
 
				 	else
			
 
				 	  lj_err_arg(L, n+1, LJ_ERR_INVFMT);
			
@@ -230,19 +231,11 @@ static int io_file_write(lua_State *L, FILE *fp, int start)
 
				   cTValue *tv;
			
 
				   int status = 1;
			
 
				   for (tv = L->base+start; tv < L->top; tv++) {
			
 
				-    if (tvisstr(tv)) {
			
 
				-      MSize len = strV(tv)->len;
			
 
				-      status = status && (fwrite(strVdata(tv), 1, len, fp) == len);
			
 
				-    } else if (tvisint(tv)) {
			
 
				-      char buf[LJ_STR_INTBUF];
			
 
				-      char *p = lj_str_bufint(buf, intV(tv));
			
 
				-      size_t len = (size_t)(buf+LJ_STR_INTBUF-p);
			
 
				-      status = status && (fwrite(p, 1, len, fp) == len);
			
 
				-    } else if (tvisnum(tv)) {
			
 
				-      status = status && (fprintf(fp, LUA_NUMBER_FMT, numV(tv)) > 0);
			
 
				-    } else {
			
 
				+    MSize len;
			
 
				+    const char *p = lj_strfmt_wstrnum(L, tv, &len);
			
 
				+    if (!p)
			
 
				       lj_err_argt(L, (int)(tv - L->base) + 1, LUA_TSTRING);
			
 
				-    }
			
 
				+    status = status && (fwrite(p, 1, len, fp) == len);
			
 
				   }
			
 
				   if (LJ_52 && status) {
			
 
				     L->top = L->base+1;
			
@@ -413,7 +406,7 @@ LJLIB_CF(io_open)
 
				 
			
 
				 LJLIB_CF(io_popen)
			
 
				 {
			
 
				-#if LJ_TARGET_POSIX || LJ_TARGET_WINDOWS
			
 
				+#if LJ_TARGET_POSIX || (LJ_TARGET_WINDOWS && !LJ_TARGET_XBOXONE && !LJ_TARGET_UWP)
			
 
				   const char *fname = strdata(lj_lib_checkstr(L, 1));
			
 
				   GCstr *s = lj_lib_optstr(L, 2);
			
 
				   const char *mode = s ? strdata(s) : "r";
			
--- a/luajit.mod/luajit/src/lib_jit.c
+++ b/luajit.mod/luajit/src/lib_jit.c
@@ -10,13 +10,17 @@
 
				 #include "lauxlib.h"
			
 
				 #include "lualib.h"
			
 
				 
			
 
				-#include "lj_arch.h"
			
 
				 #include "lj_obj.h"
			
 
				+#include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				 #include "lj_debug.h"
			
 
				 #include "lj_str.h"
			
 
				 #include "lj_tab.h"
			
 
				+#include "lj_state.h"
			
 
				 #include "lj_bc.h"
			
 
				+#if LJ_HASFFI
			
 
				+#include "lj_ctype.h"
			
 
				+#endif
			
 
				 #if LJ_HASJIT
			
 
				 #include "lj_ir.h"
			
 
				 #include "lj_jit.h"
			
@@ -24,6 +28,7 @@
 
				 #include "lj_iropt.h"
			
 
				 #include "lj_target.h"
			
 
				 #endif
			
 
				+#include "lj_trace.h"
			
 
				 #include "lj_dispatch.h"
			
 
				 #include "lj_vm.h"
			
 
				 #include "lj_vmevent.h"
			
@@ -280,7 +285,7 @@ static GCtrace *jit_checktrace(lua_State *L)
 
				 /* Names of link types. ORDER LJ_TRLINK */
			
 
				 static const char *const jit_trlinkname[] = {
			
 
				   "none", "root", "loop", "tail-recursion", "up-recursion", "down-recursion",
			
 
				-  "interpreter", "return"
			
 
				+  "interpreter", "return", "stitch"
			
 
				 };
			
 
				 
			
 
				 /* local info = jit.util.traceinfo(tr) */
			
@@ -333,6 +338,13 @@ LJLIB_CF(jit_util_tracek)
 
				       slot = ir->op2;
			
 
				       ir = &T->ir[ir->op1];
			
 
				     }
			
 
				+#if LJ_HASFFI
			
 
				+    if (ir->o == IR_KINT64 && !ctype_ctsG(G(L))) {
			
 
				+      ptrdiff_t oldtop = savestack(L, L->top);
			
 
				+      luaopen_ffi(L);  /* Load FFI library on-demand. */
			
 
				+      L->top = restorestack(L, oldtop);
			
 
				+    }
			
 
				+#endif
			
 
				     lj_ir_kvalue(L, L->top-2, ir);
			
 
				     setintV(L->top-1, (int32_t)irt_type(ir->t));
			
 
				     if (slot == -1)
			
@@ -417,6 +429,12 @@ LJLIB_CF(jit_util_ircalladdr)
 
				 
			
 
				 #include "lj_libdef.h"
			
 
				 
			
 
				+static int luaopen_jit_util(lua_State *L)
			
 
				+{
			
 
				+  LJ_LIB_REG(L, NULL, jit_util);
			
 
				+  return 1;
			
 
				+}
			
 
				+
			
 
				 /* -- jit.opt module ------------------------------------------------------ */
			
 
				 
			
 
				 #if LJ_HASJIT
			
@@ -514,6 +532,104 @@ LJLIB_CF(jit_opt_start)
 
				 
			
 
				 #endif
			
 
				 
			
 
				+/* -- jit.profile module -------------------------------------------------- */
			
 
				+
			
 
				+#if LJ_HASPROFILE
			
 
				+
			
 
				+#define LJLIB_MODULE_jit_profile
			
 
				+
			
 
				+/* Not loaded by default, use: local profile = require("jit.profile") */
			
 
				+
			
 
				+static const char KEY_PROFILE_THREAD = 't';
			
 
				+static const char KEY_PROFILE_FUNC = 'f';
			
 
				+
			
 
				+static void jit_profile_callback(lua_State *L2, lua_State *L, int samples,
			
 
				+				 int vmstate)
			
 
				+{
			
 
				+  TValue key;
			
 
				+  cTValue *tv;
			
 
				+  setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
			
 
				+  tv = lj_tab_get(L, tabV(registry(L)), &key);
			
 
				+  if (tvisfunc(tv)) {
			
 
				+    char vmst = (char)vmstate;
			
 
				+    int status;
			
 
				+    setfuncV(L2, L2->top++, funcV(tv));
			
 
				+    setthreadV(L2, L2->top++, L);
			
 
				+    setintV(L2->top++, samples);
			
 
				+    setstrV(L2, L2->top++, lj_str_new(L2, &vmst, 1));
			
 
				+    status = lua_pcall(L2, 3, 0, 0);  /* callback(thread, samples, vmstate) */
			
 
				+    if (status) {
			
 
				+      if (G(L2)->panic) G(L2)->panic(L2);
			
 
				+      exit(EXIT_FAILURE);
			
 
				+    }
			
 
				+    lj_trace_abort(G(L2));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* profile.start(mode, cb) */
			
 
				+LJLIB_CF(jit_profile_start)
			
 
				+{
			
 
				+  GCtab *registry = tabV(registry(L));
			
 
				+  GCstr *mode = lj_lib_optstr(L, 1);
			
 
				+  GCfunc *func = lj_lib_checkfunc(L, 2);
			
 
				+  lua_State *L2 = lua_newthread(L);  /* Thread that runs profiler callback. */
			
 
				+  TValue key;
			
 
				+  /* Anchor thread and function in registry. */
			
 
				+  setlightudV(&key, (void *)&KEY_PROFILE_THREAD);
			
 
				+  setthreadV(L, lj_tab_set(L, registry, &key), L2);
			
 
				+  setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
			
 
				+  setfuncV(L, lj_tab_set(L, registry, &key), func);
			
 
				+  lj_gc_anybarriert(L, registry);
			
 
				+  luaJIT_profile_start(L, mode ? strdata(mode) : "",
			
 
				+		       (luaJIT_profile_callback)jit_profile_callback, L2);
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* profile.stop() */
			
 
				+LJLIB_CF(jit_profile_stop)
			
 
				+{
			
 
				+  GCtab *registry;
			
 
				+  TValue key;
			
 
				+  luaJIT_profile_stop(L);
			
 
				+  registry = tabV(registry(L));
			
 
				+  setlightudV(&key, (void *)&KEY_PROFILE_THREAD);
			
 
				+  setnilV(lj_tab_set(L, registry, &key));
			
 
				+  setlightudV(&key, (void *)&KEY_PROFILE_FUNC);
			
 
				+  setnilV(lj_tab_set(L, registry, &key));
			
 
				+  lj_gc_anybarriert(L, registry);
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* dump = profile.dumpstack([thread,] fmt, depth) */
			
 
				+LJLIB_CF(jit_profile_dumpstack)
			
 
				+{
			
 
				+  lua_State *L2 = L;
			
 
				+  int arg = 0;
			
 
				+  size_t len;
			
 
				+  int depth;
			
 
				+  GCstr *fmt;
			
 
				+  const char *p;
			
 
				+  if (L->top > L->base && tvisthread(L->base)) {
			
 
				+    L2 = threadV(L->base);
			
 
				+    arg = 1;
			
 
				+  }
			
 
				+  fmt = lj_lib_checkstr(L, arg+1);
			
 
				+  depth = lj_lib_checkint(L, arg+2);
			
 
				+  p = luaJIT_profile_dumpstack(L2, strdata(fmt), depth, &len);
			
 
				+  lua_pushlstring(L, p, len);
			
 
				+  return 1;
			
 
				+}
			
 
				+
			
 
				+#include "lj_libdef.h"
			
 
				+
			
 
				+static int luaopen_jit_profile(lua_State *L)
			
 
				+{
			
 
				+  LJ_LIB_REG(L, NULL, jit_profile);
			
 
				+  return 1;
			
 
				+}
			
 
				+
			
 
				+#endif
			
 
				+
			
 
				 /* -- JIT compiler initialization ----------------------------------------- */
			
 
				 
			
 
				 #if LJ_HASJIT
			
@@ -539,38 +655,31 @@ static uint32_t jit_cpudetect(lua_State *L)
 
				   uint32_t features[4];
			
 
				   if (lj_vm_cpuid(0, vendor) && lj_vm_cpuid(1, features)) {
			
 
				 #if !LJ_HASJIT
			
 
				-#define JIT_F_CMOV	1
			
 
				 #define JIT_F_SSE2	2
			
 
				 #endif
			
 
				-    flags |= ((features[3] >> 15)&1) * JIT_F_CMOV;
			
 
				     flags |= ((features[3] >> 26)&1) * JIT_F_SSE2;
			
 
				 #if LJ_HASJIT
			
 
				     flags |= ((features[2] >> 0)&1) * JIT_F_SSE3;
			
 
				     flags |= ((features[2] >> 19)&1) * JIT_F_SSE4_1;
			
 
				     if (vendor[2] == 0x6c65746e) {  /* Intel. */
			
 
				-      if ((features[0] & 0x0ff00f00) == 0x00000f00)  /* P4. */
			
 
				-	flags |= JIT_F_P4;  /* Currently unused. */
			
 
				-      else if ((features[0] & 0x0fff0ff0) == 0x000106c0)  /* Atom. */
			
 
				+      if ((features[0] & 0x0fff0ff0) == 0x000106c0)  /* Atom. */
			
 
				 	flags |= JIT_F_LEA_AGU;
			
 
				     } else if (vendor[2] == 0x444d4163) {  /* AMD. */
			
 
				       uint32_t fam = (features[0] & 0x0ff00f00);
			
 
				-      if (fam == 0x00000f00)  /* K8. */
			
 
				-	flags |= JIT_F_SPLIT_XMM;
			
 
				       if (fam >= 0x00000f00)  /* K8, K10. */
			
 
				 	flags |= JIT_F_PREFER_IMUL;
			
 
				     }
			
 
				+    if (vendor[0] >= 7) {
			
 
				+      uint32_t xfeatures[4];
			
 
				+      lj_vm_cpuid(7, xfeatures);
			
 
				+      flags |= ((xfeatures[1] >> 8)&1) * JIT_F_BMI2;
			
 
				+    }
			
 
				 #endif
			
 
				   }
			
 
				   /* Check for required instruction set support on x86 (unnecessary on x64). */
			
 
				 #if LJ_TARGET_X86
			
 
				-#if !defined(LUAJIT_CPU_NOCMOV)
			
 
				-  if (!(flags & JIT_F_CMOV))
			
 
				-    luaL_error(L, "CPU not supported");
			
 
				-#endif
			
 
				-#if defined(LUAJIT_CPU_SSE2)
			
 
				   if (!(flags & JIT_F_SSE2))
			
 
				-    luaL_error(L, "CPU does not support SSE2 (recompile without -DLUAJIT_CPU_SSE2)");
			
 
				-#endif
			
 
				+    luaL_error(L, "CPU with SSE2 required");
			
 
				 #endif
			
 
				 #elif LJ_TARGET_ARM
			
 
				 #if LJ_HASJIT
			
@@ -592,6 +701,8 @@ static uint32_t jit_cpudetect(lua_State *L)
 
				 	   ver >= 60 ? JIT_F_ARMV6_ : 0;
			
 
				   flags |= LJ_ARCH_HASFPU == 0 ? 0 : ver >= 70 ? JIT_F_VFPV3 : JIT_F_VFPV2;
			
 
				 #endif
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+  /* No optional CPU features to detect (for now). */
			
 
				 #elif LJ_TARGET_PPC
			
 
				 #if LJ_HASJIT
			
 
				 #if LJ_ARCH_SQRT
			
@@ -601,21 +712,23 @@ static uint32_t jit_cpudetect(lua_State *L)
 
				   flags |= JIT_F_ROUND;
			
 
				 #endif
			
 
				 #endif
			
 
				-#elif LJ_TARGET_PPCSPE
			
 
				-  /* Nothing to do. */
			
 
				 #elif LJ_TARGET_MIPS
			
 
				 #if LJ_HASJIT
			
 
				   /* Compile-time MIPS CPU detection. */
			
 
				 #if LJ_ARCH_VERSION >= 20
			
 
				-  flags |= JIT_F_MIPS32R2;
			
 
				+  flags |= JIT_F_MIPSXXR2;
			
 
				 #endif
			
 
				   /* Runtime MIPS CPU detection. */
			
 
				 #if defined(__GNUC__)
			
 
				-  if (!(flags & JIT_F_MIPS32R2)) {
			
 
				+  if (!(flags & JIT_F_MIPSXXR2)) {
			
 
				     int x;
			
 
				+#ifdef __mips16
			
 
				+    x = 0;  /* Runtime detection is difficult. Ensure optimal -march flags. */
			
 
				+#else
			
 
				     /* On MIPS32R1 rotr is treated as srl. rotr r2,r2,1 -> srl r2,r2,1. */
			
 
				     __asm__("li $2, 1\n\t.long 0x00221042\n\tmove %0, $2" : "=r"(x) : : "$2");
			
 
				-    if (x) flags |= JIT_F_MIPS32R2;  /* Either 0x80000000 (R2) or 0 (R1). */
			
 
				+#endif
			
 
				+    if (x) flags |= JIT_F_MIPSXXR2;  /* Either 0x80000000 (R2) or 0 (R1). */
			
 
				   }
			
 
				 #endif
			
 
				 #endif
			
@@ -632,11 +745,7 @@ static void jit_init(lua_State *L)
 
				   uint32_t flags = jit_cpudetect(L);
			
 
				 #if LJ_HASJIT
			
 
				   jit_State *J = L2J(L);
			
 
				-#if LJ_TARGET_X86
			
 
				-  /* Silently turn off the JIT compiler on CPUs without SSE2. */
			
 
				-  if ((flags & JIT_F_SSE2))
			
 
				-#endif
			
 
				-    J->flags = flags | JIT_F_ON | JIT_F_OPT_DEFAULT;
			
 
				+  J->flags = flags | JIT_F_ON | JIT_F_OPT_DEFAULT;
			
 
				   memcpy(J->param, jit_param_default, sizeof(J->param));
			
 
				   lj_dispatch_update(G(L));
			
 
				 #else
			
@@ -646,19 +755,23 @@ static void jit_init(lua_State *L)
 
				 
			
 
				 LUALIB_API int luaopen_jit(lua_State *L)
			
 
				 {
			
 
				+  jit_init(L);
			
 
				   lua_pushliteral(L, LJ_OS_NAME);
			
 
				   lua_pushliteral(L, LJ_ARCH_NAME);
			
 
				   lua_pushinteger(L, LUAJIT_VERSION_NUM);
			
 
				   lua_pushliteral(L, LUAJIT_VERSION);
			
 
				   LJ_LIB_REG(L, LUA_JITLIBNAME, jit);
			
 
				+#if LJ_HASPROFILE
			
 
				+  lj_lib_prereg(L, LUA_JITLIBNAME ".profile", luaopen_jit_profile,
			
 
				+		tabref(L->env));
			
 
				+#endif
			
 
				 #ifndef LUAJIT_DISABLE_JITUTIL
			
 
				-  LJ_LIB_REG(L, "jit.util", jit_util);
			
 
				+  lj_lib_prereg(L, LUA_JITLIBNAME ".util", luaopen_jit_util, tabref(L->env));
			
 
				 #endif
			
 
				 #if LJ_HASJIT
			
 
				   LJ_LIB_REG(L, "jit.opt", jit_opt);
			
 
				 #endif
			
 
				   L->top -= 2;
			
 
				-  jit_init(L);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lib_math.c
+++ b/luajit.mod/luajit/src/lib_math.c
@@ -47,12 +47,6 @@ LJLIB_ASM_(math_tanh)		LJLIB_REC(math_htrig IRCALL_tanh)
 
				 LJLIB_ASM_(math_frexp)
			
 
				 LJLIB_ASM_(math_modf)		LJLIB_REC(.)
			
 
				 
			
 
				-LJLIB_PUSH(57.29577951308232)
			
 
				-LJLIB_ASM_(math_deg)		LJLIB_REC(math_degrad)
			
 
				-
			
 
				-LJLIB_PUSH(0.017453292519943295)
			
 
				-LJLIB_ASM_(math_rad)		LJLIB_REC(math_degrad)
			
 
				-
			
 
				 LJLIB_ASM(math_log)		LJLIB_REC(math_log)
			
 
				 {
			
 
				   double x = lj_lib_checknum(L, 1);
			
@@ -63,12 +57,15 @@ LJLIB_ASM(math_log)		LJLIB_REC(math_log)
 
				 #else
			
 
				     x = lj_vm_log2(x); y = 1.0 / lj_vm_log2(y);
			
 
				 #endif
			
 
				-    setnumV(L->base-1, x*y);  /* Do NOT join the expression to x / y. */
			
 
				+    setnumV(L->base-1-LJ_FR2, x*y);  /* Do NOT join the expression to x / y. */
			
 
				     return FFH_RES(1);
			
 
				   }
			
 
				   return FFH_RETRY;
			
 
				 }
			
 
				 
			
 
				+LJLIB_LUA(math_deg) /* function(x) return x * 57.29577951308232 end */
			
 
				+LJLIB_LUA(math_rad) /* function(x) return x * 0.017453292519943295 end */
			
 
				+
			
 
				 LJLIB_ASM(math_atan2)		LJLIB_REC(.)
			
 
				 {
			
 
				   lj_lib_checknum(L, 1);
			
@@ -224,10 +221,6 @@ LUALIB_API int luaopen_math(lua_State *L)
 
				   rs = (RandomState *)lua_newuserdata(L, sizeof(RandomState));
			
 
				   rs->valid = 0;  /* Use lazy initialization to save some time on startup. */
			
 
				   LJ_LIB_REG(L, LUA_MATHLIBNAME, math);
			
 
				-#if defined(LUA_COMPAT_MOD) && !LJ_52
			
 
				-  lua_getfield(L, -1, "fmod");
			
 
				-  lua_setfield(L, -2, "mod");
			
 
				-#endif
			
 
				   return 1;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lib_os.c
+++ b/luajit.mod/luajit/src/lib_os.c
@@ -17,7 +17,10 @@
 
				 #include "lualib.h"
			
 
				 
			
 
				 #include "lj_obj.h"
			
 
				+#include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				+#include "lj_str.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
 
				 #if LJ_TARGET_POSIX
			
@@ -188,7 +191,7 @@ LJLIB_CF(os_date)
 
				 #endif
			
 
				   }
			
 
				   if (stm == NULL) {  /* Invalid date? */
			
 
				-    setnilV(L->top-1);
			
 
				+    setnilV(L->top++);
			
 
				   } else if (strcmp(s, "*t") == 0) {
			
 
				     lua_createtable(L, 0, 9);  /* 9 = number of fields */
			
 
				     setfield(L, "sec", stm->tm_sec);
			
@@ -200,23 +203,25 @@ LJLIB_CF(os_date)
 
				     setfield(L, "wday", stm->tm_wday+1);
			
 
				     setfield(L, "yday", stm->tm_yday+1);
			
 
				     setboolfield(L, "isdst", stm->tm_isdst);
			
 
				-  } else {
			
 
				-    char cc[3];
			
 
				-    luaL_Buffer b;
			
 
				-    cc[0] = '%'; cc[2] = '\0';
			
 
				-    luaL_buffinit(L, &b);
			
 
				-    for (; *s; s++) {
			
 
				-      if (*s != '%' || *(s + 1) == '\0') {  /* No conversion specifier? */
			
 
				-	luaL_addchar(&b, *s);
			
 
				-      } else {
			
 
				-	size_t reslen;
			
 
				-	char buff[200];  /* Should be big enough for any conversion result. */
			
 
				-	cc[1] = *(++s);
			
 
				-	reslen = strftime(buff, sizeof(buff), cc, stm);
			
 
				-	luaL_addlstring(&b, buff, reslen);
			
 
				+  } else if (*s) {
			
 
				+    SBuf *sb = &G(L)->tmpbuf;
			
 
				+    MSize sz = 0, retry = 4;
			
 
				+    const char *q;
			
 
				+    for (q = s; *q; q++)
			
 
				+      sz += (*q == '%') ? 30 : 1;  /* Overflow doesn't matter. */
			
 
				+    setsbufL(sb, L);
			
 
				+    while (retry--) {  /* Limit growth for invalid format or empty result. */
			
 
				+      char *buf = lj_buf_need(sb, sz);
			
 
				+      size_t len = strftime(buf, sbufsz(sb), s, stm);
			
 
				+      if (len) {
			
 
				+	setstrV(L, L->top++, lj_str_new(L, buf, len));
			
 
				+	lj_gc_check(L);
			
 
				+	break;
			
 
				       }
			
 
				+      sz += (sz|1);
			
 
				     }
			
 
				-    luaL_pushresult(&b);
			
 
				+  } else {
			
 
				+    setstrV(L, L->top++, &G(L)->strempty);
			
 
				   }
			
 
				   return 1;
			
 
				 }
			
--- a/luajit.mod/luajit/src/lib_package.c
+++ b/luajit.mod/luajit/src/lib_package.c
@@ -76,6 +76,20 @@ static const char *ll_bcsym(void *lib, const char *sym)
 
				 BOOL WINAPI GetModuleHandleExA(DWORD, LPCSTR, HMODULE*);
			
 
				 #endif
			
 
				 
			
 
				+#if LJ_TARGET_UWP
			
 
				+void *LJ_WIN_LOADLIBA(const char *path)
			
 
				+{
			
 
				+  DWORD err = GetLastError();
			
 
				+  wchar_t wpath[256];
			
 
				+  HANDLE lib = NULL;
			
 
				+  if (MultiByteToWideChar(CP_ACP, 0, path, -1, wpath, 256) > 0) {
			
 
				+    lib = LoadPackagedLibrary(wpath, 0);
			
 
				+  }
			
 
				+  SetLastError(err);
			
 
				+  return lib;
			
 
				+}
			
 
				+#endif
			
 
				+
			
 
				 #undef setprogdir
			
 
				 
			
 
				 static void setprogdir(lua_State *L)
			
@@ -96,9 +110,17 @@ static void setprogdir(lua_State *L)
 
				 static void pusherror(lua_State *L)
			
 
				 {
			
 
				   DWORD error = GetLastError();
			
 
				+#if LJ_TARGET_XBOXONE
			
 
				+  wchar_t wbuffer[128];
			
 
				+  char buffer[128*2];
			
 
				+  if (FormatMessageW(FORMAT_MESSAGE_IGNORE_INSERTS | FORMAT_MESSAGE_FROM_SYSTEM,
			
 
				+      NULL, error, 0, wbuffer, sizeof(wbuffer)/sizeof(wchar_t), NULL) &&
			
 
				+      WideCharToMultiByte(CP_ACP, 0, wbuffer, 128, buffer, 128*2, NULL, NULL))
			
 
				+#else
			
 
				   char buffer[128];
			
 
				   if (FormatMessageA(FORMAT_MESSAGE_IGNORE_INSERTS | FORMAT_MESSAGE_FROM_SYSTEM,
			
 
				       NULL, error, 0, buffer, sizeof(buffer), NULL))
			
 
				+#endif
			
 
				     lua_pushstring(L, buffer);
			
 
				   else
			
 
				     lua_pushfstring(L, "system error %d\n", error);
			
@@ -111,7 +133,7 @@ static void ll_unloadlib(void *lib)
 
				 
			
 
				 static void *ll_load(lua_State *L, const char *path, int gl)
			
 
				 {
			
 
				-  HINSTANCE lib = LoadLibraryA(path);
			
 
				+  HINSTANCE lib = LJ_WIN_LOADLIBA(path);
			
 
				   if (lib == NULL) pusherror(L);
			
 
				   UNUSED(gl);
			
 
				   return lib;
			
@@ -124,17 +146,25 @@ static lua_CFunction ll_sym(lua_State *L, void *lib, const char *sym)
 
				   return f;
			
 
				 }
			
 
				 
			
 
				+#if LJ_TARGET_UWP
			
 
				+EXTERN_C IMAGE_DOS_HEADER __ImageBase;
			
 
				+#endif
			
 
				+
			
 
				 static const char *ll_bcsym(void *lib, const char *sym)
			
 
				 {
			
 
				   if (lib) {
			
 
				     return (const char *)GetProcAddress((HINSTANCE)lib, sym);
			
 
				   } else {
			
 
				+#if LJ_TARGET_UWP
			
 
				+    return (const char *)GetProcAddress((HINSTANCE)&__ImageBase, sym);
			
 
				+#else
			
 
				     HINSTANCE h = GetModuleHandleA(NULL);
			
 
				     const char *p = (const char *)GetProcAddress(h, sym);
			
 
				     if (p == NULL && GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS|GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT,
			
 
				 					(const char *)ll_bcsym, &h))
			
 
				       p = (const char *)GetProcAddress(h, sym);
			
 
				     return p;
			
 
				+#endif
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -185,8 +215,7 @@ static void **ll_register(lua_State *L, const char *path)
 
				     lua_pop(L, 1);
			
 
				     plib = (void **)lua_newuserdata(L, sizeof(void *));
			
 
				     *plib = NULL;
			
 
				-    luaL_getmetatable(L, "_LOADLIB");
			
 
				-    lua_setmetatable(L, -2);
			
 
				+    luaL_setmetatable(L, "_LOADLIB");
			
 
				     lua_pushfstring(L, "LOADLIB: %s", path);
			
 
				     lua_pushvalue(L, -2);
			
 
				     lua_settable(L, LUA_REGISTRYINDEX);
			
@@ -226,7 +255,7 @@ static int ll_loadfunc(lua_State *L, const char *path, const char *name, int r)
 
				       const char *bcdata = ll_bcsym(*reg, mksymname(L, name, SYMPREFIX_BC));
			
 
				       lua_pop(L, 1);
			
 
				       if (bcdata) {
			
 
				-	if (luaL_loadbuffer(L, bcdata, ~(size_t)0, name) != 0)
			
 
				+	if (luaL_loadbuffer(L, bcdata, LJ_MAX_BUF, name) != 0)
			
 
				 	  return PACKAGE_ERR_LOAD;
			
 
				 	return 0;
			
 
				       }
			
@@ -383,7 +412,7 @@ static int lj_cf_package_loader_preload(lua_State *L)
 
				   if (lua_isnil(L, -1)) {  /* Not found? */
			
 
				     const char *bcname = mksymname(L, name, SYMPREFIX_BC);
			
 
				     const char *bcdata = ll_bcsym(NULL, bcname);
			
 
				-    if (bcdata == NULL || luaL_loadbuffer(L, bcdata, ~(size_t)0, name) != 0)
			
 
				+    if (bcdata == NULL || luaL_loadbuffer(L, bcdata, LJ_MAX_BUF, name) != 0)
			
 
				       lua_pushfstring(L, "\n\tno field package.preload['%s']", name);
			
 
				   }
			
 
				   return 1;
			
@@ -391,8 +420,7 @@ static int lj_cf_package_loader_preload(lua_State *L)
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				-static const int sentinel_ = 0;
			
 
				-#define sentinel	((void *)&sentinel_)
			
 
				+#define sentinel	((void *)0x4004)
			
 
				 
			
 
				 static int lj_cf_package_require(lua_State *L)
			
 
				 {
			
@@ -482,29 +510,19 @@ static void modinit(lua_State *L, const char *modname)
 
				 static int lj_cf_package_module(lua_State *L)
			
 
				 {
			
 
				   const char *modname = luaL_checkstring(L, 1);
			
 
				-  int loaded = lua_gettop(L) + 1;  /* index of _LOADED table */
			
 
				-  lua_getfield(L, LUA_REGISTRYINDEX, "_LOADED");
			
 
				-  lua_getfield(L, loaded, modname);  /* get _LOADED[modname] */
			
 
				-  if (!lua_istable(L, -1)) {  /* not found? */
			
 
				-    lua_pop(L, 1);  /* remove previous result */
			
 
				-    /* try global variable (and create one if it does not exist) */
			
 
				-    if (luaL_findtable(L, LUA_GLOBALSINDEX, modname, 1) != NULL)
			
 
				-      lj_err_callerv(L, LJ_ERR_BADMODN, modname);
			
 
				-    lua_pushvalue(L, -1);
			
 
				-    lua_setfield(L, loaded, modname);  /* _LOADED[modname] = new table */
			
 
				-  }
			
 
				-  /* check whether table already has a _NAME field */
			
 
				+  int lastarg = (int)(L->top - L->base);
			
 
				+  luaL_pushmodule(L, modname, 1);
			
 
				   lua_getfield(L, -1, "_NAME");
			
 
				-  if (!lua_isnil(L, -1)) {  /* is table an initialized module? */
			
 
				+  if (!lua_isnil(L, -1)) {  /* Module already initialized? */
			
 
				     lua_pop(L, 1);
			
 
				-  } else {  /* no; initialize it */
			
 
				+  } else {
			
 
				     lua_pop(L, 1);
			
 
				     modinit(L, modname);
			
 
				   }
			
 
				   lua_pushvalue(L, -1);
			
 
				   setfenv(L);
			
 
				-  dooptions(L, loaded - 1);
			
 
				-  return 0;
			
 
				+  dooptions(L, lastarg);
			
 
				+  return LJ_52;
			
 
				 }
			
 
				 
			
 
				 static int lj_cf_package_seeall(lua_State *L)
			
@@ -575,13 +593,16 @@ LUALIB_API int luaopen_package(lua_State *L)
 
				   lj_lib_pushcf(L, lj_cf_package_unloadlib, 1);
			
 
				   lua_setfield(L, -2, "__gc");
			
 
				   luaL_register(L, LUA_LOADLIBNAME, package_lib);
			
 
				-  lua_pushvalue(L, -1);
			
 
				-  lua_replace(L, LUA_ENVIRONINDEX);
			
 
				+  lua_copy(L, -1, LUA_ENVIRONINDEX);
			
 
				   lua_createtable(L, sizeof(package_loaders)/sizeof(package_loaders[0])-1, 0);
			
 
				   for (i = 0; package_loaders[i] != NULL; i++) {
			
 
				     lj_lib_pushcf(L, package_loaders[i], 1);
			
 
				     lua_rawseti(L, -2, i+1);
			
 
				   }
			
 
				+#if LJ_52
			
 
				+  lua_pushvalue(L, -1);
			
 
				+  lua_setfield(L, -3, "searchers");
			
 
				+#endif
			
 
				   lua_setfield(L, -2, "loaders");
			
 
				   lua_getfield(L, LUA_REGISTRYINDEX, "LUA_NOENV");
			
 
				   noenv = lua_toboolean(L, -1);
			
--- a/luajit.mod/luajit/src/lib_string.c
+++ b/luajit.mod/luajit/src/lib_string.c
@@ -6,8 +6,6 @@
 
				 ** Copyright (C) 1994-2008 Lua.org, PUC-Rio. See Copyright Notice in lua.h
			
 
				 */
			
 
				 
			
 
				-#include <stdio.h>
			
 
				-
			
 
				 #define lib_string_c
			
 
				 #define LUA_LIB
			
 
				 
			
@@ -18,6 +16,7 @@
 
				 #include "lj_obj.h"
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				 #include "lj_str.h"
			
 
				 #include "lj_tab.h"
			
 
				 #include "lj_meta.h"
			
@@ -25,17 +24,19 @@
 
				 #include "lj_ff.h"
			
 
				 #include "lj_bcdump.h"
			
 
				 #include "lj_char.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				 #define LJLIB_MODULE_string
			
 
				 
			
 
				-LJLIB_ASM(string_len)		LJLIB_REC(.)
			
 
				-{
			
 
				-  lj_lib_checkstr(L, 1);
			
 
				-  return FFH_RETRY;
			
 
				-}
			
 
				+LJLIB_LUA(string_len) /*
			
 
				+  function(s)
			
 
				+    CHECK_str(s)
			
 
				+    return #s
			
 
				+  end
			
 
				+*/
			
 
				 
			
 
				 LJLIB_ASM(string_byte)		LJLIB_REC(string_range 0)
			
 
				 {
			
@@ -57,21 +58,21 @@ LJLIB_ASM(string_byte)		LJLIB_REC(string_range 0)
 
				   lj_state_checkstack(L, (MSize)n);
			
 
				   p = (const unsigned char *)strdata(s) + start;
			
 
				   for (i = 0; i < n; i++)
			
 
				-    setintV(L->base + i-1, p[i]);
			
 
				+    setintV(L->base + i-1-LJ_FR2, p[i]);
			
 
				   return FFH_RES(n);
			
 
				 }
			
 
				 
			
 
				-LJLIB_ASM(string_char)
			
 
				+LJLIB_ASM(string_char)		LJLIB_REC(.)
			
 
				 {
			
 
				   int i, nargs = (int)(L->top - L->base);
			
 
				-  char *buf = lj_str_needbuf(L, &G(L)->tmpbuf, (MSize)nargs);
			
 
				+  char *buf = lj_buf_tmp(L, (MSize)nargs);
			
 
				   for (i = 1; i <= nargs; i++) {
			
 
				     int32_t k = lj_lib_checkint(L, i);
			
 
				     if (!checku8(k))
			
 
				       lj_err_arg(L, i, LJ_ERR_BADVAL);
			
 
				     buf[i-1] = (char)k;
			
 
				   }
			
 
				-  setstrV(L, L->base-1, lj_str_new(L, buf, (size_t)nargs));
			
 
				+  setstrV(L, L->base-1-LJ_FR2, lj_str_new(L, buf, (size_t)nargs));
			
 
				   return FFH_RES(1);
			
 
				 }
			
 
				 
			
@@ -83,68 +84,38 @@ LJLIB_ASM(string_sub)		LJLIB_REC(string_range 1)
 
				   return FFH_RETRY;
			
 
				 }
			
 
				 
			
 
				-LJLIB_ASM(string_rep)
			
 
				+LJLIB_CF(string_rep)		LJLIB_REC(.)
			
 
				 {
			
 
				   GCstr *s = lj_lib_checkstr(L, 1);
			
 
				-  int32_t k = lj_lib_checkint(L, 2);
			
 
				+  int32_t rep = lj_lib_checkint(L, 2);
			
 
				   GCstr *sep = lj_lib_optstr(L, 3);
			
 
				-  int32_t len = (int32_t)s->len;
			
 
				-  global_State *g = G(L);
			
 
				-  int64_t tlen;
			
 
				-  const char *src;
			
 
				-  char *buf;
			
 
				-  if (k <= 0) {
			
 
				-  empty:
			
 
				-    setstrV(L, L->base-1, &g->strempty);
			
 
				-    return FFH_RES(1);
			
 
				-  }
			
 
				-  if (sep) {
			
 
				-    tlen = (int64_t)len + sep->len;
			
 
				-    if (tlen > LJ_MAX_STR)
			
 
				-      lj_err_caller(L, LJ_ERR_STROV);
			
 
				-    tlen *= k;
			
 
				-    if (tlen > LJ_MAX_STR)
			
 
				-      lj_err_caller(L, LJ_ERR_STROV);
			
 
				-  } else {
			
 
				-    tlen = (int64_t)k * len;
			
 
				-    if (tlen > LJ_MAX_STR)
			
 
				-      lj_err_caller(L, LJ_ERR_STROV);
			
 
				-  }
			
 
				-  if (tlen == 0) goto empty;
			
 
				-  buf = lj_str_needbuf(L, &g->tmpbuf, (MSize)tlen);
			
 
				-  src = strdata(s);
			
 
				-  if (sep) {
			
 
				-    tlen -= sep->len;  /* Ignore trailing separator. */
			
 
				-    if (k > 1) {  /* Paste one string and one separator. */
			
 
				-      int32_t i;
			
 
				-      i = 0; while (i < len) *buf++ = src[i++];
			
 
				-      src = strdata(sep); len = sep->len;
			
 
				-      i = 0; while (i < len) *buf++ = src[i++];
			
 
				-      src = g->tmpbuf.buf; len += s->len; k--;  /* Now copy that k-1 times. */
			
 
				-    }
			
 
				+  SBuf *sb = lj_buf_tmp_(L);
			
 
				+  if (sep && rep > 1) {
			
 
				+    GCstr *s2 = lj_buf_cat2str(L, sep, s);
			
 
				+    lj_buf_reset(sb);
			
 
				+    lj_buf_putstr(sb, s);
			
 
				+    s = s2;
			
 
				+    rep--;
			
 
				   }
			
 
				-  do {
			
 
				-    int32_t i = 0;
			
 
				-    do { *buf++ = src[i++]; } while (i < len);
			
 
				-  } while (--k > 0);
			
 
				-  setstrV(L, L->base-1, lj_str_new(L, g->tmpbuf.buf, (size_t)tlen));
			
 
				-  return FFH_RES(1);
			
 
				+  sb = lj_buf_putstr_rep(sb, s, rep);
			
 
				+  setstrV(L, L->top-1, lj_buf_str(L, sb));
			
 
				+  lj_gc_check(L);
			
 
				+  return 1;
			
 
				 }
			
 
				 
			
 
				-LJLIB_ASM(string_reverse)
			
 
				+LJLIB_ASM(string_reverse)  LJLIB_REC(string_op IRCALL_lj_buf_putstr_reverse)
			
 
				 {
			
 
				-  GCstr *s = lj_lib_checkstr(L, 1);
			
 
				-  lj_str_needbuf(L, &G(L)->tmpbuf, s->len);
			
 
				+  lj_lib_checkstr(L, 1);
			
 
				   return FFH_RETRY;
			
 
				 }
			
 
				-LJLIB_ASM_(string_lower)
			
 
				-LJLIB_ASM_(string_upper)
			
 
				+LJLIB_ASM_(string_lower)  LJLIB_REC(string_op IRCALL_lj_buf_putstr_lower)
			
 
				+LJLIB_ASM_(string_upper)  LJLIB_REC(string_op IRCALL_lj_buf_putstr_upper)
			
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				-static int writer_buf(lua_State *L, const void *p, size_t size, void *b)
			
 
				+static int writer_buf(lua_State *L, const void *p, size_t size, void *sb)
			
 
				 {
			
 
				-  luaL_addlstring((luaL_Buffer *)b, (const char *)p, size);
			
 
				+  lj_buf_putmem((SBuf *)sb, p, (MSize)size);
			
 
				   UNUSED(L);
			
 
				   return 0;
			
 
				 }
			
@@ -153,12 +124,12 @@ LJLIB_CF(string_dump)
 
				 {
			
 
				   GCfunc *fn = lj_lib_checkfunc(L, 1);
			
 
				   int strip = L->base+1 < L->top && tvistruecond(L->base+1);
			
 
				-  luaL_Buffer b;
			
 
				+  SBuf *sb = lj_buf_tmp_(L);  /* Assumes lj_bcwrite() doesn't use tmpbuf. */
			
 
				   L->top = L->base+1;
			
 
				-  luaL_buffinit(L, &b);
			
 
				-  if (!isluafunc(fn) || lj_bcwrite(L, funcproto(fn), writer_buf, &b, strip))
			
 
				+  if (!isluafunc(fn) || lj_bcwrite(L, funcproto(fn), writer_buf, sb, strip))
			
 
				     lj_err_caller(L, LJ_ERR_STRDUMP);
			
 
				-  luaL_pushresult(&b);
			
 
				+  setstrV(L, L->top-1, lj_buf_str(L, sb));
			
 
				+  lj_gc_check(L);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
@@ -183,7 +154,6 @@ typedef struct MatchState {
 
				 } MatchState;
			
 
				 
			
 
				 #define L_ESC		'%'
			
 
				-#define SPECIALS	"^$*+?.([%-"
			
 
				 
			
 
				 static int check_capture(MatchState *ms, int l)
			
 
				 {
			
@@ -450,30 +420,6 @@ static const char *match(MatchState *ms, const char *s, const char *p)
 
				   return s;
			
 
				 }
			
 
				 
			
 
				-static const char *lmemfind(const char *s1, size_t l1,
			
 
				-			    const char *s2, size_t l2)
			
 
				-{
			
 
				-  if (l2 == 0) {
			
 
				-    return s1;  /* empty strings are everywhere */
			
 
				-  } else if (l2 > l1) {
			
 
				-    return NULL;  /* avoids a negative `l1' */
			
 
				-  } else {
			
 
				-    const char *init;  /* to search for a `*s2' inside `s1' */
			
 
				-    l2--;  /* 1st char will be checked by `memchr' */
			
 
				-    l1 = l1-l2;  /* `s2' cannot be found after that */
			
 
				-    while (l1 > 0 && (init = (const char *)memchr(s1, *s2, l1)) != NULL) {
			
 
				-      init++;   /* 1st char is already checked */
			
 
				-      if (memcmp(init, s2+1, l2) == 0) {
			
 
				-	return init-1;
			
 
				-      } else {  /* correct `l1' and `s1' to try again */
			
 
				-	l1 -= (size_t)(init-s1);
			
 
				-	s1 = init;
			
 
				-      }
			
 
				-    }
			
 
				-    return NULL;  /* not found */
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				 static void push_onecapture(MatchState *ms, int i, const char *s, const char *e)
			
 
				 {
			
 
				   if (i >= ms->level) {
			
@@ -501,64 +447,60 @@ static int push_captures(MatchState *ms, const char *s, const char *e)
 
				   return nlevels;  /* number of strings pushed */
			
 
				 }
			
 
				 
			
 
				-static ptrdiff_t posrelat(ptrdiff_t pos, size_t len)
			
 
				-{
			
 
				-  /* relative string position: negative means back from end */
			
 
				-  if (pos < 0) pos += (ptrdiff_t)len + 1;
			
 
				-  return (pos >= 0) ? pos : 0;
			
 
				-}
			
 
				-
			
 
				 static int str_find_aux(lua_State *L, int find)
			
 
				 {
			
 
				-  size_t l1, l2;
			
 
				-  const char *s = luaL_checklstring(L, 1, &l1);
			
 
				-  const char *p = luaL_checklstring(L, 2, &l2);
			
 
				-  ptrdiff_t init = posrelat(luaL_optinteger(L, 3, 1), l1) - 1;
			
 
				-  if (init < 0) {
			
 
				-    init = 0;
			
 
				-  } else if ((size_t)(init) > l1) {
			
 
				+  GCstr *s = lj_lib_checkstr(L, 1);
			
 
				+  GCstr *p = lj_lib_checkstr(L, 2);
			
 
				+  int32_t start = lj_lib_optint(L, 3, 1);
			
 
				+  MSize st;
			
 
				+  if (start < 0) start += (int32_t)s->len; else start--;
			
 
				+  if (start < 0) start = 0;
			
 
				+  st = (MSize)start;
			
 
				+  if (st > s->len) {
			
 
				 #if LJ_52
			
 
				     setnilV(L->top-1);
			
 
				     return 1;
			
 
				 #else
			
 
				-    init = (ptrdiff_t)l1;
			
 
				+    st = s->len;
			
 
				 #endif
			
 
				   }
			
 
				-  if (find && (lua_toboolean(L, 4) ||  /* explicit request? */
			
 
				-      strpbrk(p, SPECIALS) == NULL)) {  /* or no special characters? */
			
 
				-    /* do a plain search */
			
 
				-    const char *s2 = lmemfind(s+init, l1-(size_t)init, p, l2);
			
 
				-    if (s2) {
			
 
				-      lua_pushinteger(L, s2-s+1);
			
 
				-      lua_pushinteger(L, s2-s+(ptrdiff_t)l2);
			
 
				+  if (find && ((L->base+3 < L->top && tvistruecond(L->base+3)) ||
			
 
				+	       !lj_str_haspattern(p))) {  /* Search for fixed string. */
			
 
				+    const char *q = lj_str_find(strdata(s)+st, strdata(p), s->len-st, p->len);
			
 
				+    if (q) {
			
 
				+      setintV(L->top-2, (int32_t)(q-strdata(s)) + 1);
			
 
				+      setintV(L->top-1, (int32_t)(q-strdata(s)) + (int32_t)p->len);
			
 
				       return 2;
			
 
				     }
			
 
				-  } else {
			
 
				+  } else {  /* Search for pattern. */
			
 
				     MatchState ms;
			
 
				-    int anchor = (*p == '^') ? (p++, 1) : 0;
			
 
				-    const char *s1=s+init;
			
 
				+    const char *pstr = strdata(p);
			
 
				+    const char *sstr = strdata(s) + st;
			
 
				+    int anchor = 0;
			
 
				+    if (*pstr == '^') { pstr++; anchor = 1; }
			
 
				     ms.L = L;
			
 
				-    ms.src_init = s;
			
 
				-    ms.src_end = s+l1;
			
 
				-    do {
			
 
				-      const char *res;
			
 
				+    ms.src_init = strdata(s);
			
 
				+    ms.src_end = strdata(s) + s->len;
			
 
				+    do {  /* Loop through string and try to match the pattern. */
			
 
				+      const char *q;
			
 
				       ms.level = ms.depth = 0;
			
 
				-      if ((res=match(&ms, s1, p)) != NULL) {
			
 
				+      q = match(&ms, sstr, pstr);
			
 
				+      if (q) {
			
 
				 	if (find) {
			
 
				-	  lua_pushinteger(L, s1-s+1);  /* start */
			
 
				-	  lua_pushinteger(L, res-s);   /* end */
			
 
				-	  return push_captures(&ms, NULL, 0) + 2;
			
 
				+	  setintV(L->top++, (int32_t)(sstr-(strdata(s)-1)));
			
 
				+	  setintV(L->top++, (int32_t)(q-strdata(s)));
			
 
				+	  return push_captures(&ms, NULL, NULL) + 2;
			
 
				 	} else {
			
 
				-	  return push_captures(&ms, s1, res);
			
 
				+	  return push_captures(&ms, sstr, q);
			
 
				 	}
			
 
				       }
			
 
				-    } while (s1++ < ms.src_end && !anchor);
			
 
				+    } while (sstr++ < ms.src_end && !anchor);
			
 
				   }
			
 
				-  lua_pushnil(L);  /* not found */
			
 
				+  setnilV(L->top-1);  /* Not found. */
			
 
				   return 1;
			
 
				 }
			
 
				 
			
 
				-LJLIB_CF(string_find)
			
 
				+LJLIB_CF(string_find)		LJLIB_REC(.)
			
 
				 {
			
 
				   return str_find_aux(L, 1);
			
 
				 }
			
@@ -698,221 +640,91 @@ LJLIB_CF(string_gsub)
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				-/* maximum size of each formatted item (> len(format('%99.99f', -1e308))) */
			
 
				-#define MAX_FMTITEM	512
			
 
				-/* valid flags in a format specification */
			
 
				-#define FMT_FLAGS	"-+ #0"
			
 
				-/*
			
 
				-** maximum size of each format specification (such as '%-099.99d')
			
 
				-** (+10 accounts for %99.99x plus margin of error)
			
 
				-*/
			
 
				-#define MAX_FMTSPEC	(sizeof(FMT_FLAGS) + sizeof(LUA_INTFRMLEN) + 10)
			
 
				-
			
 
				-static void addquoted(lua_State *L, luaL_Buffer *b, int arg)
			
 
				-{
			
 
				-  GCstr *str = lj_lib_checkstr(L, arg);
			
 
				-  int32_t len = (int32_t)str->len;
			
 
				-  const char *s = strdata(str);
			
 
				-  luaL_addchar(b, '"');
			
 
				-  while (len--) {
			
 
				-    uint32_t c = uchar(*s);
			
 
				-    if (c == '"' || c == '\\' || c == '\n') {
			
 
				-      luaL_addchar(b, '\\');
			
 
				-    } else if (lj_char_iscntrl(c)) {  /* This can only be 0-31 or 127. */
			
 
				-      uint32_t d;
			
 
				-      luaL_addchar(b, '\\');
			
 
				-      if (c >= 100 || lj_char_isdigit(uchar(s[1]))) {
			
 
				-	luaL_addchar(b, '0'+(c >= 100)); if (c >= 100) c -= 100;
			
 
				-	goto tens;
			
 
				-      } else if (c >= 10) {
			
 
				-      tens:
			
 
				-	d = (c * 205) >> 11; c -= d * 10; luaL_addchar(b, '0'+d);
			
 
				-      }
			
 
				-      c += '0';
			
 
				-    }
			
 
				-    luaL_addchar(b, c);
			
 
				-    s++;
			
 
				-  }
			
 
				-  luaL_addchar(b, '"');
			
 
				-}
			
 
				-
			
 
				-static const char *scanformat(lua_State *L, const char *strfrmt, char *form)
			
 
				-{
			
 
				-  const char *p = strfrmt;
			
 
				-  while (*p != '\0' && strchr(FMT_FLAGS, *p) != NULL) p++;  /* skip flags */
			
 
				-  if ((size_t)(p - strfrmt) >= sizeof(FMT_FLAGS))
			
 
				-    lj_err_caller(L, LJ_ERR_STRFMTR);
			
 
				-  if (lj_char_isdigit(uchar(*p))) p++;  /* skip width */
			
 
				-  if (lj_char_isdigit(uchar(*p))) p++;  /* (2 digits at most) */
			
 
				-  if (*p == '.') {
			
 
				-    p++;
			
 
				-    if (lj_char_isdigit(uchar(*p))) p++;  /* skip precision */
			
 
				-    if (lj_char_isdigit(uchar(*p))) p++;  /* (2 digits at most) */
			
 
				-  }
			
 
				-  if (lj_char_isdigit(uchar(*p)))
			
 
				-    lj_err_caller(L, LJ_ERR_STRFMTW);
			
 
				-  *(form++) = '%';
			
 
				-  strncpy(form, strfrmt, (size_t)(p - strfrmt + 1));
			
 
				-  form += p - strfrmt + 1;
			
 
				-  *form = '\0';
			
 
				-  return p;
			
 
				-}
			
 
				-
			
 
				-static void addintlen(char *form)
			
 
				-{
			
 
				-  size_t l = strlen(form);
			
 
				-  char spec = form[l - 1];
			
 
				-  strcpy(form + l - 1, LUA_INTFRMLEN);
			
 
				-  form[l + sizeof(LUA_INTFRMLEN) - 2] = spec;
			
 
				-  form[l + sizeof(LUA_INTFRMLEN) - 1] = '\0';
			
 
				-}
			
 
				-
			
 
				-static unsigned LUA_INTFRM_T num2intfrm(lua_State *L, int arg)
			
 
				-{
			
 
				-  if (sizeof(LUA_INTFRM_T) == 4) {
			
 
				-    return (LUA_INTFRM_T)lj_lib_checkbit(L, arg);
			
 
				-  } else {
			
 
				-    cTValue *o;
			
 
				-    lj_lib_checknumber(L, arg);
			
 
				-    o = L->base+arg-1;
			
 
				-    if (tvisint(o))
			
 
				-      return (LUA_INTFRM_T)intV(o);
			
 
				-    else
			
 
				-      return (LUA_INTFRM_T)numV(o);
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				-static unsigned LUA_INTFRM_T num2uintfrm(lua_State *L, int arg)
			
 
				-{
			
 
				-  if (sizeof(LUA_INTFRM_T) == 4) {
			
 
				-    return (unsigned LUA_INTFRM_T)lj_lib_checkbit(L, arg);
			
 
				-  } else {
			
 
				-    cTValue *o;
			
 
				-    lj_lib_checknumber(L, arg);
			
 
				-    o = L->base+arg-1;
			
 
				-    if (tvisint(o))
			
 
				-      return (unsigned LUA_INTFRM_T)intV(o);
			
 
				-    else if ((int32_t)o->u32.hi < 0)
			
 
				-      return (unsigned LUA_INTFRM_T)(LUA_INTFRM_T)numV(o);
			
 
				-    else
			
 
				-      return (unsigned LUA_INTFRM_T)numV(o);
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				-static GCstr *meta_tostring(lua_State *L, int arg)
			
 
				+/* Emulate tostring() inline. */
			
 
				+static GCstr *string_fmt_tostring(lua_State *L, int arg, int retry)
			
 
				 {
			
 
				   TValue *o = L->base+arg-1;
			
 
				   cTValue *mo;
			
 
				   lua_assert(o < L->top);  /* Caller already checks for existence. */
			
 
				   if (LJ_LIKELY(tvisstr(o)))
			
 
				     return strV(o);
			
 
				-  if (!tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
			
 
				+  if (retry != 2 && !tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
			
 
				     copyTV(L, L->top++, mo);
			
 
				     copyTV(L, L->top++, o);
			
 
				     lua_call(L, 1, 1);
			
 
				-    L->top--;
			
 
				-    if (tvisstr(L->top))
			
 
				-      return strV(L->top);
			
 
				-    o = L->base+arg-1;
			
 
				-    copyTV(L, o, L->top);
			
 
				-  }
			
 
				-  if (tvisnumber(o)) {
			
 
				-    return lj_str_fromnumber(L, o);
			
 
				-  } else if (tvisnil(o)) {
			
 
				-    return lj_str_newlit(L, "nil");
			
 
				-  } else if (tvisfalse(o)) {
			
 
				-    return lj_str_newlit(L, "false");
			
 
				-  } else if (tvistrue(o)) {
			
 
				-    return lj_str_newlit(L, "true");
			
 
				-  } else {
			
 
				-    if (tvisfunc(o) && isffunc(funcV(o)))
			
 
				-      lj_str_pushf(L, "function: builtin#%d", funcV(o)->c.ffid);
			
 
				-    else
			
 
				-      lj_str_pushf(L, "%s: %p", lj_typename(o), lua_topointer(L, arg));
			
 
				-    L->top--;
			
 
				-    return strV(L->top);
			
 
				+    copyTV(L, L->base+arg-1, --L->top);
			
 
				+    return NULL;  /* Buffer may be overwritten, retry. */
			
 
				   }
			
 
				-}
			
 
				-
			
 
				-LJLIB_CF(string_format)
			
 
				-{
			
 
				-  int arg = 1, top = (int)(L->top - L->base);
			
 
				-  GCstr *fmt = lj_lib_checkstr(L, arg);
			
 
				-  const char *strfrmt = strdata(fmt);
			
 
				-  const char *strfrmt_end = strfrmt + fmt->len;
			
 
				-  luaL_Buffer b;
			
 
				-  luaL_buffinit(L, &b);
			
 
				-  while (strfrmt < strfrmt_end) {
			
 
				-    if (*strfrmt != L_ESC) {
			
 
				-      luaL_addchar(&b, *strfrmt++);
			
 
				-    } else if (*++strfrmt == L_ESC) {
			
 
				-      luaL_addchar(&b, *strfrmt++);  /* %% */
			
 
				-    } else { /* format item */
			
 
				-      char form[MAX_FMTSPEC];  /* to store the format (`%...') */
			
 
				-      char buff[MAX_FMTITEM];  /* to store the formatted item */
			
 
				+  return lj_strfmt_obj(L, o);
			
 
				+}
			
 
				+
			
 
				+LJLIB_CF(string_format)		LJLIB_REC(.)
			
