 yury
|
4357caaad8
* Removed unused local vars.
|
6 years ago |
 florian
|
cdce68451a
- x86 align does not use a register anymore, code removed
|
6 years ago |
 tg74
|
60a75a2277
delete testcode
|
6 years ago |
|
tg74
|
1ef9cc01e6
avx512 disp8*N
|
6 years ago |
|
tg74
|
fba72b280b
avx512 broadcast vcvt...,vfpclass...
|
6 years ago |
|
tg74
|
4265f4d6a5
avx512 broadcast for special opcodes vfpclass.., vcvt...
|
6 years ago |
|
tg74
|
3759eac608
bugfix avx512 add ZMMReg for gather/scatter in intelOS32
|
7 years ago |
|
tg74
|
8b9d7be8e5
bugfix OS32
|
7 years ago |
|
tg74
|
1d9cbb4dcb
new AVX512 opcodes
|
7 years ago |
|
tg74
|
5e1dc1a807
bugfix for test tasm9 32bit
|
7 years ago |
|
tg74
|
6f64177c6a
bugfix avx512 OS32bit EVEX-coding
|
7 years ago |
|
tg74
|
29690162b8
bugfix OS32
|
7 years ago |
|
tg74
|
7c5cefad36
bugfix OS32
|
7 years ago |
|
tg74
|
7b8715184a
bugfix OS32
|
7 years ago |
|
tg74
|
2b1da37d66
new avx512 instructions and bugfixes avx512
|
7 years ago |
|
tg74
|
867d145e50
support vector operand bcst,{sae},{er} + k-register
|
7 years ago |
|
tg74
|
4dc5442fa5
support vector operand writemask,zeroflag
|
7 years ago |
|
tg74
|
31e4d4ef5e
AVX512 support for MMRegister xmm16..31 and ymm16..31, zmm0..31, vpaddsb support AVX512
|
7 years ago |
 marco
|
f0042a4719
* vcmppd hardcoded primitives like vcmpeqpd.
|
7 years ago |
|
florian
|
f66a91499d
* FPC uses meanwhile more mov instructions, so extended taicpu.is_same_reg_move to support them
|
7 years ago |
|
florian
|
31f78ea2b6
+ implementation of the vectorcall calling convention by J. Gareth Moreton
|
7 years ago |
|
florian
|
e6a6938787
* make fpu/mmx/xmm/ymm registers numbers instead of flags to have enough space for zmm/bnd/k registers
|
7 years ago |
 nickysn
|
c464f7fa56
* fixed the SEG inline asm directive when used with 32-bit registers on the i8086 target
|
7 years ago |
|
nickysn
|
8a0d8f025b
* fixed another i8086 inline asm 32-bit constant bug (e.g. in 'or eax, 80000001h')
|
7 years ago |
|
nickysn
|
325e66287c
* fix for inline asm of instructions with 32-bit constant operands on i8086
|
7 years ago |
|
nickysn
|
d7e4b50068
+ also optimize x86_64 references by switching [rbp+reg64] to [reg64+rbp],
|
7 years ago |
|
nickysn
|
5ae32a0ec5
+ always do the x86_64 reference optimizations as if SS=DS, because the CPU
|
7 years ago |
|
nickysn
|
875339993f
* strip segment overrides, for segments, which should be equal in the current
|
7 years ago |
|
nickysn
|
80226e3af4
+ added an optimization pass, that optimizes x86 references
|
7 years ago |
|
nickysn
|
67a0e9bdae
+ added x86 helper function get_default_segment_of_ref, which returns the
|
7 years ago |
