| .. |
|
aoptcpu.pas
|
e7b6a08eae
* x86: New peephole optimisation for improving newly inserted (V)MOVD/(V)MOVQ instructions
|
3 月之前 |
|
aoptcpub.pas
|
9b0ff05ee8
- get rid of MaxOps, it is redundant with max_operands
|
6 年之前 |
|
aoptcpud.pas
|
3c2dab9878
* i386 peephole assembler uses largely the common peephole optimizer infrastructure, the resulting code is besides a few improvements the same
|
9 年之前 |
|
cgcpu.pas
|
4e8ef448ed
* i386: Bug fix where EBP was marked as preserved when it wasn't, causing faulty optimisations
|
2 年之前 |
|
cpubase.inc
|
518cdf9674
* replaced the saved_XXX_registers arrays with virtual methods inside
|
7 年之前 |
|
cpuelf.pas
|
637976e83f
* patch by Marģers to unify internal error numbers, resolves #37888
|
4 年之前 |
|
cpuinfo.pas
|
c433b740e4
Add/actualize CPUX86_HAS_SSSE3.
|
5 月之前 |
|
cpunode.pas
|
2de19f9e66
* x86: Reimplemented TAsmNode XML dumping using new framework
|
1 年之前 |
|
cpupara.pas
|
6404478ea4
* cleanup of VER3_0 defines
|
3 年之前 |
|
cpupi.pas
|
54d3c26e27
* renamed abi_linux386_sysv to abi_i386_dynalignedstack, and also use it
|
5 年之前 |
|
cputarg.pas
|
2d8a50556c
Fix exclusion of ogelf unit using $ifndef NOOGELF
|
1 年之前 |
|
hlcgcpu.pas
|
637976e83f
* patch by Marģers to unify internal error numbers, resolves #37888
|
4 年之前 |
|
i386att.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386atts.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386int.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386nop.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386op.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386prop.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
i386tab.inc
|
53cbfe3f85
Add x86 instruction set AVX102 for targets i386 and i8086.
|
2 月之前 |
|
n386add.pas
|
99b01f6634
* second mul might get called with a zero operand if the other has a side effect,
|
2 年之前 |
|
n386cal.pas
|
2d3ef8b4bc
* pushf/push cs on all platforms when directly calling an interrupt procedure, with the
|
5 年之前 |
|
n386flw.pas
|
4d82e814a0
* patch by Do-wan Kim: fix loop unrolling for try .. finally blocks in win32,
|
4 年之前 |
|
n386inl.pas
|
a8d496ade7
* i386 uses generic x86 abs(long/int64) code now
|
1 年之前 |
|
n386ld.pas
|
3c6aa91a96
* factored out the loading of threadvars in its own method, and put the
|
10 年之前 |
|
n386mat.pas
|
6f24c8b4ef
* x86: Code generation fixes where FLAGS
|
3 年之前 |
|
n386mem.pas
|
1bba83cebb
* Flags specific to TVecNode have been moved to their own field
|
1 年之前 |
|
n386set.pas
|
07bd4ba517
* let all the case code generation work with tconstexprint instead of aint,
|
6 年之前 |
|
r386ari.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386att.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386con.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386dwrf.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386int.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386iri.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386nasm.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386nor.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386nri.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386num.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386ot.inc
|
867d145e50
support vector operand bcst,{sae},{er} + k-register
|
7 年之前 |
|
r386rni.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386sri.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386stab.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
r386std.inc
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 年之前 |
|
ra386att.pas
|
757ed4e8d3
* standard assembler reader for i386
|
20 年之前 |
|
ra386int.pas
|
6c6bf452ca
* Fixed level 2 comment warnings.
|
17 年之前 |
|
rgcpu.pas
|
9d55e3aa0e
* x86: Fixed bug where XMM7 wasn't used by the register allocator
|
2 月之前 |
|
symcpu.pas
|
7f3a5eb9ab
* extend tabstractprocdef.getcopyas by a parameter to control whether the copy should be registered or not
|
3 年之前 |
|
tripletcpu.pas
|
75c16b6126
llvm: support for getting the llvm rtl (rt) architecture name
|
3 年之前 |
J. Gareth "Curious Kit" Moreton
9d55e3aa0e
* x86: Fixed bug where XMM7 wasn't used by the register allocator