 
				+{
			
 
				+  int arg, top = (int)(L->top - L->base);
			
 
				+  GCstr *fmt;
			
 
				+  SBuf *sb;
			
 
				+  FormatState fs;
			
 
				+  SFormat sf;
			
 
				+  int retry = 0;
			
 
				+again:
			
 
				+  arg = 1;
			
 
				+  sb = lj_buf_tmp_(L);
			
 
				+  fmt = lj_lib_checkstr(L, arg);
			
 
				+  lj_strfmt_init(&fs, strdata(fmt), fmt->len);
			
 
				+  while ((sf = lj_strfmt_parse(&fs)) != STRFMT_EOF) {
			
 
				+    if (sf == STRFMT_LIT) {
			
 
				+      lj_buf_putmem(sb, fs.str, fs.len);
			
 
				+    } else if (sf == STRFMT_ERR) {
			
 
				+      lj_err_callerv(L, LJ_ERR_STRFMT, strdata(lj_str_new(L, fs.str, fs.len)));
			
 
				+    } else {
			
 
				       if (++arg > top)
			
 
				 	luaL_argerror(L, arg, lj_obj_typename[0]);
			
 
				-      strfrmt = scanformat(L, strfrmt, form);
			
 
				-      switch (*strfrmt++) {
			
 
				-      case 'c':
			
 
				-	sprintf(buff, form, lj_lib_checkint(L, arg));
			
 
				+      switch (STRFMT_TYPE(sf)) {
			
 
				+      case STRFMT_INT:
			
 
				+	if (tvisint(L->base+arg-1)) {
			
 
				+	  int32_t k = intV(L->base+arg-1);
			
 
				+	  if (sf == STRFMT_INT)
			
 
				+	    lj_strfmt_putint(sb, k);  /* Shortcut for plain %d. */
			
 
				+	  else
			
 
				+	    lj_strfmt_putfxint(sb, sf, k);
			
 
				+	} else {
			
 
				+	  lj_strfmt_putfnum_int(sb, sf, lj_lib_checknum(L, arg));
			
 
				+	}
			
 
				 	break;
			
 
				-      case 'd':  case 'i':
			
 
				-	addintlen(form);
			
 
				-	sprintf(buff, form, num2intfrm(L, arg));
			
 
				+      case STRFMT_UINT:
			
 
				+	if (tvisint(L->base+arg-1))
			
 
				+	  lj_strfmt_putfxint(sb, sf, intV(L->base+arg-1));
			
 
				+	else
			
 
				+	  lj_strfmt_putfnum_uint(sb, sf, lj_lib_checknum(L, arg));
			
 
				 	break;
			
 
				-      case 'o':  case 'u':  case 'x':  case 'X':
			
 
				-	addintlen(form);
			
 
				-	sprintf(buff, form, num2uintfrm(L, arg));
			
 
				+      case STRFMT_NUM:
			
 
				+	lj_strfmt_putfnum(sb, sf, lj_lib_checknum(L, arg));
			
 
				 	break;
			
 
				-      case 'e':  case 'E': case 'f': case 'g': case 'G': case 'a': case 'A': {
			
 
				-	TValue tv;
			
 
				-	tv.n = lj_lib_checknum(L, arg);
			
 
				-	if (LJ_UNLIKELY((tv.u32.hi << 1) >= 0xffe00000)) {
			
 
				-	  /* Canonicalize output of non-finite values. */
			
 
				-	  char *p, nbuf[LJ_STR_NUMBUF];
			
 
				-	  size_t len = lj_str_bufnum(nbuf, &tv);
			
 
				-	  if (strfrmt[-1] < 'a') {
			
 
				-	    nbuf[len-3] = nbuf[len-3] - 0x20;
			
 
				-	    nbuf[len-2] = nbuf[len-2] - 0x20;
			
 
				-	    nbuf[len-1] = nbuf[len-1] - 0x20;
			
 
				-	  }
			
 
				-	  nbuf[len] = '\0';
			
 
				-	  for (p = form; *p < 'A' && *p != '.'; p++) ;
			
 
				-	  *p++ = 's'; *p = '\0';
			
 
				-	  sprintf(buff, form, nbuf);
			
 
				-	  break;
			
 
				-	}
			
 
				-	sprintf(buff, form, (double)tv.n);
			
 
				+      case STRFMT_STR: {
			
 
				+	GCstr *str = string_fmt_tostring(L, arg, retry);
			
 
				+	if (str == NULL)
			
 
				+	  retry = 1;
			
 
				+	else if ((sf & STRFMT_T_QUOTED))
			
 
				+	  lj_strfmt_putquoted(sb, str);  /* No formatting. */
			
 
				+	else
			
 
				+	  lj_strfmt_putfstr(sb, sf, str);
			
 
				 	break;
			
 
				 	}
			
 
				-      case 'q':
			
 
				-	addquoted(L, &b, arg);
			
 
				-	continue;
			
 
				-      case 'p':
			
 
				-	lj_str_pushf(L, "%p", lua_topointer(L, arg));
			
 
				-	luaL_addvalue(&b);
			
 
				-	continue;
			
 
				-      case 's': {
			
 
				-	GCstr *str = meta_tostring(L, arg);
			
 
				-	if (!strchr(form, '.') && str->len >= 100) {
			
 
				-	  /* no precision and string is too long to be formatted;
			
 
				-	     keep original string */
			
 
				-	  setstrV(L, L->top++, str);
			
 
				-	  luaL_addvalue(&b);
			
 
				-	  continue;
			
 
				-	}
			
 
				-	sprintf(buff, form, strdata(str));
			
 
				+      case STRFMT_CHAR:
			
 
				+	lj_strfmt_putfchar(sb, sf, lj_lib_checkint(L, arg));
			
 
				+	break;
			
 
				+      case STRFMT_PTR:  /* No formatting. */
			
 
				+	lj_strfmt_putptr(sb, lj_obj_ptr(L->base+arg-1));
			
 
				 	break;
			
 
				-	}
			
 
				       default:
			
 
				-	lj_err_callerv(L, LJ_ERR_STRFMTO, *(strfrmt -1));
			
 
				+	lua_assert(0);
			
 
				 	break;
			
 
				       }
			
 
				-      luaL_addlstring(&b, buff, strlen(buff));
			
 
				     }
			
 
				   }
			
 
				-  luaL_pushresult(&b);
			
 
				+  if (retry++ == 1) goto again;
			
 
				+  setstrV(L, L->top-1, lj_buf_str(L, sb));
			
 
				+  lj_gc_check(L);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
@@ -925,10 +737,6 @@ LUALIB_API int luaopen_string(lua_State *L)
 
				   GCtab *mt;
			
 
				   global_State *g;
			
 
				   LJ_LIB_REG(L, LUA_STRLIBNAME, string);
			
 
				-#if defined(LUA_COMPAT_GFIND) && !LJ_52
			
 
				-  lua_getfield(L, -1, "gmatch");
			
 
				-  lua_setfield(L, -2, "gfind");
			
 
				-#endif
			
 
				   mt = lj_tab_new(L, 0, 1);
			
 
				   /* NOBARRIER: basemt is a GC root. */
			
 
				   g = G(L);
			
--- a/luajit.mod/luajit/src/lib_table.c
+++ b/luajit.mod/luajit/src/lib_table.c
@@ -16,57 +16,43 @@
 
				 #include "lj_obj.h"
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				 #include "lj_tab.h"
			
 
				+#include "lj_ff.h"
			
 
				 #include "lj_lib.h"
			
 
				 
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				 #define LJLIB_MODULE_table
			
 
				 
			
 
				-LJLIB_CF(table_foreachi)
			
 
				-{
			
 
				-  GCtab *t = lj_lib_checktab(L, 1);
			
 
				-  GCfunc *func = lj_lib_checkfunc(L, 2);
			
 
				-  MSize i, n = lj_tab_len(t);
			
 
				-  for (i = 1; i <= n; i++) {
			
 
				-    cTValue *val;
			
 
				-    setfuncV(L, L->top, func);
			
 
				-    setintV(L->top+1, i);
			
 
				-    val = lj_tab_getint(t, (int32_t)i);
			
 
				-    if (val) { copyTV(L, L->top+2, val); } else { setnilV(L->top+2); }
			
 
				-    L->top += 3;
			
 
				-    lua_call(L, 2, 1);
			
 
				-    if (!tvisnil(L->top-1))
			
 
				-      return 1;
			
 
				-    L->top--;
			
 
				-  }
			
 
				-  return 0;
			
 
				-}
			
 
				+LJLIB_LUA(table_foreachi) /*
			
 
				+  function(t, f)
			
 
				+    CHECK_tab(t)
			
 
				+    CHECK_func(f)
			
 
				+    for i=1,#t do
			
 
				+      local r = f(i, t[i])
			
 
				+      if r ~= nil then return r end
			
 
				+    end
			
 
				+  end
			
 
				+*/
			
 
				 
			
 
				-LJLIB_CF(table_foreach)
			
 
				-{
			
 
				-  GCtab *t = lj_lib_checktab(L, 1);
			
 
				-  GCfunc *func = lj_lib_checkfunc(L, 2);
			
 
				-  L->top = L->base+3;
			
 
				-  setnilV(L->top-1);
			
 
				-  while (lj_tab_next(L, t, L->top-1)) {
			
 
				-    copyTV(L, L->top+2, L->top);
			
 
				-    copyTV(L, L->top+1, L->top-1);
			
 
				-    setfuncV(L, L->top, func);
			
 
				-    L->top += 3;
			
 
				-    lua_call(L, 2, 1);
			
 
				-    if (!tvisnil(L->top-1))
			
 
				-      return 1;
			
 
				-    L->top--;
			
 
				-  }
			
 
				-  return 0;
			
 
				-}
			
 
				+LJLIB_LUA(table_foreach) /*
			
 
				+  function(t, f)
			
 
				+    CHECK_tab(t)
			
 
				+    CHECK_func(f)
			
 
				+    for k, v in PAIRS(t) do
			
 
				+      local r = f(k, v)
			
 
				+      if r ~= nil then return r end
			
 
				+    end
			
 
				+  end
			
 
				+*/
			
 
				 
			
 
				-LJLIB_ASM(table_getn)		LJLIB_REC(.)
			
 
				-{
			
 
				-  lj_lib_checktab(L, 1);
			
 
				-  return FFH_UNREACHABLE;
			
 
				-}
			
 
				+LJLIB_LUA(table_getn) /*
			
 
				+  function(t)
			
 
				+    CHECK_tab(t)
			
 
				+    return #t
			
 
				+  end
			
 
				+*/
			
 
				 
			
 
				 LJLIB_CF(table_maxn)
			
 
				 {
			
@@ -119,52 +105,67 @@ LJLIB_CF(table_insert)		LJLIB_REC(.)
 
				   return 0;
			
 
				 }
			
 
				 
			
 
				-LJLIB_CF(table_remove)		LJLIB_REC(.)
			
 
				-{
			
 
				-  GCtab *t = lj_lib_checktab(L, 1);
			
 
				-  int32_t e = (int32_t)lj_tab_len(t);
			
 
				-  int32_t pos = lj_lib_optint(L, 2, e);
			
 
				-  if (!(1 <= pos && pos <= e))  /* Nothing to remove? */
			
 
				-    return 0;
			
 
				-  lua_rawgeti(L, 1, pos);  /* Get previous value. */
			
 
				-  /* NOBARRIER: This just moves existing elements around. */
			
 
				-  for (; pos < e; pos++) {
			
 
				-    cTValue *src = lj_tab_getint(t, pos+1);
			
 
				-    TValue *dst = lj_tab_setint(L, t, pos);
			
 
				-    if (src) {
			
 
				-      copyTV(L, dst, src);
			
 
				-    } else {
			
 
				-      setnilV(dst);
			
 
				-    }
			
 
				-  }
			
 
				-  setnilV(lj_tab_setint(L, t, e));  /* Remove (last) value. */
			
 
				-  return 1;  /* Return previous value. */
			
 
				-}
			
 
				+LJLIB_LUA(table_remove) /*
			
 
				+  function(t, pos)
			
 
				+    CHECK_tab(t)
			
 
				+    local len = #t
			
 
				+    if pos == nil then
			
 
				+      if len ~= 0 then
			
 
				+	local old = t[len]
			
 
				+	t[len] = nil
			
 
				+	return old
			
 
				+      end
			
 
				+    else
			
 
				+      CHECK_int(pos)
			
 
				+      if pos >= 1 and pos <= len then
			
 
				+	local old = t[pos]
			
 
				+	for i=pos+1,len do
			
 
				+	  t[i-1] = t[i]
			
 
				+	end
			
 
				+	t[len] = nil
			
 
				+	return old
			
 
				+      end
			
 
				+    end
			
 
				+  end
			
 
				+*/
			
 
				+
			
 
				+LJLIB_LUA(table_move) /*
			
 
				+  function(a1, f, e, t, a2)
			
 
				+    CHECK_tab(a1)
			
 
				+    CHECK_int(f)
			
 
				+    CHECK_int(e)
			
 
				+    CHECK_int(t)
			
 
				+    if a2 == nil then a2 = a1 end
			
 
				+    CHECK_tab(a2)
			
 
				+    if e >= f then
			
 
				+      local d = t - f
			
 
				+      if t > e or t <= f or a2 ~= a1 then
			
 
				+	for i=f,e do a2[i+d] = a1[i] end
			
 
				+      else
			
 
				+	for i=e,f,-1 do a2[i+d] = a1[i] end
			
 
				+      end
			
 
				+    end
			
 
				+    return a2
			
 
				+  end
			
 
				+*/
			
 
				 
			
 
				-LJLIB_CF(table_concat)
			
 
				+LJLIB_CF(table_concat)		LJLIB_REC(.)
			
 
				 {
			
 
				-  luaL_Buffer b;
			
 
				   GCtab *t = lj_lib_checktab(L, 1);
			
 
				   GCstr *sep = lj_lib_optstr(L, 2);
			
 
				-  MSize seplen = sep ? sep->len : 0;
			
 
				   int32_t i = lj_lib_optint(L, 3, 1);
			
 
				   int32_t e = (L->base+3 < L->top && !tvisnil(L->base+3)) ?
			
 
				 	      lj_lib_checkint(L, 4) : (int32_t)lj_tab_len(t);
			
 
				-  luaL_buffinit(L, &b);
			
 
				-  if (i <= e) {
			
 
				-    for (;;) {
			
 
				-      cTValue *o;
			
 
				-      lua_rawgeti(L, 1, i);
			
 
				-      o = L->top-1;
			
 
				-      if (!(tvisstr(o) || tvisnumber(o)))
			
 
				-	lj_err_callerv(L, LJ_ERR_TABCAT, lj_typename(o), i);
			
 
				-      luaL_addvalue(&b);
			
 
				-      if (i++ == e) break;
			
 
				-      if (seplen)
			
 
				-	luaL_addlstring(&b, strdata(sep), seplen);
			
 
				-    }
			
 
				+  SBuf *sb = lj_buf_tmp_(L);
			
 
				+  SBuf *sbx = lj_buf_puttab(sb, t, sep, i, e);
			
 
				+  if (LJ_UNLIKELY(!sbx)) {  /* Error: bad element type. */
			
 
				+    int32_t idx = (int32_t)(intptr_t)sbufP(sb);
			
 
				+    cTValue *o = lj_tab_getint(t, idx);
			
 
				+    lj_err_callerv(L, LJ_ERR_TABCAT,
			
 
				+		   lj_obj_itypename[o ? itypemap(o) : ~LJ_TNIL], idx);
			
 
				   }
			
 
				-  luaL_pushresult(&b);
			
 
				+  setstrV(L, L->top-1, lj_buf_str(L, sbx));
			
 
				+  lj_gc_check(L);
			
 
				   return 1;
			
 
				 }
			
 
				 
			
@@ -284,6 +285,30 @@ LJLIB_CF(table_pack)
 
				 }
			
 
				 #endif
			
 
				 
			
 
				+LJLIB_NOREG LJLIB_CF(table_new)		LJLIB_REC(.)
			
 
				+{
			
 
				+  int32_t a = lj_lib_checkint(L, 1);
			
 
				+  int32_t h = lj_lib_checkint(L, 2);
			
 
				+  lua_createtable(L, a, h);
			
 
				+  return 1;
			
 
				+}
			
 
				+
			
 
				+LJLIB_NOREG LJLIB_CF(table_clear)	LJLIB_REC(.)
			
 
				+{
			
 
				+  lj_tab_clear(lj_lib_checktab(L, 1));
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+static int luaopen_table_new(lua_State *L)
			
 
				+{
			
 
				+  return lj_lib_postreg(L, lj_cf_table_new, FF_table_new, "new");
			
 
				+}
			
 
				+
			
 
				+static int luaopen_table_clear(lua_State *L)
			
 
				+{
			
 
				+  return lj_lib_postreg(L, lj_cf_table_clear, FF_table_clear, "clear");
			
 
				+}
			
 
				+
			
 
				 /* ------------------------------------------------------------------------ */
			
 
				 
			
 
				 #include "lj_libdef.h"
			
@@ -295,6 +320,8 @@ LUALIB_API int luaopen_table(lua_State *L)
 
				   lua_getglobal(L, "unpack");
			
 
				   lua_setfield(L, -2, "unpack");
			
 
				 #endif
			
 
				+  lj_lib_prereg(L, LUA_TABLIBNAME ".new", luaopen_table_new, tabV(L->top-1));
			
 
				+  lj_lib_prereg(L, LUA_TABLIBNAME ".clear", luaopen_table_clear, tabV(L->top-1));
			
 
				   return 1;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lj_alloc.c
+++ b/luajit.mod/luajit/src/lj_alloc.c
@@ -72,13 +72,56 @@
 
				 
			
 
				 #define IS_DIRECT_BIT		(SIZE_T_ONE)
			
 
				 
			
 
				+
			
 
				+/* Determine system-specific block allocation method. */
			
 
				 #if LJ_TARGET_WINDOWS
			
 
				 
			
 
				 #define WIN32_LEAN_AND_MEAN
			
 
				 #include <windows.h>
			
 
				 
			
 
				+#define LJ_ALLOC_VIRTUALALLOC	1
			
 
				+
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				+#define LJ_ALLOC_NTAVM		1
			
 
				+#endif
			
 
				+
			
 
				+#else
			
 
				+
			
 
				+#include <errno.h>
			
 
				+/* If this include fails, then rebuild with: -DLUAJIT_USE_SYSMALLOC */
			
 
				+#include <sys/mman.h>
			
 
				+
			
 
				+#define LJ_ALLOC_MMAP		1
			
 
				+
			
 
				 #if LJ_64
			
 
				 
			
 
				+#define LJ_ALLOC_MMAP_PROBE	1
			
 
				+
			
 
				+#if LJ_GC64
			
 
				+#define LJ_ALLOC_MBITS		47	/* 128 TB in LJ_GC64 mode. */
			
 
				+#elif LJ_TARGET_X64 && LJ_HASJIT
			
 
				+/* Due to limitations in the x64 compiler backend. */
			
 
				+#define LJ_ALLOC_MBITS		31	/* 2 GB on x64 with !LJ_GC64. */
			
 
				+#else
			
 
				+#define LJ_ALLOC_MBITS		32	/* 4 GB on other archs with !LJ_GC64. */
			
 
				+#endif
			
 
				+
			
 
				+#endif
			
 
				+
			
 
				+#if LJ_64 && !LJ_GC64 && defined(MAP_32BIT)
			
 
				+#define LJ_ALLOC_MMAP32		1
			
 
				+#endif
			
 
				+
			
 
				+#if LJ_TARGET_LINUX
			
 
				+#define LJ_ALLOC_MREMAP		1
			
 
				+#endif
			
 
				+
			
 
				+#endif
			
 
				+
			
 
				+
			
 
				+#if LJ_ALLOC_VIRTUALALLOC
			
 
				+
			
 
				+#if LJ_ALLOC_NTAVM
			
 
				 /* Undocumented, but hey, that's what we all love so much about Windows. */
			
 
				 typedef long (*PNTAVM)(HANDLE handle, void **addr, ULONG zbits,
			
 
				 		       size_t *size, ULONG alloctype, ULONG prot);
			
@@ -89,14 +132,15 @@ static PNTAVM ntavm;
 
				 */
			
 
				 #define NTAVM_ZEROBITS		1
			
 
				 
			
 
				-static void INIT_MMAP(void)
			
 
				+static void init_mmap(void)
			
 
				 {
			
 
				   ntavm = (PNTAVM)GetProcAddress(GetModuleHandleA("ntdll.dll"),
			
 
				 				 "NtAllocateVirtualMemory");
			
 
				 }
			
 
				+#define INIT_MMAP()	init_mmap()
			
 
				 
			
 
				 /* Win64 32 bit MMAP via NtAllocateVirtualMemory. */
			
 
				-static LJ_AINLINE void *CALL_MMAP(size_t size)
			
 
				+static void *CALL_MMAP(size_t size)
			
 
				 {
			
 
				   DWORD olderr = GetLastError();
			
 
				   void *ptr = NULL;
			
@@ -107,7 +151,7 @@ static LJ_AINLINE void *CALL_MMAP(size_t size)
 
				 }
			
 
				 
			
 
				 /* For direct MMAP, use MEM_TOP_DOWN to minimize interference */
			
 
				-static LJ_AINLINE void *DIRECT_MMAP(size_t size)
			
 
				+static void *DIRECT_MMAP(size_t size)
			
 
				 {
			
 
				   DWORD olderr = GetLastError();
			
 
				   void *ptr = NULL;
			
@@ -119,23 +163,21 @@ static LJ_AINLINE void *DIRECT_MMAP(size_t size)
 
				 
			
 
				 #else
			
 
				 
			
 
				-#define INIT_MMAP()		((void)0)
			
 
				-
			
 
				 /* Win32 MMAP via VirtualAlloc */
			
 
				-static LJ_AINLINE void *CALL_MMAP(size_t size)
			
 
				+static void *CALL_MMAP(size_t size)
			
 
				 {
			
 
				   DWORD olderr = GetLastError();
			
 
				-  void *ptr = VirtualAlloc(0, size, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
			
 
				+  void *ptr = LJ_WIN_VALLOC(0, size, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
			
 
				   SetLastError(olderr);
			
 
				   return ptr ? ptr : MFAIL;
			
 
				 }
			
 
				 
			
 
				 /* For direct MMAP, use MEM_TOP_DOWN to minimize interference */
			
 
				-static LJ_AINLINE void *DIRECT_MMAP(size_t size)
			
 
				+static void *DIRECT_MMAP(size_t size)
			
 
				 {
			
 
				   DWORD olderr = GetLastError();
			
 
				-  void *ptr = VirtualAlloc(0, size, MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN,
			
 
				-			   PAGE_READWRITE);
			
 
				+  void *ptr = LJ_WIN_VALLOC(0, size, MEM_RESERVE|MEM_COMMIT|MEM_TOP_DOWN,
			
 
				+			    PAGE_READWRITE);
			
 
				   SetLastError(olderr);
			
 
				   return ptr ? ptr : MFAIL;
			
 
				 }
			
@@ -143,7 +185,7 @@ static LJ_AINLINE void *DIRECT_MMAP(size_t size)
 
				 #endif
			
 
				 
			
 
				 /* This function supports releasing coalesed segments */
			
 
				-static LJ_AINLINE int CALL_MUNMAP(void *ptr, size_t size)
			
 
				+static int CALL_MUNMAP(void *ptr, size_t size)
			
 
				 {
			
 
				   DWORD olderr = GetLastError();
			
 
				   MEMORY_BASIC_INFORMATION minfo;
			
@@ -163,10 +205,7 @@ static LJ_AINLINE int CALL_MUNMAP(void *ptr, size_t size)
 
				   return 0;
			
 
				 }
			
 
				 
			
 
				-#else
			
 
				-
			
 
				-#include <errno.h>
			
 
				-#include <sys/mman.h>
			
 
				+#elif LJ_ALLOC_MMAP
			
 
				 
			
 
				 #define MMAP_PROT		(PROT_READ|PROT_WRITE)
			
 
				 #if !defined(MAP_ANONYMOUS) && defined(MAP_ANON)
			
@@ -174,105 +213,152 @@ static LJ_AINLINE int CALL_MUNMAP(void *ptr, size_t size)
 
				 #endif
			
 
				 #define MMAP_FLAGS		(MAP_PRIVATE|MAP_ANONYMOUS)
			
 
				 
			
 
				-#if LJ_64
			
 
				-/* 64 bit mode needs special support for allocating memory in the lower 2GB. */
			
 
				-
			
 
				-#if defined(MAP_32BIT)
			
 
				+#if LJ_ALLOC_MMAP_PROBE
			
 
				 
			
 
				-#if defined(__sun__)
			
 
				-#define MMAP_REGION_START	((uintptr_t)0x1000)
			
 
				+#ifdef MAP_TRYFIXED
			
 
				+#define MMAP_FLAGS_PROBE	(MMAP_FLAGS|MAP_TRYFIXED)
			
 
				 #else
			
 
				-/* Actually this only gives us max. 1GB in current Linux kernels. */
			
 
				-#define MMAP_REGION_START	((uintptr_t)0)
			
 
				+#define MMAP_FLAGS_PROBE	MMAP_FLAGS
			
 
				 #endif
			
 
				 
			
 
				-static LJ_AINLINE void *CALL_MMAP(size_t size)
			
 
				-{
			
 
				-  int olderr = errno;
			
 
				-  void *ptr = mmap((void *)MMAP_REGION_START, size, MMAP_PROT, MAP_32BIT|MMAP_FLAGS, -1, 0);
			
 
				-  errno = olderr;
			
 
				-  return ptr;
			
 
				-}
			
 
				+#define LJ_ALLOC_MMAP_PROBE_MAX		30
			
 
				+#define LJ_ALLOC_MMAP_PROBE_LINEAR	5
			
 
				 
			
 
				-#elif LJ_TARGET_OSX || LJ_TARGET_PS4 || defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) || defined(__sun__) || LJ_TARGET_CYGWIN
			
 
				+#define LJ_ALLOC_MMAP_PROBE_LOWER	((uintptr_t)0x4000)
			
 
				 
			
 
				-/* OSX and FreeBSD mmap() use a naive first-fit linear search.
			
 
				-** That's perfect for us. Except that -pagezero_size must be set for OSX,
			
 
				-** otherwise the lower 4GB are blocked. And the 32GB RLIMIT_DATA needs
			
 
				-** to be reduced to 250MB on FreeBSD.
			
 
				+/* No point in a giant ifdef mess. Just try to open /dev/urandom.
			
 
				+** It doesn't really matter if this fails, since we get some ASLR bits from
			
 
				+** every unsuitable allocation, too. And we prefer linear allocation, anyway.
			
 
				 */
			
 
				-#if LJ_TARGET_OSX || defined(__DragonFly__)
			
 
				-#define MMAP_REGION_START	((uintptr_t)0x10000)
			
 
				-#elif LJ_TARGET_PS4
			
 
				-#define MMAP_REGION_START	((uintptr_t)0x4000)
			
 
				-#else
			
 
				-#define MMAP_REGION_START	((uintptr_t)0x10000000)
			
 
				-#endif
			
 
				-#define MMAP_REGION_END		((uintptr_t)0x80000000)
			
 
				+#include <fcntl.h>
			
 
				+#include <unistd.h>
			
 
				 
			
 
				-#if (defined(__FreeBSD__) || defined(__FreeBSD_kernel__)) && !LJ_TARGET_PS4
			
 
				-#include <sys/resource.h>
			
 
				-#endif
			
 
				+static uintptr_t mmap_probe_seed(void)
			
 
				+{
			
 
				+  uintptr_t val;
			
 
				+  int fd = open("/dev/urandom", O_RDONLY);
			
 
				+  if (fd != -1) {
			
 
				+    int ok = ((size_t)read(fd, &val, sizeof(val)) == sizeof(val));
			
 
				+    (void)close(fd);
			
 
				+    if (ok) return val;
			
 
				+  }
			
 
				+  return 1;  /* Punt. */
			
 
				+}
			
 
				 
			
 
				-static LJ_AINLINE void *CALL_MMAP(size_t size)
			
 
				+static void *mmap_probe(size_t size)
			
 
				 {
			
 
				-  int olderr = errno;
			
 
				   /* Hint for next allocation. Doesn't need to be thread-safe. */
			
 
				-  static uintptr_t alloc_hint = MMAP_REGION_START;
			
 
				-  int retry = 0;
			
 
				-#if (defined(__FreeBSD__) || defined(__FreeBSD_kernel__)) && !LJ_TARGET_PS4
			
 
				-  static int rlimit_modified = 0;
			
 
				-  if (LJ_UNLIKELY(rlimit_modified == 0)) {
			
 
				-    struct rlimit rlim;
			
 
				-    rlim.rlim_cur = rlim.rlim_max = MMAP_REGION_START;
			
 
				-    setrlimit(RLIMIT_DATA, &rlim);  /* Ignore result. May fail below. */
			
 
				-    rlimit_modified = 1;
			
 
				-  }
			
 
				-#endif
			
 
				-  for (;;) {
			
 
				-    void *p = mmap((void *)alloc_hint, size, MMAP_PROT, MMAP_FLAGS, -1, 0);
			
 
				-    if ((uintptr_t)p >= MMAP_REGION_START &&
			
 
				-	(uintptr_t)p + size < MMAP_REGION_END) {
			
 
				-      alloc_hint = (uintptr_t)p + size;
			
 
				+  static uintptr_t hint_addr = 0;
			
 
				+  static uintptr_t hint_prng = 0;
			
 
				+  int olderr = errno;
			
 
				+  int retry;
			
 
				+  for (retry = 0; retry < LJ_ALLOC_MMAP_PROBE_MAX; retry++) {
			
 
				+    void *p = mmap((void *)hint_addr, size, MMAP_PROT, MMAP_FLAGS_PROBE, -1, 0);
			
 
				+    uintptr_t addr = (uintptr_t)p;
			
 
				+    if ((addr >> LJ_ALLOC_MBITS) == 0 && addr >= LJ_ALLOC_MMAP_PROBE_LOWER &&
			
 
				+	((addr + size) >> LJ_ALLOC_MBITS) == 0) {
			
 
				+      /* We got a suitable address. Bump the hint address. */
			
 
				+      hint_addr = addr + size;
			
 
				       errno = olderr;
			
 
				       return p;
			
 
				     }
			
 
				-    if (p != CMFAIL) munmap(p, size);
			
 
				-#if defined(__sun__) || defined(__DragonFly__)
			
 
				-    alloc_hint += 0x1000000;  /* Need near-exhaustive linear scan. */
			
 
				-    if (alloc_hint + size < MMAP_REGION_END) continue;
			
 
				-#endif
			
 
				-    if (retry) break;
			
 
				-    retry = 1;
			
 
				-    alloc_hint = MMAP_REGION_START;
			
 
				+    if (p != MFAIL) {
			
 
				+      munmap(p, size);
			
 
				+    } else if (errno == ENOMEM) {
			
 
				+      return MFAIL;
			
 
				+    }
			
 
				+    if (hint_addr) {
			
 
				+      /* First, try linear probing. */
			
 
				+      if (retry < LJ_ALLOC_MMAP_PROBE_LINEAR) {
			
 
				+	hint_addr += 0x1000000;
			
 
				+	if (((hint_addr + size) >> LJ_ALLOC_MBITS) != 0)
			
 
				+	  hint_addr = 0;
			
 
				+	continue;
			
 
				+      } else if (retry == LJ_ALLOC_MMAP_PROBE_LINEAR) {
			
 
				+	/* Next, try a no-hint probe to get back an ASLR address. */
			
 
				+	hint_addr = 0;
			
 
				+	continue;
			
 
				+      }
			
 
				+    }
			
 
				+    /* Finally, try pseudo-random probing. */
			
 
				+    if (LJ_UNLIKELY(hint_prng == 0)) {
			
 
				+      hint_prng = mmap_probe_seed();
			
 
				+    }
			
 
				+    /* The unsuitable address we got has some ASLR PRNG bits. */
			
 
				+    hint_addr ^= addr & ~((uintptr_t)(LJ_PAGESIZE-1));
			
 
				+    do {  /* The PRNG itself is very weak, but see above. */
			
 
				+      hint_prng = hint_prng * 1103515245 + 12345;
			
 
				+      hint_addr ^= hint_prng * (uintptr_t)LJ_PAGESIZE;
			
 
				+      hint_addr &= (((uintptr_t)1 << LJ_ALLOC_MBITS)-1);
			
 
				+    } while (hint_addr < LJ_ALLOC_MMAP_PROBE_LOWER);
			
 
				   }
			
 
				   errno = olderr;
			
 
				-  return CMFAIL;
			
 
				+  return MFAIL;
			
 
				 }
			
 
				 
			
 
				+#endif
			
 
				+
			
 
				+#if LJ_ALLOC_MMAP32
			
 
				+
			
 
				+#if defined(__sun__)
			
 
				+#define LJ_ALLOC_MMAP32_START	((uintptr_t)0x1000)
			
 
				 #else
			
 
				+#define LJ_ALLOC_MMAP32_START	((uintptr_t)0)
			
 
				+#endif
			
 
				 
			
 
				-#error "NYI: need an equivalent of MAP_32BIT for this 64 bit OS"
			
 
				+static void *mmap_map32(size_t size)
			
 
				+{
			
 
				+#if LJ_ALLOC_MMAP_PROBE
			
 
				+  static int fallback = 0;
			
 
				+  if (fallback)
			
 
				+    return mmap_probe(size);
			
 
				+#endif
			
 
				+  {
			
 
				+    int olderr = errno;
			
 
				+    void *ptr = mmap((void *)LJ_ALLOC_MMAP32_START, size, MMAP_PROT, MAP_32BIT|MMAP_FLAGS, -1, 0);
			
 
				+    errno = olderr;
			
 
				+    /* This only allows 1GB on Linux. So fallback to probing to get 2GB. */
			
 
				+#if LJ_ALLOC_MMAP_PROBE
			
 
				+    if (ptr == MFAIL) {
			
 
				+      fallback = 1;
			
 
				+      return mmap_probe(size);
			
 
				+    }
			
 
				+#endif
			
 
				+    return ptr;
			
 
				+  }
			
 
				+}
			
 
				 
			
 
				 #endif
			
 
				 
			
 
				+#if LJ_ALLOC_MMAP32
			
 
				+#define CALL_MMAP(size)		mmap_map32(size)
			
 
				+#elif LJ_ALLOC_MMAP_PROBE
			
 
				+#define CALL_MMAP(size)		mmap_probe(size)
			
 
				 #else
			
 
				-
			
 
				-/* 32 bit mode is easy. */
			
 
				-static LJ_AINLINE void *CALL_MMAP(size_t size)
			
 
				+static void *CALL_MMAP(size_t size)
			
 
				 {
			
 
				   int olderr = errno;
			
 
				   void *ptr = mmap(NULL, size, MMAP_PROT, MMAP_FLAGS, -1, 0);
			
 
				   errno = olderr;
			
 
				   return ptr;
			
 
				 }
			
 
				-
			
 
				 #endif
			
 
				 
			
 
				-#define INIT_MMAP()		((void)0)
			
 
				-#define DIRECT_MMAP(s)		CALL_MMAP(s)
			
 
				+#if LJ_64 && !LJ_GC64 && ((defined(__FreeBSD__) && __FreeBSD__ < 10) || defined(__FreeBSD_kernel__)) && !LJ_TARGET_PS4
			
 
				+
			
 
				+#include <sys/resource.h>
			
 
				+
			
 
				+static void init_mmap(void)
			
 
				+{
			
 
				+  struct rlimit rlim;
			
 
				+  rlim.rlim_cur = rlim.rlim_max = 0x10000;
			
 
				+  setrlimit(RLIMIT_DATA, &rlim);  /* Ignore result. May fail later. */
			
 
				+}
			
 
				+#define INIT_MMAP()	init_mmap()
			
 
				 
			
 
				-static LJ_AINLINE int CALL_MUNMAP(void *ptr, size_t size)
			
 
				+#endif
			
 
				+
			
 
				+static int CALL_MUNMAP(void *ptr, size_t size)
			
 
				 {
			
 
				   int olderr = errno;
			
 
				   int ret = munmap(ptr, size);
			
@@ -280,10 +366,9 @@ static LJ_AINLINE int CALL_MUNMAP(void *ptr, size_t size)
 
				   return ret;
			
 
				 }
			
 
				 
			
 
				-#if LJ_TARGET_LINUX
			
 
				+#if LJ_ALLOC_MREMAP
			
 
				 /* Need to define _GNU_SOURCE to get the mremap prototype. */
			
 
				-static LJ_AINLINE void *CALL_MREMAP_(void *ptr, size_t osz, size_t nsz,
			
 
				-				     int flags)
			
 
				+static void *CALL_MREMAP_(void *ptr, size_t osz, size_t nsz, int flags)
			
 
				 {
			
 
				   int olderr = errno;
			
 
				   ptr = mremap(ptr, osz, nsz, flags);
			
@@ -294,7 +379,7 @@ static LJ_AINLINE void *CALL_MREMAP_(void *ptr, size_t osz, size_t nsz,
 
				 #define CALL_MREMAP(addr, osz, nsz, mv) CALL_MREMAP_((addr), (osz), (nsz), (mv))
			
 
				 #define CALL_MREMAP_NOMOVE	0
			
 
				 #define CALL_MREMAP_MAYMOVE	1
			
 
				-#if LJ_64
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				 #define CALL_MREMAP_MV		CALL_MREMAP_NOMOVE
			
 
				 #else
			
 
				 #define CALL_MREMAP_MV		CALL_MREMAP_MAYMOVE
			
@@ -303,6 +388,15 @@ static LJ_AINLINE void *CALL_MREMAP_(void *ptr, size_t osz, size_t nsz,
 
				 
			
 
				 #endif
			
 
				 
			
 
				+
			
 
				+#ifndef INIT_MMAP
			
 
				+#define INIT_MMAP()		((void)0)
			
 
				+#endif
			
 
				+
			
 
				+#ifndef DIRECT_MMAP
			
 
				+#define DIRECT_MMAP(s)		CALL_MMAP(s)
			
 
				+#endif
			
 
				+
			
 
				 #ifndef CALL_MREMAP
			
 
				 #define CALL_MREMAP(addr, osz, nsz, mv) ((void)osz, MFAIL)
			
 
				 #endif
			
--- a/luajit.mod/luajit/src/lj_api.c
+++ b/luajit.mod/luajit/src/lj_api.c
@@ -24,6 +24,7 @@
 
				 #include "lj_trace.h"
			
 
				 #include "lj_vm.h"
			
 
				 #include "lj_strscan.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 
			
 
				 /* -- Common helper functions --------------------------------------------- */
			
 
				 
			
@@ -111,6 +112,13 @@ LUA_API void lua_xmove(lua_State *from, lua_State *to, int n)
 
				   from->top = f;
			
 
				 }
			
 
				 
			
 
				+LUA_API const lua_Number *lua_version(lua_State *L)
			
 
				+{
			
 
				+  static const lua_Number version = LUA_VERSION_NUM;
			
 
				+  UNUSED(L);
			
 
				+  return &version;
			
 
				+}
			
 
				+
			
 
				 /* -- Stack manipulation -------------------------------------------------- */
			
 
				 
			
 
				 LUA_API int lua_gettop(lua_State *L)
			
@@ -151,30 +159,40 @@ LUA_API void lua_insert(lua_State *L, int idx)
 
				   copyTV(L, p, L->top);
			
 
				 }
			
 
				 
			
 
				-LUA_API void lua_replace(lua_State *L, int idx)
			
 
				+static void copy_slot(lua_State *L, TValue *f, int idx)
			
 
				 {
			
 
				-  api_checknelems(L, 1);
			
 
				   if (idx == LUA_GLOBALSINDEX) {
			
 
				-    api_check(L, tvistab(L->top-1));
			
 
				+    api_check(L, tvistab(f));
			
 
				     /* NOBARRIER: A thread (i.e. L) is never black. */
			
 
				-    setgcref(L->env, obj2gco(tabV(L->top-1)));
			
 
				+    setgcref(L->env, obj2gco(tabV(f)));
			
 
				   } else if (idx == LUA_ENVIRONINDEX) {
			
 
				     GCfunc *fn = curr_func(L);
			
 
				     if (fn->c.gct != ~LJ_TFUNC)
			
 
				       lj_err_msg(L, LJ_ERR_NOENV);
			
 
				-    api_check(L, tvistab(L->top-1));
			
 
				-    setgcref(fn->c.env, obj2gco(tabV(L->top-1)));
			
 
				-    lj_gc_barrier(L, fn, L->top-1);
			
 
				+    api_check(L, tvistab(f));
			
 
				+    setgcref(fn->c.env, obj2gco(tabV(f)));
			
 
				+    lj_gc_barrier(L, fn, f);
			
 
				   } else {
			
 
				     TValue *o = index2adr(L, idx);
			
 
				     api_checkvalidindex(L, o);
			
 
				-    copyTV(L, o, L->top-1);
			
 
				+    copyTV(L, o, f);
			
 
				     if (idx < LUA_GLOBALSINDEX)  /* Need a barrier for upvalues. */
			
 
				-      lj_gc_barrier(L, curr_func(L), L->top-1);
			
 
				+      lj_gc_barrier(L, curr_func(L), f);
			
 
				   }
			
 
				+}
			
 
				+
			
 
				+LUA_API void lua_replace(lua_State *L, int idx)
			
 
				+{
			
 
				+  api_checknelems(L, 1);
			
 
				+  copy_slot(L, L->top - 1, idx);
			
 
				   L->top--;
			
 
				 }
			
 
				 
			
 
				+LUA_API void lua_copy(lua_State *L, int fromidx, int toidx)
			
 
				+{
			
 
				+  copy_slot(L, index2adr(L, fromidx), toidx);
			
 
				+}
			
 
				+
			
 
				 LUA_API void lua_pushvalue(lua_State *L, int idx)
			
 
				 {
			
 
				   copyTV(L, L->top, index2adr(L, idx));
			
@@ -188,7 +206,7 @@ LUA_API int lua_type(lua_State *L, int idx)
 
				   cTValue *o = index2adr(L, idx);
			
 
				   if (tvisnumber(o)) {
			
 
				     return LUA_TNUMBER;
			
 
				-#if LJ_64
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				   } else if (tvislightud(o)) {
			
 
				     return LUA_TLIGHTUSERDATA;
			
 
				 #endif
			
@@ -268,7 +286,7 @@ LUA_API int lua_equal(lua_State *L, int idx1, int idx2)
 
				     return 0;
			
 
				   } else if (tvispri(o1)) {
			
 
				     return o1 != niltv(L) && o2 != niltv(L);
			
 
				-#if LJ_64
			
 
				+#if LJ_64 && !LJ_GC64
			
 
				   } else if (tvislightud(o1)) {
			
 
				     return o1->u64 == o2->u64;
			
 
				 #endif
			
@@ -283,8 +301,8 @@ LUA_API int lua_equal(lua_State *L, int idx1, int idx2)
 
				     } else {
			
 
				       L->top = base+2;
			
 
				       lj_vm_call(L, base, 1+1);
			
 
				-      L->top -= 2;
			
 
				-      return tvistruecond(L->top+1);
			
 
				+      L->top -= 2+LJ_FR2;
			
 
				+      return tvistruecond(L->top+1+LJ_FR2);
			
 
				     }
			
 
				   }
			
 
				 }
			
@@ -306,8 +324,8 @@ LUA_API int lua_lessthan(lua_State *L, int idx1, int idx2)
 
				     } else {
			
 
				       L->top = base+2;
			
 
				       lj_vm_call(L, base, 1+1);
			
 
				-      L->top -= 2;
			
 
				-      return tvistruecond(L->top+1);
			
 
				+      L->top -= 2+LJ_FR2;
			
 
				+      return tvistruecond(L->top+1+LJ_FR2);
			
 
				     }
			
 
				   }
			
 
				 }
			
@@ -324,6 +342,22 @@ LUA_API lua_Number lua_tonumber(lua_State *L, int idx)
 
				     return 0;
			
 
				 }
			
 
				 
			
 
				+LUA_API lua_Number lua_tonumberx(lua_State *L, int idx, int *ok)
			
 
				+{
			
 
				+  cTValue *o = index2adr(L, idx);
			
 
				+  TValue tmp;
			
 
				+  if (LJ_LIKELY(tvisnumber(o))) {
			
 
				+    if (ok) *ok = 1;
			
 
				+    return numberVnum(o);
			
 
				+  } else if (tvisstr(o) && lj_strscan_num(strV(o), &tmp)) {
			
 
				+    if (ok) *ok = 1;
			
 
				+    return numV(&tmp);
			
 
				+  } else {
			
 
				+    if (ok) *ok = 0;
			
 
				+    return 0;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				 LUALIB_API lua_Number luaL_checknumber(lua_State *L, int idx)
			
 
				 {
			
 
				   cTValue *o = index2adr(L, idx);
			
@@ -361,9 +395,38 @@ LUA_API lua_Integer lua_tointeger(lua_State *L, int idx)
 
				     if (!(tvisstr(o) && lj_strscan_number(strV(o), &tmp)))
			
 
				       return 0;
			
 
				     if (tvisint(&tmp))
			
 
				-      return (lua_Integer)intV(&tmp);
			
 
				+      return intV(&tmp);
			
 
				+    n = numV(&tmp);
			
 
				+  }
			
 
				+#if LJ_64
			
 
				+  return (lua_Integer)n;
			
 
				+#else
			
 
				+  return lj_num2int(n);
			
 
				+#endif
			
 
				+}
			
 
				+
			
 
				+LUA_API lua_Integer lua_tointegerx(lua_State *L, int idx, int *ok)
			
 
				+{
			
 
				+  cTValue *o = index2adr(L, idx);
			
 
				+  TValue tmp;
			
 
				+  lua_Number n;
			
 
				+  if (LJ_LIKELY(tvisint(o))) {
			
 
				+    if (ok) *ok = 1;
			
 
				+    return intV(o);
			
 
				+  } else if (LJ_LIKELY(tvisnum(o))) {
			
 
				+    n = numV(o);
			
 
				+  } else {
			
 
				+    if (!(tvisstr(o) && lj_strscan_number(strV(o), &tmp))) {
			
 
				+      if (ok) *ok = 0;
			
 
				+      return 0;
			
 
				+    }
			
 
				+    if (tvisint(&tmp)) {
			
 
				+      if (ok) *ok = 1;
			
 
				+      return intV(&tmp);
			
 
				+    }
			
 
				     n = numV(&tmp);
			
 
				   }
			
 
				+  if (ok) *ok = 1;
			
 
				 #if LJ_64
			
 
				   return (lua_Integer)n;
			
 
				 #else
			
@@ -434,7 +497,7 @@ LUA_API const char *lua_tolstring(lua_State *L, int idx, size_t *len)
 
				   } else if (tvisnumber(o)) {
			
 
				     lj_gc_check(L);
			
 
				     o = index2adr(L, idx);  /* GC may move the stack. */
			
 
				-    s = lj_str_fromnumber(L, o);
			
 
				+    s = lj_strfmt_number(L, o);
			
 
				     setstrV(L, o, s);
			
 
				   } else {
			
 
				     if (len != NULL) *len = 0;
			
@@ -453,7 +516,7 @@ LUALIB_API const char *luaL_checklstring(lua_State *L, int idx, size_t *len)
 
				   } else if (tvisnumber(o)) {
			
 
				     lj_gc_check(L);
			
 
				     o = index2adr(L, idx);  /* GC may move the stack. */
			
 
				-    s = lj_str_fromnumber(L, o);
			
 
				+    s = lj_strfmt_number(L, o);
			
 
				     setstrV(L, o, s);
			
 
				   } else {
			
 
				     lj_err_argt(L, idx, LUA_TSTRING);
			
@@ -475,7 +538,7 @@ LUALIB_API const char *luaL_optlstring(lua_State *L, int idx,
 
				   } else if (tvisnumber(o)) {
			
 
				     lj_gc_check(L);
			
 
				     o = index2adr(L, idx);  /* GC may move the stack. */
			
 
				-    s = lj_str_fromnumber(L, o);
			
 
				+    s = lj_strfmt_number(L, o);
			
 
				     setstrV(L, o, s);
			
 
				   } else {
			
 
				     lj_err_argt(L, idx, LUA_TSTRING);
			
@@ -507,7 +570,7 @@ LUA_API size_t lua_objlen(lua_State *L, int idx)
 
				   } else if (tvisudata(o)) {
			
 
				     return udataV(o)->len;
			
 
				   } else if (tvisnumber(o)) {
			
 
				-    GCstr *s = lj_str_fromnumber(L, o);
			
 
				+    GCstr *s = lj_strfmt_number(L, o);
			
 
				     setstrV(L, o, s);
			
 
				     return s->len;
			
 
				   } else {
			
@@ -545,17 +608,7 @@ LUA_API lua_State *lua_tothread(lua_State *L, int idx)
 
				 
			
 
				 LUA_API const void *lua_topointer(lua_State *L, int idx)
			
 
				 {
			
 
				-  cTValue *o = index2adr(L, idx);
			
 
				-  if (tvisudata(o))
			
 
				-    return uddata(udataV(o));
			
 
				-  else if (tvislightud(o))
			
 
				-    return lightudV(o);
			
 
				-  else if (tviscdata(o))
			
 
				-    return cdataptr(cdataV(o));
			
 
				-  else if (tvisgcv(o))
			
 
				-    return gcV(o);
			
 
				-  else
			
 
				-    return NULL;
			
 
				+  return lj_obj_ptr(index2adr(L, idx));
			
 
				 }
			
 
				 
			
 
				 /* -- Stack setters (object creation) ------------------------------------- */
			
@@ -606,7 +659,7 @@ LUA_API const char *lua_pushvfstring(lua_State *L, const char *fmt,
 
				 				     va_list argp)
			
 
				 {
			
 
				   lj_gc_check(L);
			
 
				-  return lj_str_pushvf(L, fmt, argp);
			
 
				+  return lj_strfmt_pushvf(L, fmt, argp);
			
 
				 }
			
 
				 
			
 
				 LUA_API const char *lua_pushfstring(lua_State *L, const char *fmt, ...)
			
