sanchayanmaity/gem5
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

X86: Implement the X86 sse2 haddpd instruction

Browse source 
This patch implements the haddpd instruction.

It fixes the problem in the previous version (pointed out by Gabe Black)
where an incorrect result would happen if you issue the instruction
with the same argument twice, i.e. "haddpd %xmm0,%xmm0"

This instruction is used by many spec2k benchmarks.
This commit is contained in:
Vince Weaver 2009-10-30 14:19:06 -04:00
parent cf269025f9
commit b2067840a6
2 changed files with 22 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/arch/x86/isa/decoder/two_byte_opcodes.isa
View file

@ -707,7 +707,7 @@
}
// operand size (0x66)
0x1: decode OPCODE_OP_BOTTOM3 {
0x4: WarnUnimpl::haddpd_Vo_Wo();
0x4: HADDPD(Vo,Wo);
0x5: WarnUnimpl::hsubpd_Vo_Wo();
0x6: WarnUnimpl::movd_Ed_Vd();
0x7: MOVDQA(Wo,Vo);

22
src/arch/x86/isa/insts/simd128/floating_point/arithmetic/horizontal_addition.py
View file

@ -55,5 +55,25 @@
microcode = '''
# HADDPS
# HADDPD
def macroop HADDPD_XMM_XMM {
maddf ufp1, xmmh , xmml, size=8, ext=1
maddf xmmh, xmmlm, xmmhm, size=8, ext=1
movfp xmml, ufp1
};
def macroop HADDPD_XMM_M {
ldfp ufp1, seg, sib, disp, dataSize=8
ldfp ufp2, seg, sib, "DISPLACEMENT+8", dataSize=8
maddf xmml, xmmh, xmml, size=8, ext=1
maddf xmmh, ufp1, ufp2, size=8, ext=1
};
def macroop HADDPD_XMM_P {
rdip t7
ldfp ufp1, seg, riprel, disp, dataSize=8
ldfp ufp2, seg, riprel, "DISPLACEMENT+8", dataSize=8
maddf xmml, xmmh, xmml, size=8, ext=1
maddf xmmh, ufp1, ufp2, size=8, ext=1
};
'''