| .. |
|
aoptcpu.pas
|
2fa066b003
* optimize vmovaps/vmovapd after avx instructions
|
%!s(int64=11) %!d(string=hai) anos |
|
aoptcpub.pas
|
2f5ce095ce
* RefsHaveIndexReg -> cpurefshaveindexreg
|
%!s(int64=13) %!d(string=hai) anos |
|
aoptcpud.pas
|
790a4fe2d3
* log and id tags removed
|
%!s(int64=20) %!d(string=hai) anos |
|
cgcpu.pas
|
61e4a1b811
+ added tasmlist parameter to getintparaloc() (needed for llvm)
|
%!s(int64=10) %!d(string=hai) anos |
|
cpubase.inc
|
bfbb0c5b9d
* optimize mov/lea
|
%!s(int64=12) %!d(string=hai) anos |
|
cpuelf.pas
|
7ca62febcd
AROS/x86_64: initial AROS/x86_64 support and implement syscalls for AROS/x86_64
|
%!s(int64=10) %!d(string=hai) anos |
|
cpuinfo.pas
|
de17e9fa1c
+ cpu capability CPUX86_HAS_CMOV
|
%!s(int64=10) %!d(string=hai) anos |
|
cpunode.pas
|
bacd303208
* synchronized with trunk up to r27758
|
%!s(int64=11) %!d(string=hai) anos |
|
cpupara.pas
|
f40ea04540
* changed getarraydef() into a tarraydef.getreusable() class method
|
%!s(int64=10) %!d(string=hai) anos |
|
cpupi.pas
|
70dda94474
* x86_64-win64: don't allocate outgoing parameter area in nostackframe procedures, it fails compilation if range/overflow/etc checking is enabled (which always sets pi_do_call) due to check introduced in r22677.
|
%!s(int64=12) %!d(string=hai) anos |
|
cputarg.pas
|
7ca62febcd
AROS/x86_64: initial AROS/x86_64 support and implement syscalls for AROS/x86_64
|
%!s(int64=10) %!d(string=hai) anos |
|
hlcgcpu.pas
|
67b8aceaee
* synchronized with privatetrunk till r30095
|
%!s(int64=10) %!d(string=hai) anos |
|
nx64add.pas
|
2459518bdd
* use IMUL even for unsigned multiplication on x86_64, when overflow checking is
|
%!s(int64=11) %!d(string=hai) anos |
|
nx64cal.pas
|
f0ccbc87f2
AROS/x86_64: enable syscalls code in the parser, fix EAX->RAX reference base while calculating syscall offset
|
%!s(int64=10) %!d(string=hai) anos |
|
nx64cnv.pas
|
687bb15299
* renamed getdatalabel() to getglobaldatalabel
|
%!s(int64=10) %!d(string=hai) anos |
|
nx64flw.pas
|
ed4e876f4f
* Generate exception filters data on i386-win32 and x86_64-win64 without using global labels.
|
%!s(int64=10) %!d(string=hai) anos |
|
nx64inl.pas
|
1bcc276dcf
* remove implicit typecast to extended automatically inserted for trunc/round
|
%!s(int64=10) %!d(string=hai) anos |
|
nx64mat.pas
|
b594eee70b
* Moved x86_64 mod/div code to x86, with minimal changes to ensure it compiles on i386/i8086. Merging optimized division-by-const code from i386 is pending...
|
%!s(int64=11) %!d(string=hai) anos |
|
nx64set.pas
|
859676d7d3
* fixed r26519 for darwin/x86-64, see comments (mantis #25644)
|
%!s(int64=11) %!d(string=hai) anos |
|
r8664ari.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664att.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664con.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664dwrf.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664int.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664iri.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664nasm.inc
|
2ec5a649d7
* set Ch_* for more operations
|
%!s(int64=11) %!d(string=hai) anos |
|
r8664nor.inc
|
283ff05127
* merged avx support in inline assembler developed by Torsten Grundke
|
%!s(int64=13) %!d(string=hai) anos |
|
r8664num.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664ot.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664rni.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664sri.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664stab.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
r8664std.inc
|
5af873ee5b
* x86 targets: Profiling shows that quite a bit of time is spent in findreg_by_number(), despite it uses binary search. Worse, it is repeated for every piece of register information. Trying to get rid of some of these calls: rearranged registers so that their "opcode" matches 3 LSBs of superregister number (with a few exceptions described at the beginning of x86reg.dat). This allows to lookup opcodes in regval() with O(1) complexity, and removes need in rXXXop.inc files.
|
%!s(int64=12) %!d(string=hai) anos |
|
rax64att.pas
|
8f05f8c839
- Forgot to commit with r29081
|
%!s(int64=10) %!d(string=hai) anos |
|
rax64int.pas
|
f726e1691b
* Fixed warnings and notes.
|
%!s(int64=16) %!d(string=hai) anos |
|
rgcpu.pas
|
a3f58e84be
* rbp can be used for normal purpose under certain conditions so it shouldn't interfere with all other registers
|
%!s(int64=11) %!d(string=hai) anos |
|
symcpu.pas
|
7ca62febcd
AROS/x86_64: initial AROS/x86_64 support and implement syscalls for AROS/x86_64
|
%!s(int64=10) %!d(string=hai) anos |
|
win64unw.pas
|
6a3fe72de9
+ Support .rva directive in AT&T reader. Put it into base class because it generally applies to all targets with COFF output, but enabled for Windows targets only (others need additional testing).
|
%!s(int64=14) %!d(string=hai) anos |
|
x8664ats.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664att.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664int.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664nop.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664op.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664pro.inc
|
99635658ec
* corrects change flags for VSQRTSD
|
%!s(int64=10) %!d(string=hai) anos |
|
x8664tab.inc
|
d6e4af8279
+ applied remaining patches of Torsten Grundke: adds gather instructions of avx2
|
%!s(int64=10) %!d(string=hai) anos |
Jonas Maebe
1bcc276dcf
* remove implicit typecast to extended automatically inserted for trunc/round