@@ -615,7 +668,7 @@ LUA_API const char *lua_pushfstring(lua_State *L, const char *fmt, ...)
 
				   va_list argp;
			
 
				   lj_gc_check(L);
			
 
				   va_start(argp, fmt);
			
 
				-  ret = lj_str_pushvf(L, fmt, argp);
			
 
				+  ret = lj_strfmt_pushvf(L, fmt, argp);
			
 
				   va_end(argp);
			
 
				   return ret;
			
 
				 }
			
@@ -649,10 +702,8 @@ LUA_API void lua_pushlightuserdata(lua_State *L, void *p)
 
				 
			
 
				 LUA_API void lua_createtable(lua_State *L, int narray, int nrec)
			
 
				 {
			
 
				-  GCtab *t;
			
 
				   lj_gc_check(L);
			
 
				-  t = lj_tab_new(L, (uint32_t)(narray > 0 ? narray+1 : 0), hsize2hbits(nrec));
			
 
				-  settabV(L, L->top, t);
			
 
				+  settabV(L, L->top, lj_tab_new_ah(L, narray, nrec));
			
 
				   incr_top(L);
			
 
				 }
			
 
				 
			
@@ -715,8 +766,8 @@ LUA_API void lua_concat(lua_State *L, int n)
 
				       n -= (int)(L->top - top);
			
 
				       L->top = top+2;
			
 
				       lj_vm_call(L, top, 1+1);
			
 
				-      L->top--;
			
 
				-      copyTV(L, L->top-1, L->top);
			
 
				+      L->top -= 1+LJ_FR2;
			
 
				+      copyTV(L, L->top-1, L->top+LJ_FR2);
			
 
				     } while (--n > 0);
			
 
				   } else if (n == 0) {  /* Push empty string. */
			
 
				     setstrV(L, L->top, &G(L)->strempty);
			
@@ -735,8 +786,8 @@ LUA_API void lua_gettable(lua_State *L, int idx)
 
				   if (v == NULL) {
			
 
				     L->top += 2;
			
 
				     lj_vm_call(L, L->top-2, 1+1);
			
 
				-    L->top -= 2;
			
 
				-    v = L->top+1;
			
 
				+    L->top -= 2+LJ_FR2;
			
 
				+    v = L->top+1+LJ_FR2;
			
 
				   }
			
 
				   copyTV(L, L->top-1, v);
			
 
				 }
			
@@ -751,8 +802,8 @@ LUA_API void lua_getfield(lua_State *L, int idx, const char *k)
 
				   if (v == NULL) {
			
 
				     L->top += 2;
			
 
				     lj_vm_call(L, L->top-2, 1+1);
			
 
				-    L->top -= 2;
			
 
				-    v = L->top+1;
			
 
				+    L->top -= 2+LJ_FR2;
			
 
				+    v = L->top+1+LJ_FR2;
			
 
				   }
			
 
				   copyTV(L, L->top, v);
			
 
				   incr_top(L);
			
@@ -869,7 +920,7 @@ LUA_API void lua_upvaluejoin(lua_State *L, int idx1, int n1, int idx2, int n2)
 
				   lj_gc_objbarrier(L, fn1, gcref(fn1->l.uvptr[n1]));
			
 
				 }
			
 
				 
			
 
				-LUALIB_API void *luaL_checkudata(lua_State *L, int idx, const char *tname)
			
 
				+LUALIB_API void *luaL_testudata(lua_State *L, int idx, const char *tname)
			
 
				 {
			
 
				   cTValue *o = index2adr(L, idx);
			
 
				   if (tvisudata(o)) {
			
@@ -878,8 +929,14 @@ LUALIB_API void *luaL_checkudata(lua_State *L, int idx, const char *tname)
 
				     if (tv && tvistab(tv) && tabV(tv) == tabref(ud->metatable))
			
 
				       return uddata(ud);
			
 
				   }
			
 
				-  lj_err_argtype(L, idx, tname);
			
 
				-  return NULL;  /* unreachable */
			
 
				+  return NULL;  /* value is not a userdata with a metatable */
			
 
				+}
			
 
				+
			
 
				+LUALIB_API void *luaL_checkudata(lua_State *L, int idx, const char *tname)
			
 
				+{
			
 
				+  void *p = luaL_testudata(L, idx, tname);
			
 
				+  if (!p) lj_err_argtype(L, idx, tname);
			
 
				+  return p;
			
 
				 }
			
 
				 
			
 
				 /* -- Object setters ------------------------------------------------------ */
			
@@ -893,13 +950,14 @@ LUA_API void lua_settable(lua_State *L, int idx)
 
				   o = lj_meta_tset(L, t, L->top-2);
			
 
				   if (o) {
			
 
				     /* NOBARRIER: lj_meta_tset ensures the table is not black. */
			
 
				-    copyTV(L, o, L->top-1);
			
 
				     L->top -= 2;
			
 
				+    copyTV(L, o, L->top+1);
			
 
				   } else {
			
 
				-    L->top += 3;
			
 
				-    copyTV(L, L->top-1, L->top-6);
			
 
				-    lj_vm_call(L, L->top-3, 0+1);
			
 
				-    L->top -= 3;
			
 
				+    TValue *base = L->top;
			
 
				+    copyTV(L, base+2, base-3-2*LJ_FR2);
			
 
				+    L->top = base+3;
			
 
				+    lj_vm_call(L, base, 0+1);
			
 
				+    L->top -= 3+LJ_FR2;
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -913,14 +971,14 @@ LUA_API void lua_setfield(lua_State *L, int idx, const char *k)
 
				   setstrV(L, &key, lj_str_newz(L, k));
			
 
				   o = lj_meta_tset(L, t, &key);
			
 
				   if (o) {
			
 
				-    L->top--;
			
 
				     /* NOBARRIER: lj_meta_tset ensures the table is not black. */
			
 
				-    copyTV(L, o, L->top);
			
 
				+    copyTV(L, o, --L->top);
			
 
				   } else {
			
 
				-    L->top += 3;
			
 
				-    copyTV(L, L->top-1, L->top-6);
			
 
				-    lj_vm_call(L, L->top-3, 0+1);
			
 
				-    L->top -= 2;
			
 
				+    TValue *base = L->top;
			
 
				+    copyTV(L, base+2, base-3-2*LJ_FR2);
			
 
				+    L->top = base+3;
			
 
				+    lj_vm_call(L, base, 0+1);
			
 
				+    L->top -= 2+LJ_FR2;
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -987,6 +1045,12 @@ LUA_API int lua_setmetatable(lua_State *L, int idx)
 
				   return 1;
			
 
				 }
			
 
				 
			
 
				+LUALIB_API void luaL_setmetatable(lua_State *L, const char *tname)
			
 
				+{
			
 
				+  lua_getfield(L, LUA_REGISTRYINDEX, tname);
			
 
				+  lua_setmetatable(L, -2);
			
 
				+}
			
 
				+
			
 
				 LUA_API int lua_setfenv(lua_State *L, int idx)
			
 
				 {
			
 
				   cTValue *o = index2adr(L, idx);
			
@@ -1027,11 +1091,24 @@ LUA_API const char *lua_setupvalue(lua_State *L, int idx, int n)
 
				 
			
 
				 /* -- Calls --------------------------------------------------------------- */
			
 
				 
			
 
				+#if LJ_FR2
			
 
				+static TValue *api_call_base(lua_State *L, int nargs)
			
 
				+{
			
 
				+  TValue *o = L->top, *base = o - nargs;
			
 
				+  L->top = o+1;
			
 
				+  for (; o > base; o--) copyTV(L, o, o-1);
			
 
				+  setnilV(o);
			
 
				+  return o+1;
			
 
				+}
			
 
				+#else
			
 
				+#define api_call_base(L, nargs)	(L->top - (nargs))
			
 
				+#endif
			
 
				+
			
 
				 LUA_API void lua_call(lua_State *L, int nargs, int nresults)
			
 
				 {
			
 
				-  api_check(L, L->status == 0 || L->status == LUA_ERRERR);
			
 
				+  api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
			
 
				   api_checknelems(L, nargs+1);
			
 
				-  lj_vm_call(L, L->top - nargs, nresults+1);
			
 
				+  lj_vm_call(L, api_call_base(L, nargs), nresults+1);
			
 
				 }
			
 
				 
			
 
				 LUA_API int lua_pcall(lua_State *L, int nargs, int nresults, int errfunc)
			
@@ -1040,7 +1117,7 @@ LUA_API int lua_pcall(lua_State *L, int nargs, int nresults, int errfunc)
 
				   uint8_t oldh = hook_save(g);
			
 
				   ptrdiff_t ef;
			
 
				   int status;
			
 
				-  api_check(L, L->status == 0 || L->status == LUA_ERRERR);
			
 
				+  api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
			
 
				   api_checknelems(L, nargs+1);
			
 
				   if (errfunc == 0) {
			
 
				     ef = 0;
			
@@ -1049,7 +1126,7 @@ LUA_API int lua_pcall(lua_State *L, int nargs, int nresults, int errfunc)
 
				     api_checkvalidindex(L, o);
			
 
				     ef = savestack(L, o);
			
 
				   }
			
 
				-  status = lj_vm_pcall(L, L->top - nargs, nresults+1, ef);
			
 
				+  status = lj_vm_pcall(L, api_call_base(L, nargs), nresults+1, ef);
			
 
				   if (status) hook_restore(g, oldh);
			
 
				   return status;
			
 
				 }
			
@@ -1057,12 +1134,14 @@ LUA_API int lua_pcall(lua_State *L, int nargs, int nresults, int errfunc)
 
				 static TValue *cpcall(lua_State *L, lua_CFunction func, void *ud)
			
 
				 {
			
 
				   GCfunc *fn = lj_func_newC(L, 0, getcurrenv(L));
			
 
				+  TValue *top = L->top;
			
 
				   fn->c.f = func;
			
 
				-  setfuncV(L, L->top, fn);
			
 
				-  setlightudV(L->top+1, checklightudptr(L, ud));
			
 
				+  setfuncV(L, top++, fn);
			
 
				+  if (LJ_FR2) setnilV(top++);
			
 
				+  setlightudV(top++, checklightudptr(L, ud));
			
 
				   cframe_nres(L->cframe) = 1+0;  /* Zero results. */
			
 
				-  L->top += 2;
			
 
				-  return L->top-1;  /* Now call the newly allocated C function. */
			
 
				+  L->top = top;
			
 
				+  return top-1;  /* Now call the newly allocated C function. */
			
 
				 }
			
 
				 
			
 
				 LUA_API int lua_cpcall(lua_State *L, lua_CFunction func, void *ud)
			
@@ -1070,7 +1149,7 @@ LUA_API int lua_cpcall(lua_State *L, lua_CFunction func, void *ud)
 
				   global_State *g = G(L);
			
 
				   uint8_t oldh = hook_save(g);
			
 
				   int status;
			
 
				-  api_check(L, L->status == 0 || L->status == LUA_ERRERR);
			
 
				+  api_check(L, L->status == LUA_OK || L->status == LUA_ERRERR);
			
 
				   status = lj_vm_cpcall(L, func, ud, cpcall);
			
 
				   if (status) hook_restore(g, oldh);
			
 
				   return status;
			
@@ -1079,10 +1158,11 @@ LUA_API int lua_cpcall(lua_State *L, lua_CFunction func, void *ud)
 
				 LUALIB_API int luaL_callmeta(lua_State *L, int idx, const char *field)
			
 
				 {
			
 
				   if (luaL_getmetafield(L, idx, field)) {
			
 
				-    TValue *base = L->top--;
			
 
				-    copyTV(L, base, index2adr(L, idx));
			
 
				-    L->top = base+1;
			
 
				-    lj_vm_call(L, base, 1+1);
			
 
				+    TValue *top = L->top--;
			
 
				+    if (LJ_FR2) setnilV(top++);
			
 
				+    copyTV(L, top++, index2adr(L, idx));
			
 
				+    L->top = top;
			
 
				+    lj_vm_call(L, top-1, 1+1);
			
 
				     return 1;
			
 
				   }
			
 
				   return 0;
			
@@ -1090,6 +1170,11 @@ LUALIB_API int luaL_callmeta(lua_State *L, int idx, const char *field)
 
				 
			
 
				 /* -- Coroutine yield and resume ------------------------------------------ */
			
 
				 
			
 
				+LUA_API int lua_isyieldable(lua_State *L)
			
 
				+{
			
 
				+  return cframe_canyield(L->cframe);
			
 
				+}
			
 
				+
			
 
				 LUA_API int lua_yield(lua_State *L, int nresults)
			
 
				 {
			
 
				   void *cf = L->cframe;
			
@@ -1109,12 +1194,14 @@ LUA_API int lua_yield(lua_State *L, int nresults)
 
				     } else {  /* Yield from hook: add a pseudo-frame. */
			
 
				       TValue *top = L->top;
			
 
				       hook_leave(g);
			
 
				-      top->u64 = cframe_multres(cf);
			
 
				-      setcont(top+1, lj_cont_hook);
			
 
				-      setframe_pc(top+1, cframe_pc(cf)-1);
			
 
				-      setframe_gc(top+2, obj2gco(L));
			
 
				-      setframe_ftsz(top+2, (int)((char *)(top+3)-(char *)L->base)+FRAME_CONT);
			
 
				-      L->top = L->base = top+3;
			
 
				+      (top++)->u64 = cframe_multres(cf);
			
 
				+      setcont(top, lj_cont_hook);
			
 
				+      if (LJ_FR2) top++;
			
 
				+      setframe_pc(top, cframe_pc(cf)-1);
			
 
				+      if (LJ_FR2) top++;
			
 
				+      setframe_gc(top, obj2gco(L), LJ_TTHREAD);
			
 
				+      setframe_ftsz(top, ((char *)(top+1)-(char *)L->base)+FRAME_CONT);
			
 
				+      L->top = L->base = top+1;
			
 
				 #if LJ_TARGET_X64
			
 
				       lj_err_throw(L, LUA_YIELD);
			
 
				 #else
			
@@ -1131,7 +1218,9 @@ LUA_API int lua_yield(lua_State *L, int nresults)
 
				 LUA_API int lua_resume(lua_State *L, int nargs)
			
 
				 {
			
 
				   if (L->cframe == NULL && L->status <= LUA_YIELD)
			
 
				-    return lj_vm_resume(L, L->top - nargs, 0, 0);
			
 
				+    return lj_vm_resume(L,
			
 
				+      L->status == LUA_OK ? api_call_base(L, nargs) : L->top - nargs,
			
 
				+      0, 0);
			
 
				   L->top = L->base;
			
 
				   setstrV(L, L->top, lj_err_str(L, LJ_ERR_COSUSP));
			
 
				   incr_top(L);
			
@@ -1161,7 +1250,7 @@ LUA_API int lua_gc(lua_State *L, int what, int data)
 
				     res = (int)(g->gc.total & 0x3ff);
			
 
				     break;
			
 
				   case LUA_GCSTEP: {
			
 
				-    MSize a = (MSize)data << 10;
			
 
				+    GCSize a = (GCSize)data << 10;
			
 
				     g->gc.threshold = (a <= g->gc.total) ? (g->gc.total - a) : 0;
			
 
				     while (g->gc.total >= g->gc.threshold)
			
 
				       if (lj_gc_step(L) > 0) {
			
@@ -1178,6 +1267,9 @@ LUA_API int lua_gc(lua_State *L, int what, int data)
 
				     res = (int)(g->gc.stepmul);
			
 
				     g->gc.stepmul = (MSize)data;
			
 
				     break;
			
 
				+  case LUA_GCISRUNNING:
			
 
				+    res = (g->gc.threshold != LJ_MAX_MEM);
			
 
				+    break;
			
 
				   default:
			
 
				     res = -1;  /* Invalid option. */
			
 
				   }
			
--- a/luajit.mod/luajit/src/lj_arch.h
+++ b/luajit.mod/luajit/src/lj_arch.h
@@ -19,12 +19,16 @@
 
				 #define LUAJIT_ARCH_x64		2
			
 
				 #define LUAJIT_ARCH_ARM		3
			
 
				 #define LUAJIT_ARCH_arm		3
			
 
				-#define LUAJIT_ARCH_PPC		4
			
 
				-#define LUAJIT_ARCH_ppc		4
			
 
				-#define LUAJIT_ARCH_PPCSPE	5
			
 
				-#define LUAJIT_ARCH_ppcspe	5
			
 
				+#define LUAJIT_ARCH_ARM64	4
			
 
				+#define LUAJIT_ARCH_arm64	4
			
 
				+#define LUAJIT_ARCH_PPC		5
			
 
				+#define LUAJIT_ARCH_ppc		5
			
 
				 #define LUAJIT_ARCH_MIPS	6
			
 
				 #define LUAJIT_ARCH_mips	6
			
 
				+#define LUAJIT_ARCH_MIPS32	6
			
 
				+#define LUAJIT_ARCH_mips32	6
			
 
				+#define LUAJIT_ARCH_MIPS64	7
			
 
				+#define LUAJIT_ARCH_mips64	7
			
 
				 
			
 
				 /* Target OS. */
			
 
				 #define LUAJIT_OS_OTHER		0
			
@@ -43,14 +47,14 @@
 
				 #define LUAJIT_TARGET	LUAJIT_ARCH_X64
			
 
				 #elif defined(__arm__) || defined(__arm) || defined(__ARM__) || defined(__ARM)
			
 
				 #define LUAJIT_TARGET	LUAJIT_ARCH_ARM
			
 
				+#elif defined(__aarch64__)
			
 
				+#define LUAJIT_TARGET	LUAJIT_ARCH_ARM64
			
 
				 #elif defined(__ppc__) || defined(__ppc) || defined(__PPC__) || defined(__PPC) || defined(__powerpc__) || defined(__powerpc) || defined(__POWERPC__) || defined(__POWERPC) || defined(_M_PPC)
			
 
				-#ifdef __NO_FPRS__
			
 
				-#define LUAJIT_TARGET	LUAJIT_ARCH_PPCSPE
			
 
				-#else
			
 
				 #define LUAJIT_TARGET	LUAJIT_ARCH_PPC
			
 
				-#endif
			
 
				+#elif defined(__mips64__) || defined(__mips64) || defined(__MIPS64__) || defined(__MIPS64)
			
 
				+#define LUAJIT_TARGET	LUAJIT_ARCH_MIPS64
			
 
				 #elif defined(__mips__) || defined(__mips) || defined(__MIPS__) || defined(__MIPS)
			
 
				-#define LUAJIT_TARGET	LUAJIT_ARCH_MIPS
			
 
				+#define LUAJIT_TARGET	LUAJIT_ARCH_MIPS32
			
 
				 #else
			
 
				 #error "No support for this architecture (yet)"
			
 
				 #endif
			
@@ -70,7 +74,7 @@
 
				        defined(__NetBSD__) || defined(__OpenBSD__) || \
			
 
				        defined(__DragonFly__)) && !defined(__ORBIS__)
			
 
				 #define LUAJIT_OS	LUAJIT_OS_BSD
			
 
				-#elif (defined(__sun__) && defined(__svr4__))
			
 
				+#elif (defined(__sun__) && defined(__svr4__)) || defined(__HAIKU__)
			
 
				 #define LUAJIT_OS	LUAJIT_OS_POSIX
			
 
				 #elif defined(__CYGWIN__)
			
 
				 #define LJ_TARGET_CYGWIN	1
			
@@ -99,7 +103,7 @@
 
				 #define LJ_TARGET_WINDOWS	(LUAJIT_OS == LUAJIT_OS_WINDOWS)
			
 
				 #define LJ_TARGET_LINUX		(LUAJIT_OS == LUAJIT_OS_LINUX)
			
 
				 #define LJ_TARGET_OSX		(LUAJIT_OS == LUAJIT_OS_OSX)
			
 
				-#define LJ_TARGET_IOS		(LJ_TARGET_OSX && LUAJIT_TARGET == LUAJIT_ARCH_ARM)
			
 
				+#define LJ_TARGET_IOS		(LJ_TARGET_OSX && (LUAJIT_TARGET == LUAJIT_ARCH_ARM || LUAJIT_TARGET == LUAJIT_ARCH_ARM64))
			
 
				 #define LJ_TARGET_POSIX		(LUAJIT_OS > LUAJIT_OS_WINDOWS)
			
 
				 #define LJ_TARGET_DLOPEN	LJ_TARGET_POSIX
			
 
				 
			
@@ -125,6 +129,19 @@
 
				 #define LJ_TARGET_CONSOLE	1
			
 
				 #endif
			
 
				 
			
 
				+#ifdef _DURANGO
			
 
				+#define LJ_TARGET_XBOXONE	1
			
 
				+#define LJ_TARGET_CONSOLE	1
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#endif
			
 
				+
			
 
				+#ifdef _UWP
			
 
				+#define LJ_TARGET_UWP		1
			
 
				+#if LUAJIT_TARGET == LUAJIT_ARCH_X64
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				 #define LJ_NUMMODE_SINGLE	0	/* Single-number mode only. */
			
 
				 #define LJ_NUMMODE_SINGLE_DUAL	1	/* Default to single-number mode. */
			
 
				 #define LJ_NUMMODE_DUAL		2	/* Dual-number mode only. */
			
@@ -167,6 +184,9 @@
 
				 #define LJ_TARGET_MASKROT	1
			
 
				 #define LJ_TARGET_UNALIGNED	1
			
 
				 #define LJ_ARCH_NUMMODE		LJ_NUMMODE_SINGLE_DUAL
			
 
				+#ifdef LUAJIT_ENABLE_GC64
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#endif
			
 
				 
			
 
				 #elif LUAJIT_TARGET == LUAJIT_ARCH_ARM
			
 
				 
			
@@ -188,7 +208,7 @@
 
				 #define LJ_TARGET_UNIFYROT	2	/* Want only IR_BROR. */
			
 
				 #define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL
			
 
				 
			
 
				-#if __ARM_ARCH____ARM_ARCH_8__ || __ARM_ARCH_8A__
			
 
				+#if __ARM_ARCH_8__ || __ARM_ARCH_8A__
			
 
				 #define LJ_ARCH_VERSION		80
			
 
				 #elif __ARM_ARCH_7__ || __ARM_ARCH_7A__ || __ARM_ARCH_7R__ || __ARM_ARCH_7S__ || __ARM_ARCH_7VE__
			
 
				 #define LJ_ARCH_VERSION		70
			
@@ -200,22 +220,86 @@
 
				 #define LJ_ARCH_VERSION		50
			
 
				 #endif
			
 
				 
			
 
				+#elif LUAJIT_TARGET == LUAJIT_ARCH_ARM64
			
 
				+
			
 
				+#define LJ_ARCH_BITS		64
			
 
				+#if defined(__AARCH64EB__)
			
 
				+#define LJ_ARCH_NAME		"arm64be"
			
 
				+#define LJ_ARCH_ENDIAN		LUAJIT_BE
			
 
				+#else
			
 
				+#define LJ_ARCH_NAME		"arm64"
			
 
				+#define LJ_ARCH_ENDIAN		LUAJIT_LE
			
 
				+#endif
			
 
				+#define LJ_TARGET_ARM64		1
			
 
				+#define LJ_TARGET_EHRETREG	0
			
 
				+#define LJ_TARGET_JUMPRANGE	27	/* +-2^27 = +-128MB */
			
 
				+#define LJ_TARGET_MASKSHIFT	1
			
 
				+#define LJ_TARGET_MASKROT	1
			
 
				+#define LJ_TARGET_UNIFYROT	2	/* Want only IR_BROR. */
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL
			
 
				+
			
 
				+#define LJ_ARCH_VERSION		80
			
 
				+
			
 
				 #elif LUAJIT_TARGET == LUAJIT_ARCH_PPC
			
 
				 
			
 
				-#define LJ_ARCH_NAME		"ppc"
			
 
				+#ifndef LJ_ARCH_ENDIAN
			
 
				+#if __BYTE_ORDER__ != __ORDER_BIG_ENDIAN__
			
 
				+#define LJ_ARCH_ENDIAN		LUAJIT_LE
			
 
				+#else
			
 
				+#define LJ_ARCH_ENDIAN		LUAJIT_BE
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				 #if _LP64
			
 
				 #define LJ_ARCH_BITS		64
			
 
				+#if LJ_ARCH_ENDIAN == LUAJIT_LE
			
 
				+#define LJ_ARCH_NAME		"ppc64le"
			
 
				+#else
			
 
				+#define LJ_ARCH_NAME		"ppc64"
			
 
				+#endif
			
 
				 #else
			
 
				 #define LJ_ARCH_BITS		32
			
 
				+#define LJ_ARCH_NAME		"ppc"
			
 
				+
			
 
				+#if !defined(LJ_ARCH_HASFPU)
			
 
				+#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
			
 
				+#define LJ_ARCH_HASFPU		0
			
 
				+#else
			
 
				+#define LJ_ARCH_HASFPU		1
			
 
				 #endif
			
 
				-#define LJ_ARCH_ENDIAN		LUAJIT_BE
			
 
				+#endif
			
 
				+
			
 
				+#if !defined(LJ_ABI_SOFTFP)
			
 
				+#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
			
 
				+#define LJ_ABI_SOFTFP		1
			
 
				+#else
			
 
				+#define LJ_ABI_SOFTFP		0
			
 
				+#endif
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+#define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL
			
 
				+#else
			
 
				+#define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL_SINGLE
			
 
				+#endif
			
 
				+
			
 
				 #define LJ_TARGET_PPC		1
			
 
				 #define LJ_TARGET_EHRETREG	3
			
 
				 #define LJ_TARGET_JUMPRANGE	25	/* +-2^25 = +-32MB */
			
 
				 #define LJ_TARGET_MASKSHIFT	0
			
 
				 #define LJ_TARGET_MASKROT	1
			
 
				 #define LJ_TARGET_UNIFYROT	1	/* Want only IR_BROL. */
			
 
				-#define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL_SINGLE
			
 
				+
			
 
				+#if LJ_TARGET_CONSOLE
			
 
				+#define LJ_ARCH_PPC32ON64	1
			
 
				+#define LJ_ARCH_NOFFI		1
			
 
				+#elif LJ_ARCH_BITS == 64
			
 
				+#define LJ_ARCH_PPC64		1
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#define LJ_ARCH_NOJIT		1	/* NYI */
			
 
				+#endif
			
 
				 
			
 
				 #if _ARCH_PWR7
			
 
				 #define LJ_ARCH_VERSION		70
			
@@ -230,10 +314,6 @@
 
				 #else
			
 
				 #define LJ_ARCH_VERSION		0
			
 
				 #endif
			
 
				-#if __PPC64__ || __powerpc64__ || LJ_TARGET_CONSOLE
			
 
				-#define LJ_ARCH_PPC64		1
			
 
				-#define LJ_ARCH_NOFFI		1
			
 
				-#endif
			
 
				 #if _ARCH_PPCSQ
			
 
				 #define LJ_ARCH_SQRT		1
			
 
				 #endif
			
@@ -247,44 +327,57 @@
 
				 #define LJ_ARCH_XENON		1
			
 
				 #endif
			
 
				 
			
 
				-#elif LUAJIT_TARGET == LUAJIT_ARCH_PPCSPE
			
 
				-
			
 
				-#define LJ_ARCH_NAME		"ppcspe"
			
 
				-#define LJ_ARCH_BITS		32
			
 
				-#define LJ_ARCH_ENDIAN		LUAJIT_BE
			
 
				-#ifndef LJ_ABI_SOFTFP
			
 
				-#define LJ_ABI_SOFTFP		1
			
 
				-#endif
			
 
				-#define LJ_ABI_EABI		1
			
 
				-#define LJ_TARGET_PPCSPE	1
			
 
				-#define LJ_TARGET_EHRETREG	3
			
 
				-#define LJ_TARGET_JUMPRANGE	25	/* +-2^25 = +-32MB */
			
 
				-#define LJ_TARGET_MASKSHIFT	0
			
 
				-#define LJ_TARGET_MASKROT	1
			
 
				-#define LJ_TARGET_UNIFYROT	1	/* Want only IR_BROL. */
			
 
				-#define LJ_ARCH_NUMMODE		LJ_NUMMODE_SINGLE
			
 
				-#define LJ_ARCH_NOFFI		1	/* NYI: comparisons, calls. */
			
 
				-#define LJ_ARCH_NOJIT		1
			
 
				-
			
 
				-#elif LUAJIT_TARGET == LUAJIT_ARCH_MIPS
			
 
				+#elif LUAJIT_TARGET == LUAJIT_ARCH_MIPS32 || LUAJIT_TARGET == LUAJIT_ARCH_MIPS64
			
 
				 
			
 
				 #if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL)
			
 
				+#if LUAJIT_TARGET == LUAJIT_ARCH_MIPS32
			
 
				 #define LJ_ARCH_NAME		"mipsel"
			
 
				+#else
			
 
				+#define LJ_ARCH_NAME		"mips64el"
			
 
				+#endif
			
 
				 #define LJ_ARCH_ENDIAN		LUAJIT_LE
			
 
				 #else
			
 
				+#if LUAJIT_TARGET == LUAJIT_ARCH_MIPS32
			
 
				 #define LJ_ARCH_NAME		"mips"
			
 
				+#else
			
 
				+#define LJ_ARCH_NAME		"mips64"
			
 
				+#endif
			
 
				 #define LJ_ARCH_ENDIAN		LUAJIT_BE
			
 
				 #endif
			
 
				+
			
 
				+#if !defined(LJ_ARCH_HASFPU)
			
 
				+#ifdef __mips_soft_float
			
 
				+#define LJ_ARCH_HASFPU		0
			
 
				+#else
			
 
				+#define LJ_ARCH_HASFPU		1
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				+#if !defined(LJ_ABI_SOFTFP)
			
 
				+#ifdef __mips_soft_float
			
 
				+#define LJ_ABI_SOFTFP		1
			
 
				+#else
			
 
				+#define LJ_ABI_SOFTFP		0
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				+#if LUAJIT_TARGET == LUAJIT_ARCH_MIPS32
			
 
				 #define LJ_ARCH_BITS		32
			
 
				+#define LJ_TARGET_MIPS32	1
			
 
				+#else
			
 
				+#define LJ_ARCH_BITS		64
			
 
				+#define LJ_TARGET_MIPS64	1
			
 
				+#define LJ_TARGET_GC64		1
			
 
				+#endif
			
 
				 #define LJ_TARGET_MIPS		1
			
 
				 #define LJ_TARGET_EHRETREG	4
			
 
				 #define LJ_TARGET_JUMPRANGE	27	/* 2*2^27 = 256MB-aligned region */
			
 
				 #define LJ_TARGET_MASKSHIFT	1
			
 
				 #define LJ_TARGET_MASKROT	1
			
 
				 #define LJ_TARGET_UNIFYROT	2	/* Want only IR_BROR. */
			
 
				-#define LJ_ARCH_NUMMODE		LJ_NUMMODE_SINGLE
			
 
				+#define LJ_ARCH_NUMMODE		LJ_NUMMODE_DUAL
			
 
				 
			
 
				-#if _MIPS_ARCH_MIPS32R2
			
 
				+#if _MIPS_ARCH_MIPS32R2 || _MIPS_ARCH_MIPS64R2
			
 
				 #define LJ_ARCH_VERSION		20
			
 
				 #else
			
 
				 #define LJ_ARCH_VERSION		10
			
@@ -312,6 +405,16 @@
 
				 #if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 2)
			
 
				 #error "Need at least GCC 4.2 or newer"
			
 
				 #endif
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+#if __clang__
			
 
				+#if ((__clang_major__ < 3) || ((__clang_major__ == 3) && __clang_minor__ < 5)) && !defined(__NX_TOOLCHAIN_MAJOR__)
			
 
				+#error "Need at least Clang 3.5 or newer"
			
 
				+#endif
			
 
				+#else
			
 
				+#if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 8)
			
 
				+#error "Need at least GCC 4.8 or newer"
			
 
				+#endif
			
 
				+#endif
			
 
				 #elif !LJ_TARGET_PS3
			
 
				 #if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 3)
			
 
				 #error "Need at least GCC 4.3 or newer"
			
@@ -335,22 +438,24 @@
 
				 #if !(__ARM_EABI__ || LJ_TARGET_IOS)
			
 
				 #error "Only ARM EABI or iOS 3.0+ ABI is supported"
			
 
				 #endif
			
 
				-#elif LJ_TARGET_PPC || LJ_TARGET_PPCSPE
			
 
				-#if defined(_SOFT_FLOAT) || defined(_SOFT_DOUBLE)
			
 
				-#error "No support for PowerPC CPUs without double-precision FPU"
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+#if defined(_ILP32)
			
 
				+#error "No support for ILP32 model on ARM64"
			
 
				 #endif
			
 
				-#if defined(_LITTLE_ENDIAN)
			
 
				-#error "No support for little-endian PowerPC"
			
 
				+#elif LJ_TARGET_PPC
			
 
				+#if !LJ_ARCH_PPC64 && (defined(_LITTLE_ENDIAN) && (!defined(_BYTE_ORDER) || (_BYTE_ORDER == _LITTLE_ENDIAN)))
			
 
				+#error "No support for little-endian PPC32"
			
 
				 #endif
			
 
				-#if defined(_LP64)
			
 
				-#error "No support for PowerPC 64 bit mode"
			
 
				+#if defined(__NO_FPRS__) && !defined(_SOFT_FLOAT)
			
 
				+#error "No support for PPC/e500 anymore (use LuaJIT 2.0)"
			
 
				 #endif
			
 
				-#elif LJ_TARGET_MIPS
			
 
				-#if defined(__mips_soft_float)
			
 
				-#error "No support for MIPS CPUs without FPU"
			
 
				+#elif LJ_TARGET_MIPS32
			
 
				+#if !((defined(_MIPS_SIM_ABI32) && _MIPS_SIM == _MIPS_SIM_ABI32) || (defined(_ABIO32) && _MIPS_SIM == _ABIO32))
			
 
				+#error "Only o32 ABI supported for MIPS32"
			
 
				 #endif
			
 
				-#if defined(_LP64)
			
 
				-#error "No support for MIPS64"
			
 
				+#elif LJ_TARGET_MIPS64
			
 
				+#if !((defined(_MIPS_SIM_ABI64) && _MIPS_SIM == _MIPS_SIM_ABI64) || (defined(_ABI64) && _MIPS_SIM == _ABI64))
			
 
				+#error "Only n64 ABI supported for MIPS64"
			
 
				 #endif
			
 
				 #endif
			
 
				 #endif
			
@@ -376,6 +481,20 @@
 
				 #endif
			
 
				 #endif
			
 
				 
			
 
				+/* 64 bit GC references. */
			
 
				+#if LJ_TARGET_GC64
			
 
				+#define LJ_GC64			1
			
 
				+#else
			
 
				+#define LJ_GC64			0
			
 
				+#endif
			
 
				+
			
 
				+/* 2-slot frame info. */
			
 
				+#if LJ_GC64
			
 
				+#define LJ_FR2			1
			
 
				+#else
			
 
				+#define LJ_FR2			0
			
 
				+#endif
			
 
				+
			
 
				 /* Disable or enable the JIT compiler. */
			
 
				 #if defined(LUAJIT_DISABLE_JIT) || defined(LJ_ARCH_NOJIT) || defined(LJ_OS_NOJIT)
			
 
				 #define LJ_HASJIT		0
			
@@ -390,6 +509,21 @@
 
				 #define LJ_HASFFI		1
			
 
				 #endif
			
 
				 
			
 
				+#if defined(LUAJIT_DISABLE_PROFILE)
			
 
				+#define LJ_HASPROFILE		0
			
 
				+#elif LJ_TARGET_POSIX
			
 
				+#define LJ_HASPROFILE		1
			
 
				+#define LJ_PROFILE_SIGPROF	1
			
 
				+#elif LJ_TARGET_PS3
			
 
				+#define LJ_HASPROFILE		1
			
 
				+#define LJ_PROFILE_PTHREAD	1
			
 
				+#elif LJ_TARGET_WINDOWS || LJ_TARGET_XBOX360
			
 
				+#define LJ_HASPROFILE		1
			
 
				+#define LJ_PROFILE_WTHREAD	1
			
 
				+#else
			
 
				+#define LJ_HASPROFILE		0
			
 
				+#endif
			
 
				+
			
 
				 #ifndef LJ_ARCH_HASFPU
			
 
				 #define LJ_ARCH_HASFPU		1
			
 
				 #endif
			
@@ -397,6 +531,7 @@
 
				 #define LJ_ABI_SOFTFP		0
			
 
				 #endif
			
 
				 #define LJ_SOFTFP		(!LJ_ARCH_HASFPU)
			
 
				+#define LJ_SOFTFP32		(LJ_SOFTFP && LJ_32)
			
 
				 
			
 
				 #if LJ_ARCH_ENDIAN == LUAJIT_BE
			
 
				 #define LJ_LE			0
			
@@ -422,11 +557,11 @@
 
				 #define LJ_TARGET_UNALIGNED	0
			
 
				 #endif
			
 
				 
			
 
				-/* Various workarounds for embedded operating systems. */
			
 
				-#if (defined(__ANDROID__) && !defined(LJ_TARGET_X86ORX64)) || defined(__symbian__) || LJ_TARGET_XBOX360
			
 
				+/* Various workarounds for embedded operating systems or weak C runtimes. */
			
 
				+#if defined(__ANDROID__) || defined(__symbian__) || LJ_TARGET_XBOX360 || LJ_TARGET_WINDOWS
			
 
				 #define LUAJIT_NO_LOG2
			
 
				 #endif
			
 
				-#if defined(__symbian__)
			
 
				+#if defined(__symbian__) || LJ_TARGET_WINDOWS
			
 
				 #define LUAJIT_NO_EXP2
			
 
				 #endif
			
 
				 #if LJ_TARGET_CONSOLE || (LJ_TARGET_IOS && __IPHONE_OS_VERSION_MIN_REQUIRED >= __IPHONE_8_0)
			
@@ -442,6 +577,18 @@
 
				 #define LJ_NO_UNWIND		1
			
 
				 #endif
			
 
				 
			
 
				+#if LJ_TARGET_WINDOWS
			
 
				+#if LJ_TARGET_UWP
			
 
				+#define LJ_WIN_VALLOC	VirtualAllocFromApp
			
 
				+#define LJ_WIN_VPROTECT	VirtualProtectFromApp
			
 
				+extern void *LJ_WIN_LOADLIBA(const char *path);
			
 
				+#else
			
 
				+#define LJ_WIN_VALLOC	VirtualAlloc
			
 
				+#define LJ_WIN_VPROTECT	VirtualProtect
			
 
				+#define LJ_WIN_LOADLIBA(path)	LoadLibraryExA((path), NULL, 0)
			
 
				+#endif
			
 
				+#endif
			
 
				+
			
 
				 /* Compatibility with Lua 5.1 vs. 5.2. */
			
 
				 #ifdef LUAJIT_ENABLE_LUA52COMPAT
			
 
				 #define LJ_52			1
			
--- a/luajit.mod/luajit/src/lj_asm.c
+++ b/luajit.mod/luajit/src/lj_asm.c
--- a/luajit.mod/luajit/src/lj_asm_arm.h
+++ b/luajit.mod/luajit/src/lj_asm_arm.h
@@ -338,7 +338,7 @@ static int asm_fusemadd(ASMState *as, IRIns *ir, ARMIns ai, ARMIns air)
 
				 /* Generate a call to a C function. */
			
 
				 static void asm_gencall(ASMState *as, const CCallInfo *ci, IRRef *args)
			
 
				 {
			
 
				-  uint32_t n, nargs = CCI_NARGS(ci);
			
 
				+  uint32_t n, nargs = CCI_XNARGS(ci);
			
 
				   int32_t ofs = 0;
			
 
				 #if LJ_SOFTFP
			
 
				   Reg gpr = REGARG_FIRSTGPR;
			
@@ -453,15 +453,6 @@ static void asm_setupresult(ASMState *as, IRIns *ir, const CCallInfo *ci)
 
				   UNUSED(ci);
			
 
				 }
			
 
				 
			
 
				-static void asm_call(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  IRRef args[CCI_NARGS_MAX];
			
 
				-  const CCallInfo *ci = &lj_ir_callinfo[ir->op2];
			
 
				-  asm_collectargs(as, ir, ci, args);
			
 
				-  asm_setupresult(as, ir, ci);
			
 
				-  asm_gencall(as, ci, args);
			
 
				-}
			
 
				-
			
 
				 static void asm_callx(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   IRRef args[CCI_NARGS_MAX*2];
			
@@ -490,7 +481,7 @@ static void asm_retf(ASMState *as, IRIns *ir)
 
				 {
			
 
				   Reg base = ra_alloc1(as, REF_BASE, RSET_GPR);
			
 
				   void *pc = ir_kptr(IR(ir->op2));
			
 
				-  int32_t delta = 1+bc_a(*((const BCIns *)pc - 1));
			
 
				+  int32_t delta = 1+LJ_FR2+bc_a(*((const BCIns *)pc - 1));
			
 
				   as->topslot -= (BCReg)delta;
			
 
				   if ((int32_t)as->topslot < 0) as->topslot = 0;
			
 
				   irt_setmark(IR(REF_BASE)->t);  /* Children must not coalesce with BASE reg. */
			
@@ -601,31 +592,6 @@ static void asm_conv(ASMState *as, IRIns *ir)
 
				   }
			
 
				 }
			
 
				 
			
 
				-#if !LJ_SOFTFP && LJ_HASFFI
			
 
				-static void asm_conv64(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  IRType st = (IRType)((ir-1)->op2 & IRCONV_SRCMASK);
			
 
				-  IRType dt = (((ir-1)->op2 & IRCONV_DSTMASK) >> IRCONV_DSH);
			
 
				-  IRCallID id;
			
 
				-  CCallInfo ci;
			
 
				-  IRRef args[2];
			
 
				-  args[0] = (ir-1)->op1;
			
 
				-  args[1] = ir->op1;
			
 
				-  if (st == IRT_NUM || st == IRT_FLOAT) {
			
 
				-    id = IRCALL_fp64_d2l + ((st == IRT_FLOAT) ? 2 : 0) + (dt - IRT_I64);
			
 
				-    ir--;
			
 
				-  } else {
			
 
				-    id = IRCALL_fp64_l2d + ((dt == IRT_FLOAT) ? 2 : 0) + (st - IRT_I64);
			
 
				-  }
			
 
				-  ci = lj_ir_callinfo[id];
			
 
				-#if !LJ_ABI_SOFTFP
			
 
				-  ci.flags |= CCI_VARARG;  /* These calls don't use the hard-float ABI! */
			
 
				-#endif
			
 
				-  asm_setupresult(as, ir, &ci);
			
 
				-  asm_gencall(as, &ci, args);
			
 
				-}
			
 
				-#endif
			
 
				-
			
 
				 static void asm_strto(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_strscan_num];
			
@@ -689,6 +655,8 @@ static void asm_strto(ASMState *as, IRIns *ir)
 
				     emit_opk(as, ARMI_ADD, tmp, RID_SP, ofs, RSET_GPR);
			
 
				 }
			
 
				 
			
 
				+/* -- Memory references --------------------------------------------------- */
			
 
				+
			
 
				 /* Get pointer to TValue. */
			
 
				 static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
			
 
				 {
			
@@ -714,7 +682,7 @@ static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
 
				       Reg src = ra_alloc1(as, ref, allow);
			
 
				       emit_lso(as, ARMI_STR, src, RID_SP, 0);
			
 
				     }
			
 
				-    if ((ir+1)->o == IR_HIOP)
			
 
				+    if (LJ_SOFTFP && (ir+1)->o == IR_HIOP)
			
 
				       type = ra_alloc1(as, ref+1, allow);
			
 
				     else
			
 
				       type = ra_allock(as, irt_toitype(ir->t), allow);
			
@@ -722,27 +690,6 @@ static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
 
				   }
			
 
				 }
			
 
				 
			
 
				-static void asm_tostr(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  IRRef args[2];
			
 
				-  args[0] = ASMREF_L;
			
 
				-  as->gcsteps++;
			
 
				-  if (irt_isnum(IR(ir->op1)->t) || (ir+1)->o == IR_HIOP) {
			
 
				-    const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_str_fromnum];
			
 
				-    args[1] = ASMREF_TMP1;  /* const lua_Number * */
			
 
				-    asm_setupresult(as, ir, ci);  /* GCstr * */
			
 
				-    asm_gencall(as, ci, args);
			
 
				-    asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op1);
			
 
				-  } else {
			
 
				-    const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_str_fromint];
			
 
				-    args[1] = ir->op1;  /* int32_t k */
			
 
				-    asm_setupresult(as, ir, ci);  /* GCstr * */
			
 
				-    asm_gencall(as, ci, args);
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				-/* -- Memory references --------------------------------------------------- */
			
 
				-
			
 
				 static void asm_aref(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   Reg dest = ra_dest(as, ir, RSET_GPR);
			
@@ -960,20 +907,6 @@ static void asm_hrefk(ASMState *as, IRIns *ir)
 
				     emit_opk(as, ARMI_ADD, dest, node, ofs, RSET_GPR);
			
 
				 }
			
 
				 
			
 
				-static void asm_newref(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_tab_newkey];
			
 
				-  IRRef args[3];
			
 
				-  if (ir->r == RID_SINK)
			
 
				-    return;
			
 
				-  args[0] = ASMREF_L;     /* lua_State *L */
			
 
				-  args[1] = ir->op1;      /* GCtab *t     */
			
 
				-  args[2] = ASMREF_TMP1;  /* cTValue *key */
			
 
				-  asm_setupresult(as, ir, ci);  /* TValue * */
			
 
				-  asm_gencall(as, ci, args);
			
 
				-  asm_tvptr(as, ra_releasetmp(as, ASMREF_TMP1), ir->op2);
			
 
				-}
			
 
				-
			
 
				 static void asm_uref(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   Reg dest = ra_dest(as, ir, RSET_GPR);
			
@@ -1064,22 +997,26 @@ static ARMIns asm_fxstoreins(IRIns *ir)
 
				 
			
 
				 static void asm_fload(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				-  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				-  Reg idx = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				-  ARMIns ai = asm_fxloadins(ir);
			
 
				-  int32_t ofs;
			
 
				-  if (ir->op2 == IRFL_TAB_ARRAY) {
			
 
				-    ofs = asm_fuseabase(as, ir->op1);
			
 
				-    if (ofs) {  /* Turn the t->array load into an add for colocated arrays. */
			
 
				-      emit_dn(as, ARMI_ADD|ARMI_K12|ofs, dest, idx);
			
 
				-      return;
			
 
				+  if (ir->op1 == REF_NIL) {
			
 
				+    lua_assert(!ra_used(ir));  /* We can end up here if DCE is turned off. */
			
 
				+  } else {
			
 
				+    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    Reg idx = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+    ARMIns ai = asm_fxloadins(ir);
			
 
				+    int32_t ofs;
			
 
				+    if (ir->op2 == IRFL_TAB_ARRAY) {
			
 
				+      ofs = asm_fuseabase(as, ir->op1);
			
 
				+      if (ofs) {  /* Turn the t->array load into an add for colocated arrays. */
			
 
				+	emit_dn(as, ARMI_ADD|ARMI_K12|ofs, dest, idx);
			
 
				+	return;
			
 
				+      }
			
 
				     }
			
 
				+    ofs = field_ofs[ir->op2];
			
 
				+    if ((ai & 0x04000000))
			
 
				+      emit_lso(as, ai, dest, idx, ofs);
			
 
				+    else
			
 
				+      emit_lsox(as, ai, dest, idx, ofs);
			
 
				   }
			
 
				-  ofs = field_ofs[ir->op2];
			
 
				-  if ((ai & 0x04000000))
			
 
				-    emit_lso(as, ai, dest, idx, ofs);
			
 
				-  else
			
 
				-    emit_lsox(as, ai, dest, idx, ofs);
			
 
				 }
			
 
				 
			
 
				 static void asm_fstore(ASMState *as, IRIns *ir)
			
@@ -1105,7 +1042,7 @@ static void asm_xload(ASMState *as, IRIns *ir)
 
				   asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR, 0);
			
 
				 }
			
 
				 
			
 
				-static void asm_xstore(ASMState *as, IRIns *ir, int32_t ofs)
			
 
				+static void asm_xstore_(ASMState *as, IRIns *ir, int32_t ofs)
			
 
				 {
			
 
				   if (ir->r != RID_SINK) {
			
 
				     Reg src = ra_alloc1(as, ir->op2,
			
@@ -1115,6 +1052,8 @@ static void asm_xstore(ASMState *as, IRIns *ir, int32_t ofs)
 
				   }
			
 
				 }
			
 
				 
			
 
				+#define asm_xstore(as, ir)	asm_xstore_(as, ir, 0)
			
 
				+
			
 
				 static void asm_ahuvload(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   int hiop = (LJ_SOFTFP && (ir+1)->o == IR_HIOP);
			
@@ -1272,19 +1211,16 @@ dotypecheck:
 
				 static void asm_cnew(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   CTState *cts = ctype_ctsG(J2G(as->J));
			
 
				-  CTypeID ctypeid = (CTypeID)IR(ir->op1)->i;
			
 
				-  CTSize sz = (ir->o == IR_CNEWI || ir->op2 == REF_NIL) ?
			
 
				-	      lj_ctype_size(cts, ctypeid) : (CTSize)IR(ir->op2)->i;
			
 
				+  CTypeID id = (CTypeID)IR(ir->op1)->i;
			
 
				+  CTSize sz;
			
 
				+  CTInfo info = lj_ctype_info(cts, id, &sz);
			
 
				   const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
			
 
				-  IRRef args[2];
			
 
				+  IRRef args[4];
			
 
				   RegSet allow = (RSET_GPR & ~RSET_SCRATCH);
			
 
				   RegSet drop = RSET_SCRATCH;
			
 
				-  lua_assert(sz != CTSIZE_INVALID);
			
 
				+  lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
			
 
				 
			
 
				-  args[0] = ASMREF_L;     /* lua_State *L */
			
 
				-  args[1] = ASMREF_TMP1;  /* MSize size   */
			
 
				   as->gcsteps++;
			
 
				-
			
 
				   if (ra_hasreg(ir->r))
			
 
				     rset_clear(drop, ir->r);  /* Dest reg handled below. */
			
 
				   ra_evictset(as, drop);
			
@@ -1306,16 +1242,28 @@ static void asm_cnew(ASMState *as, IRIns *ir)
 
				       if (ofs == sizeof(GCcdata)) break;
			
 
				       ofs -= 4; ir--;
			
 
				     }
			
 
				+  } else if (ir->op2 != REF_NIL) {  /* Create VLA/VLS/aligned cdata. */
			
 
				+    ci = &lj_ir_callinfo[IRCALL_lj_cdata_newv];
			
 
				+    args[0] = ASMREF_L;     /* lua_State *L */
			
 
				+    args[1] = ir->op1;      /* CTypeID id   */
			
 
				+    args[2] = ir->op2;      /* CTSize sz    */
			
 
				+    args[3] = ASMREF_TMP1;  /* CTSize align */
			
 
				+    asm_gencall(as, ci, args);
			
 
				+    emit_loadi(as, ra_releasetmp(as, ASMREF_TMP1), (int32_t)ctype_align(info));
			
 
				+    return;
			
 
				   }
			
 
				+
			
 
				   /* Initialize gct and ctypeid. lj_mem_newgco() already sets marked. */
			
 
				   {
			
 
				-    uint32_t k = emit_isk12(ARMI_MOV, ctypeid);
			
 
				-    Reg r = k ? RID_R1 : ra_allock(as, ctypeid, allow);
			
 
				+    uint32_t k = emit_isk12(ARMI_MOV, id);
			
 
				+    Reg r = k ? RID_R1 : ra_allock(as, id, allow);
			
 
				     emit_lso(as, ARMI_STRB, RID_TMP, RID_RET, offsetof(GCcdata, gct));
			
 
				     emit_lsox(as, ARMI_STRH, r, RID_RET, offsetof(GCcdata, ctypeid));
			
 
				     emit_d(as, ARMI_MOV|ARMI_K12|~LJ_TCDATA, RID_TMP);
			
 
				     if (k) emit_d(as, ARMI_MOV^k, RID_R1);
			
 
				   }
			
 
				+  args[0] = ASMREF_L;     /* lua_State *L */
			
 
				+  args[1] = ASMREF_TMP1;  /* MSize size   */
			
 
				   asm_gencall(as, ci, args);
			
 
				   ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
			
 
				 	       ra_releasetmp(as, ASMREF_TMP1));
			
@@ -1392,23 +1340,38 @@ static void asm_fpunary(ASMState *as, IRIns *ir, ARMIns ai)
 
				   emit_dm(as, ai, (dest & 15), (left & 15));
			
 
				 }
			
 
				 
			
 
				-static int asm_fpjoin_pow(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  IRIns *irp = IR(ir->op1);
			
 
				-  if (irp == ir-1 && irp->o == IR_MUL && !ra_used(irp)) {
			
 
				-    IRIns *irpp = IR(irp->op1);
			
 
				-    if (irpp == ir-2 && irpp->o == IR_FPMATH &&
			
 
				-	irpp->op2 == IRFPM_LOG2 && !ra_used(irpp)) {
			
 
				-      const CCallInfo *ci = &lj_ir_callinfo[IRCALL_pow];
			
 
				-      IRRef args[2];
			
 
				-      args[0] = irpp->op1;
			
 
				-      args[1] = irp->op2;
			
 
				-      asm_setupresult(as, ir, ci);
			
 
				-      asm_gencall(as, ci, args);
			
 
				-      return 1;
			
 
				-    }
			
 
				-  }
			
 
				-  return 0;
			
 
				+static void asm_callround(ASMState *as, IRIns *ir, int id)
			
 
				+{
			
 
				+  /* The modified regs must match with the *.dasc implementation. */
			
 
				+  RegSet drop = RID2RSET(RID_R0)|RID2RSET(RID_R1)|RID2RSET(RID_R2)|
			
 
				+		RID2RSET(RID_R3)|RID2RSET(RID_R12);
			
 
				+  RegSet of;
			
 
				+  Reg dest, src;
			
 
				+  ra_evictset(as, drop);
			
 
				+  dest = ra_dest(as, ir, RSET_FPR);
			
 
				+  emit_dnm(as, ARMI_VMOV_D_RR, RID_RETLO, RID_RETHI, (dest & 15));
			
 
				+  emit_call(as, id == IRFPM_FLOOR ? (void *)lj_vm_floor_sf :
			
 
				+		id == IRFPM_CEIL ? (void *)lj_vm_ceil_sf :
			
 
				+				   (void *)lj_vm_trunc_sf);
			
 
				+  /* Workaround to protect argument GPRs from being used for remat. */
			
 
				+  of = as->freeset;
			
 
				+  as->freeset &= ~RSET_RANGE(RID_R0, RID_R1+1);
			
 
				+  as->cost[RID_R0] = as->cost[RID_R1] = REGCOST(~0u, ASMREF_L);
			
 
				+  src = ra_alloc1(as, ir->op1, RSET_FPR);  /* May alloc GPR to remat FPR. */
			
 
				+  as->freeset |= (of & RSET_RANGE(RID_R0, RID_R1+1));
			
 
				+  emit_dnm(as, ARMI_VMOV_RR_D, RID_R0, RID_R1, (src & 15));
			
 
				+}
			
 
				+
			
 
				+static void asm_fpmath(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (ir->op2 == IRFPM_EXP2 && asm_fpjoin_pow(as, ir))
			
 
				+    return;
			
 
				+  if (ir->op2 <= IRFPM_TRUNC)
			
 
				+    asm_callround(as, ir, ir->op2);
			
 
				+  else if (ir->op2 == IRFPM_SQRT)
			
 
				+    asm_fpunary(as, ir, ARMI_VSQRT_D);
			
 
				+  else
			
 
				+    asm_callid(as, ir, IRCALL_lj_vm_floor + ir->op2);
			
 
				 }
			
 
				 #endif
			
 
				 
			
@@ -1459,32 +1422,6 @@ static void asm_intop_s(ASMState *as, IRIns *ir, ARMIns ai)
 
				   asm_intop(as, ir, ai);
			
 
				 }
			
 
				 
			
 
				-static void asm_bitop(ASMState *as, IRIns *ir, ARMIns ai)
			
 
				-{
			
 
				-  if (as->flagmcp == as->mcp) {  /* Try to drop cmp r, #0. */
			
 
				-    uint32_t cc = (as->mcp[1] >> 28);
			
 
				-    as->flagmcp = NULL;
			
 
				-    if (cc <= CC_NE) {
			
 
				-      as->mcp++;
			
 
				-      ai |= ARMI_S;
			
 
				-    } else if (cc == CC_GE) {
			
 
				-      *++as->mcp ^= ((CC_GE^CC_PL) << 28);
			
 
				-      ai |= ARMI_S;
			
 
				-    } else if (cc == CC_LT) {
			
 
				-      *++as->mcp ^= ((CC_LT^CC_MI) << 28);
			
 
				-      ai |= ARMI_S;
			
 
				-    }  /* else: other conds don't work with bit ops. */
			
 
				-  }
			
 
				-  if (ir->op2 == 0) {
			
 
				-    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				-    uint32_t m = asm_fuseopm(as, ai, ir->op1, RSET_GPR);
			
 
				-    emit_d(as, ai^m, dest);
			
 
				-  } else {
			
 
				-    /* NYI: Turn BAND !k12 into uxtb, uxth or bfc or shl+shr. */
			
 
				-    asm_intop(as, ir, ai);
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				 static void asm_intneg(ASMState *as, IRIns *ir, ARMIns ai)
			
 
				 {
			
 
				   Reg dest = ra_dest(as, ir, RSET_GPR);
			
@@ -1550,6 +1487,20 @@ static void asm_mul(ASMState *as, IRIns *ir)
 
				   asm_intmul(as, ir);
			
 
				 }
			
 
				 
			
 
				+#define asm_addov(as, ir)	asm_add(as, ir)
			
 
				+#define asm_subov(as, ir)	asm_sub(as, ir)
			
 
				+#define asm_mulov(as, ir)	asm_mul(as, ir)
			
 
				+
			
 
				+#if !LJ_SOFTFP
			
 
				+#define asm_div(as, ir)		asm_fparith(as, ir, ARMI_VDIV_D)
			
 
				+#define asm_pow(as, ir)		asm_callid(as, ir, IRCALL_lj_vm_powi)
			
 
				+#define asm_abs(as, ir)		asm_fpunary(as, ir, ARMI_VABS_D)
			
 
				+#define asm_atan2(as, ir)	asm_callid(as, ir, IRCALL_atan2)
			
 
				+#define asm_ldexp(as, ir)	asm_callid(as, ir, IRCALL_ldexp)
			
 
				+#endif
			
 
				+
			
 
				+#define asm_mod(as, ir)		asm_callid(as, ir, IRCALL_lj_vm_modi)
			
 
				+
			
 
				 static void asm_neg(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				 #if !LJ_SOFTFP
			
@@ -1561,41 +1512,35 @@ static void asm_neg(ASMState *as, IRIns *ir)
 
				   asm_intneg(as, ir, ARMI_RSB);
			
 
				 }
			
 
				 
			
 
				-static void asm_callid(ASMState *as, IRIns *ir, IRCallID id)
			
 
				+static void asm_bitop(ASMState *as, IRIns *ir, ARMIns ai)
			
 
				 {
			
 
				-  const CCallInfo *ci = &lj_ir_callinfo[id];
			
 
				-  IRRef args[2];
			
 
				-  args[0] = ir->op1;
			
 
				-  args[1] = ir->op2;
			
 
				-  asm_setupresult(as, ir, ci);
			
 
				-  asm_gencall(as, ci, args);
			
 
				+  if (as->flagmcp == as->mcp) {  /* Try to drop cmp r, #0. */
			
 
				+    uint32_t cc = (as->mcp[1] >> 28);
			
 
				+    as->flagmcp = NULL;
			
 
				+    if (cc <= CC_NE) {
			
 
				+      as->mcp++;
			
 
				+      ai |= ARMI_S;
			
 
				+    } else if (cc == CC_GE) {
			
 
				+      *++as->mcp ^= ((CC_GE^CC_PL) << 28);
			
 
				+      ai |= ARMI_S;
			
 
				+    } else if (cc == CC_LT) {
			
 
				+      *++as->mcp ^= ((CC_LT^CC_MI) << 28);
			
 
				+      ai |= ARMI_S;
			
 
				+    }  /* else: other conds don't work with bit ops. */
			
 
				+  }
			
 
				+  if (ir->op2 == 0) {
			
 
				+    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    uint32_t m = asm_fuseopm(as, ai, ir->op1, RSET_GPR);
			
 
				+    emit_d(as, ai^m, dest);
			
 
				+  } else {
			
 
				+    /* NYI: Turn BAND !k12 into uxtb, uxth or bfc or shl+shr. */
			
 
				+    asm_intop(as, ir, ai);
			
 
				+  }
			
 
				 }
			
 
				 
			
 
				-#if !LJ_SOFTFP
			
 
				-static void asm_callround(ASMState *as, IRIns *ir, int id)
			
 
				-{
			
 
				-  /* The modified regs must match with the *.dasc implementation. */
			
 
				-  RegSet drop = RID2RSET(RID_R0)|RID2RSET(RID_R1)|RID2RSET(RID_R2)|
			
 
				-		RID2RSET(RID_R3)|RID2RSET(RID_R12);
			
 
				-  RegSet of;
			
 
				-  Reg dest, src;
			
 
				-  ra_evictset(as, drop);
			
 
				-  dest = ra_dest(as, ir, RSET_FPR);
			
 
				-  emit_dnm(as, ARMI_VMOV_D_RR, RID_RETLO, RID_RETHI, (dest & 15));
			
 
				-  emit_call(as, id == IRFPM_FLOOR ? (void *)lj_vm_floor_sf :
			
 
				-		id == IRFPM_CEIL ? (void *)lj_vm_ceil_sf :
			
 
				-				   (void *)lj_vm_trunc_sf);
			
 
				-  /* Workaround to protect argument GPRs from being used for remat. */
			
 
				-  of = as->freeset;
			
 
				-  as->freeset &= ~RSET_RANGE(RID_R0, RID_R1+1);
			
 
				-  as->cost[RID_R0] = as->cost[RID_R1] = REGCOST(~0u, ASMREF_L);
			
 
				-  src = ra_alloc1(as, ir->op1, RSET_FPR);  /* May alloc GPR to remat FPR. */
			
 
				-  as->freeset |= (of & RSET_RANGE(RID_R0, RID_R1+1));
			
 
				-  emit_dnm(as, ARMI_VMOV_RR_D, RID_R0, RID_R1, (src & 15));
			
 
				-}
			
 
				-#endif
			
 
				+#define asm_bnot(as, ir)	asm_bitop(as, ir, ARMI_MVN)
			
 
				 
			
 
				-static void asm_bitswap(ASMState *as, IRIns *ir)
			
 
				+static void asm_bswap(ASMState *as, IRIns *ir)
			
 
				 {
			
 
				   Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				   Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
			
@@ -1612,6 +1557,10 @@ static void asm_bitswap(ASMState *as, IRIns *ir)
 
				   }
			
 
				 }
			
 
				 
			
 
				+#define asm_band(as, ir)	asm_bitop(as, ir, ARMI_AND)
			
 
				+#define asm_bor(as, ir)		asm_bitop(as, ir, ARMI_ORR)
			
 
				+#define asm_bxor(as, ir)	asm_bitop(as, ir, ARMI_EOR)
			
 
				+
			
 
				 static void asm_bitshift(ASMState *as, IRIns *ir, ARMShift sh)
			
 
				 {
			
 
				   if (irref_isk(ir->op2)) {  /* Constant shifts. */
			
@@ -1629,6 +1578,12 @@ static void asm_bitshift(ASMState *as, IRIns *ir, ARMShift sh)
 
				   }
			
 
				 }
			
 
				 
			
 
				+#define asm_bshl(as, ir)	asm_bitshift(as, ir, ARMSH_LSL)
			
 
				+#define asm_bshr(as, ir)	asm_bitshift(as, ir, ARMSH_LSR)
			
 
				+#define asm_bsar(as, ir)	asm_bitshift(as, ir, ARMSH_ASR)
			
 
				+#define asm_bror(as, ir)	asm_bitshift(as, ir, ARMSH_ROR)
			
 
				+#define asm_brol(as, ir)	lua_assert(0)
			
 
				+
			
 
				 static void asm_intmin_max(ASMState *as, IRIns *ir, int cc)
			
 
				 {
			
 
				   uint32_t kcmp = 0, kmov = 0;
			
@@ -1702,6 +1657,9 @@ static void asm_min_max(ASMState *as, IRIns *ir, int cc, int fcc)
 
				     asm_intmin_max(as, ir, cc);
			
 
				 }
			
 
				 
			
 
				+#define asm_min(as, ir)		asm_min_max(as, ir, CC_GT, CC_HI)
			
 
				+#define asm_max(as, ir)		asm_min_max(as, ir, CC_LT, CC_LO)
			
 
				+
			
 
				 /* -- Comparisons --------------------------------------------------------- */
			
 
				 
			
 
				 /* Map of comparisons to flags. ORDER IR. */
			
@@ -1817,6 +1775,18 @@ notst:
 
				     as->flagmcp = as->mcp;  /* Allow elimination of the compare. */
			
 
				 }
			
 
				 
			
 
				+static void asm_comp(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+#if !LJ_SOFTFP
			
 
				+  if (irt_isnum(ir->t))
			
 
				+    asm_fpcomp(as, ir);
			
 
				+  else
			
 
				+#endif
			
 
				+    asm_intcomp(as, ir);
			
 
				+}
			
 
				+
			
 
				+#define asm_equal(as, ir)	asm_comp(as, ir)
			
 
				+
			
 
				 #if LJ_HASFFI
			
 
				 /* 64 bit integer comparisons. */
			
 
				 static void asm_int64comp(ASMState *as, IRIns *ir)
			
@@ -1891,7 +1861,7 @@ static void asm_hiop(ASMState *as, IRIns *ir)
 
				 #endif
			
 
				   } else if ((ir-1)->o == IR_XSTORE) {
			
 
				     if ((ir-1)->r != RID_SINK)
			
 
				-      asm_xstore(as, ir, 4);
			
 
				+      asm_xstore_(as, ir, 4);
			
 
				     return;
			
 
				   }
			
 
				   if (!usehi) return;  /* Skip unused hiword op for all remaining ops. */
			
@@ -1939,6 +1909,16 @@ static void asm_hiop(ASMState *as, IRIns *ir)
 
				 #endif
			
 
				 }
			
 
				 
			
 
				+/* -- Profiling ----------------------------------------------------------- */
			
 
				+
			
 
				+static void asm_prof(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  UNUSED(ir);
			
 
				+  asm_guardcc(as, CC_NE);
			
 
				+  emit_n(as, ARMI_TST|ARMI_K12|HOOK_PROFILE, RID_TMP);
			
 
				+  emit_lsptr(as, ARMI_LDRB, RID_TMP, (void *)&J2G(as->J)->hookmask);
			
 
				+}
			
 
				+
			
 
				 /* -- Stack handling ------------------------------------------------------ */
			
 
				 
			
 
				 /* Check Lua stack size for overflow. Use exit handler as fallback. */
			
@@ -1968,7 +1948,7 @@ static void asm_stack_check(ASMState *as, BCReg topslot,
 
				   emit_lso(as, ARMI_LDR, RID_TMP, RID_TMP,
			
 
				 	   (int32_t)offsetof(lua_State, maxstack));
			
 
				   if (irp) {  /* Must not spill arbitrary registers in head of side trace. */
			
 
				-    int32_t i = i32ptr(&J2G(as->J)->jit_L);
			
 
				+    int32_t i = i32ptr(&J2G(as->J)->cur_L);
			
 
				     if (ra_hasspill(irp->s))
			
 
				       emit_lso(as, ARMI_LDR, pbase, RID_SP, sps_scale(irp->s));
			
 
				     emit_lso(as, ARMI_LDR, RID_TMP, RID_TMP, (i & 4095));
			
@@ -1976,7 +1956,7 @@ static void asm_stack_check(ASMState *as, BCReg topslot,
 
				       emit_lso(as, ARMI_STR, RID_RET, RID_SP, 0);  /* Save temp. register. */
			
 
				     emit_loadi(as, RID_TMP, (i & ~4095));
			
 
				   } else {
			
 
				-    emit_getgl(as, RID_TMP, jit_L);
			
 
				+    emit_getgl(as, RID_TMP, cur_L);
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -2085,13 +2065,13 @@ static void asm_loop_fixup(ASMState *as)
 
				 
			
 
				 /* -- Head of trace ------------------------------------------------------- */
			
 
				 
			
 
				-/* Reload L register from g->jit_L. */
			
 
				+/* Reload L register from g->cur_L. */
			
 
				 static void asm_head_lreg(ASMState *as)
			
 
				 {
			
 
				   IRIns *ir = IR(ASMREF_L);
			
 
				   if (ra_used(ir)) {
			
 
				     Reg r = ra_dest(as, ir, RSET_GPR);
			
 
				-    emit_getgl(as, r, jit_L);
			
 
				+    emit_getgl(as, r, cur_L);
			
 
				     ra_evictk(as);
			
 
				   }
			
 
				 }
			
@@ -2162,143 +2142,13 @@ static void asm_tail_prep(ASMState *as)
 
				   *p = 0;  /* Prevent load/store merging. */
			
 
				 }
			
 
				 
			
 
				-/* -- Instruction dispatch ------------------------------------------------ */
			
 
				-
			
 
				-/* Assemble a single instruction. */
			
 
				-static void asm_ir(ASMState *as, IRIns *ir)
			
 
				-{
			
 
				-  switch ((IROp)ir->o) {
			
 
				-  /* Miscellaneous ops. */
			
 
				-  case IR_LOOP: asm_loop(as); break;
			
 
				-  case IR_NOP: case IR_XBAR: lua_assert(!ra_used(ir)); break;
			
 
				-  case IR_USE:
			
 
				-    ra_alloc1(as, ir->op1, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR); break;
			
 
				-  case IR_PHI: asm_phi(as, ir); break;
			
 
				-  case IR_HIOP: asm_hiop(as, ir); break;
			
 
				-  case IR_GCSTEP: asm_gcstep(as, ir); break;
			
 
				-
			
 
				-  /* Guarded assertions. */
			
 
				-  case IR_EQ: case IR_NE:
			
 
				-    if ((ir-1)->o == IR_HREF && ir->op1 == as->curins-1) {
			
 
				-      as->curins--;
			
 
				-      asm_href(as, ir-1, (IROp)ir->o);
			
 
				-      break;
			
 
				-    }
			
 
				-    /* fallthrough */
			
 
				-  case IR_LT: case IR_GE: case IR_LE: case IR_GT:
			
 
				-  case IR_ULT: case IR_UGE: case IR_ULE: case IR_UGT:
			
 
				-  case IR_ABC:
			
 
				-#if !LJ_SOFTFP
			
 
				-    if (irt_isnum(ir->t)) { asm_fpcomp(as, ir); break; }
			
 
				-#endif
			
 
				-    asm_intcomp(as, ir);
			
 
				-    break;
			
 
				-
			
 
				-  case IR_RETF: asm_retf(as, ir); break;
			
 
				-
			
 
				-  /* Bit ops. */
			
 
				-  case IR_BNOT: asm_bitop(as, ir, ARMI_MVN); break;
			
 
				-  case IR_BSWAP: asm_bitswap(as, ir); break;
			
 
				-
			
 
				-  case IR_BAND: asm_bitop(as, ir, ARMI_AND); break;
			
 
				-  case IR_BOR:  asm_bitop(as, ir, ARMI_ORR); break;
			
 
				-  case IR_BXOR: asm_bitop(as, ir, ARMI_EOR); break;
			
 
				-
			
 
				-  case IR_BSHL: asm_bitshift(as, ir, ARMSH_LSL); break;
			
 
				-  case IR_BSHR: asm_bitshift(as, ir, ARMSH_LSR); break;
			
 
				-  case IR_BSAR: asm_bitshift(as, ir, ARMSH_ASR); break;
			
 
				-  case IR_BROR: asm_bitshift(as, ir, ARMSH_ROR); break;
			
 
				-  case IR_BROL: lua_assert(0); break;
			
 
				-
			
 
				-  /* Arithmetic ops. */
			
 
				-  case IR_ADD: case IR_ADDOV: asm_add(as, ir); break;
			
 
				-  case IR_SUB: case IR_SUBOV: asm_sub(as, ir); break;
			
 
				-  case IR_MUL: case IR_MULOV: asm_mul(as, ir); break;
			
 
				-  case IR_MOD: asm_callid(as, ir, IRCALL_lj_vm_modi); break;
			
 
				-  case IR_NEG: asm_neg(as, ir); break;
			
 
				-
			
 
				-#if LJ_SOFTFP
			
 
				-  case IR_DIV: case IR_POW: case IR_ABS:
			
 
				-  case IR_ATAN2: case IR_LDEXP: case IR_FPMATH: case IR_TOBIT:
			
 
				-    lua_assert(0);  /* Unused for LJ_SOFTFP. */
			
 
				-    break;
			
 
				-#else
			
 
				-  case IR_DIV: asm_fparith(as, ir, ARMI_VDIV_D); break;
			
 
				-  case IR_POW: asm_callid(as, ir, IRCALL_lj_vm_powi); break;
			
 
				-  case IR_ABS: asm_fpunary(as, ir, ARMI_VABS_D); break;
			
 
				-  case IR_ATAN2: asm_callid(as, ir, IRCALL_atan2); break;
			
 
				-  case IR_LDEXP: asm_callid(as, ir, IRCALL_ldexp); break;
			
 
				-  case IR_FPMATH:
			
 
				-    if (ir->op2 == IRFPM_EXP2 && asm_fpjoin_pow(as, ir))
			
 
				-      break;
			
 
				-    if (ir->op2 <= IRFPM_TRUNC)
			
 
				-      asm_callround(as, ir, ir->op2);
			
 
				-    else if (ir->op2 == IRFPM_SQRT)
			
 
				-      asm_fpunary(as, ir, ARMI_VSQRT_D);
			
 
				-    else
			
 
				-      asm_callid(as, ir, IRCALL_lj_vm_floor + ir->op2);
			
 
				-    break;
			
 
				-  case IR_TOBIT: asm_tobit(as, ir); break;
			
 
				-#endif
			
 
				-
			
 
				-  case IR_MIN: asm_min_max(as, ir, CC_GT, CC_HI); break;
			
 
				-  case IR_MAX: asm_min_max(as, ir, CC_LT, CC_LO); break;
			
 
				-
			
 
				-  /* Memory references. */
			
 
				-  case IR_AREF: asm_aref(as, ir); break;
			
 
				-  case IR_HREF: asm_href(as, ir, 0); break;
			
 
				-  case IR_HREFK: asm_hrefk(as, ir); break;
			
 
				-  case IR_NEWREF: asm_newref(as, ir); break;
			
 
				-  case IR_UREFO: case IR_UREFC: asm_uref(as, ir); break;
			
 
				-  case IR_FREF: asm_fref(as, ir); break;
			
 
				-  case IR_STRREF: asm_strref(as, ir); break;
			
 
				-
			
 
				-  /* Loads and stores. */
			
 
				-  case IR_ALOAD: case IR_HLOAD: case IR_ULOAD: case IR_VLOAD:
			
 
				-    asm_ahuvload(as, ir);
			
 
				-    break;
			
 
				-  case IR_FLOAD: asm_fload(as, ir); break;
			
 
				-  case IR_XLOAD: asm_xload(as, ir); break;
			
 
				-  case IR_SLOAD: asm_sload(as, ir); break;
			
 
				-
			
 
				-  case IR_ASTORE: case IR_HSTORE: case IR_USTORE: asm_ahustore(as, ir); break;
			
 
				-  case IR_FSTORE: asm_fstore(as, ir); break;
			
 
				-  case IR_XSTORE: asm_xstore(as, ir, 0); break;
			
 
				-
			
 
				-  /* Allocations. */
			
 
				-  case IR_SNEW: case IR_XSNEW: asm_snew(as, ir); break;
			
 
				-  case IR_TNEW: asm_tnew(as, ir); break;
			
 
				-  case IR_TDUP: asm_tdup(as, ir); break;
			
 
				-  case IR_CNEW: case IR_CNEWI: asm_cnew(as, ir); break;
			
 
				-
			
 
				-  /* Write barriers. */
			
 
				-  case IR_TBAR: asm_tbar(as, ir); break;
			
 
				-  case IR_OBAR: asm_obar(as, ir); break;
			
 
				-
			
 
				-  /* Type conversions. */
			
 
				-  case IR_CONV: asm_conv(as, ir); break;
			
 
				-  case IR_TOSTR: asm_tostr(as, ir); break;
			
 
				-  case IR_STRTO: asm_strto(as, ir); break;
			
 
				-
			
 
				-  /* Calls. */
			
 
				-  case IR_CALLN: case IR_CALLL: case IR_CALLS: asm_call(as, ir); break;
			
 
				-  case IR_CALLXS: asm_callx(as, ir); break;
			
 
				-  case IR_CARG: break;
			
 
				-
			
 
				-  default:
			
 
				-    setintV(&as->J->errinfo, ir->o);
			
 
				-    lj_trace_err_info(as->J, LJ_TRERR_NYIIR);
			
 
				-    break;
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				 /* -- Trace setup --------------------------------------------------------- */
			
 
				 
			
 
				 /* Ensure there are enough stack slots for call arguments. */
			
 
				 static Reg asm_setup_call_slots(ASMState *as, IRIns *ir, const CCallInfo *ci)
			
 
				 {
			
 
				   IRRef args[CCI_NARGS_MAX*2];
			
 
				-  uint32_t i, nargs = (int)CCI_NARGS(ci);
			
 
				+  uint32_t i, nargs = CCI_XNARGS(ci);
			
 
				   int nslots = 0, ngpr = REGARG_NUMGPR, nfpr = REGARG_NUMFPR, fprodd = 0;
			
 
				   asm_collectargs(as, ir, ci, args);
			
 
				   for (i = 0; i < nargs; i++) {
			
--- a/luajit.mod/luajit/src/lj_asm_arm64.h
+++ b/luajit.mod/luajit/src/lj_asm_arm64.h
@@ -0,0 +1,2031 @@
 
				+/*
			
 
				+** ARM64 IR assembler (SSA IR -> machine code).
			
 
				+** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
			
 
				+**
			
 
				+** Contributed by Djordje Kovacevic and Stefan Pejic from RT-RK.com.
			
 
				+** Sponsored by Cisco Systems, Inc.
			
 
				+*/
			
 
				+
			
 
				+/* -- Register allocator extensions --------------------------------------- */
			
 
				+
			
 
				+/* Allocate a register with a hint. */
			
 
				+static Reg ra_hintalloc(ASMState *as, IRRef ref, Reg hint, RegSet allow)
			
 
				+{
			
 
				+  Reg r = IR(ref)->r;
			
 
				+  if (ra_noreg(r)) {
			
 
				+    if (!ra_hashint(r) && !iscrossref(as, ref))
			
 
				+      ra_sethint(IR(ref)->r, hint);  /* Propagate register hint. */
			
 
				+    r = ra_allocref(as, ref, allow);
			
 
				+  }
			
 
				+  ra_noweak(as, r);
			
 
				+  return r;
			
 
				+}
			
 
				+
			
 
				+/* Allocate two source registers for three-operand instructions. */
			
 
				+static Reg ra_alloc2(ASMState *as, IRIns *ir, RegSet allow)
			
 
				+{
			
 
				+  IRIns *irl = IR(ir->op1), *irr = IR(ir->op2);
			
 
				+  Reg left = irl->r, right = irr->r;
			
 
				+  if (ra_hasreg(left)) {
			
 
				+    ra_noweak(as, left);
			
 
				+    if (ra_noreg(right))
			
 
				+      right = ra_allocref(as, ir->op2, rset_exclude(allow, left));
			
 
				+    else
			
 
				+      ra_noweak(as, right);
			
 
				+  } else if (ra_hasreg(right)) {
			
 
				+    ra_noweak(as, right);
			
 
				+    left = ra_allocref(as, ir->op1, rset_exclude(allow, right));
			
 
				+  } else if (ra_hashint(right)) {
			
 
				+    right = ra_allocref(as, ir->op2, allow);
			
 
				+    left = ra_alloc1(as, ir->op1, rset_exclude(allow, right));
			
 
				+  } else {
			
 
				+    left = ra_allocref(as, ir->op1, allow);
			
 
				+    right = ra_alloc1(as, ir->op2, rset_exclude(allow, left));
			
 
				+  }
			
 
				+  return left | (right << 8);
			
 
				+}
			
 
				+
			
 
				+/* -- Guard handling ------------------------------------------------------ */
			
 
				+
			
 
				+/* Setup all needed exit stubs. */
			
 
				+static void asm_exitstub_setup(ASMState *as, ExitNo nexits)
			
 
				+{
			
 
				+  ExitNo i;
			
 
				+  MCode *mxp = as->mctop;
			
 
				+  if (mxp - (nexits + 3 + MCLIM_REDZONE) < as->mclim)
			
 
				+    asm_mclimit(as);
			
 
				+  /* 1: str lr,[sp]; bl ->vm_exit_handler; movz w0,traceno; bl <1; bl <1; ... */
			
 
				+  for (i = nexits-1; (int32_t)i >= 0; i--)
			
 
				+    *--mxp = A64I_LE(A64I_BL | A64F_S26(-3-i));
			
 
				+  *--mxp = A64I_LE(A64I_MOVZw | A64F_U16(as->T->traceno));
			
 
				+  mxp--;
			
 
				+  *mxp = A64I_LE(A64I_BL | A64F_S26(((MCode *)(void *)lj_vm_exit_handler-mxp)));
			
 
				+  *--mxp = A64I_LE(A64I_STRx | A64F_D(RID_LR) | A64F_N(RID_SP));
			
 
				+  as->mctop = mxp;
			
 
				+}
			
 
				+
			
 
				+static MCode *asm_exitstub_addr(ASMState *as, ExitNo exitno)
			
 
				+{
			
 
				+  /* Keep this in-sync with exitstub_trace_addr(). */
			
 
				+  return as->mctop + exitno + 3;
			
 
				+}
			
 
				+
			
 
				+/* Emit conditional branch to exit for guard. */
			
 
				+static void asm_guardcc(ASMState *as, A64CC cc)
			
 
				+{
			
 
				+  MCode *target = asm_exitstub_addr(as, as->snapno);
			
 
				+  MCode *p = as->mcp;
			
 
				+  if (LJ_UNLIKELY(p == as->invmcp)) {
			
 
				+    as->loopinv = 1;
			
 
				+    *p = A64I_B | A64F_S26(target-p);
			
 
				+    emit_cond_branch(as, cc^1, p-1);
			
 
				+    return;
			
 
				+  }
			
 
				+  emit_cond_branch(as, cc, target);
			
 
				+}
			
 
				+
			
 
				+/* Emit test and branch instruction to exit for guard. */
			
 
				+static void asm_guardtnb(ASMState *as, A64Ins ai, Reg r, uint32_t bit)
			
 
				+{
			
 
				+  MCode *target = asm_exitstub_addr(as, as->snapno);
			
 
				+  MCode *p = as->mcp;
			
 
				+  if (LJ_UNLIKELY(p == as->invmcp)) {
			
 
				+    as->loopinv = 1;
			
 
				+    *p = A64I_B | A64F_S26(target-p);
			
 
				+    emit_tnb(as, ai^0x01000000u, r, bit, p-1);
			
 
				+    return;
			
 
				+  }
			
 
				+  emit_tnb(as, ai, r, bit, target);
			
 
				+}
			
 
				+
			
 
				+/* Emit compare and branch instruction to exit for guard. */
			
 
				+static void asm_guardcnb(ASMState *as, A64Ins ai, Reg r)
			
 
				+{
			
 
				+  MCode *target = asm_exitstub_addr(as, as->snapno);
			
 
				+  MCode *p = as->mcp;
			
 
				+  if (LJ_UNLIKELY(p == as->invmcp)) {
			
 
				+    as->loopinv = 1;
			
 
				+    *p = A64I_B | A64F_S26(target-p);
			
 
				+    emit_cnb(as, ai^0x01000000u, r, p-1);
			
 
				+    return;
			
 
				+  }
			
 
				+  emit_cnb(as, ai, r, target);
			
 
				+}
			
 
				+
			
 
				+/* -- Operand fusion ------------------------------------------------------ */
			
 
				+
			
 
				+/* Limit linear search to this distance. Avoids O(n^2) behavior. */
			
 
				+#define CONFLICT_SEARCH_LIM	31
			
 
				+
			
 
				+static int asm_isk32(ASMState *as, IRRef ref, int32_t *k)
			
 
				+{
			
 
				+  if (irref_isk(ref)) {
			
 
				+    IRIns *ir = IR(ref);
			
 
				+    if (ir->o == IR_KNULL || !irt_is64(ir->t)) {
			
 
				+      *k = ir->i;
			
 
				+      return 1;
			
 
				+    } else if (checki32((int64_t)ir_k64(ir)->u64)) {
			
 
				+      *k = (int32_t)ir_k64(ir)->u64;
			
 
				+      return 1;
			
 
				+    }
			
 
				+  }
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* Check if there's no conflicting instruction between curins and ref. */
			
 
				+static int noconflict(ASMState *as, IRRef ref, IROp conflict)
			
 
				+{
			
 
				+  IRIns *ir = as->ir;
			
 
				+  IRRef i = as->curins;
			
 
				+  if (i > ref + CONFLICT_SEARCH_LIM)
			
 
				+    return 0;  /* Give up, ref is too far away. */
			
 
				+  while (--i > ref)
			
 
				+    if (ir[i].o == conflict)
			
 
				+      return 0;  /* Conflict found. */
			
 
				+  return 1;  /* Ok, no conflict. */
			
 
				+}
			
 
				+
			
 
				+/* Fuse the array base of colocated arrays. */
			
 
				+static int32_t asm_fuseabase(ASMState *as, IRRef ref)
			
 
				+{
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  if (ir->o == IR_TNEW && ir->op1 <= LJ_MAX_COLOSIZE &&
			
 
				+      !neverfuse(as) && noconflict(as, ref, IR_NEWREF))
			
 
				+    return (int32_t)sizeof(GCtab);
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+#define FUSE_REG	0x40000000
			
 
				+
			
 
				+/* Fuse array/hash/upvalue reference into register+offset operand. */
			
 
				+static Reg asm_fuseahuref(ASMState *as, IRRef ref, int32_t *ofsp, RegSet allow,
			
 
				+			  A64Ins ins)
			
 
				+{
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  if (ra_noreg(ir->r)) {
			
 
				+    if (ir->o == IR_AREF) {
			
 
				+      if (mayfuse(as, ref)) {
			
 
				+	if (irref_isk(ir->op2)) {
			
 
				+	  IRRef tab = IR(ir->op1)->op1;
			
 
				+	  int32_t ofs = asm_fuseabase(as, tab);
			
 
				+	  IRRef refa = ofs ? tab : ir->op1;
			
 
				+	  ofs += 8*IR(ir->op2)->i;
			
 
				+	  if (emit_checkofs(ins, ofs)) {
			
 
				+	    *ofsp = ofs;
			
 
				+	    return ra_alloc1(as, refa, allow);
			
 
				+	  }
			
 
				+	} else {
			
 
				+	  Reg base = ra_alloc1(as, ir->op1, allow);
			
 
				+	  *ofsp = FUSE_REG|ra_alloc1(as, ir->op2, rset_exclude(allow, base));
			
 
				+	  return base;
			
 
				+	}
			
 
				+      }
			
 
				+    } else if (ir->o == IR_HREFK) {
			
 
				+      if (mayfuse(as, ref)) {
			
 
				+	int32_t ofs = (int32_t)(IR(ir->op2)->op2 * sizeof(Node));
			
 
				+	if (emit_checkofs(ins, ofs)) {
			
 
				+	  *ofsp = ofs;
			
 
				+	  return ra_alloc1(as, ir->op1, allow);
			
 
				+	}
			
 
				+      }
			
 
				+    } else if (ir->o == IR_UREFC) {
			
 
				+      if (irref_isk(ir->op1)) {
			
 
				+	GCfunc *fn = ir_kfunc(IR(ir->op1));
			
 
				+	GCupval *uv = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv;
			
 
				+	int64_t ofs = glofs(as, &uv->tv);
			
 
				+	if (emit_checkofs(ins, ofs)) {
			
 
				+	  *ofsp = (int32_t)ofs;
			
 
				+	  return RID_GL;
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  *ofsp = 0;
			
 
				+  return ra_alloc1(as, ref, allow);
			
 
				+}
			
 
				+
			
 
				+/* Fuse m operand into arithmetic/logic instructions. */
			
 
				+static uint32_t asm_fuseopm(ASMState *as, A64Ins ai, IRRef ref, RegSet allow)
			
 
				+{
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  if (ra_hasreg(ir->r)) {
			
 
				+    ra_noweak(as, ir->r);
			
 
				+    return A64F_M(ir->r);
			
 
				+  } else if (irref_isk(ref)) {
			
 
				+    uint32_t m;
			
 
				+    int64_t k = get_k64val(ir);
			
 
				+    if ((ai & 0x1f000000) == 0x0a000000)
			
 
				+      m = emit_isk13(k, irt_is64(ir->t));
			
 
				+    else
			
 
				+      m = emit_isk12(k);
			
 
				+    if (m)
			
 
				+      return m;
			
 
				+  } else if (mayfuse(as, ref)) {
			
 
				+    if ((ir->o >= IR_BSHL && ir->o <= IR_BSAR && irref_isk(ir->op2)) ||
			
 
				+	(ir->o == IR_ADD && ir->op1 == ir->op2)) {
			
 
				+      A64Shift sh = ir->o == IR_BSHR ? A64SH_LSR :
			
 
				+		    ir->o == IR_BSAR ? A64SH_ASR : A64SH_LSL;
			
 
				+      int shift = ir->o == IR_ADD ? 1 :
			
 
				+		    (IR(ir->op2)->i & (irt_is64(ir->t) ? 63 : 31));
			
 
				+      IRIns *irl = IR(ir->op1);
			
 
				+      if (sh == A64SH_LSL &&
			
 
				+	  irl->o == IR_CONV &&
			
 
				+	  irl->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT) &&
			
 
				+	  shift <= 4 &&
			
 
				+	  canfuse(as, irl)) {
			
 
				+	Reg m = ra_alloc1(as, irl->op1, allow);
			
 
				+	return A64F_M(m) | A64F_EXSH(A64EX_SXTW, shift);
			
 
				+      } else {
			
 
				+	Reg m = ra_alloc1(as, ir->op1, allow);
			
 
				+	return A64F_M(m) | A64F_SH(sh, shift);
			
 
				+      }
			
 
				+    } else if (ir->o == IR_CONV &&
			
 
				+	       ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)) {
			
 
				+      Reg m = ra_alloc1(as, ir->op1, allow);
			
 
				+      return A64F_M(m) | A64F_EX(A64EX_SXTW);
			
 
				+    }
			
 
				+  }
			
 
				+  return A64F_M(ra_allocref(as, ref, allow));
			
 
				+}
			
 
				+
			
 
				+/* Fuse XLOAD/XSTORE reference into load/store operand. */
			
 
				+static void asm_fusexref(ASMState *as, A64Ins ai, Reg rd, IRRef ref,
			
 
				+			 RegSet allow)
			
 
				+{
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  Reg base;
			
 
				+  int32_t ofs = 0;
			
 
				+  if (ra_noreg(ir->r) && canfuse(as, ir)) {
			
 
				+    if (ir->o == IR_ADD) {
			
 
				+      if (asm_isk32(as, ir->op2, &ofs) && emit_checkofs(ai, ofs)) {
			
 
				+	ref = ir->op1;
			
 
				+      } else {
			
 
				+	Reg rn, rm;
			
 
				+	IRRef lref = ir->op1, rref = ir->op2;
			
 
				+	IRIns *irl = IR(lref);
			
 
				+	if (mayfuse(as, irl->op1)) {
			
 
				+	  unsigned int shift = 4;
			
 
				+	  if (irl->o == IR_BSHL && irref_isk(irl->op2)) {
			
 
				+	    shift = (IR(irl->op2)->i & 63);
			
 
				+	  } else if (irl->o == IR_ADD && irl->op1 == irl->op2) {
			
 
				+	    shift = 1;
			
 
				+	  }
			
 
				+	  if ((ai >> 30) == shift) {
			
 
				+	    lref = irl->op1;
			
 
				+	    irl = IR(lref);
			
 
				+	    ai |= A64I_LS_SH;
			
 
				+	  }
			
 
				+	}
			
 
				+	if (irl->o == IR_CONV &&
			
 
				+	    irl->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT) &&
			
 
				+	    canfuse(as, irl)) {
			
 
				+	  lref = irl->op1;
			
 
				+	  ai |= A64I_LS_SXTWx;
			
 
				+	} else {
			
 
				+	  ai |= A64I_LS_LSLx;
			
 
				+	}
			
 
				+	rm = ra_alloc1(as, lref, allow);
			
 
				+	rn = ra_alloc1(as, rref, rset_exclude(allow, rm));
			
 
				+	emit_dnm(as, (ai^A64I_LS_R), (rd & 31), rn, rm);
			
 
				+	return;
			
 
				+      }
			
 
				+    } else if (ir->o == IR_STRREF) {
			
 
				+      if (asm_isk32(as, ir->op2, &ofs)) {
			
 
				+	ref = ir->op1;
			
 
				+      } else if (asm_isk32(as, ir->op1, &ofs)) {
			
 
				+	ref = ir->op2;
			
 
				+      } else {
			
 
				+	Reg rn = ra_alloc1(as, ir->op1, allow);
			
 
				+	IRIns *irr = IR(ir->op2);
			
 
				+	uint32_t m;
			
 
				+	if (irr+1 == ir && !ra_used(irr) &&
			
 
				+	    irr->o == IR_ADD && irref_isk(irr->op2)) {
			
 
				+	  ofs = sizeof(GCstr) + IR(irr->op2)->i;
			
 
				+	  if (emit_checkofs(ai, ofs)) {
			
 
				+	    Reg rm = ra_alloc1(as, irr->op1, rset_exclude(allow, rn));
			
 
				+	    m = A64F_M(rm) | A64F_EX(A64EX_SXTW);
			
 
				+	    goto skipopm;
			
 
				+	  }
			
 
				+	}
			
 
				+	m = asm_fuseopm(as, 0, ir->op2, rset_exclude(allow, rn));
			
 
				+	ofs = sizeof(GCstr);
			
 
				+      skipopm:
			
 
				+	emit_lso(as, ai, rd, rd, ofs);
			
 
				+	emit_dn(as, A64I_ADDx^m, rd, rn);
			
 
				+	return;
			
 
				+      }
			
 
				+      ofs += sizeof(GCstr);
			
 
				+      if (!emit_checkofs(ai, ofs)) {
			
 
				+	Reg rn = ra_alloc1(as, ref, allow);
			
 
				+	Reg rm = ra_allock(as, ofs, rset_exclude(allow, rn));
			
 
				+	emit_dnm(as, (ai^A64I_LS_R)|A64I_LS_UXTWx, rd, rn, rm);
			
 
				+	return;
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  base = ra_alloc1(as, ref, allow);
			
 
				+  emit_lso(as, ai, (rd & 31), base, ofs);
			
 
				+}
			
 
				+
			
 
				+/* Fuse FP multiply-add/sub. */
			
 
				+static int asm_fusemadd(ASMState *as, IRIns *ir, A64Ins ai, A64Ins air)
			
 
				+{
			
 
				+  IRRef lref = ir->op1, rref = ir->op2;
			
 
				+  IRIns *irm;
			
 
				+  if (lref != rref &&
			
 
				+      ((mayfuse(as, lref) && (irm = IR(lref), irm->o == IR_MUL) &&
			
 
				+       ra_noreg(irm->r)) ||
			
 
				+       (mayfuse(as, rref) && (irm = IR(rref), irm->o == IR_MUL) &&
			
 
				+       (rref = lref, ai = air, ra_noreg(irm->r))))) {
			
 
				+    Reg dest = ra_dest(as, ir, RSET_FPR);
			
 
				+    Reg add = ra_hintalloc(as, rref, dest, RSET_FPR);
			
 
				+    Reg left = ra_alloc2(as, irm,
			
 
				+			 rset_exclude(rset_exclude(RSET_FPR, dest), add));
			
 
				+    Reg right = (left >> 8); left &= 255;
			
 
				+    emit_dnma(as, ai, (dest & 31), (left & 31), (right & 31), (add & 31));
			
 
				+    return 1;
			
 
				+  }
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* Fuse BAND + BSHL/BSHR into UBFM. */
			
 
				+static int asm_fuseandshift(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRIns *irl = IR(ir->op1);
			
 
				+  lua_assert(ir->o == IR_BAND);
			
 
				+  if (canfuse(as, irl) && irref_isk(ir->op2)) {
			
 
				+    uint64_t mask = get_k64val(IR(ir->op2));
			
 
				+    if (irref_isk(irl->op2) && (irl->o == IR_BSHR || irl->o == IR_BSHL)) {
			
 
				+      int32_t shmask = irt_is64(irl->t) ? 63 : 31;
			
 
				+      int32_t shift = (IR(irl->op2)->i & shmask);
			
 
				+      int32_t imms = shift;
			
 
				+      if (irl->o == IR_BSHL) {
			
 
				+	mask >>= shift;
			
 
				+	shift = (shmask-shift+1) & shmask;
			
 
				+	imms = 0;
			
 
				+      }
			
 
				+      if (mask && !((mask+1) & mask)) {  /* Contiguous 1-bits at the bottom. */
			
 
				+	Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+	Reg left = ra_alloc1(as, irl->op1, RSET_GPR);
			
 
				+	A64Ins ai = shmask == 63 ? A64I_UBFMx : A64I_UBFMw;
			
 
				+	imms += 63 - emit_clz64(mask);
			
 
				+	if (imms > shmask) imms = shmask;
			
 
				+	emit_dn(as, ai | A64F_IMMS(imms) | A64F_IMMR(shift), dest, left);
			
 
				+	return 1;
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* Fuse BOR(BSHL, BSHR) into EXTR/ROR. */
			
 
				+static int asm_fuseorshift(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRIns *irl = IR(ir->op1), *irr = IR(ir->op2);
			
 
				+  lua_assert(ir->o == IR_BOR);
			
 
				+  if (canfuse(as, irl) && canfuse(as, irr) &&
			
 
				+      ((irl->o == IR_BSHR && irr->o == IR_BSHL) ||
			
 
				+       (irl->o == IR_BSHL && irr->o == IR_BSHR))) {
			
 
				+    if (irref_isk(irl->op2) && irref_isk(irr->op2)) {
			
 
				+      IRRef lref = irl->op1, rref = irr->op1;
			
 
				+      uint32_t lshift = IR(irl->op2)->i, rshift = IR(irr->op2)->i;
			
 
				+      if (irl->o == IR_BSHR) {  /* BSHR needs to be the right operand. */
			
 
				+	uint32_t tmp2;
			
 
				+	IRRef tmp1 = lref; lref = rref; rref = tmp1;
			
 
				+	tmp2 = lshift; lshift = rshift; rshift = tmp2;
			
 
				+      }
			
 
				+      if (rshift + lshift == (irt_is64(ir->t) ? 64 : 32)) {
			
 
				+	A64Ins ai = irt_is64(ir->t) ? A64I_EXTRx : A64I_EXTRw;
			
 
				+	Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+	Reg left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+	Reg right = ra_alloc1(as, rref, rset_exclude(RSET_GPR, left));
			
 
				+	emit_dnm(as, ai | A64F_IMMS(rshift), dest, left, right);
			
 
				+	return 1;
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  return 0;
			
 
				+}
			
 
				+
			
 
				+/* -- Calls --------------------------------------------------------------- */
			
 
				+
			
 
				+/* Generate a call to a C function. */
			
 
				+static void asm_gencall(ASMState *as, const CCallInfo *ci, IRRef *args)
			
 
				+{
			
 
				+  uint32_t n, nargs = CCI_XNARGS(ci);
			
 
				+  int32_t ofs = 0;
			
 
				+  Reg gpr, fpr = REGARG_FIRSTFPR;
			
 
				+  if ((void *)ci->func)
			
 
				+    emit_call(as, (void *)ci->func);
			
 
				+  for (gpr = REGARG_FIRSTGPR; gpr <= REGARG_LASTGPR; gpr++)
			
 
				+    as->cost[gpr] = REGCOST(~0u, ASMREF_L);
			
 
				+  gpr = REGARG_FIRSTGPR;
			
 
				+  for (n = 0; n < nargs; n++) { /* Setup args. */
			
 
				+    IRRef ref = args[n];
			
 
				+    IRIns *ir = IR(ref);
			
 
				+    if (ref) {
			
 
				+      if (irt_isfp(ir->t)) {
			
 
				+	if (fpr <= REGARG_LASTFPR) {
			
 
				+	  lua_assert(rset_test(as->freeset, fpr)); /* Must have been evicted. */
			
 
				+	  ra_leftov(as, fpr, ref);
			
 
				+	  fpr++;
			
 
				+	} else {
			
 
				+	  Reg r = ra_alloc1(as, ref, RSET_FPR);
			
 
				+	  emit_spstore(as, ir, r, ofs + ((LJ_BE && !irt_isnum(ir->t)) ? 4 : 0));
			
 
				+	  ofs += 8;
			
 
				+	}
			
 
				+      } else {
			
 
				+	if (gpr <= REGARG_LASTGPR) {
			
 
				+	  lua_assert(rset_test(as->freeset, gpr)); /* Must have been evicted. */
			
 
				+	  ra_leftov(as, gpr, ref);
			
 
				+	  gpr++;
			
 
				+	} else {
			
 
				+	  Reg r = ra_alloc1(as, ref, RSET_GPR);
			
 
				+	  emit_spstore(as, ir, r, ofs + ((LJ_BE && !irt_is64(ir->t)) ? 4 : 0));
			
 
				+	  ofs += 8;
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* Setup result reg/sp for call. Evict scratch regs. */
			
 
				+static void asm_setupresult(ASMState *as, IRIns *ir, const CCallInfo *ci)
			
 
				+{
			
 
				+  RegSet drop = RSET_SCRATCH;
			
 
				+  if (ra_hasreg(ir->r))
			
 
				+    rset_clear(drop, ir->r); /* Dest reg handled below. */
			
 
				+  ra_evictset(as, drop); /* Evictions must be performed first. */
			
 
				+  if (ra_used(ir)) {
			
 
				+    lua_assert(!irt_ispri(ir->t));
			
 
				+    if (irt_isfp(ir->t)) {
			
 
				+      if (ci->flags & CCI_CASTU64) {
			
 
				+	Reg dest = ra_dest(as, ir, RSET_FPR) & 31;
			
 
				+	emit_dn(as, irt_isnum(ir->t) ? A64I_FMOV_D_R : A64I_FMOV_S_R,
			
 
				+		dest, RID_RET);
			
 
				+      } else {
			
 
				+	ra_destreg(as, ir, RID_FPRET);
			
 
				+      }
			
 
				+    } else {
			
 
				+      ra_destreg(as, ir, RID_RET);
			
 
				+    }
			
 
				+  }
			
 
				+  UNUSED(ci);
			
 
				+}
			
 
				+
			
 
				+static void asm_callx(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRRef args[CCI_NARGS_MAX*2];
			
 
				+  CCallInfo ci;
			
 
				+  IRRef func;
			
 
				+  IRIns *irf;
			
 
				+  ci.flags = asm_callx_flags(as, ir);
			
 
				+  asm_collectargs(as, ir, &ci, args);
			
 
				+  asm_setupresult(as, ir, &ci);
			
 
				+  func = ir->op2; irf = IR(func);
			
 
				+  if (irf->o == IR_CARG) { func = irf->op1; irf = IR(func); }
			
 
				+  if (irref_isk(func)) {  /* Call to constant address. */
			
 
				+    ci.func = (ASMFunction)(ir_k64(irf)->u64);
			
 
				+  } else {  /* Need a non-argument register for indirect calls. */
			
 
				+    Reg freg = ra_alloc1(as, func, RSET_RANGE(RID_X8, RID_MAX_GPR)-RSET_FIXED);
			
 
				+    emit_n(as, A64I_BLR, freg);
			
 
				+    ci.func = (ASMFunction)(void *)0;
			
 
				+  }
			
 
				+  asm_gencall(as, &ci, args);
			
 
				+}
			
 
				+
			
 
				+/* -- Returns ------------------------------------------------------------- */
			
 
				+
			
 
				+/* Return to lower frame. Guard that it goes to the right spot. */
			
 
				+static void asm_retf(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg base = ra_alloc1(as, REF_BASE, RSET_GPR);
			
 
				+  void *pc = ir_kptr(IR(ir->op2));
			
 
				+  int32_t delta = 1+LJ_FR2+bc_a(*((const BCIns *)pc - 1));
			
 
				+  as->topslot -= (BCReg)delta;
			
 
				+  if ((int32_t)as->topslot < 0) as->topslot = 0;
			
 
				+  irt_setmark(IR(REF_BASE)->t);  /* Children must not coalesce with BASE reg. */
			
 
				+  /* Need to force a spill on REF_BASE now to update the stack slot. */
			
 
				+  emit_lso(as, A64I_STRx, base, RID_SP, ra_spill(as, IR(REF_BASE)));
			
 
				+  emit_setgl(as, base, jit_base);
			
 
				+  emit_addptr(as, base, -8*delta);
			
 
				+  asm_guardcc(as, CC_NE);
			
 
				+  emit_nm(as, A64I_CMPx, RID_TMP,
			
 
				+	  ra_allock(as, i64ptr(pc), rset_exclude(RSET_GPR, base)));
			
 
				+  emit_lso(as, A64I_LDRx, RID_TMP, base, -8);
			
 
				+}
			
 
				+
			
 
				+/* -- Type conversions ---------------------------------------------------- */
			
 
				+
			
 
				+static void asm_tointg(ASMState *as, IRIns *ir, Reg left)
			
 
				+{
			
 
				+  Reg tmp = ra_scratch(as, rset_exclude(RSET_FPR, left));
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  asm_guardcc(as, CC_NE);
			
 
				+  emit_nm(as, A64I_FCMPd, (tmp & 31), (left & 31));
			
 
				+  emit_dn(as, A64I_FCVT_F64_S32, (tmp & 31), dest);
			
 
				+  emit_dn(as, A64I_FCVT_S32_F64, dest, (left & 31));
			
 
				+}
			
 
				+
			
 
				+static void asm_tobit(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  RegSet allow = RSET_FPR;
			
 
				+  Reg left = ra_alloc1(as, ir->op1, allow);
			
 
				+  Reg right = ra_alloc1(as, ir->op2, rset_clear(allow, left));
			
 
				+  Reg tmp = ra_scratch(as, rset_clear(allow, right));
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  emit_dn(as, A64I_FMOV_R_S, dest, (tmp & 31));
			
 
				+  emit_dnm(as, A64I_FADDd, (tmp & 31), (left & 31), (right & 31));
			
 
				+}
			
 
				+
			
 
				+static void asm_conv(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRType st = (IRType)(ir->op2 & IRCONV_SRCMASK);
			
 
				+  int st64 = (st == IRT_I64 || st == IRT_U64 || st == IRT_P64);
			
 
				+  int stfp = (st == IRT_NUM || st == IRT_FLOAT);
			
 
				+  IRRef lref = ir->op1;
			
 
				+  lua_assert(irt_type(ir->t) != st);
			
 
				+  if (irt_isfp(ir->t)) {
			
 
				+    Reg dest = ra_dest(as, ir, RSET_FPR);
			
 
				+    if (stfp) {  /* FP to FP conversion. */
			
 
				+      emit_dn(as, st == IRT_NUM ? A64I_FCVT_F32_F64 : A64I_FCVT_F64_F32,
			
 
				+	      (dest & 31), (ra_alloc1(as, lref, RSET_FPR) & 31));
			
 
				+    } else {  /* Integer to FP conversion. */
			
 
				+      Reg left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+      A64Ins ai = irt_isfloat(ir->t) ?
			
 
				+	(((IRT_IS64 >> st) & 1) ?
			
 
				+	 (st == IRT_I64 ? A64I_FCVT_F32_S64 : A64I_FCVT_F32_U64) :
			
 
				+	 (st == IRT_INT ? A64I_FCVT_F32_S32 : A64I_FCVT_F32_U32)) :
			
 
				+	(((IRT_IS64 >> st) & 1) ?
			
 
				+	 (st == IRT_I64 ? A64I_FCVT_F64_S64 : A64I_FCVT_F64_U64) :
			
 
				+	 (st == IRT_INT ? A64I_FCVT_F64_S32 : A64I_FCVT_F64_U32));
			
 
				+      emit_dn(as, ai, (dest & 31), left);
			
 
				+    }
			
 
				+  } else if (stfp) {  /* FP to integer conversion. */
			
 
				+    if (irt_isguard(ir->t)) {
			
 
				+      /* Checked conversions are only supported from number to int. */
			
 
				+      lua_assert(irt_isint(ir->t) && st == IRT_NUM);
			
 
				+      asm_tointg(as, ir, ra_alloc1(as, lref, RSET_FPR));
			
 
				+    } else {
			
 
				+      Reg left = ra_alloc1(as, lref, RSET_FPR);
			
 
				+      Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+      A64Ins ai = irt_is64(ir->t) ?
			
 
				+	(st == IRT_NUM ?
			
 
				+	 (irt_isi64(ir->t) ? A64I_FCVT_S64_F64 : A64I_FCVT_U64_F64) :
			
 
				+	 (irt_isi64(ir->t) ? A64I_FCVT_S64_F32 : A64I_FCVT_U64_F32)) :
			
 
				+	(st == IRT_NUM ?
			
 
				+	 (irt_isint(ir->t) ? A64I_FCVT_S32_F64 : A64I_FCVT_U32_F64) :
			
 
				+	 (irt_isint(ir->t) ? A64I_FCVT_S32_F32 : A64I_FCVT_U32_F32));
			
 
				+      emit_dn(as, ai, dest, (left & 31));
			
 
				+    }
			
 
				+  } else if (st >= IRT_I8 && st <= IRT_U16) { /* Extend to 32 bit integer. */
			
 
				+    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    Reg left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+    A64Ins ai = st == IRT_I8 ? A64I_SXTBw :
			
 
				+		st == IRT_U8 ? A64I_UXTBw :
			
 
				+		st == IRT_I16 ? A64I_SXTHw : A64I_UXTHw;
			
 
				+    lua_assert(irt_isint(ir->t) || irt_isu32(ir->t));
			
 
				+    emit_dn(as, ai, dest, left);
			
 
				+  } else {
			
 
				+    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    if (irt_is64(ir->t)) {
			
 
				+      if (st64 || !(ir->op2 & IRCONV_SEXT)) {
			
 
				+	/* 64/64 bit no-op (cast) or 32 to 64 bit zero extension. */
			
 
				+	ra_leftov(as, dest, lref);  /* Do nothing, but may need to move regs. */
			
 
				+      } else {  /* 32 to 64 bit sign extension. */
			
 
				+	Reg left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+	emit_dn(as, A64I_SXTW, dest, left);
			
 
				+      }
			
 
				+    } else {
			
 
				+      if (st64) {
			
 
				+	/* This is either a 32 bit reg/reg mov which zeroes the hiword
			
 
				+	** or a load of the loword from a 64 bit address.
			
 
				+	*/
			
 
				+	Reg left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+	emit_dm(as, A64I_MOVw, dest, left);
			
 
				+      } else {  /* 32/32 bit no-op (cast). */
			
 
				+	ra_leftov(as, dest, lref);  /* Do nothing, but may need to move regs. */
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_strto(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_strscan_num];
			
 
				+  IRRef args[2];
			
 
				+  Reg dest = 0, tmp;
			
 
				+  int destused = ra_used(ir);
			
 
				+  int32_t ofs = 0;
			
 
				+  ra_evictset(as, RSET_SCRATCH);
			
 
				+  if (destused) {
			
 
				+    if (ra_hasspill(ir->s)) {
			
 
				+      ofs = sps_scale(ir->s);
			
 
				+      destused = 0;
			
 
				+      if (ra_hasreg(ir->r)) {
			
 
				+	ra_free(as, ir->r);
			
 
				+	ra_modified(as, ir->r);
			
 
				+	emit_spload(as, ir, ir->r, ofs);
			
 
				+      }
			
 
				+    } else {
			
 
				+      dest = ra_dest(as, ir, RSET_FPR);
			
 
				+    }
			
 
				+  }
			
 
				+  if (destused)
			
 
				+    emit_lso(as, A64I_LDRd, (dest & 31), RID_SP, 0);
			
 
				+  asm_guardcnb(as, A64I_CBZ, RID_RET);
			
 
				+  args[0] = ir->op1; /* GCstr *str */
			
 
				+  args[1] = ASMREF_TMP1; /* TValue *n  */
			
 
				+  asm_gencall(as, ci, args);
			
 
				+  tmp = ra_releasetmp(as, ASMREF_TMP1);
			
 
				+  emit_opk(as, A64I_ADDx, tmp, RID_SP, ofs, RSET_GPR);
			
 
				+}
			
 
				+
			
 
				+/* -- Memory references --------------------------------------------------- */
			
 
				+
			
 
				+/* Store tagged value for ref at base+ofs. */
			
 
				+static void asm_tvstore64(ASMState *as, Reg base, int32_t ofs, IRRef ref)
			
 
				+{
			
 
				+  RegSet allow = rset_exclude(RSET_GPR, base);
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  lua_assert(irt_ispri(ir->t) || irt_isaddr(ir->t) || irt_isinteger(ir->t));
			
 
				+  if (irref_isk(ref)) {
			
 
				+    TValue k;
			
 
				+    lj_ir_kvalue(as->J->L, &k, ir);
			
 
				+    emit_lso(as, A64I_STRx, ra_allock(as, k.u64, allow), base, ofs);
			
 
				+  } else {
			
 
				+    Reg src = ra_alloc1(as, ref, allow);
			
 
				+    rset_clear(allow, src);
			
 
				+    if (irt_isinteger(ir->t)) {
			
 
				+      Reg type = ra_allock(as, (int64_t)irt_toitype(ir->t) << 47, allow);
			
 
				+      emit_lso(as, A64I_STRx, RID_TMP, base, ofs);
			
 
				+      emit_dnm(as, A64I_ADDx | A64F_EX(A64EX_UXTW), RID_TMP, type, src);
			
 
				+    } else {
			
 
				+      Reg type = ra_allock(as, (int32_t)irt_toitype(ir->t), allow);
			
 
				+      emit_lso(as, A64I_STRx, RID_TMP, base, ofs);
			
 
				+      emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 47), RID_TMP, src, type);
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* Get pointer to TValue. */
			
 
				+static void asm_tvptr(ASMState *as, Reg dest, IRRef ref)
			
 
				+{
			
 
				+  IRIns *ir = IR(ref);
			
 
				+  if (irt_isnum(ir->t)) {
			
 
				+    if (irref_isk(ref)) {
			
 
				+      /* Use the number constant itself as a TValue. */
			
 
				+      ra_allockreg(as, i64ptr(ir_knum(ir)), dest);
			
 
				+    } else {
			
 
				+      /* Otherwise force a spill and use the spill slot. */
			
 
				+      emit_opk(as, A64I_ADDx, dest, RID_SP, ra_spill(as, ir), RSET_GPR);
			
 
				+    }
			
 
				+  } else {
			
 
				+    /* Otherwise use g->tmptv to hold the TValue. */
			
 
				+    asm_tvstore64(as, dest, 0, ref);
			
 
				+    ra_allockreg(as, i64ptr(&J2G(as->J)->tmptv), dest);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_aref(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg idx, base;
			
 
				+  if (irref_isk(ir->op2)) {
			
 
				+    IRRef tab = IR(ir->op1)->op1;
			
 
				+    int32_t ofs = asm_fuseabase(as, tab);
			
 
				+    IRRef refa = ofs ? tab : ir->op1;
			
 
				+    uint32_t k = emit_isk12(ofs + 8*IR(ir->op2)->i);
			
 
				+    if (k) {
			
 
				+      base = ra_alloc1(as, refa, RSET_GPR);
			
 
				+      emit_dn(as, A64I_ADDx^k, dest, base);
			
 
				+      return;
			
 
				+    }
			
 
				+  }
			
 
				+  base = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+  idx = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, base));
			
 
				+  emit_dnm(as, A64I_ADDx | A64F_EXSH(A64EX_UXTW, 3), dest, base, idx);
			
 
				+}
			
 
				+
			
 
				+/* Inlined hash lookup. Specialized for key type and for const keys.
			
 
				+** The equivalent C code is:
			
 
				+**   Node *n = hashkey(t, key);
			
 
				+**   do {
			
 
				+**     if (lj_obj_equal(&n->key, key)) return &n->val;
			
 
				+**   } while ((n = nextnode(n)));
			
 
				+**   return niltv(L);
			
 
				+*/
			
 
				+static void asm_href(ASMState *as, IRIns *ir, IROp merge)
			
 
				+{
			
 
				+  RegSet allow = RSET_GPR;
			
 
				+  int destused = ra_used(ir);
			
 
				+  Reg dest = ra_dest(as, ir, allow);
			
 
				+  Reg tab = ra_alloc1(as, ir->op1, rset_clear(allow, dest));
			
 
				+  Reg key = 0, tmp = RID_TMP;
			
 
				+  IRRef refkey = ir->op2;
			
 
				+  IRIns *irkey = IR(refkey);
			
 
				+  int isk = irref_isk(ir->op2);
			
 
				+  IRType1 kt = irkey->t;
			
 
				+  uint32_t k = 0;
			
 
				+  uint32_t khash;
			
 
				+  MCLabel l_end, l_loop, l_next;
			
 
				+  rset_clear(allow, tab);
			
 
				+
			
 
				+  if (!isk) {
			
 
				+    key = ra_alloc1(as, ir->op2, irt_isnum(kt) ? RSET_FPR : allow);
			
 
				+    rset_clear(allow, key);
			
 
				+    if (!irt_isstr(kt)) {
			
 
				+      tmp = ra_scratch(as, allow);
			
 
				+      rset_clear(allow, tmp);
			
 
				+    }
			
 
				+  } else if (irt_isnum(kt)) {
			
 
				+    int64_t val = (int64_t)ir_knum(irkey)->u64;
			
 
				+    if (!(k = emit_isk12(val))) {
			
 
				+      key = ra_allock(as, val, allow);
			
 
				+      rset_clear(allow, key);
			
 
				+    }
			
 
				+  } else if (!irt_ispri(kt)) {
			
 
				+    if (!(k = emit_isk12(irkey->i))) {
			
 
				+      key = ra_alloc1(as, refkey, allow);
			
 
				+      rset_clear(allow, key);
			
 
				+    }
			
 
				+  }
			
 
				+
			
 
				+  /* Key not found in chain: jump to exit (if merged) or load niltv. */
			
 
				+  l_end = emit_label(as);
			
 
				+  as->invmcp = NULL;
			
 
				+  if (merge == IR_NE)
			
 
				+    asm_guardcc(as, CC_AL);
			
 
				+  else if (destused)
			
 
				+    emit_loada(as, dest, niltvg(J2G(as->J)));
			
 
				+
			
 
				+  /* Follow hash chain until the end. */
			
 
				+  l_loop = --as->mcp;
			
 
				+  emit_n(as, A64I_CMPx^A64I_K12^0, dest);
			
 
				+  emit_lso(as, A64I_LDRx, dest, dest, offsetof(Node, next));
			
 
				+  l_next = emit_label(as);
			
 
				+
			
 
				+  /* Type and value comparison. */
			
 
				+  if (merge == IR_EQ)
			
 
				+    asm_guardcc(as, CC_EQ);
			
 
				+  else
			
 
				+    emit_cond_branch(as, CC_EQ, l_end);
			
 
				+
			
 
				+  if (irt_isnum(kt)) {
			
 
				+    if (isk) {
			
 
				+      /* Assumes -0.0 is already canonicalized to +0.0. */
			
 
				+      if (k)
			
 
				+	emit_n(as, A64I_CMPx^k, tmp);
			
 
				+      else
			
 
				+	emit_nm(as, A64I_CMPx, key, tmp);
			
 
				+      emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.u64));
			
 
				+    } else {
			
 
				+      Reg tisnum = ra_allock(as, LJ_TISNUM << 15, allow);
			
 
				+      Reg ftmp = ra_scratch(as, rset_exclude(RSET_FPR, key));
			
 
				+      rset_clear(allow, tisnum);
			
 
				+      emit_nm(as, A64I_FCMPd, key, ftmp);
			
 
				+      emit_dn(as, A64I_FMOV_D_R, (ftmp & 31), (tmp & 31));
			
 
				+      emit_cond_branch(as, CC_LO, l_next);
			
 
				+      emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32), tisnum, tmp);
			
 
				+      emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.n));
			
 
				+    }
			
 
				+  } else if (irt_isaddr(kt)) {
			
 
				+    Reg scr;
			
 
				+    if (isk) {
			
 
				+      int64_t kk = ((int64_t)irt_toitype(irkey->t) << 47) | irkey[1].tv.u64;
			
 
				+      scr = ra_allock(as, kk, allow);
			
 
				+      emit_nm(as, A64I_CMPx, scr, tmp);
			
 
				+      emit_lso(as, A64I_LDRx, tmp, dest, offsetof(Node, key.u64));
			
 
				+    } else {
			
 
				+      scr = ra_scratch(as, allow);
			
 
				+      emit_nm(as, A64I_CMPx, tmp, scr);
			
 
				+      emit_lso(as, A64I_LDRx, scr, dest, offsetof(Node, key.u64));
			
 
				+    }
			
 
				+    rset_clear(allow, scr);
			
 
				+  } else {
			
 
				+    Reg type, scr;
			
 
				+    lua_assert(irt_ispri(kt) && !irt_isnil(kt));
			
 
				+    type = ra_allock(as, ~((int64_t)~irt_toitype(ir->t) << 47), allow);
			
 
				+    scr = ra_scratch(as, rset_clear(allow, type));
			
 
				+    rset_clear(allow, scr);
			
 
				+    emit_nm(as, A64I_CMPw, scr, type);
			
 
				+    emit_lso(as, A64I_LDRx, scr, dest, offsetof(Node, key));
			
 
				+  }
			
 
				+
			
 
				+  *l_loop = A64I_BCC | A64F_S19(as->mcp - l_loop) | CC_NE;
			
 
				+  if (!isk && irt_isaddr(kt)) {
			
 
				+    Reg type = ra_allock(as, (int32_t)irt_toitype(kt), allow);
			
 
				+    emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 47), tmp, key, type);
			
 
				+    rset_clear(allow, type);
			
 
				+  }
			
 
				+  /* Load main position relative to tab->node into dest. */
			
 
				+  khash = isk ? ir_khash(irkey) : 1;
			
 
				+  if (khash == 0) {
			
 
				+    emit_lso(as, A64I_LDRx, dest, tab, offsetof(GCtab, node));
			
 
				+  } else {
			
 
				+    emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 3), dest, tmp, dest);
			
 
				+    emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 1), dest, dest, dest);
			
 
				+    emit_lso(as, A64I_LDRx, tmp, tab, offsetof(GCtab, node));
			
 
				+    if (isk) {
			
 
				+      Reg tmphash = ra_allock(as, khash, allow);
			
 
				+      emit_dnm(as, A64I_ANDw, dest, dest, tmphash);
			
 
				+      emit_lso(as, A64I_LDRw, dest, tab, offsetof(GCtab, hmask));
			
 
				+    } else if (irt_isstr(kt)) {
			
 
				+      /* Fetch of str->hash is cheaper than ra_allock. */
			
 
				+      emit_dnm(as, A64I_ANDw, dest, dest, tmp);
			
 
				+      emit_lso(as, A64I_LDRw, tmp, key, offsetof(GCstr, hash));
			
 
				+      emit_lso(as, A64I_LDRw, dest, tab, offsetof(GCtab, hmask));
			
 
				+    } else {  /* Must match with hash*() in lj_tab.c. */
			
 
				+      emit_dnm(as, A64I_ANDw, dest, dest, tmp);
			
 
				+      emit_lso(as, A64I_LDRw, tmp, tab, offsetof(GCtab, hmask));
			
 
				+      emit_dnm(as, A64I_SUBw, dest, dest, tmp);
			
 
				+      emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT3)), tmp, tmp, tmp);
			
 
				+      emit_dnm(as, A64I_EORw, dest, dest, tmp);
			
 
				+      emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT2)), dest, dest, dest);
			
 
				+      emit_dnm(as, A64I_SUBw, tmp, tmp, dest);
			
 
				+      emit_dnm(as, A64I_EXTRw | (A64F_IMMS(32-HASH_ROT1)), dest, dest, dest);
			
 
				+      emit_dnm(as, A64I_EORw, tmp, tmp, dest);
			
 
				+      if (irt_isnum(kt)) {
			
 
				+	emit_dnm(as, A64I_ADDw, dest, dest, dest);
			
 
				+	emit_dn(as, A64I_LSRx | A64F_IMMR(32)|A64F_IMMS(32), dest, dest);
			
 
				+	emit_dm(as, A64I_MOVw, tmp, dest);
			
 
				+	emit_dn(as, A64I_FMOV_R_D, dest, (key & 31));
			
 
				+      } else {
			
 
				+	checkmclim(as);
			
 
				+	emit_dm(as, A64I_MOVw, tmp, key);
			
 
				+	emit_dnm(as, A64I_EORw, dest, dest,
			
 
				+		 ra_allock(as, irt_toitype(kt) << 15, allow));
			
 
				+	emit_dn(as, A64I_LSRx | A64F_IMMR(32)|A64F_IMMS(32), dest, dest);
			
 
				+	emit_dm(as, A64I_MOVx, dest, key);
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_hrefk(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRIns *kslot = IR(ir->op2);
			
 
				+  IRIns *irkey = IR(kslot->op1);
			
 
				+  int32_t ofs = (int32_t)(kslot->op2 * sizeof(Node));
			
 
				+  int32_t kofs = ofs + (int32_t)offsetof(Node, key);
			
 
				+  int bigofs = !emit_checkofs(A64I_LDRx, ofs);
			
 
				+  Reg dest = (ra_used(ir) || bigofs) ? ra_dest(as, ir, RSET_GPR) : RID_NONE;
			
 
				+  Reg node = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+  Reg key, idx = node;
			
 
				+  RegSet allow = rset_exclude(RSET_GPR, node);
			
 
				+  uint64_t k;
			
 
				+  lua_assert(ofs % sizeof(Node) == 0);
			
 
				+  if (bigofs) {
			
 
				+    idx = dest;
			
 
				+    rset_clear(allow, dest);
			
 
				+    kofs = (int32_t)offsetof(Node, key);
			
 
				+  } else if (ra_hasreg(dest)) {
			
 
				+    emit_opk(as, A64I_ADDx, dest, node, ofs, allow);
			
 
				+  }
			
 
				+  asm_guardcc(as, CC_NE);
			
 
				+  if (irt_ispri(irkey->t)) {
			
 
				+    k = ~((int64_t)~irt_toitype(irkey->t) << 47);
			
 
				+  } else if (irt_isnum(irkey->t)) {
			
 
				+    k = ir_knum(irkey)->u64;
			
 
				+  } else {
			
 
				+    k = ((uint64_t)irt_toitype(irkey->t) << 47) | (uint64_t)ir_kgc(irkey);
			
 
				+  }
			
 
				+  key = ra_scratch(as, allow);
			
 
				+  emit_nm(as, A64I_CMPx, key, ra_allock(as, k, rset_exclude(allow, key)));
			
 
				+  emit_lso(as, A64I_LDRx, key, idx, kofs);
			
 
				+  if (bigofs)
			
 
				+    emit_opk(as, A64I_ADDx, dest, node, ofs, RSET_GPR);
			
 
				+}
			
 
				+
			
 
				+static void asm_uref(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  if (irref_isk(ir->op1)) {
			
 
				+    GCfunc *fn = ir_kfunc(IR(ir->op1));
			
 
				+    MRef *v = &gcref(fn->l.uvptr[(ir->op2 >> 8)])->uv.v;
			
 
				+    emit_lsptr(as, A64I_LDRx, dest, v);
			
 
				+  } else {
			
 
				+    Reg uv = ra_scratch(as, RSET_GPR);
			
 
				+    Reg func = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+    if (ir->o == IR_UREFC) {
			
 
				+      asm_guardcc(as, CC_NE);
			
 
				+      emit_n(as, (A64I_CMPx^A64I_K12) | A64F_U12(1), RID_TMP);
			
 
				+      emit_opk(as, A64I_ADDx, dest, uv,
			
 
				+	       (int32_t)offsetof(GCupval, tv), RSET_GPR);
			
 
				+      emit_lso(as, A64I_LDRB, RID_TMP, uv, (int32_t)offsetof(GCupval, closed));
			
 
				+    } else {
			
 
				+      emit_lso(as, A64I_LDRx, dest, uv, (int32_t)offsetof(GCupval, v));
			
 
				+    }
			
 
				+    emit_lso(as, A64I_LDRx, uv, func,
			
 
				+	     (int32_t)offsetof(GCfuncL, uvptr) + 8*(int32_t)(ir->op2 >> 8));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_fref(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  UNUSED(as); UNUSED(ir);
			
 
				+  lua_assert(!ra_used(ir));
			
 
				+}
			
 
				+
			
 
				+static void asm_strref(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  RegSet allow = RSET_GPR;
			
 
				+  Reg dest = ra_dest(as, ir, allow);
			
 
				+  Reg base = ra_alloc1(as, ir->op1, allow);
			
 
				+  IRIns *irr = IR(ir->op2);
			
 
				+  int32_t ofs = sizeof(GCstr);
			
 
				+  uint32_t m;
			
 
				+  rset_clear(allow, base);
			
 
				+  if (irref_isk(ir->op2) && (m = emit_isk12(ofs + irr->i))) {
			
 
				+    emit_dn(as, A64I_ADDx^m, dest, base);
			
 
				+  } else {
			
 
				+    emit_dn(as, (A64I_ADDx^A64I_K12) | A64F_U12(ofs), dest, dest);
			
 
				+    emit_dnm(as, A64I_ADDx, dest, base, ra_alloc1(as, ir->op2, allow));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* -- Loads and stores ---------------------------------------------------- */
			
 
				+
			
 
				+static A64Ins asm_fxloadins(IRIns *ir)
			
 
				+{
			
 
				+  switch (irt_type(ir->t)) {
			
 
				+  case IRT_I8: return A64I_LDRB ^ A64I_LS_S;
			
 
				+  case IRT_U8: return A64I_LDRB;
			
 
				+  case IRT_I16: return A64I_LDRH ^ A64I_LS_S;
			
 
				+  case IRT_U16: return A64I_LDRH;
			
 
				+  case IRT_NUM: return A64I_LDRd;
			
 
				+  case IRT_FLOAT: return A64I_LDRs;
			
 
				+  default: return irt_is64(ir->t) ? A64I_LDRx : A64I_LDRw;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static A64Ins asm_fxstoreins(IRIns *ir)
			
 
				+{
			
 
				+  switch (irt_type(ir->t)) {
			
 
				+  case IRT_I8: case IRT_U8: return A64I_STRB;
			
 
				+  case IRT_I16: case IRT_U16: return A64I_STRH;
			
 
				+  case IRT_NUM: return A64I_STRd;
			
 
				+  case IRT_FLOAT: return A64I_STRs;
			
 
				+  default: return irt_is64(ir->t) ? A64I_STRx : A64I_STRw;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_fload(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg idx;
			
 
				+  A64Ins ai = asm_fxloadins(ir);
			
 
				+  int32_t ofs;
			
 
				+  if (ir->op1 == REF_NIL) {
			
 
				+    idx = RID_GL;
			
 
				+    ofs = (ir->op2 << 2) - GG_OFS(g);
			
 
				+  } else {
			
 
				+    idx = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+    if (ir->op2 == IRFL_TAB_ARRAY) {
			
 
				+      ofs = asm_fuseabase(as, ir->op1);
			
 
				+      if (ofs) {  /* Turn the t->array load into an add for colocated arrays. */
			
 
				+	emit_dn(as, (A64I_ADDx^A64I_K12) | A64F_U12(ofs), dest, idx);
			
 
				+	return;
			
 
				+      }
			
 
				+    }
			
 
				+    ofs = field_ofs[ir->op2];
			
 
				+  }
			
 
				+  emit_lso(as, ai, (dest & 31), idx, ofs);
			
 
				+}
			
 
				+
			
 
				+static void asm_fstore(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (ir->r != RID_SINK) {
			
 
				+    Reg src = ra_alloc1(as, ir->op2, RSET_GPR);
			
 
				+    IRIns *irf = IR(ir->op1);
			
 
				+    Reg idx = ra_alloc1(as, irf->op1, rset_exclude(RSET_GPR, src));
			
 
				+    int32_t ofs = field_ofs[irf->op2];
			
 
				+    emit_lso(as, asm_fxstoreins(ir), (src & 31), idx, ofs);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_xload(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR);
			
 
				+  lua_assert(!(ir->op2 & IRXLOAD_UNALIGNED));
			
 
				+  asm_fusexref(as, asm_fxloadins(ir), dest, ir->op1, RSET_GPR);
			
 
				+}
			
 
				+
			
 
				+static void asm_xstore(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (ir->r != RID_SINK) {
			
 
				+    Reg src = ra_alloc1(as, ir->op2, irt_isfp(ir->t) ? RSET_FPR : RSET_GPR);
			
 
				+    asm_fusexref(as, asm_fxstoreins(ir), src, ir->op1,
			
 
				+		 rset_exclude(RSET_GPR, src));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_ahuvload(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg idx, tmp, type;
			
 
				+  int32_t ofs = 0;
			
 
				+  RegSet gpr = RSET_GPR, allow = irt_isnum(ir->t) ? RSET_FPR : RSET_GPR;
			
 
				+  lua_assert(irt_isnum(ir->t) || irt_ispri(ir->t) || irt_isaddr(ir->t) ||
			
 
				+	     irt_isint(ir->t));
			
 
				+  if (ra_used(ir)) {
			
 
				+    Reg dest = ra_dest(as, ir, allow);
			
 
				+    tmp = irt_isnum(ir->t) ? ra_scratch(as, rset_clear(gpr, dest)) : dest;
			
 
				+    if (irt_isaddr(ir->t)) {
			
 
				+      emit_dn(as, A64I_ANDx^emit_isk13(LJ_GCVMASK, 1), dest, dest);
			
 
				+    } else if (irt_isnum(ir->t)) {
			
 
				+      emit_dn(as, A64I_FMOV_D_R, (dest & 31), tmp);
			
 
				+    } else if (irt_isint(ir->t)) {
			
 
				+      emit_dm(as, A64I_MOVw, dest, dest);
			
 
				+    }
			
 
				+  } else {
			
 
				+    tmp = ra_scratch(as, gpr);
			
 
				+  }
			
 
				+  type = ra_scratch(as, rset_clear(gpr, tmp));
			
 
				+  idx = asm_fuseahuref(as, ir->op1, &ofs, rset_clear(gpr, type), A64I_LDRx);
			
 
				+  /* Always do the type check, even if the load result is unused. */
			
 
				+  asm_guardcc(as, irt_isnum(ir->t) ? CC_LS : CC_NE);
			
 
				+  if (irt_type(ir->t) >= IRT_NUM) {
			
 
				+    lua_assert(irt_isinteger(ir->t) || irt_isnum(ir->t));
			
 
				+    emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
			
 
				+	    ra_allock(as, LJ_TISNUM << 15, rset_exclude(gpr, idx)), tmp);
			
 
				+  } else if (irt_isaddr(ir->t)) {
			
 
				+    emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(ir->t)), type);
			
 
				+    emit_dn(as, A64I_ASRx | A64F_IMMR(47), type, tmp);
			
 
				+  } else if (irt_isnil(ir->t)) {
			
 
				+    emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(1), tmp);
			
 
				+  } else {
			
 
				+    emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
			
 
				+	    ra_allock(as, (irt_toitype(ir->t) << 15) | 0x7fff, allow), tmp);
			
 
				+  }
			
 
				+  if (ofs & FUSE_REG)
			
 
				+    emit_dnm(as, (A64I_LDRx^A64I_LS_R)|A64I_LS_UXTWx|A64I_LS_SH, tmp, idx, (ofs & 31));
			
 
				+  else
			
 
				+    emit_lso(as, A64I_LDRx, tmp, idx, ofs);
			
 
				+}
			
 
				+
			
 
				+static void asm_ahustore(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (ir->r != RID_SINK) {
			
 
				+    RegSet allow = RSET_GPR;
			
 
				+    Reg idx, src = RID_NONE, tmp = RID_TMP, type = RID_NONE;
			
 
				+    int32_t ofs = 0;
			
 
				+    if (irt_isnum(ir->t)) {
			
 
				+      src = ra_alloc1(as, ir->op2, RSET_FPR);
			
 
				+      idx = asm_fuseahuref(as, ir->op1, &ofs, allow, A64I_STRd);
			
 
				+      if (ofs & FUSE_REG)
			
 
				+	emit_dnm(as, (A64I_STRd^A64I_LS_R)|A64I_LS_UXTWx|A64I_LS_SH, (src & 31), idx, (ofs &31));
			
 
				+      else
			
 
				+	emit_lso(as, A64I_STRd, (src & 31), idx, ofs);
			
 
				+    } else {
			
 
				+      if (!irt_ispri(ir->t)) {
			
 
				+	src = ra_alloc1(as, ir->op2, allow);
			
 
				+	rset_clear(allow, src);
			
 
				+	if (irt_isinteger(ir->t))
			
 
				+	  type = ra_allock(as, (uint64_t)(int32_t)LJ_TISNUM << 47, allow);
			
 
				+	else
			
 
				+	  type = ra_allock(as, irt_toitype(ir->t), allow);
			
 
				+      } else {
			
 
				+	tmp = type = ra_allock(as, ~((int64_t)~irt_toitype(ir->t)<<47), allow);
			
 
				+      }
			
 
				+      idx = asm_fuseahuref(as, ir->op1, &ofs, rset_exclude(allow, type),
			
 
				+			   A64I_STRx);
			
 
				+      if (ofs & FUSE_REG)
			
 
				+	emit_dnm(as, (A64I_STRx^A64I_LS_R)|A64I_LS_UXTWx|A64I_LS_SH, tmp, idx, (ofs & 31));
			
 
				+      else
			
 
				+	emit_lso(as, A64I_STRx, tmp, idx, ofs);
			
 
				+      if (ra_hasreg(src)) {
			
 
				+	if (irt_isinteger(ir->t)) {
			
 
				+	  emit_dnm(as, A64I_ADDx | A64F_EX(A64EX_UXTW), tmp, type, src);
			
 
				+	} else {
			
 
				+	  emit_dnm(as, A64I_ADDx | A64F_SH(A64SH_LSL, 47), tmp, src, type);
			
 
				+	}
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_sload(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  int32_t ofs = 8*((int32_t)ir->op1-2);
			
 
				+  IRType1 t = ir->t;
			
 
				+  Reg dest = RID_NONE, base;
			
 
				+  RegSet allow = RSET_GPR;
			
 
				+  lua_assert(!(ir->op2 & IRSLOAD_PARENT));  /* Handled by asm_head_side(). */
			
 
				+  lua_assert(irt_isguard(t) || !(ir->op2 & IRSLOAD_TYPECHECK));
			
 
				+  if ((ir->op2 & IRSLOAD_CONVERT) && irt_isguard(t) && irt_isint(t)) {
			
 
				+    dest = ra_scratch(as, RSET_FPR);
			
 
				+    asm_tointg(as, ir, dest);
			
 
				+    t.irt = IRT_NUM;  /* Continue with a regular number type check. */
			
 
				+  } else if (ra_used(ir)) {
			
 
				+    Reg tmp = RID_NONE;
			
 
				+    if ((ir->op2 & IRSLOAD_CONVERT))
			
 
				+      tmp = ra_scratch(as, irt_isint(t) ? RSET_FPR : RSET_GPR);
			
 
				+    lua_assert((irt_isnum(t)) || irt_isint(t) || irt_isaddr(t));
			
 
				+    dest = ra_dest(as, ir, irt_isnum(t) ? RSET_FPR : allow);
			
 
				+    base = ra_alloc1(as, REF_BASE, rset_clear(allow, dest));
			
 
				+    if (irt_isaddr(t)) {
			
 
				+      emit_dn(as, A64I_ANDx^emit_isk13(LJ_GCVMASK, 1), dest, dest);
			
 
				+    } else if ((ir->op2 & IRSLOAD_CONVERT)) {
			
 
				+      if (irt_isint(t)) {
			
 
				+	emit_dn(as, A64I_FCVT_S32_F64, dest, (tmp & 31));
			
 
				+	/* If value is already loaded for type check, move it to FPR. */
			
 
				+	if ((ir->op2 & IRSLOAD_TYPECHECK))
			
 
				+	  emit_dn(as, A64I_FMOV_D_R, (tmp & 31), dest);
			
 
				+	else
			
 
				+	  dest = tmp;
			
 
				+	t.irt = IRT_NUM;  /* Check for original type. */
			
 
				+      } else {
			
 
				+	emit_dn(as, A64I_FCVT_F64_S32, (dest & 31), tmp);
			
 
				+	dest = tmp;
			
 
				+	t.irt = IRT_INT;  /* Check for original type. */
			
 
				+      }
			
 
				+    } else if (irt_isint(t) && (ir->op2 & IRSLOAD_TYPECHECK)) {
			
 
				+      emit_dm(as, A64I_MOVw, dest, dest);
			
 
				+    }
			
 
				+    goto dotypecheck;
			
 
				+  }
			
 
				+  base = ra_alloc1(as, REF_BASE, allow);
			
 
				+dotypecheck:
			
 
				+  rset_clear(allow, base);
			
 
				+  if ((ir->op2 & IRSLOAD_TYPECHECK)) {
			
 
				+    Reg tmp;
			
 
				+    if (ra_hasreg(dest) && rset_test(RSET_GPR, dest)) {
			
 
				+      tmp = dest;
			
 
				+    } else {
			
 
				+      tmp = ra_scratch(as, allow);
			
 
				+      rset_clear(allow, tmp);
			
 
				+    }
			
 
				+    if (irt_isnum(t) && !(ir->op2 & IRSLOAD_CONVERT))
			
 
				+      emit_dn(as, A64I_FMOV_D_R, (dest & 31), tmp);
			
 
				+    /* Need type check, even if the load result is unused. */
			
 
				+    asm_guardcc(as, irt_isnum(t) ? CC_LS : CC_NE);
			
 
				+    if (irt_type(t) >= IRT_NUM) {
			
 
				+      lua_assert(irt_isinteger(t) || irt_isnum(t));
			
 
				+      emit_nm(as, A64I_CMPx | A64F_SH(A64SH_LSR, 32),
			
 
				+	      ra_allock(as, LJ_TISNUM << 15, allow), tmp);
			
 
				+    } else if (irt_isnil(t)) {
			
 
				+      emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(1), tmp);
			
 
				+    } else if (irt_ispri(t)) {
			
 
				+      emit_nm(as, A64I_CMPx,
			
 
				+	      ra_allock(as, ~((int64_t)~irt_toitype(t) << 47) , allow), tmp);
			
 
				+    } else {
			
 
				+      Reg type = ra_scratch(as, allow);
			
 
				+      emit_n(as, (A64I_CMNx^A64I_K12) | A64F_U12(-irt_toitype(t)), type);
			
 
				+      emit_dn(as, A64I_ASRx | A64F_IMMR(47), type, tmp);
			
 
				+    }
			
 
				+    emit_lso(as, A64I_LDRx, tmp, base, ofs);
			
 
				+    return;
			
 
				+  }
			
 
				+  if (ra_hasreg(dest)) {
			
 
				+    emit_lso(as, irt_isnum(t) ? A64I_LDRd :
			
 
				+	     (irt_isint(t) ? A64I_LDRw : A64I_LDRx), (dest & 31), base,
			
 
				+	     ofs ^ ((LJ_BE && irt_isint(t) ? 4 : 0)));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* -- Allocations --------------------------------------------------------- */
			
 
				+
			
 
				+#if LJ_HASFFI
			
 
				+static void asm_cnew(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  CTState *cts = ctype_ctsG(J2G(as->J));
			
 
				+  CTypeID id = (CTypeID)IR(ir->op1)->i;
			
 
				+  CTSize sz;
			
 
				+  CTInfo info = lj_ctype_info(cts, id, &sz);
			
 
				+  const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_mem_newgco];
			
 
				+  IRRef args[4];
			
 
				+  RegSet allow = (RSET_GPR & ~RSET_SCRATCH);
			
 
				+  lua_assert(sz != CTSIZE_INVALID || (ir->o == IR_CNEW && ir->op2 != REF_NIL));
			
 
				+
			
 
				+  as->gcsteps++;
			
 
				+  asm_setupresult(as, ir, ci);  /* GCcdata * */
			
 
				+  /* Initialize immutable cdata object. */
			
 
				+  if (ir->o == IR_CNEWI) {
			
 
				+    int32_t ofs = sizeof(GCcdata);
			
 
				+    Reg r = ra_alloc1(as, ir->op2, allow);
			
 
				+    lua_assert(sz == 4 || sz == 8);
			
 
				+    emit_lso(as, sz == 8 ? A64I_STRx : A64I_STRw, r, RID_RET, ofs);
			
 
				+  } else if (ir->op2 != REF_NIL) {  /* Create VLA/VLS/aligned cdata. */
			
 
				+    ci = &lj_ir_callinfo[IRCALL_lj_cdata_newv];
			
 
				+    args[0] = ASMREF_L;     /* lua_State *L */
			
 
				+    args[1] = ir->op1;      /* CTypeID id   */
			
 
				+    args[2] = ir->op2;      /* CTSize sz    */
			
 
				+    args[3] = ASMREF_TMP1;  /* CTSize align */
			
 
				+    asm_gencall(as, ci, args);
			
 
				+    emit_loadi(as, ra_releasetmp(as, ASMREF_TMP1), (int32_t)ctype_align(info));
			
 
				+    return;
			
 
				+  }
			
 
				+
			
 
				+  /* Initialize gct and ctypeid. lj_mem_newgco() already sets marked. */
			
 
				+  {
			
 
				+    Reg r = (id < 65536) ? RID_X1 : ra_allock(as, id, allow);
			
 
				+    emit_lso(as, A64I_STRB, RID_TMP, RID_RET, offsetof(GCcdata, gct));
			
 
				+    emit_lso(as, A64I_STRH, r, RID_RET, offsetof(GCcdata, ctypeid));
			
 
				+    emit_d(as, A64I_MOVZw | A64F_U16(~LJ_TCDATA), RID_TMP);
			
 
				+    if (id < 65536) emit_d(as, A64I_MOVZw | A64F_U16(id), RID_X1);
			
 
				+  }
			
 
				+  args[0] = ASMREF_L;     /* lua_State *L */
			
 
				+  args[1] = ASMREF_TMP1;  /* MSize size   */
			
 
				+  asm_gencall(as, ci, args);
			
 
				+  ra_allockreg(as, (int32_t)(sz+sizeof(GCcdata)),
			
 
				+	       ra_releasetmp(as, ASMREF_TMP1));
			
 
				+}
			
 
				+#else
			
 
				+#define asm_cnew(as, ir)	((void)0)
			
 
				+#endif
			
 
				+
			
 
				+/* -- Write barriers ------------------------------------------------------ */
			
 
				+
			
 
				+static void asm_tbar(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg tab = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+  Reg link = ra_scratch(as, rset_exclude(RSET_GPR, tab));
			
 
				+  Reg gr = ra_allock(as, i64ptr(J2G(as->J)),
			
 
				+		     rset_exclude(rset_exclude(RSET_GPR, tab), link));
			
 
				+  Reg mark = RID_TMP;
			
 
				+  MCLabel l_end = emit_label(as);
			
 
				+  emit_lso(as, A64I_STRx, link, tab, (int32_t)offsetof(GCtab, gclist));
			
 
				+  emit_lso(as, A64I_STRB, mark, tab, (int32_t)offsetof(GCtab, marked));
			
 
				+  emit_lso(as, A64I_STRx, tab, gr,
			
 
				+	   (int32_t)offsetof(global_State, gc.grayagain));
			
 
				+  emit_dn(as, A64I_ANDw^emit_isk13(~LJ_GC_BLACK, 0), mark, mark);
			
 
				+  emit_lso(as, A64I_LDRx, link, gr,
			
 
				+	   (int32_t)offsetof(global_State, gc.grayagain));
			
 
				+  emit_cond_branch(as, CC_EQ, l_end);
			
 
				+  emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_BLACK, 0), mark);
			
 
				+  emit_lso(as, A64I_LDRB, mark, tab, (int32_t)offsetof(GCtab, marked));
			
 
				+}
			
 
				+
			
 
				+static void asm_obar(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_gc_barrieruv];
			
 
				+  IRRef args[2];
			
 
				+  MCLabel l_end;
			
 
				+  RegSet allow = RSET_GPR;
			
 
				+  Reg obj, val, tmp;
			
 
				+  /* No need for other object barriers (yet). */
			
 
				+  lua_assert(IR(ir->op1)->o == IR_UREFC);
			
 
				+  ra_evictset(as, RSET_SCRATCH);
			
 
				+  l_end = emit_label(as);
			
 
				+  args[0] = ASMREF_TMP1;  /* global_State *g */
			
 
				+  args[1] = ir->op1;      /* TValue *tv      */
			
 
				+  asm_gencall(as, ci, args);
			
 
				+  ra_allockreg(as, i64ptr(J2G(as->J)), ra_releasetmp(as, ASMREF_TMP1) );
			
 
				+  obj = IR(ir->op1)->r;
			
 
				+  tmp = ra_scratch(as, rset_exclude(allow, obj));
			
 
				+  emit_cond_branch(as, CC_EQ, l_end);
			
 
				+  emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_BLACK, 0), tmp);
			
 
				+  emit_cond_branch(as, CC_EQ, l_end);
			
 
				+  emit_n(as, A64I_TSTw^emit_isk13(LJ_GC_WHITES, 0), RID_TMP);
			
 
				+  val = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, obj));
			
 
				+  emit_lso(as, A64I_LDRB, tmp, obj,
			
 
				+     (int32_t)offsetof(GCupval, marked)-(int32_t)offsetof(GCupval, tv));
			
 
				+  emit_lso(as, A64I_LDRB, RID_TMP, val, (int32_t)offsetof(GChead, marked));
			
 
				+}
			
 
				+
			
 
				+/* -- Arithmetic and logic operations ------------------------------------- */
			
 
				+
			
 
				+static void asm_fparith(ASMState *as, IRIns *ir, A64Ins ai)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_FPR);
			
 
				+  Reg right, left = ra_alloc2(as, ir, RSET_FPR);
			
 
				+  right = (left >> 8); left &= 255;
			
 
				+  emit_dnm(as, ai, (dest & 31), (left & 31), (right & 31));
			
 
				+}
			
 
				+
			
 
				+static void asm_fpunary(ASMState *as, IRIns *ir, A64Ins ai)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_FPR);
			
 
				+  Reg left = ra_hintalloc(as, ir->op1, dest, RSET_FPR);
			
 
				+  emit_dn(as, ai, (dest & 31), (left & 31));
			
 
				+}
			
 
				+
			
 
				+static void asm_fpmath(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  IRFPMathOp fpm = (IRFPMathOp)ir->op2;
			
 
				+  if (fpm == IRFPM_SQRT) {
			
 
				+    asm_fpunary(as, ir, A64I_FSQRTd);
			
 
				+  } else if (fpm <= IRFPM_TRUNC) {
			
 
				+    asm_fpunary(as, ir, fpm == IRFPM_FLOOR ? A64I_FRINTMd :
			
 
				+			fpm == IRFPM_CEIL ? A64I_FRINTPd : A64I_FRINTZd);
			
 
				+  } else if (fpm == IRFPM_EXP2 && asm_fpjoin_pow(as, ir)) {
			
 
				+    return;
			
 
				+  } else {
			
 
				+    asm_callid(as, ir, IRCALL_lj_vm_floor + fpm);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static int asm_swapops(ASMState *as, IRRef lref, IRRef rref)
			
 
				+{
			
 
				+  IRIns *ir;
			
 
				+  if (irref_isk(rref))
			
 
				+    return 0;  /* Don't swap constants to the left. */
			
 
				+  if (irref_isk(lref))
			
 
				+    return 1;  /* But swap constants to the right. */
			
 
				+  ir = IR(rref);
			
 
				+  if ((ir->o >= IR_BSHL && ir->o <= IR_BSAR) ||
			
 
				+      (ir->o == IR_ADD && ir->op1 == ir->op2) ||
			
 
				+      (ir->o == IR_CONV && ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)))
			
 
				+    return 0;  /* Don't swap fusable operands to the left. */
			
 
				+  ir = IR(lref);
			
 
				+  if ((ir->o >= IR_BSHL && ir->o <= IR_BSAR) ||
			
 
				+      (ir->o == IR_ADD && ir->op1 == ir->op2) ||
			
 
				+      (ir->o == IR_CONV && ir->op2 == ((IRT_I64<<IRCONV_DSH)|IRT_INT|IRCONV_SEXT)))
			
 
				+    return 1;  /* But swap fusable operands to the right. */
			
 
				+  return 0;  /* Otherwise don't swap. */
			
 
				+}
			
 
				+
			
 
				+static void asm_intop(ASMState *as, IRIns *ir, A64Ins ai)
			
 
				+{
			
 
				+  IRRef lref = ir->op1, rref = ir->op2;
			
 
				+  Reg left, dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  uint32_t m;
			
 
				+  if ((ai & ~A64I_S) != A64I_SUBw && asm_swapops(as, lref, rref)) {
			
 
				+    IRRef tmp = lref; lref = rref; rref = tmp;
			
 
				+  }
			
 
				+  left = ra_hintalloc(as, lref, dest, RSET_GPR);
			
 
				+  if (irt_is64(ir->t)) ai |= A64I_X;
			
 
				+  m = asm_fuseopm(as, ai, rref, rset_exclude(RSET_GPR, left));
			
 
				+  if (irt_isguard(ir->t)) {  /* For IR_ADDOV etc. */
			
 
				+    asm_guardcc(as, CC_VS);
			
 
				+    ai |= A64I_S;
			
 
				+  }
			
 
				+  emit_dn(as, ai^m, dest, left);
			
 
				+}
			
 
				+
			
 
				+static void asm_intop_s(ASMState *as, IRIns *ir, A64Ins ai)
			
 
				+{
			
 
				+  if (as->flagmcp == as->mcp) {  /* Drop cmp r, #0. */
			
 
				+    as->flagmcp = NULL;
			
 
				+    as->mcp++;
			
 
				+    ai |= A64I_S;
			
 
				+  }
			
 
				+  asm_intop(as, ir, ai);
			
 
				+}
			
 
				+
			
 
				+static void asm_intneg(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
			
 
				+  emit_dm(as, irt_is64(ir->t) ? A64I_NEGx : A64I_NEGw, dest, left);
			
 
				+}
			
 
				+
			
 
				+/* NYI: use add/shift for MUL(OV) with constants. FOLD only does 2^k. */
			
 
				+static void asm_intmul(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg left = ra_alloc1(as, ir->op1, rset_exclude(RSET_GPR, dest));
			
 
				+  Reg right = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, left));
			
 
				+  if (irt_isguard(ir->t)) {  /* IR_MULOV */
			
 
				+    asm_guardcc(as, CC_NE);
			
 
				+    emit_dm(as, A64I_MOVw, dest, dest);  /* Zero-extend. */
			
 
				+    emit_nm(as, A64I_CMPw | A64F_SH(A64SH_ASR, 31), RID_TMP, dest);
			
 
				+    emit_dn(as, A64I_ASRx | A64F_IMMR(32), RID_TMP, dest);
			
 
				+    emit_dnm(as, A64I_SMULL, dest, right, left);
			
 
				+  } else {
			
 
				+    emit_dnm(as, irt_is64(ir->t) ? A64I_MULx : A64I_MULw, dest, left, right);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_add(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t)) {
			
 
				+    if (!asm_fusemadd(as, ir, A64I_FMADDd, A64I_FMADDd))
			
 
				+      asm_fparith(as, ir, A64I_FADDd);
			
 
				+    return;
			
 
				+  }
			
 
				+  asm_intop_s(as, ir, A64I_ADDw);
			
 
				+}
			
 
				+
			
 
				+static void asm_sub(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t)) {
			
 
				+    if (!asm_fusemadd(as, ir, A64I_FNMSUBd, A64I_FMSUBd))
			
 
				+      asm_fparith(as, ir, A64I_FSUBd);
			
 
				+    return;
			
 
				+  }
			
 
				+  asm_intop_s(as, ir, A64I_SUBw);
			
 
				+}
			
 
				+
			
 
				+static void asm_mul(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t)) {
			
 
				+    asm_fparith(as, ir, A64I_FMULd);
			
 
				+    return;
			
 
				+  }
			
 
				+  asm_intmul(as, ir);
			
 
				+}
			
 
				+
			
 
				+static void asm_div(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  if (!irt_isnum(ir->t))
			
 
				+    asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_divi64 :
			
 
				+					  IRCALL_lj_carith_divu64);
			
 
				+  else
			
 
				+#endif
			
 
				+    asm_fparith(as, ir, A64I_FDIVd);
			
 
				+}
			
 
				+
			
 
				+static void asm_pow(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  if (!irt_isnum(ir->t))
			
 
				+    asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_powi64 :
			
 
				+					  IRCALL_lj_carith_powu64);
			
 
				+  else
			
 
				+#endif
			
 
				+    asm_callid(as, ir, IRCALL_lj_vm_powi);
			
 
				+}
			
 
				+
			
 
				+#define asm_addov(as, ir)	asm_add(as, ir)
			
 
				+#define asm_subov(as, ir)	asm_sub(as, ir)
			
 
				+#define asm_mulov(as, ir)	asm_mul(as, ir)
			
 
				+
			
 
				+#define asm_abs(as, ir)		asm_fpunary(as, ir, A64I_FABS)
			
 
				+#define asm_atan2(as, ir)	asm_callid(as, ir, IRCALL_atan2)
			
 
				+#define asm_ldexp(as, ir)	asm_callid(as, ir, IRCALL_ldexp)
			
 
				+
			
 
				+static void asm_mod(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+#if LJ_HASFFI
			
 
				+  if (!irt_isint(ir->t))
			
 
				+    asm_callid(as, ir, irt_isi64(ir->t) ? IRCALL_lj_carith_modi64 :
			
 
				+					  IRCALL_lj_carith_modu64);
			
 
				+  else
			
 
				+#endif
			
 
				+    asm_callid(as, ir, IRCALL_lj_vm_modi);
			
 
				+}
			
 
				+
			
 
				+static void asm_neg(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t)) {
			
 
				+    asm_fpunary(as, ir, A64I_FNEGd);
			
 
				+    return;
			
 
				+  }
			
 
				+  asm_intneg(as, ir);
			
 
				+}
			
 
				+
			
 
				+static void asm_band(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  A64Ins ai = A64I_ANDw;
			
 
				+  if (asm_fuseandshift(as, ir))
			
 
				+    return;
			
 
				+  if (as->flagmcp == as->mcp) {
			
 
				+    /* Try to drop cmp r, #0. */
			
 
				+    as->flagmcp = NULL;
			
 
				+    as->mcp++;
			
 
				+    ai = A64I_ANDSw;
			
 
				+  }
			
 
				+  asm_intop(as, ir, ai);
			
 
				+}
			
 
				+
			
 
				+static void asm_borbxor(ASMState *as, IRIns *ir, A64Ins ai)
			
 
				+{
			
 
				+  IRRef lref = ir->op1, rref = ir->op2;
			
 
				+  IRIns *irl = IR(lref), *irr = IR(rref);
			
 
				+  if ((canfuse(as, irl) && irl->o == IR_BNOT && !irref_isk(rref)) ||
			
 
				+      (canfuse(as, irr) && irr->o == IR_BNOT && !irref_isk(lref))) {
			
 
				+    Reg left, dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    uint32_t m;
			
 
				+    if (irl->o == IR_BNOT) {
			
 
				+      IRRef tmp = lref; lref = rref; rref = tmp;
			
 
				+    }
			
 
				+    left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+    ai |= A64I_ON;
			
 
				+    if (irt_is64(ir->t)) ai |= A64I_X;
			
 
				+    m = asm_fuseopm(as, ai, IR(rref)->op1, rset_exclude(RSET_GPR, left));
			
 
				+    emit_dn(as, ai^m, dest, left);
			
 
				+  } else {
			
 
				+    asm_intop(as, ir, ai);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+static void asm_bor(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (asm_fuseorshift(as, ir))
			
 
				+    return;
			
 
				+  asm_borbxor(as, ir, A64I_ORRw);
			
 
				+}
			
 
				+
			
 
				+#define asm_bxor(as, ir)	asm_borbxor(as, ir, A64I_EORw)
			
 
				+
			
 
				+static void asm_bnot(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  A64Ins ai = A64I_MVNw;
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  uint32_t m = asm_fuseopm(as, ai, ir->op1, RSET_GPR);
			
 
				+  if (irt_is64(ir->t)) ai |= A64I_X;
			
 
				+  emit_d(as, ai^m, dest);
			
 
				+}
			
 
				+
			
 
				+static void asm_bswap(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+  emit_dn(as, irt_is64(ir->t) ? A64I_REVx : A64I_REVw, dest, left);
			
 
				+}
			
 
				+
			
 
				+static void asm_bitshift(ASMState *as, IRIns *ir, A64Ins ai, A64Shift sh)
			
 
				+{
			
 
				+  int32_t shmask = irt_is64(ir->t) ? 63 : 31;
			
 
				+  if (irref_isk(ir->op2)) {  /* Constant shifts. */
			
 
				+    Reg left, dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    int32_t shift = (IR(ir->op2)->i & shmask);
			
 
				+    IRIns *irl = IR(ir->op1);
			
 
				+    if (shmask == 63) ai += A64I_UBFMx - A64I_UBFMw;
			
 
				+
			
 
				+    /* Fuse BSHL + BSHR/BSAR into UBFM/SBFM aka UBFX/SBFX/UBFIZ/SBFIZ. */
			
 
				+    if ((sh == A64SH_LSR || sh == A64SH_ASR) && canfuse(as, irl)) {
			
 
				+      if (irl->o == IR_BSHL && irref_isk(irl->op2)) {
			
 
				+	int32_t shift2 = (IR(irl->op2)->i & shmask);
			
 
				+	shift = ((shift - shift2) & shmask);
			
 
				+	shmask -= shift2;
			
 
				+	ir = irl;
			
 
				+      }
			
 
				+    }
			
 
				+
			
 
				+    left = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+    switch (sh) {
			
 
				+    case A64SH_LSL:
			
 
				+      emit_dn(as, ai | A64F_IMMS(shmask-shift) |
			
 
				+		  A64F_IMMR((shmask-shift+1)&shmask), dest, left);
			
 
				+      break;
			
 
				+    case A64SH_LSR: case A64SH_ASR:
			
 
				+      emit_dn(as, ai | A64F_IMMS(shmask) | A64F_IMMR(shift), dest, left);
			
 
				+      break;
			
 
				+    case A64SH_ROR:
			
 
				+      emit_dnm(as, ai | A64F_IMMS(shift), dest, left, left);
			
 
				+      break;
			
 
				+    }
			
 
				+  } else {  /* Variable-length shifts. */
			
 
				+    Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+    Reg left = ra_alloc1(as, ir->op1, RSET_GPR);
			
 
				+    Reg right = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, left));
			
 
				+    emit_dnm(as, (shmask == 63 ? A64I_SHRx : A64I_SHRw) | A64F_BSH(sh), dest, left, right);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+#define asm_bshl(as, ir)	asm_bitshift(as, ir, A64I_UBFMw, A64SH_LSL)
			
 
				+#define asm_bshr(as, ir)	asm_bitshift(as, ir, A64I_UBFMw, A64SH_LSR)
			
 
				+#define asm_bsar(as, ir)	asm_bitshift(as, ir, A64I_SBFMw, A64SH_ASR)
			
 
				+#define asm_bror(as, ir)	asm_bitshift(as, ir, A64I_EXTRw, A64SH_ROR)
			
 
				+#define asm_brol(as, ir)	lua_assert(0)
			
 
				+
			
 
				+static void asm_intmin_max(ASMState *as, IRIns *ir, A64CC cc)
			
 
				+{
			
 
				+  Reg dest = ra_dest(as, ir, RSET_GPR);
			
 
				+  Reg left = ra_hintalloc(as, ir->op1, dest, RSET_GPR);
			
 
				+  Reg right = ra_alloc1(as, ir->op2, rset_exclude(RSET_GPR, left));
			
 
				+  emit_dnm(as, A64I_CSELw|A64F_CC(cc), dest, left, right);
			
 
				+  emit_nm(as, A64I_CMPw, left, right);
			
 
				+}
			
 
				+
			
 
				+static void asm_fpmin_max(ASMState *as, IRIns *ir, A64CC fcc)
			
 
				+{
			
 
				+  Reg dest = (ra_dest(as, ir, RSET_FPR) & 31);
			
 
				+  Reg right, left = ra_alloc2(as, ir, RSET_FPR);
			
 
				+  right = ((left >> 8) & 31); left &= 31;
			
 
				+  emit_dnm(as, A64I_FCSELd | A64F_CC(fcc), dest, left, right);
			
 
				+  emit_nm(as, A64I_FCMPd, left, right);
			
 
				+}
			
 
				+
			
 
				+static void asm_min_max(ASMState *as, IRIns *ir, A64CC cc, A64CC fcc)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t))
			
 
				+    asm_fpmin_max(as, ir, fcc);
			
 
				+  else
			
 
				+    asm_intmin_max(as, ir, cc);
			
 
				+}
			
 
				+
			
 
				+#define asm_max(as, ir)		asm_min_max(as, ir, CC_GT, CC_HI)
			
 
				+#define asm_min(as, ir)		asm_min_max(as, ir, CC_LT, CC_LO)
			
 
				+
			
 
				+/* -- Comparisons --------------------------------------------------------- */
			
 
				+
			
 
				+/* Map of comparisons to flags. ORDER IR. */
			
 
				+static const uint8_t asm_compmap[IR_ABC+1] = {
			
 
				+  /* op  FP swp  int cc   FP cc */
			
 
				+  /* LT       */ CC_GE + (CC_HS << 4),
			
 
				+  /* GE    x  */ CC_LT + (CC_HI << 4),
			
 
				+  /* LE       */ CC_GT + (CC_HI << 4),
			
 
				+  /* GT    x  */ CC_LE + (CC_HS << 4),
			
 
				+  /* ULT   x  */ CC_HS + (CC_LS << 4),
			
 
				+  /* UGE      */ CC_LO + (CC_LO << 4),
			
 
				+  /* ULE   x  */ CC_HI + (CC_LO << 4),
			
 
				+  /* UGT      */ CC_LS + (CC_LS << 4),
			
 
				+  /* EQ       */ CC_NE + (CC_NE << 4),
			
 
				+  /* NE       */ CC_EQ + (CC_EQ << 4),
			
 
				+  /* ABC      */ CC_LS + (CC_LS << 4)  /* Same as UGT. */
			
 
				+};
			
 
				+
			
 
				+/* FP comparisons. */
			
 
				+static void asm_fpcomp(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  Reg left, right;
			
 
				+  A64Ins ai;
			
 
				+  int swp = ((ir->o ^ (ir->o >> 2)) & ~(ir->o >> 3) & 1);
			
 
				+  if (!swp && irref_isk(ir->op2) && ir_knum(IR(ir->op2))->u64 == 0) {
			
 
				+    left = (ra_alloc1(as, ir->op1, RSET_FPR) & 31);
			
 
				+    right = 0;
			
 
				+    ai = A64I_FCMPZd;
			
 
				+  } else {
			
 
				+    left = ra_alloc2(as, ir, RSET_FPR);
			
 
				+    if (swp) {
			
 
				+      right = (left & 31); left = ((left >> 8) & 31);
			
 
				+    } else {
			
 
				+      right = ((left >> 8) & 31); left &= 31;
			
 
				+    }
			
 
				+    ai = A64I_FCMPd;
			
 
				+  }
			
 
				+  asm_guardcc(as, (asm_compmap[ir->o] >> 4));
			
 
				+  emit_nm(as, ai, left, right);
			
 
				+}
			
 
				+
			
 
				+/* Integer comparisons. */
			
 
				+static void asm_intcomp(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  A64CC oldcc, cc = (asm_compmap[ir->o] & 15);
			
 
				+  A64Ins ai = irt_is64(ir->t) ? A64I_CMPx : A64I_CMPw;
			
 
				+  IRRef lref = ir->op1, rref = ir->op2;
			
 
				+  Reg left;
			
 
				+  uint32_t m;
			
 
				+  int cmpprev0 = 0;
			
 
				+  lua_assert(irt_is64(ir->t) || irt_isint(ir->t) ||
			
 
				+	     irt_isu32(ir->t) || irt_isaddr(ir->t) || irt_isu8(ir->t));
			
 
				+  if (asm_swapops(as, lref, rref)) {
			
 
				+    IRRef tmp = lref; lref = rref; rref = tmp;
			
 
				+    if (cc >= CC_GE) cc ^= 7;  /* LT <-> GT, LE <-> GE */
			
 
				+    else if (cc > CC_NE) cc ^= 11;  /* LO <-> HI, LS <-> HS */
			
 
				+  }
			
 
				+  oldcc = cc;
			
 
				+  if (irref_isk(rref) && get_k64val(IR(rref)) == 0) {
			
 
				+    IRIns *irl = IR(lref);
			
 
				+    if (cc == CC_GE) cc = CC_PL;
			
 
				+    else if (cc == CC_LT) cc = CC_MI;
			
 
				+    else if (cc > CC_NE) goto nocombine;  /* Other conds don't work with tst. */
			
 
				+    cmpprev0 = (irl+1 == ir);
			
 
				+    /* Combine and-cmp-bcc into tbz/tbnz or and-cmp into tst. */
			
 
				+    if (cmpprev0 && irl->o == IR_BAND && !ra_used(irl)) {
			
 
				+      IRRef blref = irl->op1, brref = irl->op2;
			
 
				+      uint32_t m2 = 0;
			
 
				+      Reg bleft;
			
 
				+      if (asm_swapops(as, blref, brref)) {
			
 
				+	Reg tmp = blref; blref = brref; brref = tmp;
			
 
				+      }
			
 
				+      if (irref_isk(brref)) {
			
 
				+	uint64_t k = get_k64val(IR(brref));
			
 
				+	if (k && !(k & (k-1)) && (cc == CC_EQ || cc == CC_NE)) {
			
 
				+	  asm_guardtnb(as, cc == CC_EQ ? A64I_TBZ : A64I_TBNZ,
			
 
				+		       ra_alloc1(as, blref, RSET_GPR), emit_ctz64(k));
			
 
				+	  return;
			
 
				+	}
			
 
				+	m2 = emit_isk13(k, irt_is64(irl->t));
			
 
				+      }
			
 
				+      bleft = ra_alloc1(as, blref, RSET_GPR);
			
 
				+      ai = (irt_is64(irl->t) ? A64I_TSTx : A64I_TSTw);
			
 
				+      if (!m2)
			
 
				+	m2 = asm_fuseopm(as, ai, brref, rset_exclude(RSET_GPR, bleft));
			
 
				+      asm_guardcc(as, cc);
			
 
				+      emit_n(as, ai^m2, bleft);
			
 
				+      return;
			
 
				+    }
			
 
				+    if (cc == CC_EQ || cc == CC_NE) {
			
 
				+      /* Combine cmp-bcc into cbz/cbnz. */
			
 
				+      ai = cc == CC_EQ ? A64I_CBZ : A64I_CBNZ;
			
 
				+      if (irt_is64(ir->t)) ai |= A64I_X;
			
 
				+      asm_guardcnb(as, ai, ra_alloc1(as, lref, RSET_GPR));
			
 
				+      return;
			
 
				+    }
			
 
				+  }
			
 
				+nocombine:
			
 
				+  left = ra_alloc1(as, lref, RSET_GPR);
			
 
				+  m = asm_fuseopm(as, ai, rref, rset_exclude(RSET_GPR, left));
			
 
				+  asm_guardcc(as, cc);
			
 
				+  emit_n(as, ai^m, left);
			
 
				+  /* Signed comparison with zero and referencing previous ins? */
			
 
				+  if (cmpprev0 && (oldcc <= CC_NE || oldcc >= CC_GE))
			
 
				+    as->flagmcp = as->mcp;  /* Allow elimination of the compare. */
			
 
				+}
			
 
				+
			
 
				+static void asm_comp(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  if (irt_isnum(ir->t))
			
 
				+    asm_fpcomp(as, ir);
			
 
				+  else
			
 
				+    asm_intcomp(as, ir);
			
 
				+}
			
 
				+
			
 
				+#define asm_equal(as, ir)	asm_comp(as, ir)
			
 
				+
			
 
				+/* -- Support for 64 bit ops in 32 bit mode ------------------------------- */
			
 
				+
			
 
				+/* Hiword op of a split 64 bit op. Previous op must be the loword op. */
			
 
				+static void asm_hiop(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  UNUSED(as); UNUSED(ir); lua_assert(0);  /* Unused on 64 bit. */
			
 
				+}
			
 
				+
			
 
				+/* -- Profiling ----------------------------------------------------------- */
			
 
				+
			
 
				+static void asm_prof(ASMState *as, IRIns *ir)
			
 
				+{
			
 
				+  uint32_t k = emit_isk13(HOOK_PROFILE, 0);
			
 
				+  lua_assert(k != 0);
			
 
				+  UNUSED(ir);
			
 
				+  asm_guardcc(as, CC_NE);
			
 
				+  emit_n(as, A64I_TSTw^k, RID_TMP);
			
 
				+  emit_lsptr(as, A64I_LDRB, RID_TMP, (void *)&J2G(as->J)->hookmask);
			
 
				+}
			
 
				+
			
 
				+/* -- Stack handling ------------------------------------------------------ */
			
 
				+
			
 
				+/* Check Lua stack size for overflow. Use exit handler as fallback. */
			
 
				+static void asm_stack_check(ASMState *as, BCReg topslot,
			
 
				+			    IRIns *irp, RegSet allow, ExitNo exitno)
			
 
				+{
			
 
				+  Reg pbase;
			
 
				+  uint32_t k;
			
 
				+  if (irp) {
			
 
				+    if (!ra_hasspill(irp->s)) {
			
 
				+      pbase = irp->r;
			
 
				+      lua_assert(ra_hasreg(pbase));
			
 
				+    } else if (allow) {
			
 
				+      pbase = rset_pickbot(allow);
			
 
				+    } else {
			
 
				+      pbase = RID_RET;
			
 
				+      emit_lso(as, A64I_LDRx, RID_RET, RID_SP, 0);  /* Restore temp register. */
			
 
				+    }
			
 
				+  } else {
			
 
				+    pbase = RID_BASE;
			
 
				+  }
			
 
				+  emit_cond_branch(as, CC_LS, asm_exitstub_addr(as, exitno));
			
 
				+  k = emit_isk12((8*topslot));
			
 
				+  lua_assert(k);
			
 
				+  emit_n(as, A64I_CMPx^k, RID_TMP);
			
 
				+  emit_dnm(as, A64I_SUBx, RID_TMP, RID_TMP, pbase);
			
 
				+  emit_lso(as, A64I_LDRx, RID_TMP, RID_TMP,
			
 
				+	   (int32_t)offsetof(lua_State, maxstack));
			
 
				+  if (irp) {  /* Must not spill arbitrary registers in head of side trace. */
			
 
				+    if (ra_hasspill(irp->s))
			
 
				+      emit_lso(as, A64I_LDRx, pbase, RID_SP, sps_scale(irp->s));
			
 
				+    emit_lso(as, A64I_LDRx, RID_TMP, RID_GL, glofs(as, &J2G(as->J)->cur_L));
			
 
				+    if (ra_hasspill(irp->s) && !allow)
			
 
				+      emit_lso(as, A64I_STRx, RID_RET, RID_SP, 0);  /* Save temp register. */
			
 
				+  } else {
			
 
				+    emit_getgl(as, RID_TMP, cur_L);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* Restore Lua stack from on-trace state. */
			
 
				+static void asm_stack_restore(ASMState *as, SnapShot *snap)
			
 
				+{
			
 
				+  SnapEntry *map = &as->T->snapmap[snap->mapofs];
			
 
				+#ifdef LUA_USE_ASSERT
			
 
				+  SnapEntry *flinks = &as->T->snapmap[snap_nextofs(as->T, snap)-1-LJ_FR2];
			
 
				+#endif
			
 
				+  MSize n, nent = snap->nent;
			
 
				+  /* Store the value of all modified slots to the Lua stack. */
			
 
				+  for (n = 0; n < nent; n++) {
			
 
				+    SnapEntry sn = map[n];
			
 
				+    BCReg s = snap_slot(sn);
			
 
				+    int32_t ofs = 8*((int32_t)s-1-LJ_FR2);
			
 
				+    IRRef ref = snap_ref(sn);
			
 
				+    IRIns *ir = IR(ref);
			
 
				+    if ((sn & SNAP_NORESTORE))
			
 
				+      continue;
			
 
				+    if (irt_isnum(ir->t)) {
			
 
				+      Reg src = ra_alloc1(as, ref, RSET_FPR);
			
 
				+      emit_lso(as, A64I_STRd, (src & 31), RID_BASE, ofs);
			
 
				+    } else {
			
 
				+      asm_tvstore64(as, RID_BASE, ofs, ref);
			
 
				+    }
			
 
				+    checkmclim(as);
			
 
				+  }
			
 
				+  lua_assert(map + nent == flinks);
			
 
				+}
			
 
				+
			
 
				+/* -- GC handling --------------------------------------------------------- */
			
 
				+
			
 
				+/* Check GC threshold and do one or more GC steps. */
			
 
				+static void asm_gc_check(ASMState *as)
			
 
				+{
			
 
				+  const CCallInfo *ci = &lj_ir_callinfo[IRCALL_lj_gc_step_jit];
			
 
				+  IRRef args[2];
			
 
				+  MCLabel l_end;
			
 
				+  Reg tmp1, tmp2;
			
 
				+  ra_evictset(as, RSET_SCRATCH);
			
 
				+  l_end = emit_label(as);
			
 
				+  /* Exit trace if in GCSatomic or GCSfinalize. Avoids syncing GC objects. */
			
 
				+  asm_guardcnb(as, A64I_CBNZ, RID_RET); /* Assumes asm_snap_prep() is done. */
			
 
				+  args[0] = ASMREF_TMP1;  /* global_State *g */
			
 
				+  args[1] = ASMREF_TMP2;  /* MSize steps     */
			
 
				+  asm_gencall(as, ci, args);
			
 
				+  tmp1 = ra_releasetmp(as, ASMREF_TMP1);
			
 
				+  tmp2 = ra_releasetmp(as, ASMREF_TMP2);
			
 
				+  emit_loadi(as, tmp2, as->gcsteps);
			
 
				+  /* Jump around GC step if GC total < GC threshold. */
			
 
				+  emit_cond_branch(as, CC_LS, l_end);
			
 
				+  emit_nm(as, A64I_CMPx, RID_TMP, tmp2);
			
 
				+  emit_lso(as, A64I_LDRx, tmp2, tmp1,
			
 
				+	   (int32_t)offsetof(global_State, gc.threshold));
			
 
				+  emit_lso(as, A64I_LDRx, RID_TMP, tmp1,
			
 
				+	   (int32_t)offsetof(global_State, gc.total));
			
 
				+  ra_allockreg(as, i64ptr(J2G(as->J)), tmp1);
			
 
				+  as->gcsteps = 0;
			
 
				+  checkmclim(as);
			
 
				+}
			
 
				+
			
 
				+/* -- Loop handling ------------------------------------------------------- */
			
 
				+
			
 
				+/* Fixup the loop branch. */
			
 
				+static void asm_loop_fixup(ASMState *as)
			
 
				+{
			
 
				+  MCode *p = as->mctop;
			
 
				+  MCode *target = as->mcp;
			
 
				+  if (as->loopinv) {  /* Inverted loop branch? */
			
 
				+    uint32_t mask = (p[-2] & 0x7e000000) == 0x36000000 ? 0x3fffu : 0x7ffffu;
			
 
				+    ptrdiff_t delta = target - (p - 2);
			
 
				+    /* asm_guard* already inverted the bcc/tnb/cnb and patched the final b. */
			
 
				+    p[-2] |= ((uint32_t)delta & mask) << 5;
			
 
				+  } else {
			
 
				+    ptrdiff_t delta = target - (p - 1);
			
 
				+    p[-1] = A64I_B | A64F_S26(delta);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* -- Head of trace ------------------------------------------------------- */
			
 
				+
			
 
				+/* Reload L register from g->cur_L. */
			
 
				+static void asm_head_lreg(ASMState *as)
			
 
				+{
			
 
				+  IRIns *ir = IR(ASMREF_L);
			
 
				+  if (ra_used(ir)) {
			
 
				+    Reg r = ra_dest(as, ir, RSET_GPR);
			
 
				+    emit_getgl(as, r, cur_L);
			
 
				+    ra_evictk(as);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+/* Coalesce BASE register for a root trace. */
			
 
				+static void asm_head_root_base(ASMState *as)
			
 
				+{
			
 
				+  IRIns *ir;
			
 
				+  asm_head_lreg(as);
			
 
				+  ir = IR(REF_BASE);
			
 
				+  if (ra_hasreg(ir->r) && (rset_test(as->modset, ir->r) || irt_ismarked(ir->t)))
			
 
				+    ra_spill(as, ir);
			
 
				+  ra_destreg(as, ir, RID_BASE);
			
 
				+}
			
 
				+
			
 
				+/* Coalesce BASE register for a side trace. */
			
 
				+static RegSet asm_head_side_base(ASMState *as, IRIns *irp, RegSet allow)
			
 
				+{
			
 
				+  IRIns *ir;
			
 
				+  asm_head_lreg(as);
			
 
				+  ir = IR(REF_BASE);
			
 
				+  if (ra_hasreg(ir->r) && (rset_test(as->modset, ir->r) || irt_ismarked(ir->t)))
			
 
				+    ra_spill(as, ir);
			
 
				+  if (ra_hasspill(irp->s)) {
			
 
				+    rset_clear(allow, ra_dest(as, ir, allow));
			
 
				+  } else {
			
 
				+    Reg r = irp->r;
			
 
				+    lua_assert(ra_hasreg(r));
			
 
				+    rset_clear(allow, r);
			
 
				+    if (r != ir->r && !rset_test(as->freeset, r))
			
 
				+      ra_restore(as, regcost_ref(as->cost[r]));
			
 
				+    ra_destreg(as, ir, r);
			
 
				+  }
			
 
				+  return allow;
			
 
				+}
			
 
				+
			
 
				+/* -- Tail of trace ------------------------------------------------------- */
			
 
				+
			
 
				+/* Fixup the tail code. */
			
 
				+static void asm_tail_fixup(ASMState *as, TraceNo lnk)
			
 
				+{
			
 
				+  MCode *p = as->mctop;
			
 
				+  MCode *target;
			
 
				+  /* Undo the sp adjustment in BC_JLOOP when exiting to the interpreter. */
			
 
				+  int32_t spadj = as->T->spadjust + (lnk ? 0 : sps_scale(SPS_FIXED));
			
 
				+  if (spadj == 0) {
			
 
				+    *--p = A64I_LE(A64I_NOP);
			
 
				+    as->mctop = p;
			
 
				+  } else {
			
 
				+    /* Patch stack adjustment. */
			
 
				+    uint32_t k = emit_isk12(spadj);
			
 
				+    lua_assert(k);
			
 
				+    p[-2] = (A64I_ADDx^k) | A64F_D(RID_SP) | A64F_N(RID_SP);
			
 
				+  }
			
 
				+  /* Patch exit branch. */
			
 
				+  target = lnk ? traceref(as->J, lnk)->mcode : (MCode *)lj_vm_exit_interp;
			
 
				+  p[-1] = A64I_B | A64F_S26((target-p)+1);
			
 
				+}
			
 
				+
			
 
				+/* Prepare tail of code. */
			
 
				+static void asm_tail_prep(ASMState *as)
			
 
				+{
			
 
				+  MCode *p = as->mctop - 1;  /* Leave room for exit branch. */
			
 
				+  if (as->loopref) {
			
 
				+    as->invmcp = as->mcp = p;
			
 
				+  } else {
			
 
				+    as->mcp = p-1;  /* Leave room for stack pointer adjustment. */
			
 
				+    as->invmcp = NULL;
			
 
				+  }
			
 
				+  *p = 0;  /* Prevent load/store merging. */
			
 
				+}
			
 
				+
			
 
				+/* -- Trace setup --------------------------------------------------------- */
			
 
				+
			
 
				+/* Ensure there are enough stack slots for call arguments. */
			
 
				+static Reg asm_setup_call_slots(ASMState *as, IRIns *ir, const CCallInfo *ci)
			
 
				+{
			
 
				+  IRRef args[CCI_NARGS_MAX*2];
			
 
				+  uint32_t i, nargs = CCI_XNARGS(ci);
			
 
				+  int nslots = 0, ngpr = REGARG_NUMGPR, nfpr = REGARG_NUMFPR;
			
 
				+  asm_collectargs(as, ir, ci, args);
			
 
				+  for (i = 0; i < nargs; i++) {
			
 
				+    if (args[i] && irt_isfp(IR(args[i])->t)) {
			
 
				+      if (nfpr > 0) nfpr--; else nslots += 2;
			
 
				+    } else {
			
 
				+      if (ngpr > 0) ngpr--; else nslots += 2;
			
 
				+    }
			
 
				+  }
			
 
				+  if (nslots > as->evenspill)  /* Leave room for args in stack slots. */
			
 
				+    as->evenspill = nslots;
			
 
				+  return REGSP_HINT(RID_RET);
			
 
				+}
			
 
				+
			
 
				+static void asm_setup_target(ASMState *as)
			
 
				+{
			
 
				+  /* May need extra exit for asm_stack_check on side traces. */
			
 
				+  asm_exitstub_setup(as, as->T->nsnap + (as->parent ? 1 : 0));
			
 
				+}
			
 
				+
			
 
				+#if LJ_BE
			
 
				+/* ARM64 instructions are always little-endian. Swap for ARM64BE. */
			
 
				+static void asm_mcode_fixup(MCode *mcode, MSize size)
			
 
				+{
			
 
				+  MCode *pe = (MCode *)((char *)mcode + size);
			
 
				+  while (mcode < pe) {
			
 
				+    MCode ins = *mcode;
			
 
				+    *mcode++ = lj_bswap(ins);
			
 
				+  }
			
 
				+}
			
 
				+#define LJ_TARGET_MCODE_FIXUP	1
			
 
				+#endif
			
 
				+
			
 
				+/* -- Trace patching ------------------------------------------------------ */
			
 
				+
			
 
				+/* Patch exit jumps of existing machine code to a new target. */
			
 
				+void lj_asm_patchexit(jit_State *J, GCtrace *T, ExitNo exitno, MCode *target)
			
 
				+{
			
 
				+  MCode *p = T->mcode;
			
 
				+  MCode *pe = (MCode *)((char *)p + T->szmcode);
			
 
				+  MCode *cstart = NULL;
			
 
				+  MCode *mcarea = lj_mcode_patch(J, p, 0);
			
 
				+  MCode *px = exitstub_trace_addr(T, exitno);
			
 
				+  /* Note: this assumes a trace exit is only ever patched once. */
			
 
				+  for (; p < pe; p++) {
			
 
				+    /* Look for exitstub branch, replace with branch to target. */
			
 
				+    ptrdiff_t delta = target - p;
			
 
				+    MCode ins = A64I_LE(*p);
			
 
				+    if ((ins & 0xff000000u) == 0x54000000u &&
			
 
				+	((ins ^ ((px-p)<<5)) & 0x00ffffe0u) == 0) {
			
 
				+      /* Patch bcc, if within range. */
			
 
				+      if (A64F_S_OK(delta, 19)) {
			
 
				+	*p = A64I_LE((ins & 0xff00001fu) | A64F_S19(delta));
			
 
				+	if (!cstart) cstart = p;
			
 
				+      }
			
 
				+    } else if ((ins & 0xfc000000u) == 0x14000000u &&
			
 
				+	       ((ins ^ (px-p)) & 0x03ffffffu) == 0) {
			
 
				+      /* Patch b. */
			
 
				+      lua_assert(A64F_S_OK(delta, 26));
			
 
				+      *p = A64I_LE((ins & 0xfc000000u) | A64F_S26(delta));
			
 
				+      if (!cstart) cstart = p;
			
 
				+    } else if ((ins & 0x7e000000u) == 0x34000000u &&
			
 
				+	       ((ins ^ ((px-p)<<5)) & 0x00ffffe0u) == 0) {
			
 
				+      /* Patch cbz/cbnz, if within range. */
			
 
				+      if (A64F_S_OK(delta, 19)) {
			
 
				+	*p = A64I_LE((ins & 0xff00001fu) | A64F_S19(delta));
			
 
				+	if (!cstart) cstart = p;
			
 
				+      }
			
 
				+    } else if ((ins & 0x7e000000u) == 0x36000000u &&
			
 
				+	       ((ins ^ ((px-p)<<5)) & 0x0007ffe0u) == 0) {
			
 
				+      /* Patch tbz/tbnz, if within range. */
			
 
				+      if (A64F_S_OK(delta, 14)) {
			
 
				+	*p = A64I_LE((ins & 0xfff8001fu) | A64F_S14(delta));
			
 
				+	if (!cstart) cstart = p;
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  {  /* Always patch long-range branch in exit stub itself. */
			
 
				+    ptrdiff_t delta = target - px;
			
 
				+    lua_assert(A64F_S_OK(delta, 26));
			
 
				+    *px = A64I_B | A64F_S26(delta);
			
 
				+    if (!cstart) cstart = px;
			
 
				+  }
			
 
				+  lj_mcode_sync(cstart, px+1);
			
 
				+  lj_mcode_patch(J, mcarea, 1);
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/lj_asm_mips.h
+++ b/luajit.mod/luajit/src/lj_asm_mips.h
--- a/luajit.mod/luajit/src/lj_asm_ppc.h
+++ b/luajit.mod/luajit/src/lj_asm_ppc.h
--- a/luajit.mod/luajit/src/lj_asm_x86.h
+++ b/luajit.mod/luajit/src/lj_asm_x86.h
--- a/luajit.mod/luajit/src/lj_bc.h
+++ b/luajit.mod/luajit/src/lj_bc.h
@@ -89,6 +89,8 @@
 
				   _(ISFC,	dst,	___,	var,	___) \
			
 
				   _(IST,	___,	___,	var,	___) \
			
 
				   _(ISF,	___,	___,	var,	___) \
			
 
				+  _(ISTYPE,	var,	___,	lit,	___) \
			
 
				+  _(ISNUM,	var,	___,	lit,	___) \
			
 
				   \
			
 
				   /* Unary ops. */ \
			
 
				   _(MOV,	dst,	___,	var,	___) \
			
@@ -143,10 +145,12 @@
 
				   _(TGETV,	dst,	var,	var,	index) \
			
 
				   _(TGETS,	dst,	var,	str,	index) \
			
 
				   _(TGETB,	dst,	var,	lit,	index) \
			
 
				+  _(TGETR,	dst,	var,	var,	index) \
			
 
				   _(TSETV,	var,	var,	var,	newindex) \
			
 
				   _(TSETS,	var,	var,	str,	newindex) \
			
 
				   _(TSETB,	var,	var,	lit,	newindex) \
			
 
				   _(TSETM,	base,	___,	num,	newindex) \
			
 
				+  _(TSETR,	var,	var,	var,	newindex) \
			
 
				   \
			
 
				   /* Calls and vararg handling. T = tail call. */ \
			
 
				   _(CALLM,	base,	lit,	lit,	call) \
			
--- a/luajit.mod/luajit/src/lj_bcdump.h
+++ b/luajit.mod/luajit/src/lj_bcdump.h
@@ -36,14 +36,15 @@
 
				 /* If you perform *any* kind of private modifications to the bytecode itself
			
 
				 ** or to the dump format, you *must* set BCDUMP_VERSION to 0x80 or higher.
			
 
				 */
			
 
				-#define BCDUMP_VERSION		1
			
 
				+#define BCDUMP_VERSION		2
			
 
				 
			
 
				 /* Compatibility flags. */
			
 
				 #define BCDUMP_F_BE		0x01
			
 
				 #define BCDUMP_F_STRIP		0x02
			
 
				 #define BCDUMP_F_FFI		0x04
			
 
				+#define BCDUMP_F_FR2		0x08
			
 
				 
			
 
				-#define BCDUMP_F_KNOWN		(BCDUMP_F_FFI*2-1)
			
 
				+#define BCDUMP_F_KNOWN		(BCDUMP_F_FR2*2-1)
			
 
				 
			
 
				 /* Type codes for the GC constants of a prototype. Plus length for strings. */
			
 
				 enum {
			
@@ -61,6 +62,7 @@ enum {
 
				 
			
 
				 LJ_FUNC int lj_bcwrite(lua_State *L, GCproto *pt, lua_Writer writer,
			
 
				 		       void *data, int strip);
			
 
				+LJ_FUNC GCproto *lj_bcread_proto(LexState *ls);
			
 
				 LJ_FUNC GCproto *lj_bcread(LexState *ls);
			
 
				 
			
 
				 #endif
			
--- a/luajit.mod/luajit/src/lj_bcread.c
+++ b/luajit.mod/luajit/src/lj_bcread.c
@@ -9,6 +9,7 @@
 
				 #include "lj_obj.h"
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				 #include "lj_str.h"
			
 
				 #include "lj_tab.h"
			
 
				 #include "lj_bc.h"
			
@@ -20,6 +21,7 @@
 
				 #include "lj_lex.h"
			
 
				 #include "lj_bcdump.h"
			
 
				 #include "lj_state.h"
			
 
				+#include "lj_strfmt.h"
			
 
				 
			
 
				 /* Reuse some lexer fields for our own purposes. */
			
 
				 #define bcread_flags(ls)	ls->level
			
@@ -38,84 +40,73 @@ static LJ_NOINLINE void bcread_error(LexState *ls, ErrMsg em)
 
				   const char *name = ls->chunkarg;
			
 
				   if (*name == BCDUMP_HEAD1) name = "(binary)";
			
 
				   else if (*name == '@' || *name == '=') name++;
			
 
				-  lj_str_pushf(L, "%s: %s", name, err2msg(em));
			
 
				+  lj_strfmt_pushf(L, "%s: %s", name, err2msg(em));
			
 
				   lj_err_throw(L, LUA_ERRSYNTAX);
			
 
				 }
			
 
				 
			
 
				-/* Resize input buffer. */
			
 
				-static void bcread_resize(LexState *ls, MSize len)
			
 
				-{
			
 
				-  if (ls->sb.sz < len) {
			
 
				-    MSize sz = ls->sb.sz * 2;
			
 
				-    while (len > sz) sz = sz * 2;
			
 
				-    lj_str_resizebuf(ls->L, &ls->sb, sz);
			
 
				-    /* Caveat: this may change ls->sb.buf which may affect ls->p. */
			
 
				-  }
			
 
				-}
			
 
				-
			
 
				-/* Refill buffer if needed. */
			
 
				+/* Refill buffer. */
			
 
				 static LJ_NOINLINE void bcread_fill(LexState *ls, MSize len, int need)
			
 
				 {
			
 
				   lua_assert(len != 0);
			
 
				-  if (len > LJ_MAX_MEM || ls->current < 0)
			
 
				+  if (len > LJ_MAX_BUF || ls->c < 0)
			
 
				     bcread_error(ls, LJ_ERR_BCBAD);
			
 
				   do {
			
 
				     const char *buf;
			
 
				-    size_t size;
			
 
				-    if (ls->n) {  /* Copy remainder to buffer. */
			
 
				-      if (ls->sb.n) {  /* Move down in buffer. */
			
 
				-	lua_assert(ls->p + ls->n == ls->sb.buf + ls->sb.n);
			
 
				-	if (ls->n != ls->sb.n)
			
 
				-	  memmove(ls->sb.buf, ls->p, ls->n);
			
 
				+    size_t sz;
			
 
				+    char *p = sbufB(&ls->sb);
			
 
				+    MSize n = (MSize)(ls->pe - ls->p);
			
 
				+    if (n) {  /* Copy remainder to buffer. */
			
 
				+      if (sbuflen(&ls->sb)) {  /* Move down in buffer. */
			
 
				+	lua_assert(ls->pe == sbufP(&ls->sb));
			
 
				+	if (ls->p != p) memmove(p, ls->p, n);
			
 
				       } else {  /* Copy from buffer provided by reader. */
			
 
				-	bcread_resize(ls, len);
			
 
				-	memcpy(ls->sb.buf, ls->p, ls->n);
			
 
				+	p = lj_buf_need(&ls->sb, len);
			
 
				+	memcpy(p, ls->p, n);
			
 
				       }
			
 
				-      ls->p = ls->sb.buf;
			
 
				+      ls->p = p;
			
 
				+      ls->pe = p + n;
			
 
				     }
			
 
				-    ls->sb.n = ls->n;
			
 
				-    buf = ls->rfunc(ls->L, ls->rdata, &size);  /* Get more data from reader. */
			
 
				-    if (buf == NULL || size == 0) {  /* EOF? */
			
 
				+    setsbufP(&ls->sb, p + n);
			
 
				+    buf = ls->rfunc(ls->L, ls->rdata, &sz);  /* Get more data from reader. */
			
 
				+    if (buf == NULL || sz == 0) {  /* EOF? */
			
 
				       if (need) bcread_error(ls, LJ_ERR_BCBAD);
			
 
				-      ls->current = -1;  /* Only bad if we get called again. */
			
 
				+      ls->c = -1;  /* Only bad if we get called again. */
			
 
				       break;
			
 
				     }
			
 
				-    if (ls->sb.n) {  /* Append to buffer. */
			
 
				-      MSize n = ls->sb.n + (MSize)size;
			
 
				-      bcread_resize(ls, n < len ? len : n);
			
 
				-      memcpy(ls->sb.buf + ls->sb.n, buf, size);
			
 
				-      ls->n = ls->sb.n = n;
			
 
				-      ls->p = ls->sb.buf;
			
 
				+    if (n) {  /* Append to buffer. */
			
 
				+      n += (MSize)sz;
			
 
				+      p = lj_buf_need(&ls->sb, n < len ? len : n);
			
 
				+      memcpy(sbufP(&ls->sb), buf, sz);
			
 
				+      setsbufP(&ls->sb, p + n);
			
 
				+      ls->p = p;
			
 
				+      ls->pe = p + n;
			
 
				     } else {  /* Return buffer provided by reader. */
			
 
				-      ls->n = (MSize)size;
			
 
				       ls->p = buf;
			
 
				+      ls->pe = buf + sz;
			
 
				     }
			
 
				-  } while (ls->n < len);
			
 
				+  } while (ls->p + len > ls->pe);
			
 
				 }
			
 
				 
			
 
				 /* Need a certain number of bytes. */
			
 
				 static LJ_AINLINE void bcread_need(LexState *ls, MSize len)
			
 
				 {
			
 
				-  if (LJ_UNLIKELY(ls->n < len))
			
 
				+  if (LJ_UNLIKELY(ls->p + len > ls->pe))
			
 
				     bcread_fill(ls, len, 1);
			
 
				 }
			
 
				 
			
 
				 /* Want to read up to a certain number of bytes, but may need less. */
			
 
				 static LJ_AINLINE void bcread_want(LexState *ls, MSize len)
			
 
				 {
			
 
				-  if (LJ_UNLIKELY(ls->n < len))
			
 
				+  if (LJ_UNLIKELY(ls->p + len > ls->pe))
			
 
				     bcread_fill(ls, len, 0);
			
 
				 }
			
 
				 
			
 
				-#define bcread_dec(ls)		check_exp(ls->n > 0, ls->n--)
			
 
				-#define bcread_consume(ls, len)	check_exp(ls->n >= (len), ls->n -= (len))
			
 
				-
			
 
				 /* Return memory block from buffer. */
			
 
				-static uint8_t *bcread_mem(LexState *ls, MSize len)
			
 
				+static LJ_AINLINE uint8_t *bcread_mem(LexState *ls, MSize len)
			
 
				 {
			
 
				   uint8_t *p = (uint8_t *)ls->p;
			
 
				-  bcread_consume(ls, len);
			
 
				-  ls->p = (char *)p + len;
			
 
				+  ls->p += len;
			
 
				+  lua_assert(ls->p <= ls->pe);
			
 
				   return p;
			
 
				 }
			
 
				 
			
@@ -128,25 +119,15 @@ static void bcread_block(LexState *ls, void *q, MSize len)
 
				 /* Read byte from buffer. */
			
 
				 static LJ_AINLINE uint32_t bcread_byte(LexState *ls)
			
 
				 {
			
 
				-  bcread_dec(ls);
			
 
				+  lua_assert(ls->p < ls->pe);
			
 
				   return (uint32_t)(uint8_t)*ls->p++;
			
 
				 }
			
 
				 
			
 
				 /* Read ULEB128 value from buffer. */
			
 
				-static uint32_t bcread_uleb128(LexState *ls)
			
 
				+static LJ_AINLINE uint32_t bcread_uleb128(LexState *ls)
			
 
				 {
			
 
				-  const uint8_t *p = (const uint8_t *)ls->p;
			
 
				-  uint32_t v = *p++;
			
 
				-  if (LJ_UNLIKELY(v >= 0x80)) {
			
 
				-    int sh = 0;
			
 
				-    v &= 0x7f;
			
 
				-    do {
			
 
				-     v |= ((*p & 0x7f) << (sh += 7));
			
 
				-     bcread_dec(ls);
			
 
				-   } while (*p++ >= 0x80);
			
 
				-  }
			
 
				-  bcread_dec(ls);
			
 
				-  ls->p = (char *)p;
			
 
				+  uint32_t v = lj_buf_ruleb128(&ls->p);
			
 
				+  lua_assert(ls->p <= ls->pe);
			
 
				   return v;
			
 
				 }
			
 
				 
			
@@ -160,11 +141,10 @@ static uint32_t bcread_uleb128_33(LexState *ls)
 
				     v &= 0x3f;
			
 
				     do {
			
 
				      v |= ((*p & 0x7f) << (sh += 7));
			
 
				-     bcread_dec(ls);
			
 
				    } while (*p++ >= 0x80);
			
 
				   }
			
 
				-  bcread_dec(ls);
			
 
				   ls->p = (char *)p;
			
 
				+  lua_assert(ls->p <= ls->pe);
			
 
				   return v;
			
 
				 }
			
 
				 
			
@@ -212,7 +192,7 @@ static void bcread_ktabk(LexState *ls, TValue *o)
 
				     o->u32.hi = bcread_uleb128(ls);
			
 
				   } else {
			
 
				     lua_assert(tp <= BCDUMP_KTAB_TRUE);
			
 
				-    setitype(o, ~tp);
			
 
				+    setpriV(o, ~tp);
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -326,25 +306,13 @@ static void bcread_uv(LexState *ls, GCproto *pt, MSize sizeuv)
 
				 }
			
 
				 
			
 
				 /* Read a prototype. */
			
 
				-static GCproto *bcread_proto(LexState *ls)
			
 
				+GCproto *lj_bcread_proto(LexState *ls)
			
 
				 {
			
 
				   GCproto *pt;
			
 
				   MSize framesize, numparams, flags, sizeuv, sizekgc, sizekn, sizebc, sizept;
			
 
				   MSize ofsk, ofsuv, ofsdbg;
			
 
				   MSize sizedbg = 0;
			
 
				   BCLine firstline = 0, numline = 0;
			
 
				-  MSize len, startn;
			
 
				-
			
 
				-  /* Read length. */
			
 
				-  if (ls->n > 0 && ls->p[0] == 0) {  /* Shortcut EOF. */
			
 
				-    ls->n--; ls->p++;
			
 
				-    return NULL;
			
 
				-  }
			
 
				-  bcread_want(ls, 5);
			
 
				-  len = bcread_uleb128(ls);
			
 
				-  if (!len) return NULL;  /* EOF */
			
 
				-  bcread_need(ls, len);
			
 
				-  startn = ls->n;
			
 
				 
			
 
				   /* Read prototype header. */
			
 
				   flags = bcread_byte(ls);
			
@@ -413,9 +381,6 @@ static GCproto *bcread_proto(LexState *ls)
 
				     setmref(pt->uvinfo, NULL);
			
 
				     setmref(pt->varinfo, NULL);
			
 
				   }
			
 
				-
			
 
				-  if (len != startn - ls->n)
			
 
				-    bcread_error(ls, LJ_ERR_BCBAD);
			
 
				   return pt;
			
 
				 }
			
 
				 
			
@@ -429,6 +394,7 @@ static int bcread_header(LexState *ls)
 
				       bcread_byte(ls) != BCDUMP_VERSION) return 0;
			
 
				   bcread_flags(ls) = flags = bcread_uleb128(ls);
			
 
				   if ((flags & ~(BCDUMP_F_KNOWN)) != 0) return 0;
			
 
				+  if ((flags & BCDUMP_F_FR2) != LJ_FR2*BCDUMP_F_FR2) return 0;
			
 
				   if ((flags & BCDUMP_F_FFI)) {
			
 
				 #if LJ_HASFFI
			
 
				     lua_State *L = ls->L;
			
@@ -455,19 +421,34 @@ static int bcread_header(LexState *ls)
 
				 GCproto *lj_bcread(LexState *ls)
			
 
				 {
			
 
				   lua_State *L = ls->L;
			
 
				-  lua_assert(ls->current == BCDUMP_HEAD1);
			
 
				+  lua_assert(ls->c == BCDUMP_HEAD1);
			
 
				   bcread_savetop(L, ls, L->top);
			
 
				-  lj_str_resetbuf(&ls->sb);
			
 
				+  lj_buf_reset(&ls->sb);
			
 
				   /* Check for a valid bytecode dump header. */
			
 
				   if (!bcread_header(ls))
			
 
				     bcread_error(ls, LJ_ERR_BCFMT);
			
 
				   for (;;) {  /* Process all prototypes in the bytecode dump. */
			
 
				-    GCproto *pt = bcread_proto(ls);
			
 
				-    if (!pt) break;
			
 
				+    GCproto *pt;
			
 
				+    MSize len;
			
 
				+    const char *startp;
			
 
				+    /* Read length. */
			
 
				+    if (ls->p < ls->pe && ls->p[0] == 0) {  /* Shortcut EOF. */
			
 
				+      ls->p++;
			
 
				+      break;
			
 
				+    }
			
 
				+    bcread_want(ls, 5);
			
 
				+    len = bcread_uleb128(ls);
			
 
				+    if (!len) break;  /* EOF */
			
 
				+    bcread_need(ls, len);
			
 
				+    startp = ls->p;
			
 
				+    pt = lj_bcread_proto(ls);
			
 
				+    if (ls->p != startp + len)
			
 
				+      bcread_error(ls, LJ_ERR_BCBAD);
			
 
				     setprotoV(L, L->top, pt);
			
 
				     incr_top(L);
			
 
				   }
			
 
				-  if ((int32_t)ls->n > 0 || L->top-1 != bcread_oldtop(L, ls))
			
 
				+  if ((int32_t)(2*(uint32_t)(ls->pe - ls->p)) > 0 ||
			
 
				+      L->top-1 != bcread_oldtop(L, ls))
			
 
				     bcread_error(ls, LJ_ERR_BCBAD);
			
 
				   /* Pop off last prototype. */
			
 
				   L->top--;
			
--- a/luajit.mod/luajit/src/lj_bcwrite.c
+++ b/luajit.mod/luajit/src/lj_bcwrite.c
@@ -8,7 +8,7 @@
 
				 
			
 
				 #include "lj_obj.h"
			
 
				 #include "lj_gc.h"
			
 
				-#include "lj_str.h"
			
 
				+#include "lj_buf.h"
			
 
				 #include "lj_bc.h"
			
 
				 #if LJ_HASFFI
			
 
				 #include "lj_ctype.h"
			
@@ -17,13 +17,13 @@
 
				 #include "lj_dispatch.h"
			
 
				 #include "lj_jit.h"
			
 
				 #endif
			
 
				+#include "lj_strfmt.h"
			
 
				 #include "lj_bcdump.h"
			
 
				 #include "lj_vm.h"
			
 
				 
			
 
				 /* Context for bytecode writer. */
			
 
				 typedef struct BCWriteCtx {
			
 
				   SBuf sb;			/* Output buffer. */
			
 
				-  lua_State *L;			/* Lua state. */
			
 
				   GCproto *pt;			/* Root prototype. */
			
 
				   lua_Writer wfunc;		/* Writer callback. */
			
 
				   void *wdata;			/* Writer callback data. */
			
@@ -31,85 +31,44 @@ typedef struct BCWriteCtx {
 
				   int status;			/* Status from writer callback. */
			
 
				 } BCWriteCtx;
			
 
				 
			
 
				-/* -- Output buffer handling ---------------------------------------------- */
			
 
				-
			
 
				-/* Resize buffer if needed. */
			
 
				-static LJ_NOINLINE void bcwrite_resize(BCWriteCtx *ctx, MSize len)
			
 
				-{
			
 
				-  MSize sz = ctx->sb.sz * 2;
			
 
				-  while (ctx->sb.n + len > sz) sz = sz * 2;
			
 
				-  lj_str_resizebuf(ctx->L, &ctx->sb, sz);
			
 
				-}
			
 
				-
			
 
				-/* Need a certain amount of buffer space. */
			
 
				-static LJ_AINLINE void bcwrite_need(BCWriteCtx *ctx, MSize len)
			
 
				-{
			
 
				-  if (LJ_UNLIKELY(ctx->sb.n + len > ctx->sb.sz))
			
 
				-    bcwrite_resize(ctx, len);
			
 
				-}
			
 
				-
			
 
				-/* Add memory block to buffer. */
			
 
				-static void bcwrite_block(BCWriteCtx *ctx, const void *p, MSize len)
			
 
				-{
			
 
				-  uint8_t *q = (uint8_t *)(ctx->sb.buf + ctx->sb.n);
			
 
				-  MSize i;
			
 
				-  ctx->sb.n += len;
			
 
				-  for (i = 0; i < len; i++) q[i] = ((uint8_t *)p)[i];
			
 
				-}
			
 
				-
			
 
				-/* Add byte to buffer. */
			
 
				-static LJ_AINLINE void bcwrite_byte(BCWriteCtx *ctx, uint8_t b)
			
 
				-{
			
 
				-  ctx->sb.buf[ctx->sb.n++] = b;
			
 
				-}
			
 
				-
			
 
				-/* Add ULEB128 value to buffer. */
			
 
				-static void bcwrite_uleb128(BCWriteCtx *ctx, uint32_t v)
			
 
				-{
			
 
				-  MSize n = ctx->sb.n;
			
 
				-  uint8_t *p = (uint8_t *)ctx->sb.buf;
			
 
				-  for (; v >= 0x80; v >>= 7)
			
 
				-    p[n++] = (uint8_t)((v & 0x7f) | 0x80);
			
 
				-  p[n++] = (uint8_t)v;
			
 
				-  ctx->sb.n = n;
			
 
				-}
			
 
				-
			
 
				 /* -- Bytecode writer ----------------------------------------------------- */
			
 
				 
			
 
				 /* Write a single constant key/value of a template table. */
			
 
				 static void bcwrite_ktabk(BCWriteCtx *ctx, cTValue *o, int narrow)
			
 
				 {
			
 
				-  bcwrite_need(ctx, 1+10);
			
 
				+  char *p = lj_buf_more(&ctx->sb, 1+10);
			
 
				   if (tvisstr(o)) {
			
 
				     const GCstr *str = strV(o);
			
 
				     MSize len = str->len;
			
 
				-    bcwrite_need(ctx, 5+len);
			
 
				-    bcwrite_uleb128(ctx, BCDUMP_KTAB_STR+len);
			
 
				-    bcwrite_block(ctx, strdata(str), len);
			
 
				+    p = lj_buf_more(&ctx->sb, 5+len);
			
 
				+    p = lj_strfmt_wuleb128(p, BCDUMP_KTAB_STR+len);
			
 
				+    p = lj_buf_wmem(p, strdata(str), len);
			
 
				   } else if (tvisint(o)) {
			
 
				-    bcwrite_byte(ctx, BCDUMP_KTAB_INT);
			
 
				-    bcwrite_uleb128(ctx, intV(o));
			
 
				+    *p++ = BCDUMP_KTAB_INT;
			
 
				+    p = lj_strfmt_wuleb128(p, intV(o));
			
 
				   } else if (tvisnum(o)) {
			
 
				     if (!LJ_DUALNUM && narrow) {  /* Narrow number constants to integers. */
			
 
				       lua_Number num = numV(o);
			
 
				       int32_t k = lj_num2int(num);
			
 
				       if (num == (lua_Number)k) {  /* -0 is never a constant. */
			
 
				-	bcwrite_byte(ctx, BCDUMP_KTAB_INT);
			
 
				-	bcwrite_uleb128(ctx, k);
			
 
				+	*p++ = BCDUMP_KTAB_INT;
			
 
				+	p = lj_strfmt_wuleb128(p, k);
			
 
				+	setsbufP(&ctx->sb, p);
			
 
				 	return;
			
 
				       }
			
 
				     }
			
 
				-    bcwrite_byte(ctx, BCDUMP_KTAB_NUM);
			
 
				-    bcwrite_uleb128(ctx, o->u32.lo);
			
 
				-    bcwrite_uleb128(ctx, o->u32.hi);
			
 
				+    *p++ = BCDUMP_KTAB_NUM;
			
 
				+    p = lj_strfmt_wuleb128(p, o->u32.lo);
			
 
				+    p = lj_strfmt_wuleb128(p, o->u32.hi);
			
 
				   } else {
			
 
				     lua_assert(tvispri(o));
			
 
				-    bcwrite_byte(ctx, BCDUMP_KTAB_NIL+~itype(o));
			
 
				+    *p++ = BCDUMP_KTAB_NIL+~itype(o);
			
 
				   }
			
 
				+  setsbufP(&ctx->sb, p);
			
 
				 }
			
 
				 
			
 
				 /* Write a template table. */
			
 
				-static void bcwrite_ktab(BCWriteCtx *ctx, const GCtab *t)
			
 
				+static void bcwrite_ktab(BCWriteCtx *ctx, char *p, const GCtab *t)
			
 
				 {
			
 
				   MSize narray = 0, nhash = 0;
			
 
				   if (t->asize > 0) {  /* Determine max. length of array part. */
			
@@ -127,8 +86,9 @@ static void bcwrite_ktab(BCWriteCtx *ctx, const GCtab *t)
 
				       nhash += !tvisnil(&node[i].val);
			
 
				   }
			
 
				   /* Write number of array slots and hash slots. */
			
 
				-  bcwrite_uleb128(ctx, narray);
			
 
				-  bcwrite_uleb128(ctx, nhash);
			
 
				+  p = lj_strfmt_wuleb128(p, narray);
			
 
				+  p = lj_strfmt_wuleb128(p, nhash);
			
 
				+  setsbufP(&ctx->sb, p);
			
 
				   if (narray) {  /* Write array entries (may contain nil). */
			
 
				     MSize i;
			
 
				     TValue *o = tvref(t->array);
			
@@ -155,6 +115,7 @@ static void bcwrite_kgc(BCWriteCtx *ctx, GCproto *pt)
 
				   for (i = 0; i < sizekgc; i++, kr++) {
			
 
				     GCobj *o = gcref(*kr);
			
 
				     MSize tp, need = 1;
			
 
				+    char *p;
			
 
				     /* Determine constant type and needed size. */
			
 
				     if (o->gch.gct == ~LJ_TSTR) {
			
 
				       tp = BCDUMP_KGC_STR + gco2str(o)->len;
			
@@ -181,24 +142,26 @@ static void bcwrite_kgc(BCWriteCtx *ctx, GCproto *pt)
 
				       need = 1+2*5;
			
 
				     }
			
 
				     /* Write constant type. */
			
 
				-    bcwrite_need(ctx, need);
			
 
				-    bcwrite_uleb128(ctx, tp);
			
 
				+    p = lj_buf_more(&ctx->sb, need);
			
 
				+    p = lj_strfmt_wuleb128(p, tp);
			
 
				     /* Write constant data (if any). */
			
 
				     if (tp >= BCDUMP_KGC_STR) {
			
 
				-      bcwrite_block(ctx, strdata(gco2str(o)), gco2str(o)->len);
			
 
				+      p = lj_buf_wmem(p, strdata(gco2str(o)), gco2str(o)->len);
			
 
				     } else if (tp == BCDUMP_KGC_TAB) {
			
 
				-      bcwrite_ktab(ctx, gco2tab(o));
			
 
				+      bcwrite_ktab(ctx, p, gco2tab(o));
			
 
				+      continue;
			
 
				 #if LJ_HASFFI
			
 
				     } else if (tp != BCDUMP_KGC_CHILD) {
			
 
				-      cTValue *p = (TValue *)cdataptr(gco2cd(o));
			
 
				-      bcwrite_uleb128(ctx, p[0].u32.lo);
			
 
				-      bcwrite_uleb128(ctx, p[0].u32.hi);
			
 
				+      cTValue *q = (TValue *)cdataptr(gco2cd(o));
			
 
				+      p = lj_strfmt_wuleb128(p, q[0].u32.lo);
			
 
				+      p = lj_strfmt_wuleb128(p, q[0].u32.hi);
			
 
				       if (tp == BCDUMP_KGC_COMPLEX) {
			
 
				-	bcwrite_uleb128(ctx, p[1].u32.lo);
			
 
				-	bcwrite_uleb128(ctx, p[1].u32.hi);
			
 
				+	p = lj_strfmt_wuleb128(p, q[1].u32.lo);
			
 
				+	p = lj_strfmt_wuleb128(p, q[1].u32.hi);
			
 
				       }
			
 
				 #endif
			
 
				     }
			
 
				+    setsbufP(&ctx->sb, p);
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -207,7 +170,7 @@ static void bcwrite_knum(BCWriteCtx *ctx, GCproto *pt)
 
				 {
			
 
				   MSize i, sizekn = pt->sizekn;
			
 
				   cTValue *o = mref(pt->k, TValue);
			
 
				-  bcwrite_need(ctx, 10*sizekn);
			
 
				+  char *p = lj_buf_more(&ctx->sb, 10*sizekn);
			
 
				   for (i = 0; i < sizekn; i++, o++) {
			
 
				     int32_t k;
			
 
				     if (tvisint(o)) {
			
@@ -220,58 +183,58 @@ static void bcwrite_knum(BCWriteCtx *ctx, GCproto *pt)
 
				 	k = lj_num2int(num);
			
 
				 	if (num == (lua_Number)k) {  /* -0 is never a constant. */
			
 
				 	save_int:
			
 
				-	  bcwrite_uleb128(ctx, 2*(uint32_t)k | ((uint32_t)k & 0x80000000u));
			
 
				-	  if (k < 0) {
			
 
				-	    char *p = &ctx->sb.buf[ctx->sb.n-1];
			
 
				-	    *p = (*p & 7) | ((k>>27) & 0x18);
			
 
				-	  }
			
 
				+	  p = lj_strfmt_wuleb128(p, 2*(uint32_t)k | ((uint32_t)k&0x80000000u));
			
 
				+	  if (k < 0)
			
 
				+	    p[-1] = (p[-1] & 7) | ((k>>27) & 0x18);
			
 
				 	  continue;
			
 
				 	}
			
 
				       }
			
 
				-      bcwrite_uleb128(ctx, 1+(2*o->u32.lo | (o->u32.lo & 0x80000000u)));
			
 
				-      if (o->u32.lo >= 0x80000000u) {
			
 
				-	char *p = &ctx->sb.buf[ctx->sb.n-1];
			
 
				-	*p = (*p & 7) | ((o->u32.lo>>27) & 0x18);
			
 
				-      }
			
 
				-      bcwrite_uleb128(ctx, o->u32.hi);
			
 
				+      p = lj_strfmt_wuleb128(p, 1+(2*o->u32.lo | (o->u32.lo & 0x80000000u)));
			
 
				+      if (o->u32.lo >= 0x80000000u)
			
 
				+	p[-1] = (p[-1] & 7) | ((o->u32.lo>>27) & 0x18);
			
 
				+      p = lj_strfmt_wuleb128(p, o->u32.hi);
			
 
				     }
			
 
				   }
			
 
				+  setsbufP(&ctx->sb, p);
			
 
				 }
			
 
				 
			
 
				 /* Write bytecode instructions. */
			
 
				-static void bcwrite_bytecode(BCWriteCtx *ctx, GCproto *pt)
			
 
				+static char *bcwrite_bytecode(BCWriteCtx *ctx, char *p, GCproto *pt)
			
 
				 {
			
 
				   MSize nbc = pt->sizebc-1;  /* Omit the [JI]FUNC* header. */
			
 
				 #if LJ_HASJIT
			
 
				-  uint8_t *p = (uint8_t *)&ctx->sb.buf[ctx->sb.n];
			
 
				+  uint8_t *q = (uint8_t *)p;
			
 
				 #endif
			
 
				-  bcwrite_block(ctx, proto_bc(pt)+1, nbc*(MSize)sizeof(BCIns));
			
 
				+  p = lj_buf_wmem(p, proto_bc(pt)+1, nbc*(MSize)sizeof(BCIns));
			
 
				+  UNUSED(ctx);
			
 
				 #if LJ_HASJIT
			
 
				   /* Unpatch modified bytecode containing ILOOP/JLOOP etc. */
			
 
				   if ((pt->flags & PROTO_ILOOP) || pt->trace) {
			
 
				-    jit_State *J = L2J(ctx->L);
			
 
				+    jit_State *J = L2J(sbufL(&ctx->sb));
			
 
				     MSize i;
			
 
				-    for (i = 0; i < nbc; i++, p += sizeof(BCIns)) {
			
 
				-      BCOp op = (BCOp)p[LJ_ENDIAN_SELECT(0, 3)];
			
 
				+    for (i = 0; i < nbc; i++, q += sizeof(BCIns)) {
			
 
				+      BCOp op = (BCOp)q[LJ_ENDIAN_SELECT(0, 3)];
			
 
				       if (op == BC_IFORL || op == BC_IITERL || op == BC_ILOOP ||
			
 
				 	  op == BC_JFORI) {
			
 
				-	p[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_IFORL+BC_FORL);
			
 
				+	q[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_IFORL+BC_FORL);
			
 
				       } else if (op == BC_JFORL || op == BC_JITERL || op == BC_JLOOP) {
			
 
				-	BCReg rd = p[LJ_ENDIAN_SELECT(2, 1)] + (p[LJ_ENDIAN_SELECT(3, 0)] << 8);
			
 
				+	BCReg rd = q[LJ_ENDIAN_SELECT(2, 1)] + (q[LJ_ENDIAN_SELECT(3, 0)] << 8);
			
 
				 	BCIns ins = traceref(J, rd)->startins;
			
 
				-	p[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_JFORL+BC_FORL);
			
 
				-	p[LJ_ENDIAN_SELECT(2, 1)] = bc_c(ins);
			
 
				-	p[LJ_ENDIAN_SELECT(3, 0)] = bc_b(ins);
			
 
				+	q[LJ_ENDIAN_SELECT(0, 3)] = (uint8_t)(op-BC_JFORL+BC_FORL);
			
 
				+	q[LJ_ENDIAN_SELECT(2, 1)] = bc_c(ins);
			
 
				+	q[LJ_ENDIAN_SELECT(3, 0)] = bc_b(ins);
			
 
				       }
			
 
				     }
			
 
				   }
			
 
				 #endif
			
 
				+  return p;
			
 
				 }
			
 
				 
			
 
				 /* Write prototype. */
			
 
				 static void bcwrite_proto(BCWriteCtx *ctx, GCproto *pt)
			
 
				 {
			
 
				   MSize sizedbg = 0;
			
 
				+  char *p;
			
 
				 
			
 
				   /* Recursively write children of prototype. */
			
 
				   if ((pt->flags & PROTO_CHILD)) {
			
@@ -285,31 +248,32 @@ static void bcwrite_proto(BCWriteCtx *ctx, GCproto *pt)
 
				   }
			
 
				 
			
 
				   /* Start writing the prototype info to a buffer. */
			
 
				-  lj_str_resetbuf(&ctx->sb);
			
 
				-  ctx->sb.n = 5;  /* Leave room for final size. */
			
 
				-  bcwrite_need(ctx, 4+6*5+(pt->sizebc-1)*(MSize)sizeof(BCIns)+pt->sizeuv*2);
			
 
				+  p = lj_buf_need(&ctx->sb,
			
 
				+		  5+4+6*5+(pt->sizebc-1)*(MSize)sizeof(BCIns)+pt->sizeuv*2);
			
 
				+  p += 5;  /* Leave room for final size. */
			
 
				 
			
 
				   /* Write prototype header. */
			
 
				-  bcwrite_byte(ctx, (pt->flags & (PROTO_CHILD|PROTO_VARARG|PROTO_FFI)));
			
 
				-  bcwrite_byte(ctx, pt->numparams);
			
 
				-  bcwrite_byte(ctx, pt->framesize);
			
 
				-  bcwrite_byte(ctx, pt->sizeuv);
			
 
				-  bcwrite_uleb128(ctx, pt->sizekgc);
			
 
				-  bcwrite_uleb128(ctx, pt->sizekn);
			
 
				-  bcwrite_uleb128(ctx, pt->sizebc-1);
			
 
				+  *p++ = (pt->flags & (PROTO_CHILD|PROTO_VARARG|PROTO_FFI));
			
 
				+  *p++ = pt->numparams;
			
 
				+  *p++ = pt->framesize;
			
 
				+  *p++ = pt->sizeuv;
			
 
				+  p = lj_strfmt_wuleb128(p, pt->sizekgc);
			
 
				+  p = lj_strfmt_wuleb128(p, pt->sizekn);
			
 
				+  p = lj_strfmt_wuleb128(p, pt->sizebc-1);
			
 
				   if (!ctx->strip) {
			
 
				     if (proto_lineinfo(pt))
			
 
				       sizedbg = pt->sizept - (MSize)((char *)proto_lineinfo(pt) - (char *)pt);
			
 
				-    bcwrite_uleb128(ctx, sizedbg);
			
 
				+    p = lj_strfmt_wuleb128(p, sizedbg);
			
 
				     if (sizedbg) {
			
 
				-      bcwrite_uleb128(ctx, pt->firstline);
			
 
				-      bcwrite_uleb128(ctx, pt->numline);
			
 
				+      p = lj_strfmt_wuleb128(p, pt->firstline);
			
 
				+      p = lj_strfmt_wuleb128(p, pt->numline);
			
 
				     }
			
 
				   }
			
 
				 
			
 
				   /* Write bytecode instructions and upvalue refs. */
			
 
				-  bcwrite_bytecode(ctx, pt);
			
 
				-  bcwrite_block(ctx, proto_uv(pt), pt->sizeuv*2);
			
 
				+  p = bcwrite_bytecode(ctx, p, pt);
			
 
				+  p = lj_buf_wmem(p, proto_uv(pt), pt->sizeuv*2);
			
 
				+  setsbufP(&ctx->sb, p);
			
 
				 
			
 
				   /* Write constants. */
			
 
				   bcwrite_kgc(ctx, pt);
			
@@ -317,18 +281,19 @@ static void bcwrite_proto(BCWriteCtx *ctx, GCproto *pt)
 
				 
			
 
				   /* Write debug info, if not stripped. */
			
 
				   if (sizedbg) {
			
 
				-    bcwrite_need(ctx, sizedbg);
			
 
				-    bcwrite_block(ctx, proto_lineinfo(pt), sizedbg);
			
 
				+    p = lj_buf_more(&ctx->sb, sizedbg);
			
 
				+    p = lj_buf_wmem(p, proto_lineinfo(pt), sizedbg);
			
 
				+    setsbufP(&ctx->sb, p);
			
 
				   }
			
 
				 
			
 
				   /* Pass buffer to writer function. */
			
 
				   if (ctx->status == 0) {
			
 
				-    MSize n = ctx->sb.n - 5;
			
 
				+    MSize n = sbuflen(&ctx->sb) - 5;
			
 
				     MSize nn = (lj_fls(n)+8)*9 >> 6;
			
 
				-    ctx->sb.n = 5 - nn;
			
 
				-    bcwrite_uleb128(ctx, n);  /* Fill in final size. */
			
 
				-    lua_assert(ctx->sb.n == 5);
			
 
				-    ctx->status = ctx->wfunc(ctx->L, ctx->sb.buf+5-nn, nn+n, ctx->wdata);
			
 
				+    char *q = sbufB(&ctx->sb) + (5 - nn);
			
 
				+    p = lj_strfmt_wuleb128(q, n);  /* Fill in final size. */
			
 
				+    lua_assert(p == sbufB(&ctx->sb) + 5);
			
 
				+    ctx->status = ctx->wfunc(sbufL(&ctx->sb), q, nn+n, ctx->wdata);
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -338,20 +303,21 @@ static void bcwrite_header(BCWriteCtx *ctx)
 
				   GCstr *chunkname = proto_chunkname(ctx->pt);
			
 
				   const char *name = strdata(chunkname);
			
 
				   MSize len = chunkname->len;
			
 
				-  lj_str_resetbuf(&ctx->sb);
			
 
				-  bcwrite_need(ctx, 5+5+len);
			
 
				-  bcwrite_byte(ctx, BCDUMP_HEAD1);
			
 
				-  bcwrite_byte(ctx, BCDUMP_HEAD2);
			
 
				-  bcwrite_byte(ctx, BCDUMP_HEAD3);
			
 
				-  bcwrite_byte(ctx, BCDUMP_VERSION);
			
 
				-  bcwrite_byte(ctx, (ctx->strip ? BCDUMP_F_STRIP : 0) +
			
 
				-		   (LJ_BE ? BCDUMP_F_BE : 0) +
			
 
				-		   ((ctx->pt->flags & PROTO_FFI) ? BCDUMP_F_FFI : 0));
			
 
				+  char *p = lj_buf_need(&ctx->sb, 5+5+len);
			
 
				+  *p++ = BCDUMP_HEAD1;
			
 
				+  *p++ = BCDUMP_HEAD2;
			
 
				+  *p++ = BCDUMP_HEAD3;
			
 
				+  *p++ = BCDUMP_VERSION;
			
 
				+  *p++ = (ctx->strip ? BCDUMP_F_STRIP : 0) +
			
 
				+	 LJ_BE*BCDUMP_F_BE +
			
 
				+	 ((ctx->pt->flags & PROTO_FFI) ? BCDUMP_F_FFI : 0) +
			
 
				+	 LJ_FR2*BCDUMP_F_FR2;
			
 
				   if (!ctx->strip) {
			
 
				-    bcwrite_uleb128(ctx, len);
			
 
				-    bcwrite_block(ctx, name, len);
			
 
				+    p = lj_strfmt_wuleb128(p, len);
			
 
				+    p = lj_buf_wmem(p, name, len);
			
 
				   }
			
 
				-  ctx->status = ctx->wfunc(ctx->L, ctx->sb.buf, ctx->sb.n, ctx->wdata);
			
 
				+  ctx->status = ctx->wfunc(sbufL(&ctx->sb), sbufB(&ctx->sb),
			
 
				+			   (MSize)(p - sbufB(&ctx->sb)), ctx->wdata);
			
 
				 }
			
 
				 
			
 
				 /* Write footer of bytecode dump. */
			
@@ -359,7 +325,7 @@ static void bcwrite_footer(BCWriteCtx *ctx)
 
				 {
			
 
				   if (ctx->status == 0) {
			
 
				     uint8_t zero = 0;
			
 
				-    ctx->status = ctx->wfunc(ctx->L, &zero, 1, ctx->wdata);
			
 
				+    ctx->status = ctx->wfunc(sbufL(&ctx->sb), &zero, 1, ctx->wdata);
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -367,8 +333,8 @@ static void bcwrite_footer(BCWriteCtx *ctx)
 
				 static TValue *cpwriter(lua_State *L, lua_CFunction dummy, void *ud)
			
 
				 {
			
 
				   BCWriteCtx *ctx = (BCWriteCtx *)ud;
			
 
				-  UNUSED(dummy);
			
 
				-  lj_str_resizebuf(L, &ctx->sb, 1024);  /* Avoids resize for most prototypes. */
			
 
				+  UNUSED(L); UNUSED(dummy);
			
 
				+  lj_buf_need(&ctx->sb, 1024);  /* Avoids resize for most prototypes. */
			
 
				   bcwrite_header(ctx);
			
 
				   bcwrite_proto(ctx, ctx->pt);
			
 
				   bcwrite_footer(ctx);
			
@@ -381,16 +347,15 @@ int lj_bcwrite(lua_State *L, GCproto *pt, lua_Writer writer, void *data,
 
				 {
			
 
				   BCWriteCtx ctx;
			
 
				   int status;
			
 
				-  ctx.L = L;
			
 
				   ctx.pt = pt;
			
 
				   ctx.wfunc = writer;
			
 
				   ctx.wdata = data;
			
 
				   ctx.strip = strip;
			
 
				   ctx.status = 0;
			
 
				-  lj_str_initbuf(&ctx.sb);
			
 
				+  lj_buf_init(L, &ctx.sb);
			
 
				   status = lj_vm_cpcall(L, NULL, &ctx, cpwriter);
			
 
				   if (status == 0) status = ctx.status;
			
 
				-  lj_str_freebuf(G(ctx.L), &ctx.sb);
			
 
				+  lj_buf_free(G(sbufL(&ctx.sb)), &ctx.sb);
			
 
				   return status;
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lj_buf.c
+++ b/luajit.mod/luajit/src/lj_buf.c
@@ -0,0 +1,232 @@
 
				+/*
			
 
				+** Buffer handling.
			
 
				+** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
			
 
				+*/
			
 
				+
			
 
				+#define lj_buf_c
			
 
				+#define LUA_CORE
			
 
				+
			
 
				+#include "lj_obj.h"
			
 
				+#include "lj_gc.h"
			
 
				+#include "lj_err.h"
			
 
				+#include "lj_buf.h"
			
 
				+#include "lj_str.h"
			
 
				+#include "lj_tab.h"
			
 
				+#include "lj_strfmt.h"
			
 
				+
			
 
				+/* -- Buffer management --------------------------------------------------- */
			
 
				+
			
 
				+static void buf_grow(SBuf *sb, MSize sz)
			
 
				+{
			
 
				+  MSize osz = sbufsz(sb), len = sbuflen(sb), nsz = osz;
			
 
				+  char *b;
			
 
				+  if (nsz < LJ_MIN_SBUF) nsz = LJ_MIN_SBUF;
			
 
				+  while (nsz < sz) nsz += nsz;
			
 
				+  b = (char *)lj_mem_realloc(sbufL(sb), sbufB(sb), osz, nsz);
			
 
				+  setmref(sb->b, b);
			
 
				+  setmref(sb->p, b + len);
			
 
				+  setmref(sb->e, b + nsz);
			
 
				+}
			
 
				+
			
 
				+LJ_NOINLINE char *LJ_FASTCALL lj_buf_need2(SBuf *sb, MSize sz)
			
 
				+{
			
 
				+  lua_assert(sz > sbufsz(sb));
			
 
				+  if (LJ_UNLIKELY(sz > LJ_MAX_BUF))
			
 
				+    lj_err_mem(sbufL(sb));
			
 
				+  buf_grow(sb, sz);
			
 
				+  return sbufB(sb);
			
 
				+}
			
 
				+
			
 
				+LJ_NOINLINE char *LJ_FASTCALL lj_buf_more2(SBuf *sb, MSize sz)
			
 
				+{
			
 
				+  MSize len = sbuflen(sb);
			
 
				+  lua_assert(sz > sbufleft(sb));
			
 
				+  if (LJ_UNLIKELY(sz > LJ_MAX_BUF || len + sz > LJ_MAX_BUF))
			
 
				+    lj_err_mem(sbufL(sb));
			
 
				+  buf_grow(sb, len + sz);
			
 
				+  return sbufP(sb);
			
 
				+}
			
 
				+
			
 
				+void LJ_FASTCALL lj_buf_shrink(lua_State *L, SBuf *sb)
			
 
				+{
			
 
				+  char *b = sbufB(sb);
			
 
				+  MSize osz = (MSize)(sbufE(sb) - b);
			
 
				+  if (osz > 2*LJ_MIN_SBUF) {
			
 
				+    MSize n = (MSize)(sbufP(sb) - b);
			
 
				+    b = lj_mem_realloc(L, b, osz, (osz >> 1));
			
 
				+    setmref(sb->b, b);
			
 
				+    setmref(sb->p, b + n);
			
 
				+    setmref(sb->e, b + (osz >> 1));
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+char * LJ_FASTCALL lj_buf_tmp(lua_State *L, MSize sz)
			
 
				+{
			
 
				+  SBuf *sb = &G(L)->tmpbuf;
			
 
				+  setsbufL(sb, L);
			
 
				+  return lj_buf_need(sb, sz);
			
 
				+}
			
 
				+
			
 
				+/* -- Low-level buffer put operations ------------------------------------- */
			
 
				+
			
 
				+SBuf *lj_buf_putmem(SBuf *sb, const void *q, MSize len)
			
 
				+{
			
 
				+  char *p = lj_buf_more(sb, len);
			
 
				+  p = lj_buf_wmem(p, q, len);
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf * LJ_FASTCALL lj_buf_putchar(SBuf *sb, int c)
			
 
				+{
			
 
				+  char *p = lj_buf_more(sb, 1);
			
 
				+  *p++ = (char)c;
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf * LJ_FASTCALL lj_buf_putstr(SBuf *sb, GCstr *s)
			
 
				+{
			
 
				+  MSize len = s->len;
			
 
				+  char *p = lj_buf_more(sb, len);
			
 
				+  p = lj_buf_wmem(p, strdata(s), len);
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+/* -- High-level buffer put operations ------------------------------------ */
			
 
				+
			
 
				+SBuf * LJ_FASTCALL lj_buf_putstr_reverse(SBuf *sb, GCstr *s)
			
 
				+{
			
 
				+  MSize len = s->len;
			
 
				+  char *p = lj_buf_more(sb, len), *e = p+len;
			
 
				+  const char *q = strdata(s)+len-1;
			
 
				+  while (p < e)
			
 
				+    *p++ = *q--;
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf * LJ_FASTCALL lj_buf_putstr_lower(SBuf *sb, GCstr *s)
			
 
				+{
			
 
				+  MSize len = s->len;
			
 
				+  char *p = lj_buf_more(sb, len), *e = p+len;
			
 
				+  const char *q = strdata(s);
			
 
				+  for (; p < e; p++, q++) {
			
 
				+    uint32_t c = *(unsigned char *)q;
			
 
				+#if LJ_TARGET_PPC
			
 
				+    *p = c + ((c >= 'A' && c <= 'Z') << 5);
			
 
				+#else
			
 
				+    if (c >= 'A' && c <= 'Z') c += 0x20;
			
 
				+    *p = c;
			
 
				+#endif
			
 
				+  }
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf * LJ_FASTCALL lj_buf_putstr_upper(SBuf *sb, GCstr *s)
			
 
				+{
			
 
				+  MSize len = s->len;
			
 
				+  char *p = lj_buf_more(sb, len), *e = p+len;
			
 
				+  const char *q = strdata(s);
			
 
				+  for (; p < e; p++, q++) {
			
 
				+    uint32_t c = *(unsigned char *)q;
			
 
				+#if LJ_TARGET_PPC
			
 
				+    *p = c - ((c >= 'a' && c <= 'z') << 5);
			
 
				+#else
			
 
				+    if (c >= 'a' && c <= 'z') c -= 0x20;
			
 
				+    *p = c;
			
 
				+#endif
			
 
				+  }
			
 
				+  setsbufP(sb, p);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf *lj_buf_putstr_rep(SBuf *sb, GCstr *s, int32_t rep)
			
 
				+{
			
 
				+  MSize len = s->len;
			
 
				+  if (rep > 0 && len) {
			
 
				+    uint64_t tlen = (uint64_t)rep * len;
			
 
				+    char *p;
			
 
				+    if (LJ_UNLIKELY(tlen > LJ_MAX_STR))
			
 
				+      lj_err_mem(sbufL(sb));
			
 
				+    p = lj_buf_more(sb, (MSize)tlen);
			
 
				+    if (len == 1) {  /* Optimize a common case. */
			
 
				+      uint32_t c = strdata(s)[0];
			
 
				+      do { *p++ = c; } while (--rep > 0);
			
 
				+    } else {
			
 
				+      const char *e = strdata(s) + len;
			
 
				+      do {
			
 
				+	const char *q = strdata(s);
			
 
				+	do { *p++ = *q++; } while (q < e);
			
 
				+      } while (--rep > 0);
			
 
				+    }
			
 
				+    setsbufP(sb, p);
			
 
				+  }
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+SBuf *lj_buf_puttab(SBuf *sb, GCtab *t, GCstr *sep, int32_t i, int32_t e)
			
 
				+{
			
 
				+  MSize seplen = sep ? sep->len : 0;
			
 
				+  if (i <= e) {
			
 
				+    for (;;) {
			
 
				+      cTValue *o = lj_tab_getint(t, i);
			
 
				+      char *p;
			
 
				+      if (!o) {
			
 
				+      badtype:  /* Error: bad element type. */
			
 
				+	setsbufP(sb, (void *)(intptr_t)i);  /* Store failing index. */
			
 
				+	return NULL;
			
 
				+      } else if (tvisstr(o)) {
			
 
				+	MSize len = strV(o)->len;
			
 
				+	p = lj_buf_wmem(lj_buf_more(sb, len + seplen), strVdata(o), len);
			
 
				+      } else if (tvisint(o)) {
			
 
				+	p = lj_strfmt_wint(lj_buf_more(sb, STRFMT_MAXBUF_INT+seplen), intV(o));
			
 
				+      } else if (tvisnum(o)) {
			
 
				+	p = lj_buf_more(lj_strfmt_putfnum(sb, STRFMT_G14, numV(o)), seplen);
			
 
				+      } else {
			
 
				+	goto badtype;
			
 
				+      }
			
 
				+      if (i++ == e) {
			
 
				+	setsbufP(sb, p);
			
 
				+	break;
			
 
				+      }
			
 
				+      if (seplen) p = lj_buf_wmem(p, strdata(sep), seplen);
			
 
				+      setsbufP(sb, p);
			
 
				+    }
			
 
				+  }
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+/* -- Miscellaneous buffer operations ------------------------------------- */
			
 
				+
			
 
				+GCstr * LJ_FASTCALL lj_buf_tostr(SBuf *sb)
			
 
				+{
			
 
				+  return lj_str_new(sbufL(sb), sbufB(sb), sbuflen(sb));
			
 
				+}
			
 
				+
			
 
				+/* Concatenate two strings. */
			
 
				+GCstr *lj_buf_cat2str(lua_State *L, GCstr *s1, GCstr *s2)
			
 
				+{
			
 
				+  MSize len1 = s1->len, len2 = s2->len;
			
 
				+  char *buf = lj_buf_tmp(L, len1 + len2);
			
 
				+  memcpy(buf, strdata(s1), len1);
			
 
				+  memcpy(buf+len1, strdata(s2), len2);
			
 
				+  return lj_str_new(L, buf, len1 + len2);
			
 
				+}
			
 
				+
			
 
				+/* Read ULEB128 from buffer. */
			
 
				+uint32_t LJ_FASTCALL lj_buf_ruleb128(const char **pp)
			
 
				+{
			
 
				+  const uint8_t *p = (const uint8_t *)*pp;
			
 
				+  uint32_t v = *p++;
			
 
				+  if (LJ_UNLIKELY(v >= 0x80)) {
			
 
				+    int sh = 0;
			
 
				+    v &= 0x7f;
			
 
				+    do { v |= ((*p & 0x7f) << (sh += 7)); } while (*p++ >= 0x80);
			
 
				+  }
			
 
				+  *pp = (const char *)p;
			
 
				+  return v;
			
 
				+}
			
 
				+
			
--- a/luajit.mod/luajit/src/lj_buf.h
+++ b/luajit.mod/luajit/src/lj_buf.h
@@ -0,0 +1,103 @@
 
				+/*
			
 
				+** Buffer handling.
			
 
				+** Copyright (C) 2005-2017 Mike Pall. See Copyright Notice in luajit.h
			
 
				+*/
			
 
				+
			
 
				+#ifndef _LJ_BUF_H
			
 
				+#define _LJ_BUF_H
			
 
				+
			
 
				+#include "lj_obj.h"
			
 
				+#include "lj_gc.h"
			
 
				+#include "lj_str.h"
			
 
				+
			
 
				+/* Resizable string buffers. Struct definition in lj_obj.h. */
			
 
				+#define sbufB(sb)	(mref((sb)->b, char))
			
 
				+#define sbufP(sb)	(mref((sb)->p, char))
			
 
				+#define sbufE(sb)	(mref((sb)->e, char))
			
 
				+#define sbufL(sb)	(mref((sb)->L, lua_State))
			
 
				+#define sbufsz(sb)	((MSize)(sbufE((sb)) - sbufB((sb))))
			
 
				+#define sbuflen(sb)	((MSize)(sbufP((sb)) - sbufB((sb))))
			
 
				+#define sbufleft(sb)	((MSize)(sbufE((sb)) - sbufP((sb))))
			
 
				+#define setsbufP(sb, q)	(setmref((sb)->p, (q)))
			
 
				+#define setsbufL(sb, l)	(setmref((sb)->L, (l)))
			
 
				+
			
 
				+/* Buffer management */
			
 
				+LJ_FUNC char *LJ_FASTCALL lj_buf_need2(SBuf *sb, MSize sz);
			
 
				+LJ_FUNC char *LJ_FASTCALL lj_buf_more2(SBuf *sb, MSize sz);
			
 
				+LJ_FUNC void LJ_FASTCALL lj_buf_shrink(lua_State *L, SBuf *sb);
			
 
				+LJ_FUNC char * LJ_FASTCALL lj_buf_tmp(lua_State *L, MSize sz);
			
 
				+
			
 
				+static LJ_AINLINE void lj_buf_init(lua_State *L, SBuf *sb)
			
 
				+{
			
 
				+  setsbufL(sb, L);
			
 
				+  setmref(sb->p, NULL); setmref(sb->e, NULL); setmref(sb->b, NULL);
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE void lj_buf_reset(SBuf *sb)
			
 
				+{
			
 
				+  setmrefr(sb->p, sb->b);
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE SBuf *lj_buf_tmp_(lua_State *L)
			
 
				+{
			
 
				+  SBuf *sb = &G(L)->tmpbuf;
			
 
				+  setsbufL(sb, L);
			
 
				+  lj_buf_reset(sb);
			
 
				+  return sb;
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE void lj_buf_free(global_State *g, SBuf *sb)
			
 
				+{
			
 
				+  lj_mem_free(g, sbufB(sb), sbufsz(sb));
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE char *lj_buf_need(SBuf *sb, MSize sz)
			
 
				+{
			
 
				+  if (LJ_UNLIKELY(sz > sbufsz(sb)))
			
 
				+    return lj_buf_need2(sb, sz);
			
 
				+  return sbufB(sb);
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE char *lj_buf_more(SBuf *sb, MSize sz)
			
 
				+{
			
 
				+  if (LJ_UNLIKELY(sz > sbufleft(sb)))
			
 
				+    return lj_buf_more2(sb, sz);
			
 
				+  return sbufP(sb);
			
 
				+}
			
 
				+
			
 
				+/* Low-level buffer put operations */
			
 
				+LJ_FUNC SBuf *lj_buf_putmem(SBuf *sb, const void *q, MSize len);
			
 
				+LJ_FUNC SBuf * LJ_FASTCALL lj_buf_putchar(SBuf *sb, int c);
			
 
				+LJ_FUNC SBuf * LJ_FASTCALL lj_buf_putstr(SBuf *sb, GCstr *s);
			
 
				+
			
 
				+static LJ_AINLINE char *lj_buf_wmem(char *p, const void *q, MSize len)
			
 
				+{
			
 
				+  return (char *)memcpy(p, q, len) + len;
			
 
				+}
			
 
				+
			
 
				+static LJ_AINLINE void lj_buf_putb(SBuf *sb, int c)
			
 
				+{
			
 
				+  char *p = lj_buf_more(sb, 1);
			
 
				+  *p++ = (char)c;
			
 
				+  setsbufP(sb, p);
			
 
				+}
			
 
				+
			
 
				+/* High-level buffer put operations */
			
 
				+LJ_FUNCA SBuf * LJ_FASTCALL lj_buf_putstr_reverse(SBuf *sb, GCstr *s);
			
 
				+LJ_FUNCA SBuf * LJ_FASTCALL lj_buf_putstr_lower(SBuf *sb, GCstr *s);
			
 
				+LJ_FUNCA SBuf * LJ_FASTCALL lj_buf_putstr_upper(SBuf *sb, GCstr *s);
			
 
				+LJ_FUNC SBuf *lj_buf_putstr_rep(SBuf *sb, GCstr *s, int32_t rep);
			
 
				+LJ_FUNC SBuf *lj_buf_puttab(SBuf *sb, GCtab *t, GCstr *sep,
			
 
				+			    int32_t i, int32_t e);
			
 
				+
			
 
				+/* Miscellaneous buffer operations */
			
 
				+LJ_FUNCA GCstr * LJ_FASTCALL lj_buf_tostr(SBuf *sb);
			
 
				+LJ_FUNC GCstr *lj_buf_cat2str(lua_State *L, GCstr *s1, GCstr *s2);
			
 
				+LJ_FUNC uint32_t LJ_FASTCALL lj_buf_ruleb128(const char **pp);
			
 
				+
			
 
				+static LJ_AINLINE GCstr *lj_buf_str(lua_State *L, SBuf *sb)
			
 
				+{
			
 
				+  return lj_str_new(L, sbufB(sb), sbuflen(sb));
			
 
				+}
			
 
				+
			
 
				+#endif
			
--- a/luajit.mod/luajit/src/lj_carith.c
+++ b/luajit.mod/luajit/src/lj_carith.c
@@ -11,10 +11,12 @@
 
				 #include "lj_err.h"
			
 
				 #include "lj_tab.h"
			
 
				 #include "lj_meta.h"
			
 
				+#include "lj_ir.h"
			
 
				 #include "lj_ctype.h"
			
 
				 #include "lj_cconv.h"
			
 
				 #include "lj_cdata.h"
			
 
				 #include "lj_carith.h"
			
 
				+#include "lj_strscan.h"
			
 
				 
			
 
				 /* -- C data arithmetic --------------------------------------------------- */
			
 
				 
			
@@ -272,6 +274,88 @@ int lj_carith_op(lua_State *L, MMS mm)
 
				   return lj_carith_meta(L, cts, &ca, mm);
			
 
				 }
			
 
				 
			
 
				+/* No built-in functionality for length of cdata. */
			
 
				+int lj_carith_len(lua_State *L)
			
 
				+{
			
 
				+  CTState *cts = ctype_cts(L);
			
 
				+  CDArith ca;
			
 
				+  carith_checkarg(L, cts, &ca);
			
 
				+  return lj_carith_meta(L, cts, &ca, MM_len);
			
 
				+}
			
 
				+
			
 
				+/* -- 64 bit bit operations helpers --------------------------------------- */
			
 
				+
			
 
				+#if LJ_64
			
 
				+#define B64DEF(name) \
			
 
				+  static LJ_AINLINE uint64_t lj_carith_##name(uint64_t x, int32_t sh)
			
 
				+#else
			
 
				+/* Not inlined on 32 bit archs, since some of these are quite lengthy. */
			
 
				+#define B64DEF(name) \
			
 
				+  uint64_t LJ_NOINLINE lj_carith_##name(uint64_t x, int32_t sh)
			
 
				+#endif
			
 
				+
			
 
				+B64DEF(shl64) { return x << (sh&63); }
			
 
				+B64DEF(shr64) { return x >> (sh&63); }
			
 
				+B64DEF(sar64) { return (uint64_t)((int64_t)x >> (sh&63)); }
			
 
				+B64DEF(rol64) { return lj_rol(x, (sh&63)); }
			
 
				+B64DEF(ror64) { return lj_ror(x, (sh&63)); }
			
 
				+
			
 
				+#undef B64DEF
			
 
				+
			
 
				+uint64_t lj_carith_shift64(uint64_t x, int32_t sh, int op)
			
 
				+{
			
 
				+  switch (op) {
			
 
				+  case IR_BSHL-IR_BSHL: x = lj_carith_shl64(x, sh); break;
			
 
				+  case IR_BSHR-IR_BSHL: x = lj_carith_shr64(x, sh); break;
			
 
				+  case IR_BSAR-IR_BSHL: x = lj_carith_sar64(x, sh); break;
			
 
				+  case IR_BROL-IR_BSHL: x = lj_carith_rol64(x, sh); break;
			
 
				+  case IR_BROR-IR_BSHL: x = lj_carith_ror64(x, sh); break;
			
 
				+  default: lua_assert(0); break;
			
 
				+  }
			
 
				+  return x;
			
 
				+}
			
 
				+
			
 
				+/* Equivalent to lj_lib_checkbit(), but handles cdata. */
			
 
				+uint64_t lj_carith_check64(lua_State *L, int narg, CTypeID *id)
			
 
				+{
			
 
				+  TValue *o = L->base + narg-1;
			
 
				+  if (o >= L->top) {
			
 
				+  err:
			
 
				+    lj_err_argt(L, narg, LUA_TNUMBER);
			
 
				+  } else if (LJ_LIKELY(tvisnumber(o))) {
			
 
				+    /* Handled below. */
			
 
				+  } else if (tviscdata(o)) {
			
 
				+    CTState *cts = ctype_cts(L);
			
 
				+    uint8_t *sp = (uint8_t *)cdataptr(cdataV(o));
			
 
				+    CTypeID sid = cdataV(o)->ctypeid;
			
 
				+    CType *s = ctype_get(cts, sid);
			
 
				+    uint64_t x;
			
 
				+    if (ctype_isref(s->info)) {
			
 
				+      sp = *(void **)sp;
			
 
				+      sid = ctype_cid(s->info);
			
 
				+    }
			
 
				+    s = ctype_raw(cts, sid);
			
 
				+    if (ctype_isenum(s->info)) s = ctype_child(cts, s);
			
 
				+    if ((s->info & (CTMASK_NUM|CTF_BOOL|CTF_FP|CTF_UNSIGNED)) ==
			
 
				+	CTINFO(CT_NUM, CTF_UNSIGNED) && s->size == 8)
			
 
				+      *id = CTID_UINT64;  /* Use uint64_t, since it has the highest rank. */
			
 
				+    else if (!*id)
			
 
				+      *id = CTID_INT64;  /* Use int64_t, unless already set. */
			
 
				+    lj_cconv_ct_ct(cts, ctype_get(cts, *id), s,
			
 
				+		   (uint8_t *)&x, sp, CCF_ARG(narg));
			
 
				+    return x;
			
 
				+  } else if (!(tvisstr(o) && lj_strscan_number(strV(o), o))) {
			
 
				+    goto err;
			
 
				+  }
			
 
				+  if (LJ_LIKELY(tvisint(o))) {
			
 
				+    return (uint32_t)intV(o);
			
 
				+  } else {
			
 
				+    int32_t i = lj_num2bit(numV(o));
			
 
				+    if (LJ_DUALNUM) setintV(o, i);
			
 
				+    return (uint32_t)i;
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				 /* -- 64 bit integer arithmetic helpers ----------------------------------- */
			
 
				 
			
 
				 #if LJ_32 && LJ_HASJIT
			
--- a/luajit.mod/luajit/src/lj_carith.h
+++ b/luajit.mod/luajit/src/lj_carith.h
@@ -11,6 +11,17 @@
 
				 #if LJ_HASFFI
			
 
				 
			
 
				 LJ_FUNC int lj_carith_op(lua_State *L, MMS mm);
			
 
				+LJ_FUNC int lj_carith_len(lua_State *L);
			
 
				+
			
 
				+#if LJ_32
			
 
				+LJ_FUNC uint64_t lj_carith_shl64(uint64_t x, int32_t sh);
			
 
				+LJ_FUNC uint64_t lj_carith_shr64(uint64_t x, int32_t sh);
			
 
				+LJ_FUNC uint64_t lj_carith_sar64(uint64_t x, int32_t sh);
			
 
				+LJ_FUNC uint64_t lj_carith_rol64(uint64_t x, int32_t sh);
			
 
				+LJ_FUNC uint64_t lj_carith_ror64(uint64_t x, int32_t sh);
			
 
				+#endif
			
 
				+LJ_FUNC uint64_t lj_carith_shift64(uint64_t x, int32_t sh, int op);
			
 
				+LJ_FUNC uint64_t lj_carith_check64(lua_State *L, int narg, CTypeID *id);
			
 
				 
			
 
				 #if LJ_32 && LJ_HASJIT
			
 
				 LJ_FUNC int64_t lj_carith_mul64(int64_t x, int64_t k);
			
--- a/luajit.mod/luajit/src/lj_ccall.c
+++ b/luajit.mod/luajit/src/lj_ccall.c
@@ -9,7 +9,6 @@
 
				 
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				-#include "lj_str.h"
			
 
				 #include "lj_tab.h"
			
 
				 #include "lj_ctype.h"
			
 
				 #include "lj_cconv.h"
			
@@ -291,56 +290,84 @@
 
				 #define CCALL_HANDLE_RET \
			
 
				   if ((ct->info & CTF_VARARG)) sp = (uint8_t *)&cc->gpr[0];
			
 
				 
			
 
				-#elif LJ_TARGET_PPC
			
 
				-/* -- PPC calling conventions --------------------------------------------- */
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+/* -- ARM64 calling conventions ------------------------------------------- */
			
 
				 
			
 
				 #define CCALL_HANDLE_STRUCTRET \
			
 
				-  cc->retref = 1;  /* Return all structs by reference. */ \
			
 
				-  cc->gpr[ngpr++] = (GPRArg)dp;
			
 
				+  cc->retref = !ccall_classify_struct(cts, ctr); \
			
 
				+  if (cc->retref) cc->retp = dp;
			
 
				+
			
 
				+#define CCALL_HANDLE_STRUCTRET2 \
			
 
				+  unsigned int cl = ccall_classify_struct(cts, ctr); \
			
 
				+  if ((cl & 4)) { /* Combine float HFA from separate registers. */ \
			
 
				+    CTSize i = (cl >> 8) - 1; \
			
 
				+    do { ((uint32_t *)dp)[i] = cc->fpr[i].lo; } while (i--); \
			
 
				+  } else { \
			
 
				+    if (cl > 1) sp = (uint8_t *)&cc->fpr[0]; \
			
 
				+    memcpy(dp, sp, ctr->size); \
			
 
				+  }
			
 
				 
			
 
				 #define CCALL_HANDLE_COMPLEXRET \
			
 
				-  /* Complex values are returned in 2 or 4 GPRs. */ \
			
 
				+  /* Complex values are returned in one or two FPRs. */ \
			
 
				   cc->retref = 0;
			
 
				 
			
 
				 #define CCALL_HANDLE_COMPLEXRET2 \
			
 
				-  memcpy(dp, sp, ctr->size);  /* Copy complex from GPRs. */
			
 
				+  if (ctr->size == 2*sizeof(float)) {  /* Copy complex float from FPRs. */ \
			
 
				+    ((float *)dp)[0] = cc->fpr[0].f; \
			
 
				+    ((float *)dp)[1] = cc->fpr[1].f; \
			
 
				+  } else {  /* Copy complex double from FPRs. */ \
			
 
				+    ((double *)dp)[0] = cc->fpr[0].d; \
			
 
				+    ((double *)dp)[1] = cc->fpr[1].d; \
			
 
				+  }
			
 
				 
			
 
				 #define CCALL_HANDLE_STRUCTARG \
			
 
				-  rp = cdataptr(lj_cdata_new(cts, did, sz)); \
			
 
				-  sz = CTSIZE_PTR;  /* Pass all structs by reference. */
			
 
				+  unsigned int cl = ccall_classify_struct(cts, d); \
			
 
				+  if (cl == 0) {  /* Pass struct by reference. */ \
			
 
				+    rp = cdataptr(lj_cdata_new(cts, did, sz)); \
			
 
				+    sz = CTSIZE_PTR; \
			
 
				+  } else if (cl > 1) {  /* Pass struct in FPRs or on stack. */ \
			
 
				+    isfp = (cl & 4) ? 2 : 1; \
			
 
				+  }  /* else: Pass struct in GPRs or on stack. */
			
 
				 
			
 
				 #define CCALL_HANDLE_COMPLEXARG \
			
 
				-  /* Pass complex by value in 2 or 4 GPRs. */
			
 
				+  /* Pass complex by value in separate (!) FPRs or on stack. */ \
			
 
				+  isfp = sz == 2*sizeof(float) ? 2 : 1;
			
 
				 
			
 
				 #define CCALL_HANDLE_REGARG \
			
 
				-  if (isfp) {  /* Try to pass argument in FPRs. */ \
			
 
				-    if (nfpr + 1 <= CCALL_NARG_FPR) { \
			
 
				+  if (LJ_TARGET_IOS && isva) { \
			
 
				+    /* IOS: All variadic arguments are on the stack. */ \
			
 
				+  } else if (isfp) {  /* Try to pass argument in FPRs. */ \
			
 
				+    int n2 = ctype_isvector(d->info) ? 1 : n*isfp; \
			
 
				+    if (nfpr + n2 <= CCALL_NARG_FPR) { \
			
 
				       dp = &cc->fpr[nfpr]; \
			
 
				-      nfpr += 1; \
			
 
				-      d = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */ \
			
 
				+      nfpr += n2; \
			
 
				       goto done; \
			
 
				+    } else { \
			
 
				+      nfpr = CCALL_NARG_FPR;  /* Prevent reordering. */ \
			
 
				+      if (LJ_TARGET_IOS && d->size < 8) goto err_nyi; \
			
 
				     } \
			
 
				   } else {  /* Try to pass argument in GPRs. */ \
			
 
				-    if (n > 1) { \
			
 
				-      lua_assert(n == 2 || n == 4);  /* int64_t or complex (float). */ \
			
 
				-      if (ctype_isinteger(d->info)) \
			
 
				-	ngpr = (ngpr + 1u) & ~1u;  /* Align int64_t to regpair. */ \
			
 
				-      else if (ngpr + n > maxgpr) \
			
 
				-	ngpr = maxgpr;  /* Prevent reordering. */ \
			
 
				-    } \
			
 
				+    if (!LJ_TARGET_IOS && (d->info & CTF_ALIGN) > CTALIGN_PTR) \
			
 
				+      ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				     if (ngpr + n <= maxgpr) { \
			
 
				       dp = &cc->gpr[ngpr]; \
			
 
				       ngpr += n; \
			
 
				       goto done; \
			
 
				+    } else { \
			
 
				+      ngpr = maxgpr;  /* Prevent reordering. */ \
			
 
				+      if (LJ_TARGET_IOS && d->size < 8) goto err_nyi; \
			
 
				     } \
			
 
				   }
			
 
				 
			
 
				+#if LJ_BE
			
 
				 #define CCALL_HANDLE_RET \
			
 
				   if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				-    ctr = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */
			
 
				+    sp = (uint8_t *)&cc->fpr[0].f;
			
 
				+#endif
			
 
				 
			
 
				-#elif LJ_TARGET_PPCSPE
			
 
				-/* -- PPC/SPE calling conventions ----------------------------------------- */
			
 
				+
			
 
				+#elif LJ_TARGET_PPC
			
 
				+/* -- PPC calling conventions --------------------------------------------- */
			
 
				 
			
 
				 #define CCALL_HANDLE_STRUCTRET \
			
 
				   cc->retref = 1;  /* Return all structs by reference. */ \
			
@@ -360,12 +387,12 @@
 
				 #define CCALL_HANDLE_COMPLEXARG \
			
 
				   /* Pass complex by value in 2 or 4 GPRs. */
			
 
				 
			
 
				-/* PPC/SPE has a softfp ABI. */
			
 
				-#define CCALL_HANDLE_REGARG \
			
 
				-  if (n > 1) {  /* Doesn't fit in a single GPR? */ \
			
 
				-    lua_assert(n == 2 || n == 4);  /* int64_t, double or complex (float). */ \
			
 
				-    if (n == 2) \
			
 
				-      ngpr = (ngpr + 1u) & ~1u;  /* Only align 64 bit value to regpair. */ \
			
 
				+#define CCALL_HANDLE_GPR \
			
 
				+  /* Try to pass argument in GPRs. */ \
			
 
				+  if (n > 1) { \
			
 
				+    lua_assert(n == 2 || n == 4);  /* int64_t or complex (float). */ \
			
 
				+    if (ctype_isinteger(d->info) || ctype_isfp(d->info)) \
			
 
				+      ngpr = (ngpr + 1u) & ~1u;  /* Align int64_t to regpair. */ \
			
 
				     else if (ngpr + n > maxgpr) \
			
 
				       ngpr = maxgpr;  /* Prevent reordering. */ \
			
 
				   } \
			
@@ -373,10 +400,32 @@
 
				     dp = &cc->gpr[ngpr]; \
			
 
				     ngpr += n; \
			
 
				     goto done; \
			
 
				+  } \
			
 
				+
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+#define CCALL_HANDLE_REGARG  CCALL_HANDLE_GPR
			
 
				+#else
			
 
				+#define CCALL_HANDLE_REGARG \
			
 
				+  if (isfp) {  /* Try to pass argument in FPRs. */ \
			
 
				+    if (nfpr + 1 <= CCALL_NARG_FPR) { \
			
 
				+      dp = &cc->fpr[nfpr]; \
			
 
				+      nfpr += 1; \
			
 
				+      d = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */ \
			
 
				+      goto done; \
			
 
				+    } \
			
 
				+  } else { \
			
 
				+    CCALL_HANDLE_GPR \
			
 
				   }
			
 
				+#endif
			
 
				 
			
 
				-#elif LJ_TARGET_MIPS
			
 
				-/* -- MIPS calling conventions -------------------------------------------- */
			
 
				+#if !LJ_ABI_SOFTFP
			
 
				+#define CCALL_HANDLE_RET \
			
 
				+  if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				+    ctr = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */
			
 
				+#endif
			
 
				+
			
 
				+#elif LJ_TARGET_MIPS32
			
 
				+/* -- MIPS o32 calling conventions ---------------------------------------- */
			
 
				 
			
 
				 #define CCALL_HANDLE_STRUCTRET \
			
 
				   cc->retref = 1;  /* Return all structs by reference. */ \
			
@@ -386,6 +435,18 @@
 
				   /* Complex values are returned in 1 or 2 FPRs. */ \
			
 
				   cc->retref = 0;
			
 
				 
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+#define CCALL_HANDLE_COMPLEXRET2 \
			
 
				+  if (ctr->size == 2*sizeof(float)) {  /* Copy complex float from GPRs. */ \
			
 
				+    ((intptr_t *)dp)[0] = cc->gpr[0]; \
			
 
				+    ((intptr_t *)dp)[1] = cc->gpr[1]; \
			
 
				+  } else {  /* Copy complex double from GPRs. */ \
			
 
				+    ((intptr_t *)dp)[0] = cc->gpr[0]; \
			
 
				+    ((intptr_t *)dp)[1] = cc->gpr[1]; \
			
 
				+    ((intptr_t *)dp)[2] = cc->gpr[2]; \
			
 
				+    ((intptr_t *)dp)[3] = cc->gpr[3]; \
			
 
				+  }
			
 
				+#else
			
 
				 #define CCALL_HANDLE_COMPLEXRET2 \
			
 
				   if (ctr->size == 2*sizeof(float)) {  /* Copy complex float from FPRs. */ \
			
 
				     ((float *)dp)[0] = cc->fpr[0].f; \
			
@@ -394,6 +455,7 @@
 
				     ((double *)dp)[0] = cc->fpr[0].d; \
			
 
				     ((double *)dp)[1] = cc->fpr[1].d; \
			
 
				   }
			
 
				+#endif
			
 
				 
			
 
				 #define CCALL_HANDLE_STRUCTARG \
			
 
				   /* Pass all structs by value in registers and/or on the stack. */
			
@@ -401,6 +463,22 @@
 
				 #define CCALL_HANDLE_COMPLEXARG \
			
 
				   /* Pass complex by value in 2 or 4 GPRs. */
			
 
				 
			
 
				+#define CCALL_HANDLE_GPR \
			
 
				+  if ((d->info & CTF_ALIGN) > CTALIGN_PTR) \
			
 
				+    ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				+  if (ngpr < maxgpr) { \
			
 
				+    dp = &cc->gpr[ngpr]; \
			
 
				+    if (ngpr + n > maxgpr) { \
			
 
				+     nsp += ngpr + n - maxgpr;  /* Assumes contiguous gpr/stack fields. */ \
			
 
				+     if (nsp > CCALL_MAXSTACK) goto err_nyi;  /* Too many arguments. */ \
			
 
				+     ngpr = maxgpr; \
			
 
				+    } else { \
			
 
				+     ngpr += n; \
			
 
				+    } \
			
 
				+    goto done; \
			
 
				+  }
			
 
				+
			
 
				+#if !LJ_ABI_SOFTFP	/* MIPS32 hard-float */
			
 
				 #define CCALL_HANDLE_REGARG \
			
 
				   if (isfp && nfpr < CCALL_NARG_FPR && !(ct->info & CTF_VARARG)) { \
			
 
				     /* Try to pass argument in FPRs. */ \
			
@@ -409,25 +487,91 @@
 
				     goto done; \
			
 
				   } else {  /* Try to pass argument in GPRs. */ \
			
 
				     nfpr = CCALL_NARG_FPR; \
			
 
				-    if ((d->info & CTF_ALIGN) > CTALIGN_PTR) \
			
 
				-      ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				-    if (ngpr < maxgpr) { \
			
 
				-      dp = &cc->gpr[ngpr]; \
			
 
				-      if (ngpr + n > maxgpr) { \
			
 
				-	nsp += ngpr + n - maxgpr;  /* Assumes contiguous gpr/stack fields. */ \
			
 
				-	if (nsp > CCALL_MAXSTACK) goto err_nyi;  /* Too many arguments. */ \
			
 
				-	ngpr = maxgpr; \
			
 
				-      } else { \
			
 
				-	ngpr += n; \
			
 
				-      } \
			
 
				-      goto done; \
			
 
				-    } \
			
 
				+    CCALL_HANDLE_GPR \
			
 
				+  }
			
 
				+#else			/* MIPS32 soft-float */
			
 
				+#define CCALL_HANDLE_REGARG CCALL_HANDLE_GPR
			
 
				+#endif
			
 
				+
			
 
				+#if !LJ_ABI_SOFTFP
			
 
				+/* On MIPS64 soft-float, position of float return values is endian-dependant. */
			
 
				+#define CCALL_HANDLE_RET \
			
 
				+  if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				+    sp = (uint8_t *)&cc->fpr[0].f;
			
 
				+#endif
			
 
				+
			
 
				+#elif LJ_TARGET_MIPS64
			
 
				+/* -- MIPS n64 calling conventions ---------------------------------------- */
			
 
				+
			
 
				+#define CCALL_HANDLE_STRUCTRET \
			
 
				+  cc->retref = !(sz <= 16); \
			
 
				+  if (cc->retref) cc->gpr[ngpr++] = (GPRArg)dp;
			
 
				+
			
 
				+#define CCALL_HANDLE_STRUCTRET2 \
			
 
				+  ccall_copy_struct(cc, ctr, dp, sp, ccall_classify_struct(cts, ctr, ct));
			
 
				+
			
 
				+#define CCALL_HANDLE_COMPLEXRET \
			
 
				+  /* Complex values are returned in 1 or 2 FPRs. */ \
			
 
				+  cc->retref = 0;
			
 
				+
			
 
				+#if LJ_ABI_SOFTFP	/* MIPS64 soft-float */
			
 
				+
			
 
				+#define CCALL_HANDLE_COMPLEXRET2 \
			
 
				+  if (ctr->size == 2*sizeof(float)) {  /* Copy complex float from GPRs. */ \
			
 
				+    ((intptr_t *)dp)[0] = cc->gpr[0]; \
			
 
				+  } else {  /* Copy complex double from GPRs. */ \
			
 
				+    ((intptr_t *)dp)[0] = cc->gpr[0]; \
			
 
				+    ((intptr_t *)dp)[1] = cc->gpr[1]; \
			
 
				+  }
			
 
				+
			
 
				+#define CCALL_HANDLE_COMPLEXARG \
			
 
				+  /* Pass complex by value in 2 or 4 GPRs. */
			
 
				+
			
 
				+/* Position of soft-float 'float' return value depends on endianess.  */
			
 
				+#define CCALL_HANDLE_RET \
			
 
				+  if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				+    sp = (uint8_t *)cc->gpr + LJ_ENDIAN_SELECT(0, 4);
			
 
				+
			
 
				+#else			/* MIPS64 hard-float */
			
 
				+
			
 
				+#define CCALL_HANDLE_COMPLEXRET2 \
			
 
				+  if (ctr->size == 2*sizeof(float)) {  /* Copy complex float from FPRs. */ \
			
 
				+    ((float *)dp)[0] = cc->fpr[0].f; \
			
 
				+    ((float *)dp)[1] = cc->fpr[1].f; \
			
 
				+  } else {  /* Copy complex double from FPRs. */ \
			
 
				+    ((double *)dp)[0] = cc->fpr[0].d; \
			
 
				+    ((double *)dp)[1] = cc->fpr[1].d; \
			
 
				+  }
			
 
				+
			
 
				+#define CCALL_HANDLE_COMPLEXARG \
			
 
				+  if (sz == 2*sizeof(float)) { \
			
 
				+    isfp = 2; \
			
 
				+    if (ngpr < maxgpr) \
			
 
				+      sz *= 2; \
			
 
				   }
			
 
				 
			
 
				 #define CCALL_HANDLE_RET \
			
 
				   if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				     sp = (uint8_t *)&cc->fpr[0].f;
			
 
				 
			
 
				+#endif
			
 
				+
			
 
				+#define CCALL_HANDLE_STRUCTARG \
			
 
				+  /* Pass all structs by value in registers and/or on the stack. */
			
 
				+
			
 
				+#define CCALL_HANDLE_REGARG \
			
 
				+  if (ngpr < maxgpr) { \
			
 
				+    dp = &cc->gpr[ngpr]; \
			
 
				+    if (ngpr + n > maxgpr) { \
			
 
				+      nsp += ngpr + n - maxgpr;  /* Assumes contiguous gpr/stack fields. */ \
			
 
				+      if (nsp > CCALL_MAXSTACK) goto err_nyi;  /* Too many arguments. */ \
			
 
				+      ngpr = maxgpr; \
			
 
				+    } else { \
			
 
				+      ngpr += n; \
			
 
				+    } \
			
 
				+    goto done; \
			
 
				+  }
			
 
				+
			
 
				 #else
			
 
				 #error "Missing calling convention definitions for this architecture"
			
 
				 #endif
			
@@ -621,6 +765,125 @@ noth:  /* Not a homogeneous float/double aggregate. */
 
				 
			
 
				 #endif
			
 
				 
			
 
				+/* -- ARM64 ABI struct classification ------------------------------------- */
			
 
				+
			
 
				+#if LJ_TARGET_ARM64
			
 
				+
			
 
				+/* Classify a struct based on its fields. */
			
 
				+static unsigned int ccall_classify_struct(CTState *cts, CType *ct)
			
 
				+{
			
 
				+  CTSize sz = ct->size;
			
 
				+  unsigned int r = 0, n = 0, isu = (ct->info & CTF_UNION);
			
 
				+  while (ct->sib) {
			
 
				+    CType *sct;
			
 
				+    ct = ctype_get(cts, ct->sib);
			
 
				+    if (ctype_isfield(ct->info)) {
			
 
				+      sct = ctype_rawchild(cts, ct);
			
 
				+      if (ctype_isfp(sct->info)) {
			
 
				+	r |= sct->size;
			
 
				+	if (!isu) n++; else if (n == 0) n = 1;
			
 
				+      } else if (ctype_iscomplex(sct->info)) {
			
 
				+	r |= (sct->size >> 1);
			
 
				+	if (!isu) n += 2; else if (n < 2) n = 2;
			
 
				+      } else if (ctype_isstruct(sct->info)) {
			
 
				+	goto substruct;
			
 
				+      } else {
			
 
				+	goto noth;
			
 
				+      }
			
 
				+    } else if (ctype_isbitfield(ct->info)) {
			
 
				+      goto noth;
			
 
				+    } else if (ctype_isxattrib(ct->info, CTA_SUBTYPE)) {
			
 
				+      sct = ctype_rawchild(cts, ct);
			
 
				+    substruct:
			
 
				+      if (sct->size > 0) {
			
 
				+	unsigned int s = ccall_classify_struct(cts, sct);
			
 
				+	if (s <= 1) goto noth;
			
 
				+	r |= (s & 255);
			
 
				+	if (!isu) n += (s >> 8); else if (n < (s >>8)) n = (s >> 8);
			
 
				+      }
			
 
				+    }
			
 
				+  }
			
 
				+  if ((r == 4 || r == 8) && n <= 4)
			
 
				+    return r + (n << 8);
			
 
				+noth:  /* Not a homogeneous float/double aggregate. */
			
 
				+  return (sz <= 16);  /* Return structs of size <= 16 in GPRs. */
			
 
				+}
			
 
				+
			
 
				+#endif
			
 
				+
			
 
				+/* -- MIPS64 ABI struct classification ---------------------------- */
			
 
				+
			
 
				+#if LJ_TARGET_MIPS64
			
 
				+
			
 
				+#define FTYPE_FLOAT	1
			
 
				+#define FTYPE_DOUBLE	2
			
 
				+
			
 
				+/* Classify FP fields (max. 2) and their types. */
			
 
				+static unsigned int ccall_classify_struct(CTState *cts, CType *ct, CType *ctf)
			
 
				+{
			
 
				+  int n = 0, ft = 0;
			
 
				+  if ((ctf->info & CTF_VARARG) || (ct->info & CTF_UNION))
			
 
				+    goto noth;
			
 
				+  while (ct->sib) {
			
 
				+    CType *sct;
			
 
				+    ct = ctype_get(cts, ct->sib);
			
 
				+    if (n == 2) {
			
 
				+      goto noth;
			
 
				+    } else if (ctype_isfield(ct->info)) {
			
 
				+      sct = ctype_rawchild(cts, ct);
			
 
				+      if (ctype_isfp(sct->info)) {
			
 
				+	ft |= (sct->size == 4 ? FTYPE_FLOAT : FTYPE_DOUBLE) << 2*n;
			
 
				+	n++;
			
 
				+      } else {
			
 
				+	goto noth;
			
 
				+      }
			
 
				+    } else if (ctype_isbitfield(ct->info) ||
			
 
				+	       ctype_isxattrib(ct->info, CTA_SUBTYPE)) {
			
 
				+      goto noth;
			
 
				+    }
			
 
				+  }
			
 
				+  if (n <= 2)
			
 
				+    return ft;
			
 
				+noth:  /* Not a homogeneous float/double aggregate. */
			
 
				+  return 0;  /* Struct is in GPRs. */
			
 
				+}
			
 
				+
			
 
				+static void ccall_copy_struct(CCallState *cc, CType *ctr, void *dp, void *sp,
			
 
				+			      int ft)
			
 
				+{
			
 
				+  if (LJ_ABI_SOFTFP ? ft :
			
 
				+      ((ft & 3) == FTYPE_FLOAT || (ft >> 2) == FTYPE_FLOAT)) {
			
 
				+    int i, ofs = 0;
			
 
				+    for (i = 0; ft != 0; i++, ft >>= 2) {
			
 
				+      if ((ft & 3) == FTYPE_FLOAT) {
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+	/* The 2nd FP struct result is in CARG1 (gpr[2]) and not CRET2. */
			
 
				+	memcpy((uint8_t *)dp + ofs,
			
 
				+	       (uint8_t *)&cc->gpr[2*i] + LJ_ENDIAN_SELECT(0, 4), 4);
			
 
				+#else
			
 
				+	*(float *)((uint8_t *)dp + ofs) = cc->fpr[i].f;
			
 
				+#endif
			
 
				+	ofs += 4;
			
 
				+      } else {
			
 
				+	ofs = (ofs + 7) & ~7;  /* 64 bit alignment. */
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+	*(intptr_t *)((uint8_t *)dp + ofs) = cc->gpr[2*i];
			
 
				+#else
			
 
				+	*(double *)((uint8_t *)dp + ofs) = cc->fpr[i].d;
			
 
				+#endif
			
 
				+	ofs += 8;
			
 
				+      }
			
 
				+    }
			
 
				+  } else {
			
 
				+#if !LJ_ABI_SOFTFP
			
 
				+    if (ft) sp = (uint8_t *)&cc->fpr[0];
			
 
				+#endif
			
 
				+    memcpy(dp, sp, ctr->size);
			
 
				+  }
			
 
				+}
			
 
				+
			
 
				+#endif
			
 
				+
			
 
				 /* -- Common C call handling ---------------------------------------------- */
			
 
				 
			
 
				 /* Infer the destination CTypeID for a vararg argument. */
			
@@ -788,6 +1051,19 @@ static int ccall_set_args(lua_State *L, CTState *cts, CType *ct,
 
				 	*(int32_t *)dp = d->size == 1 ? (int32_t)*(int8_t *)dp :
			
 
				 					(int32_t)*(int16_t *)dp;
			
 
				     }
			
 
				+#if LJ_TARGET_ARM64 && LJ_BE
			
 
				+    if (isfp && d->size == sizeof(float))
			
 
				+      ((float *)dp)[1] = ((float *)dp)[0];  /* Floats occupy high slot. */
			
 
				+#endif
			
 
				+#if LJ_TARGET_MIPS64 || (LJ_TARGET_ARM64 && LJ_BE)
			
 
				+    if ((ctype_isinteger_or_bool(d->info) || ctype_isenum(d->info)
			
 
				+#if LJ_TARGET_MIPS64
			
 
				+	 || (isfp && nsp == 0)
			
 
				+#endif
			
 
				+	 ) && d->size <= 4) {
			
 
				+      *(int64_t *)dp = (int64_t)*(int32_t *)dp;  /* Sign-extend to 64 bit. */
			
 
				+    }
			
 
				+#endif
			
 
				 #if LJ_TARGET_X64 && LJ_ABI_WIN
			
 
				     if (isva) {  /* Windows/x64 mirrors varargs in both register sets. */
			
 
				       if (nfpr == ngpr)
			
@@ -803,13 +1079,19 @@ static int ccall_set_args(lua_State *L, CTState *cts, CType *ct,
 
				       cc->fpr[nfpr-1].d[0] = cc->fpr[nfpr-2].d[1];  /* Split complex double. */
			
 
				       cc->fpr[nfpr-2].d[1] = 0;
			
 
				     }
			
 
				+#elif LJ_TARGET_ARM64 || (LJ_TARGET_MIPS64 && !LJ_ABI_SOFTFP)
			
 
				+    if (isfp == 2 && (uint8_t *)dp < (uint8_t *)cc->stack) {
			
 
				+      /* Split float HFA or complex float into separate registers. */
			
 
				+      CTSize i = (sz >> 2) - 1;
			
 
				+      do { ((uint64_t *)dp)[i] = ((uint32_t *)dp)[i]; } while (i--);
			
 
				+    }
			
 
				 #else
			
 
				     UNUSED(isfp);
			
 
				 #endif
			
 
				   }
			
 
				   if (fid) lj_err_caller(L, LJ_ERR_FFI_NUMARG);  /* Too few arguments. */
			
 
				 
			
 
				-#if LJ_TARGET_X64 || LJ_TARGET_PPC
			
 
				+#if LJ_TARGET_X64 || (LJ_TARGET_PPC && !LJ_ABI_SOFTFP)
			
 
				   cc->nfpr = nfpr;  /* Required for vararg functions. */
			
 
				 #endif
			
 
				   cc->nsp = nsp;
			
@@ -844,7 +1126,8 @@ static int ccall_get_results(lua_State *L, CTState *cts, CType *ct,
 
				     CCALL_HANDLE_COMPLEXRET2
			
 
				     return 1;  /* One GC step. */
			
 
				   }
			
 
				-  if (LJ_BE && ctype_isinteger_or_bool(ctr->info) && ctr->size < CTSIZE_PTR)
			
 
				+  if (LJ_BE && ctr->size < CTSIZE_PTR &&
			
 
				+      (ctype_isinteger_or_bool(ctr->info) || ctype_isenum(ctr->info)))
			
 
				     sp += (CTSIZE_PTR - ctr->size);
			
 
				 #if CCALL_NUM_FPR
			
 
				   if (ctype_isfp(ctr->info) || ctype_isvector(ctr->info))
			
--- a/luajit.mod/luajit/src/lj_ccall.h
+++ b/luajit.mod/luajit/src/lj_ccall.h
@@ -68,35 +68,56 @@ typedef union FPRArg {
 
				   float f[2];
			
 
				 } FPRArg;
			
 
				 
			
 
				-#elif LJ_TARGET_PPC
			
 
				+#elif LJ_TARGET_ARM64
			
 
				 
			
 
				 #define CCALL_NARG_GPR		8
			
 
				+#define CCALL_NRET_GPR		2
			
 
				 #define CCALL_NARG_FPR		8
			
 
				+#define CCALL_NRET_FPR		4
			
 
				+#define CCALL_SPS_FREE		0
			
 
				+
			
 
				+typedef intptr_t GPRArg;
			
 
				+typedef union FPRArg {
			
 
				+  double d;
			
 
				+  struct { LJ_ENDIAN_LOHI(float f; , float g;) };
			
 
				+  struct { LJ_ENDIAN_LOHI(uint32_t lo; , uint32_t hi;) };
			
 
				+} FPRArg;
			
 
				+
			
 
				+#elif LJ_TARGET_PPC
			
 
				+
			
 
				+#define CCALL_NARG_GPR		8
			
 
				+#define CCALL_NARG_FPR		(LJ_ABI_SOFTFP ? 0 : 8)
			
 
				 #define CCALL_NRET_GPR		4	/* For complex double. */
			
 
				-#define CCALL_NRET_FPR		1
			
 
				+#define CCALL_NRET_FPR		(LJ_ABI_SOFTFP ? 0 : 1)
			
 
				 #define CCALL_SPS_EXTRA		4
			
 
				 #define CCALL_SPS_FREE		0
			
 
				 
			
 
				 typedef intptr_t GPRArg;
			
 
				 typedef double FPRArg;
			
 
				 
			
 
				-#elif LJ_TARGET_PPCSPE
			
 
				+#elif LJ_TARGET_MIPS32
			
 
				 
			
 
				-#define CCALL_NARG_GPR		8
			
 
				-#define CCALL_NARG_FPR		0
			
 
				-#define CCALL_NRET_GPR		4	/* For softfp complex double. */
			
 
				-#define CCALL_NRET_FPR		0
			
 
				-#define CCALL_SPS_FREE		0	/* NYI */
			
 
				+#define CCALL_NARG_GPR		4
			
 
				+#define CCALL_NARG_FPR		(LJ_ABI_SOFTFP ? 0 : 2)
			
 
				+#define CCALL_NRET_GPR		(LJ_ABI_SOFTFP ? 4 : 2)
			
 
				+#define CCALL_NRET_FPR		(LJ_ABI_SOFTFP ? 0 : 2)
			
 
				+#define CCALL_SPS_EXTRA		7
			
 
				+#define CCALL_SPS_FREE		1
			
 
				 
			
 
				 typedef intptr_t GPRArg;
			
 
				+typedef union FPRArg {
			
 
				+  double d;
			
 
				+  struct { LJ_ENDIAN_LOHI(float f; , float g;) };
			
 
				+} FPRArg;
			
 
				 
			
 
				-#elif LJ_TARGET_MIPS
			
 
				+#elif LJ_TARGET_MIPS64
			
 
				 
			
 
				-#define CCALL_NARG_GPR		4
			
 
				-#define CCALL_NARG_FPR		2
			
 
				+/* FP args are positional and overlay the GPR array. */
			
 
				+#define CCALL_NARG_GPR		8
			
 
				+#define CCALL_NARG_FPR		0
			
 
				 #define CCALL_NRET_GPR		2
			
 
				-#define CCALL_NRET_FPR		2
			
 
				-#define CCALL_SPS_EXTRA		7
			
 
				+#define CCALL_NRET_FPR		(LJ_ABI_SOFTFP ? 0 : 2)
			
 
				+#define CCALL_SPS_EXTRA		3
			
 
				 #define CCALL_SPS_FREE		1
			
 
				 
			
 
				 typedef intptr_t GPRArg;
			
@@ -145,6 +166,8 @@ typedef LJ_ALIGN(CCALL_ALIGN_CALLSTATE) struct CCallState {
 
				   uint8_t nfpr;			/* Number of arguments in FPRs. */
			
 
				 #elif LJ_TARGET_X86
			
 
				   uint8_t resx87;		/* Result on x87 stack: 1:float, 2:double. */
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+  void *retp;			/* Aggregate return pointer in x8. */
			
 
				 #elif LJ_TARGET_PPC
			
 
				   uint8_t nfpr;			/* Number of arguments in FPRs. */
			
 
				 #endif
			
--- a/luajit.mod/luajit/src/lj_ccallback.c
+++ b/luajit.mod/luajit/src/lj_ccallback.c
@@ -27,7 +27,7 @@
 
				 
			
 
				 #if LJ_OS_NOJIT
			
 
				 
			
 
				-/* Disabled callback support. */
			
 
				+/* Callbacks disabled. */
			
 
				 #define CALLBACK_SLOT2OFS(slot)	(0*(slot))
			
 
				 #define CALLBACK_OFS2SLOT(ofs)	(0*(ofs))
			
 
				 #define CALLBACK_MAX_SLOT	0
			
@@ -35,7 +35,7 @@
 
				 #elif LJ_TARGET_X86ORX64
			
 
				 
			
 
				 #define CALLBACK_MCODE_HEAD	(LJ_64 ? 8 : 0)
			
 
				-#define CALLBACK_MCODE_GROUP	(-2+1+2+5+(LJ_64 ? 6 : 5))
			
 
				+#define CALLBACK_MCODE_GROUP	(-2+1+2+(LJ_GC64 ? 10 : 5)+(LJ_64 ? 6 : 5))
			
 
				 
			
 
				 #define CALLBACK_SLOT2OFS(slot) \
			
 
				   (CALLBACK_MCODE_HEAD + CALLBACK_MCODE_GROUP*((slot)/32) + 4*(slot))
			
@@ -54,23 +54,22 @@ static MSize CALLBACK_OFS2SLOT(MSize ofs)
 
				 #elif LJ_TARGET_ARM
			
 
				 
			
 
				 #define CALLBACK_MCODE_HEAD		32
			
 
				-#define CALLBACK_SLOT2OFS(slot)		(CALLBACK_MCODE_HEAD + 8*(slot))
			
 
				-#define CALLBACK_OFS2SLOT(ofs)		(((ofs)-CALLBACK_MCODE_HEAD)/8)
			
 
				-#define CALLBACK_MAX_SLOT		(CALLBACK_OFS2SLOT(CALLBACK_MCODE_SIZE))
			
 
				+
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+
			
 
				+#define CALLBACK_MCODE_HEAD		32
			
 
				 
			
 
				 #elif LJ_TARGET_PPC
			
 
				 
			
 
				 #define CALLBACK_MCODE_HEAD		24
			
 
				-#define CALLBACK_SLOT2OFS(slot)		(CALLBACK_MCODE_HEAD + 8*(slot))
			
 
				-#define CALLBACK_OFS2SLOT(ofs)		(((ofs)-CALLBACK_MCODE_HEAD)/8)
			
 
				-#define CALLBACK_MAX_SLOT		(CALLBACK_OFS2SLOT(CALLBACK_MCODE_SIZE))
			
 
				 
			
 
				-#elif LJ_TARGET_MIPS
			
 
				+#elif LJ_TARGET_MIPS32
			
 
				 
			
 
				-#define CALLBACK_MCODE_HEAD		24
			
 
				-#define CALLBACK_SLOT2OFS(slot)		(CALLBACK_MCODE_HEAD + 8*(slot))
			
 
				-#define CALLBACK_OFS2SLOT(ofs)		(((ofs)-CALLBACK_MCODE_HEAD)/8)
			
 
				-#define CALLBACK_MAX_SLOT		(CALLBACK_OFS2SLOT(CALLBACK_MCODE_SIZE))
			
 
				+#define CALLBACK_MCODE_HEAD		20
			
 
				+
			
 
				+#elif LJ_TARGET_MIPS64
			
 
				+
			
 
				+#define CALLBACK_MCODE_HEAD		52
			
 
				 
			
 
				 #else
			
 
				 
			
@@ -81,6 +80,12 @@ static MSize CALLBACK_OFS2SLOT(MSize ofs)
 
				 
			
 
				 #endif
			
 
				 
			
 
				+#ifndef CALLBACK_SLOT2OFS
			
 
				+#define CALLBACK_SLOT2OFS(slot)		(CALLBACK_MCODE_HEAD + 8*(slot))
			
 
				+#define CALLBACK_OFS2SLOT(ofs)		(((ofs)-CALLBACK_MCODE_HEAD)/8)
			
 
				+#define CALLBACK_MAX_SLOT		(CALLBACK_OFS2SLOT(CALLBACK_MCODE_SIZE))
			
 
				+#endif
			
 
				+
			
 
				 /* Convert callback slot number to callback function pointer. */
			
 
				 static void *callback_slot2ptr(CTState *cts, MSize slot)
			
 
				 {
			
@@ -119,8 +124,13 @@ static void callback_mcode_init(global_State *g, uint8_t *page)
 
				       /* push ebp/rbp; mov ah, slot>>8; mov ebp, &g. */
			
 
				       *p++ = XI_PUSH + RID_EBP;
			
 
				       *p++ = XI_MOVrib | (RID_EAX+4); *p++ = (uint8_t)(slot >> 8);
			
 
				+#if LJ_GC64
			
 
				+      *p++ = 0x48; *p++ = XI_MOVri | RID_EBP;
			
 
				+      *(uint64_t *)p = (uint64_t)(g); p += 8;
			
 
				+#else
			
 
				       *p++ = XI_MOVri | RID_EBP;
			
 
				       *(int32_t *)p = i32ptr(g); p += 4;
			
 
				+#endif
			
 
				 #if LJ_64
			
 
				       /* jmp [rip-pageofs] where lj_vm_ffi_callback is stored. */
			
 
				       *p++ = XI_GROUP5; *p++ = XM_OFS0 + (XOg_JMP<<3) + RID_EBP;
			
@@ -157,6 +167,26 @@ static void callback_mcode_init(global_State *g, uint32_t *page)
 
				   }
			
 
				   lua_assert(p - page <= CALLBACK_MCODE_SIZE);
			
 
				 }
			
 
				+#elif LJ_TARGET_ARM64
			
 
				+static void callback_mcode_init(global_State *g, uint32_t *page)
			
 
				+{
			
 
				+  uint32_t *p = page;
			
 
				+  void *target = (void *)lj_vm_ffi_callback;
			
 
				+  MSize slot;
			
 
				+  *p++ = A64I_LE(A64I_LDRLx | A64F_D(RID_X11) | A64F_S19(4));
			
 
				+  *p++ = A64I_LE(A64I_LDRLx | A64F_D(RID_X10) | A64F_S19(5));
			
 
				+  *p++ = A64I_LE(A64I_BR | A64F_N(RID_X11));
			
 
				+  *p++ = A64I_LE(A64I_NOP);
			
 
				+  ((void **)p)[0] = target;
			
 
				+  ((void **)p)[1] = g;
			
 
				+  p += 4;
			
 
				+  for (slot = 0; slot < CALLBACK_MAX_SLOT; slot++) {
			
 
				+    *p++ = A64I_LE(A64I_MOVZw | A64F_D(RID_X9) | A64F_U16(slot));
			
 
				+    *p = A64I_LE(A64I_B | A64F_S26((page-p) & 0x03ffffffu));
			
 
				+    p++;
			
 
				+  }
			
 
				+  lua_assert(p - page <= CALLBACK_MCODE_SIZE);
			
 
				+}
			
 
				 #elif LJ_TARGET_PPC
			
 
				 static void callback_mcode_init(global_State *g, uint32_t *page)
			
 
				 {
			
@@ -180,14 +210,27 @@ static void callback_mcode_init(global_State *g, uint32_t *page)
 
				 static void callback_mcode_init(global_State *g, uint32_t *page)
			
 
				 {
			
 
				   uint32_t *p = page;
			
 
				-  void *target = (void *)lj_vm_ffi_callback;
			
 
				+  uintptr_t target = (uintptr_t)(void *)lj_vm_ffi_callback;
			
 
				+  uintptr_t ug = (uintptr_t)(void *)g;
			
 
				   MSize slot;
			
 
				-  *p++ = MIPSI_SW | MIPSF_T(RID_R1)|MIPSF_S(RID_SP) | 0;
			
 
				-  *p++ = MIPSI_LUI | MIPSF_T(RID_R3) | (u32ptr(target) >> 16);
			
 
				-  *p++ = MIPSI_LUI | MIPSF_T(RID_R2) | (u32ptr(g) >> 16);
			
 
				-  *p++ = MIPSI_ORI | MIPSF_T(RID_R3)|MIPSF_S(RID_R3) |(u32ptr(target)&0xffff);
			
 
				+#if LJ_TARGET_MIPS32
			
 
				+  *p++ = MIPSI_LUI | MIPSF_T(RID_R3) | (target >> 16);
			
 
				+  *p++ = MIPSI_LUI | MIPSF_T(RID_R2) | (ug >> 16);
			
 
				+#else
			
 
				+  *p++ = MIPSI_LUI  | MIPSF_T(RID_R3) | (target >> 48);
			
 
				+  *p++ = MIPSI_LUI  | MIPSF_T(RID_R2) | (ug >> 48);
			
 
				+  *p++ = MIPSI_ORI  | MIPSF_T(RID_R3)|MIPSF_S(RID_R3) | ((target >> 32) & 0xffff);
			
 
				+  *p++ = MIPSI_ORI  | MIPSF_T(RID_R2)|MIPSF_S(RID_R2) | ((ug >> 32) & 0xffff);
			
 
				+  *p++ = MIPSI_DSLL | MIPSF_D(RID_R3)|MIPSF_T(RID_R3) | MIPSF_A(16);
			
 
				+  *p++ = MIPSI_DSLL | MIPSF_D(RID_R2)|MIPSF_T(RID_R2) | MIPSF_A(16);
			
 
				+  *p++ = MIPSI_ORI  | MIPSF_T(RID_R3)|MIPSF_S(RID_R3) | ((target >> 16) & 0xffff);
			
 
				+  *p++ = MIPSI_ORI  | MIPSF_T(RID_R2)|MIPSF_S(RID_R2) | ((ug >> 16) & 0xffff);
			
 
				+  *p++ = MIPSI_DSLL | MIPSF_D(RID_R3)|MIPSF_T(RID_R3) | MIPSF_A(16);
			
 
				+  *p++ = MIPSI_DSLL | MIPSF_D(RID_R2)|MIPSF_T(RID_R2) | MIPSF_A(16);
			
 
				+#endif
			
 
				+  *p++ = MIPSI_ORI  | MIPSF_T(RID_R3)|MIPSF_S(RID_R3) | (target & 0xffff);
			
 
				   *p++ = MIPSI_JR | MIPSF_S(RID_R3);
			
 
				-  *p++ = MIPSI_ORI | MIPSF_T(RID_R2)|MIPSF_S(RID_R2) | (u32ptr(g)&0xffff);
			
 
				+  *p++ = MIPSI_ORI | MIPSF_T(RID_R2)|MIPSF_S(RID_R2) | (ug & 0xffff);
			
 
				   for (slot = 0; slot < CALLBACK_MAX_SLOT; slot++) {
			
 
				     *p = MIPSI_B | ((page-p-1) & 0x0000ffffu);
			
 
				     p++;
			
@@ -224,7 +267,7 @@ static void callback_mcode_new(CTState *cts)
 
				   if (CALLBACK_MAX_SLOT == 0)
			
 
				     lj_err_caller(cts->L, LJ_ERR_FFI_CBACKOV);
			
 
				 #if LJ_TARGET_WINDOWS
			
 
				-  p = VirtualAlloc(NULL, sz, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
			
 
				+  p = LJ_WIN_VALLOC(NULL, sz, MEM_RESERVE|MEM_COMMIT, PAGE_READWRITE);
			
 
				   if (!p)
			
 
				     lj_err_caller(cts->L, LJ_ERR_FFI_CBACKOV);
			
 
				 #elif LJ_TARGET_POSIX
			
@@ -242,7 +285,7 @@ static void callback_mcode_new(CTState *cts)
 
				 #if LJ_TARGET_WINDOWS
			
 
				   {
			
 
				     DWORD oprot;
			
 
				-    VirtualProtect(p, sz, PAGE_EXECUTE_READ, &oprot);
			
 
				+    LJ_WIN_VPROTECT(p, sz, PAGE_EXECUTE_READ, &oprot);
			
 
				   }
			
 
				 #elif LJ_TARGET_POSIX
			
 
				   mprotect(p, sz, (PROT_READ|PROT_EXEC));
			
@@ -351,33 +394,77 @@ void lj_ccallback_mcode_free(CTState *cts)
 
				     goto done; \
			
 
				   } CALLBACK_HANDLE_REGARG_FP2
			
 
				 
			
 
				-#elif LJ_TARGET_PPC
			
 
				+#elif LJ_TARGET_ARM64
			
 
				 
			
 
				 #define CALLBACK_HANDLE_REGARG \
			
 
				   if (isfp) { \
			
 
				-    if (nfpr + 1 <= CCALL_NARG_FPR) { \
			
 
				-      sp = &cts->cb.fpr[nfpr++]; \
			
 
				-      cta = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */ \
			
 
				+    if (nfpr + n <= CCALL_NARG_FPR) { \
			
 
				+      sp = &cts->cb.fpr[nfpr]; \
			
 
				+      nfpr += n; \
			
 
				       goto done; \
			
 
				+    } else { \
			
 
				+      nfpr = CCALL_NARG_FPR;  /* Prevent reordering. */ \
			
 
				     } \
			
 
				-  } else {  /* Try to pass argument in GPRs. */ \
			
 
				-    if (n > 1) { \
			
 
				-      lua_assert(ctype_isinteger(cta->info) && n == 2);  /* int64_t. */ \
			
 
				-      ngpr = (ngpr + 1u) & ~1u;  /* Align int64_t to regpair. */ \
			
 
				-    } \
			
 
				+  } else { \
			
 
				+    if (!LJ_TARGET_IOS && n > 1) \
			
 
				+      ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				     if (ngpr + n <= maxgpr) { \
			
 
				       sp = &cts->cb.gpr[ngpr]; \
			
 
				       ngpr += n; \
			
 
				       goto done; \
			
 
				+    } else { \
			
 
				+      ngpr = CCALL_NARG_GPR;  /* Prevent reordering. */ \
			
 
				+    } \
			
 
				+  }
			
 
				+
			
 
				+#elif LJ_TARGET_PPC
			
 
				+
			
 
				+#define CALLBACK_HANDLE_GPR \
			
 
				+  if (n > 1) { \
			
 
				+    lua_assert(((LJ_ABI_SOFTFP && ctype_isnum(cta->info)) ||  /* double. */ \
			
 
				+		ctype_isinteger(cta->info)) && n == 2);  /* int64_t. */ \
			
 
				+    ngpr = (ngpr + 1u) & ~1u;  /* Align int64_t to regpair. */ \
			
 
				+  } \
			
 
				+  if (ngpr + n <= maxgpr) { \
			
 
				+    sp = &cts->cb.gpr[ngpr]; \
			
 
				+    ngpr += n; \
			
 
				+    goto done; \
			
 
				+  }
			
 
				+
			
 
				+#if LJ_ABI_SOFTFP
			
 
				+#define CALLBACK_HANDLE_REGARG \
			
 
				+  CALLBACK_HANDLE_GPR \
			
 
				+  UNUSED(isfp);
			
 
				+#else
			
 
				+#define CALLBACK_HANDLE_REGARG \
			
 
				+  if (isfp) { \
			
 
				+    if (nfpr + 1 <= CCALL_NARG_FPR) { \
			
 
				+      sp = &cts->cb.fpr[nfpr++]; \
			
 
				+      cta = ctype_get(cts, CTID_DOUBLE);  /* FPRs always hold doubles. */ \
			
 
				+      goto done; \
			
 
				     } \
			
 
				+  } else {  /* Try to pass argument in GPRs. */ \
			
 
				+    CALLBACK_HANDLE_GPR \
			
 
				   }
			
 
				+#endif
			
 
				 
			
 
				+#if !LJ_ABI_SOFTFP
			
 
				 #define CALLBACK_HANDLE_RET \
			
 
				   if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				     *(double *)dp = *(float *)dp;  /* FPRs always hold doubles. */
			
 
				+#endif
			
 
				 
			
 
				-#elif LJ_TARGET_MIPS
			
 
				+#elif LJ_TARGET_MIPS32
			
 
				 
			
 
				+#define CALLBACK_HANDLE_GPR \
			
 
				+  if (n > 1) ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				+  if (ngpr + n <= maxgpr) { \
			
 
				+    sp = &cts->cb.gpr[ngpr]; \
			
 
				+    ngpr += n; \
			
 
				+    goto done; \
			
 
				+  }
			
 
				+
			
 
				+#if !LJ_ABI_SOFTFP	/* MIPS32 hard-float */
			
 
				 #define CALLBACK_HANDLE_REGARG \
			
 
				   if (isfp && nfpr < CCALL_NARG_FPR) {  /* Try to pass argument in FPRs. */ \
			
 
				     sp = (void *)((uint8_t *)&cts->cb.fpr[nfpr] + ((LJ_BE && n==1) ? 4 : 0)); \
			
@@ -385,13 +472,36 @@ void lj_ccallback_mcode_free(CTState *cts)
 
				     goto done; \
			
 
				   } else {  /* Try to pass argument in GPRs. */ \
			
 
				     nfpr = CCALL_NARG_FPR; \
			
 
				-    if (n > 1) ngpr = (ngpr + 1u) & ~1u;  /* Align to regpair. */ \
			
 
				-    if (ngpr + n <= maxgpr) { \
			
 
				-      sp = &cts->cb.gpr[ngpr]; \
			
 
				-      ngpr += n; \
			
 
				-      goto done; \
			
 
				-    } \
			
 
				+    CALLBACK_HANDLE_GPR \
			
 
				+  }
			
 
				+#else			/* MIPS32 soft-float */
			
 
				+#define CALLBACK_HANDLE_REGARG \
			
 
				+  CALLBACK_HANDLE_GPR \
			
 
				+  UNUSED(isfp);
			
 
				+#endif
			
 
				+
			
 
				+#define CALLBACK_HANDLE_RET \
			
 
				+  if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
 
				+    ((float *)dp)[1] = *(float *)dp;
			
 
				+
			
 
				+#elif LJ_TARGET_MIPS64
			
 
				+
			
 
				+#if !LJ_ABI_SOFTFP	/* MIPS64 hard-float */
			
 
				+#define CALLBACK_HANDLE_REGARG \
			
 
				+  if (ngpr + n <= maxgpr) { \
			
 
				+    sp = isfp ? (void*) &cts->cb.fpr[ngpr] : (void*) &cts->cb.gpr[ngpr]; \
			
 
				+    ngpr += n; \
			
 
				+    goto done; \
			
 
				   }
			
 
				+#else			/* MIPS64 soft-float */
			
 
				+#define CALLBACK_HANDLE_REGARG \
			
 
				+  if (ngpr + n <= maxgpr) { \
			
 
				+    UNUSED(isfp); \
			
 
				+    sp = (void*) &cts->cb.gpr[ngpr]; \
			
 
				+    ngpr += n; \
			
 
				+    goto done; \
			
 
				+  }
			
 
				+#endif
			
 
				 
			
 
				 #define CALLBACK_HANDLE_RET \
			
 
				   if (ctype_isfp(ctr->info) && ctr->size == sizeof(float)) \
			
@@ -411,6 +521,7 @@ static void callback_conv_args(CTState *cts, lua_State *L)
 
				   int gcsteps = 0;
			
 
				   CType *ct;
			
 
				   GCfunc *fn;
			
 
				+  int fntp;
			
 
				   MSize ngpr = 0, nsp = 0, maxgpr = CCALL_NARG_GPR;
			
 
				 #if CCALL_NARG_FPR
			
 
				   MSize nfpr = 0;
			
@@ -421,18 +532,27 @@ static void callback_conv_args(CTState *cts, lua_State *L)
 
				 
			
 
				   if (slot < cts->cb.sizeid && (id = cts->cb.cbid[slot]) != 0) {
			
 
				     ct = ctype_get(cts, id);
			
 
				-    rid = ctype_cid(ct->info);
			
 
				+    rid = ctype_cid(ct->info);  /* Return type. x86: +(spadj<<16). */
			
 
				     fn = funcV(lj_tab_getint(cts->miscmap, (int32_t)slot));
			
 
				+    fntp = LJ_TFUNC;
			
 
				   } else {  /* Must set up frame first, before throwing the error. */
			
 
				     ct = NULL;
			
 
				     rid = 0;
			
 
				     fn = (GCfunc *)L;
			
 
				+    fntp = LJ_TTHREAD;
			
 
				+  }
			
 
				+  /* Continuation returns from callback. */
			
 
				+  if (LJ_FR2) {
			
 
				+    (o++)->u64 = LJ_CONT_FFI_CALLBACK;
			
 
				+    (o++)->u64 = rid;
			
 
				+    o++;
			
 
				+  } else {
			
 
				+    o->u32.lo = LJ_CONT_FFI_CALLBACK;
			
 
				+    o->u32.hi = rid;
			
 
				+    o++;
			
 
				   }
			
 
				-  o->u32.lo = LJ_CONT_FFI_CALLBACK;  /* Continuation returns from callback. */
			
 
				-  o->u32.hi = rid;  /* Return type. x86: +(spadj<<16). */
			
 
				-  o++;
			
 
				-  setframe_gc(o, obj2gco(fn));
			
 
				-  setframe_ftsz(o, (int)((char *)(o+1) - (char *)L->base) + FRAME_CONT);
			
 
				+  setframe_gc(o, obj2gco(fn), fntp);
			
 
				+  setframe_ftsz(o, ((char *)(o+1) - (char *)L->base) + FRAME_CONT);
			
 
				   L->top = L->base = ++o;
			
 
				   if (!ct)
			
 
				     lj_err_caller(cts->L, LJ_ERR_FFI_BADCBACK);
			
@@ -474,7 +594,11 @@ static void callback_conv_args(CTState *cts, lua_State *L)
 
				       nsp += n;
			
 
				 
			
 
				     done:
			
 
				-      if (LJ_BE && cta->size < CTSIZE_PTR)
			
 
				+      if (LJ_BE && cta->size < CTSIZE_PTR
			
 
				+#if LJ_TARGET_MIPS64
			
 
				+	  && !(isfp && nsp)
			
 
				+#endif
			
 
				+	 )
			
 
				 	sp = (void *)((uint8_t *)sp + CTSIZE_PTR-cta->size);
			
 
				       gcsteps += lj_cconv_tv_ct(cts, cta, 0, o++, sp);
			
 
				     }
			
@@ -483,8 +607,13 @@ static void callback_conv_args(CTState *cts, lua_State *L)
 
				   L->top = o;
			
 
				 #if LJ_TARGET_X86
			
 
				   /* Store stack adjustment for returns from non-cdecl callbacks. */
			
 
				-  if (ctype_cconv(ct->info) != CTCC_CDECL)
			
 
				+  if (ctype_cconv(ct->info) != CTCC_CDECL) {
			
 
				+#if LJ_FR2
			
 
				+    (L->base-3)->u64 |= (nsp << (16+2));
			
 
				+#else
			
 
				     (L->base-2)->u32.hi |= (nsp << (16+2));
			
 
				+#endif
			
 
				+  }
			
 
				 #endif
			
 
				   while (gcsteps-- > 0)
			
 
				     lj_gc_check(L);
			
@@ -493,7 +622,11 @@ static void callback_conv_args(CTState *cts, lua_State *L)
 
				 /* Convert Lua object to callback result. */
			
 
				 static void callback_conv_result(CTState *cts, lua_State *L, TValue *o)
			
 
				 {
			
 
				+#if LJ_FR2
			
 
				+  CType *ctr = ctype_raw(cts, (uint16_t)(L->base-3)->u64);
			
 
				+#else
			
 
				   CType *ctr = ctype_raw(cts, (uint16_t)(L->base-2)->u32.hi);
			
 
				+#endif
			
 
				 #if LJ_TARGET_X86
			
 
				   cts->cb.gpr[2] = 0;
			
 
				 #endif
			
@@ -502,6 +635,10 @@ static void callback_conv_result(CTState *cts, lua_State *L, TValue *o)
 
				 #if CCALL_NUM_FPR
			
 
				     if (ctype_isfp(ctr->info))
			
 
				       dp = (uint8_t *)&cts->cb.fpr[0];
			
 
				+#endif
			
 
				+#if LJ_TARGET_ARM64 && LJ_BE
			
 
				+    if (ctype_isfp(ctr->info) && ctr->size == sizeof(float))
			
 
				+      dp = (uint8_t *)&cts->cb.fpr[0].f[1];
			
 
				 #endif
			
 
				     lj_cconv_ct_tv(cts, ctr, dp, o, 0);
			
 
				 #ifdef CALLBACK_HANDLE_RET
			
@@ -516,6 +653,12 @@ static void callback_conv_result(CTState *cts, lua_State *L, TValue *o)
 
				 	*(int32_t *)dp = ctr->size == 1 ? (int32_t)*(int8_t *)dp :
			
 
				 					  (int32_t)*(int16_t *)dp;
			
 
				     }
			
 
				+#if LJ_TARGET_MIPS64 || (LJ_TARGET_ARM64 && LJ_BE)
			
 
				+    /* Always sign-extend results to 64 bits. Even a soft-fp 'float'. */
			
 
				+    if (ctr->size <= 4 &&
			
 
				+	(LJ_ABI_SOFTFP || ctype_isinteger_or_bool(ctr->info)))
			
 
				+      *(int64_t *)dp = (int64_t)*(int32_t *)dp;
			
 
				+#endif
			
 
				 #if LJ_TARGET_X86
			
 
				     if (ctype_isfp(ctr->info))
			
 
				       cts->cb.gpr[2] = ctr->size == sizeof(float) ? 1 : 2;
			
@@ -529,7 +672,7 @@ lua_State * LJ_FASTCALL lj_ccallback_enter(CTState *cts, void *cf)
 
				   lua_State *L = cts->L;
			
 
				   global_State *g = cts->g;
			
 
				   lua_assert(L != NULL);
			
 
				-  if (gcref(g->jit_L)) {
			
 
				+  if (tvref(g->jit_base)) {
			
 
				     setstrV(L, L->top++, lj_err_str(L, LJ_ERR_FFI_BADCBACK));
			
 
				     if (g->panic) g->panic(L);
			
 
				     exit(EXIT_FAILURE);
			
@@ -562,9 +705,9 @@ void LJ_FASTCALL lj_ccallback_leave(CTState *cts, TValue *o)
 
				   }
			
 
				   callback_conv_result(cts, L, o);
			
 
				   /* Finally drop C frame and continuation frame. */
			
 
				-  L->cframe = cframe_prev(L->cframe);
			
 
				-  L->top -= 2;
			
 
				+  L->top -= 2+2*LJ_FR2;
			
 
				   L->base = obase;
			
 
				+  L->cframe = cframe_prev(L->cframe);
			
 
				   cts->cb.slot = 0;  /* Blacklist C function that called the callback. */
			
 
				 }
			
 
				 
			
--- a/luajit.mod/luajit/src/lj_cconv.c
+++ b/luajit.mod/luajit/src/lj_cconv.c
@@ -448,8 +448,10 @@ int lj_cconv_tv_bf(CTState *cts, CType *s, TValue *o, uint8_t *sp)
 
				 	setintV(o, (int32_t)val);
			
 
				     }
			
 
				   } else {
			
 
				+    uint32_t b = (val >> pos) & 1;
			
 
				     lua_assert(bsz == 1);
			
 
				-    setboolV(o, (val >> pos) & 1);
			
 
				+    setboolV(o, b);
			
 
				+    setboolV(&cts->g->tmptv2, b);  /* Remember for trace recorder. */
			
 
				   }
			
 
				   return 0;  /* No GC step needed. */
			
 
				 }
			
--- a/luajit.mod/luajit/src/lj_cdata.c
+++ b/luajit.mod/luajit/src/lj_cdata.c
@@ -9,7 +9,6 @@
 
				 
			
 
				 #include "lj_gc.h"
			
 
				 #include "lj_err.h"
			
 
				-#include "lj_str.h"
			
 
				 #include "lj_tab.h"
			
 
				 #include "lj_ctype.h"
			
 
				 #include "lj_cconv.h"
			
@@ -27,12 +26,12 @@ GCcdata *lj_cdata_newref(CTState *cts, const void *p, CTypeID id)
 
				 }
			
 
				 
			
 
				 /* Allocate variable-sized or specially aligned C data object. */
			
 
				-GCcdata *lj_cdata_newv(CTState *cts, CTypeID id, CTSize sz, CTSize align)
			
 
				+GCcdata *lj_cdata_newv(lua_State *L, CTypeID id, CTSize sz, CTSize align)
			
 
				 {
			
 
				   global_State *g;
			
 
				   MSize extra = sizeof(GCcdataVar) + sizeof(GCcdata) +
			
 
				 		(align > CT_MEMALIGN ? (1u<<align) - (1u<<CT_MEMALIGN) : 0);
			
 
				-  char *p = lj_mem_newt(cts->L, extra + sz, char);
			
 
				+  char *p = lj_mem_newt(L, extra + sz, char);
			
 
				   uintptr_t adata = (uintptr_t)p + sizeof(GCcdataVar) + sizeof(GCcdata);
			
 
				   uintptr_t almask = (1u << align) - 1u;
			
 
				   GCcdata *cd = (GCcdata *)(((adata + almask) & ~almask) - sizeof(GCcdata));
			
@@ -40,7 +39,7 @@ GCcdata *lj_cdata_newv(CTState *cts, CTypeID id, CTSize sz, CTSize align)
 
				   cdatav(cd)->offset = (uint16_t)((char *)cd - p);
			
 
				   cdatav(cd)->extra = extra;
			
 
				   cdatav(cd)->len = sz;
			
 
				-  g = cts->g;
			
 
				+  g = G(L);
			
 
				   setgcrefr(cd->nextgc, g->gc.root);
			
 
				   setgcref(g->gc.root, obj2gco(cd));
			
 
				   newwhite(g, obj2gco(cd));
			
@@ -50,6 +49,15 @@ GCcdata *lj_cdata_newv(CTState *cts, CTypeID id, CTSize sz, CTSize align)
 
				   return cd;
			
 
				 }
			
 
				 
			
 
				+/* Allocate arbitrary C data object. */
			
 
				+GCcdata *lj_cdata_newx(CTState *cts, CTypeID id, CTSize sz, CTInfo info)
			
 
				+{
			
 
				+  if (!(info & CTF_VLA) && ctype_align(info) <= CT_MEMALIGN)
			
 
				+    return lj_cdata_new(cts, id, sz);
			
 
				+  else
			
 
				+    return lj_cdata_newv(cts->L, id, sz, ctype_align(info));
			
 
				+}
			
 
				+
			
 
				 /* Free a C data object. */
			
 
				 void LJ_FASTCALL lj_cdata_free(global_State *g, GCcdata *cd)
			
 
				 {
			
@@ -76,21 +84,22 @@ void LJ_FASTCALL lj_cdata_free(global_State *g, GCcdata *cd)
 
				   }
			
 
				 }
			
 
				 
			
 
				-TValue * LJ_FASTCALL lj_cdata_setfin(lua_State *L, GCcdata *cd)
			
 
				+void lj_cdata_setfin(lua_State *L, GCcdata *cd, GCobj *obj, uint32_t it)
			
 
				 {
			
 
				-  global_State *g = G(L);
			
 
				-  GCtab *t = ctype_ctsG(g)->finalizer;
			
 
				+  GCtab *t = ctype_ctsG(G(L))->finalizer;
			
 
				   if (gcref(t->metatable)) {
			
 
				     /* Add cdata to finalizer table, if still enabled. */
			
 
				     TValue *tv, tmp;
			
 
				     setcdataV(L, &tmp, cd);
			
 
				     lj_gc_anybarriert(L, t);
			
 
				     tv = lj_tab_set(L, t, &tmp);
			
 
				-    cd->marked |= LJ_GC_CDATA_FIN;
			
 
				-    return tv;
			
 
				-  } else {
			
 
				-    /* Otherwise return dummy TValue. */
			
 
				-    return &g->tmptv;
			
 
				+    if (it == LJ_TNIL) {
			
 
				+      setnilV(tv);
			
 
				+      cd->marked &= ~LJ_GC_CDATA_FIN;
			
 
				+    } else {
			
 
				+      setgcV(L, tv, obj, it);
			
 
				+      cd->marked |= LJ_GC_CDATA_FIN;
			
 
				+    }
			
 
				   }
			
 
				 }
			
 
				 
			
@@ -123,7 +132,12 @@ collect_attrib:
 
				     idx = (ptrdiff_t)intV(key);
			
 
				     goto integer_key;
			
 
				   } else if (tvisnum(key)) {  /* Numeric key. */
			
 
				-    idx = LJ_64 ? (ptrdiff_t)numV(key) : (ptrdiff_t)lj_num2int(numV(key));
			
 
				+#ifdef _MSC_VER
			
 
				+    /* Workaround for MSVC bug. */
			
 
				+    volatile
			
 
				+#endif
			
 
				+    lua_Number n = numV(key);
			
 
				+    idx = LJ_64 ? (ptrdiff_t)n : (ptrdiff_t)lj_num2int(n);
			
 
				   integer_key:
			
 
				     if (ctype_ispointer(ct->info)) {
			
 
				       CTSize sz = lj_ctype_size(cts, ctype_cid(ct->info));  /* Element size. */