Skip to content

AVX-NE-CONVERT fix #139

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Oct 8, 2025
Merged

AVX-NE-CONVERT fix #139

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 20 additions & 0 deletions test/avx-ne-convert-64.asm
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
BITS 64
vbcstnebf16ps xmm1, [rax]
vbcstnebf16ps ymm1, [rax]
vbcstnebf162ps xmm1, [rax]
vbcstnebf162ps ymm1, [rax]
vbcstnesh2ps xmm1, [rax]
vbcstnesh2ps ymm1, [rax]
vcvtneebf162ps xmm1, oword [rbx]
vcvtneebf162ps ymm1, yword [rcx]
vcvtneeph2ps xmm1, oword [rbx]
vcvtneeph2ps ymm1, yword [rcx]
vcvtneobf162ps xmm1, oword [rbx]
vcvtneobf162ps ymm1, yword [rcx]
vcvtneoph2ps xmm1, oword [rbx]
vcvtneoph2ps ymm1, yword [rcx]
cpu latevex
vcvtneps2bf16 xmm1, xmm2
vcvtneps2bf16 xmm1, ymm2
vcvtneps2bf16 xmm1, oword [rbx]
vcvtneps2bf16 xmm1, yword [rbx]
20 changes: 20 additions & 0 deletions test/avx-ne-convert.asm
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
BITS 32
vbcstnebf16ps xmm1, [eax]
vbcstnebf16ps ymm1, [eax]
vbcstnebf162ps xmm1, [eax]
vbcstnebf162ps ymm1, [eax]
vbcstnesh2ps xmm1, [eax]
vbcstnesh2ps ymm1, [eax]
vcvtneebf162ps xmm1, oword [ebx]
vcvtneebf162ps ymm1, yword [ecx]
vcvtneeph2ps xmm1, oword [ebx]
vcvtneeph2ps ymm1, yword [ecx]
vcvtneobf162ps xmm1, oword [ebx]
vcvtneobf162ps ymm1, yword [ecx]
vcvtneoph2ps xmm1, oword [ebx]
vcvtneoph2ps ymm1, yword [ecx]
cpu latevex
vcvtneps2bf16 xmm1, xmm2
vcvtneps2bf16 xmm1, ymm2
vcvtneps2bf16 xmm1, oword [ebx]
vcvtneps2bf16 xmm1, yword [ebx]
30 changes: 15 additions & 15 deletions x86/insns.dat
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2983,22 +2983,22 @@ VSM4RNDS4 zmmreg,zmmreg,zmmrm512 [rvm: evex.nds.512.f2.0f38.w

;# AVX no exception conversions
; Must precede AVX-512 versions
VBCSTNEBF16PS xmmreg,mem16 [rm: vex.128.f3.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VBCSTNEBF16PS ymmreg,mem16 [rm: vex.256.f3.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VBCSTNEBF162PS xmmreg,mem16 [rm: vex.128.f3.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VBCSTNEBF162PS ymmreg,mem16 [rm: vex.256.f3.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VBCSTNESH2PS xmmreg,mem16 [rm: vex.128.66.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VBCSTNESH2PS ymmreg,mem16 [rm: vex.256.66.0f38.w0 b1 /r] AVXNECONVERT,LATEVEX,SW
VCVTNEEBF162PS xmmreg,mem128 [rm: vex.128.f3.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SO
VCVTNEEBF162PS ymmreg,mem256 [rm: vex.256.f3.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SY
VCVTNEEPH2PS xmmreg,mem128 [rm: vex.128.66.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SO
VCVTNEEPH2PS ymmreg,mem256 [rm: vex.256.66.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SY
VCVTNEOBF162PS xmmreg,mem128 [rm: vex.128.f2.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SO
VCVTNEOBF162PS ymmreg,mem256 [rm: vex.256.f2.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SY
VCVTNEOPH2PS xmmreg,mem128 [rm: vex.128.np.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SO
VCVTNEOPH2PS ymmreg,mem256 [rm: vex.256.np.0f38.w0 b0 /r] AVXNECONVERT,LATEVEX,SY
VBCSTNEBF16PS xmmreg,mem16 [rm: vex.128.f3.0f38.w0 b1 /r] AVXNECONVERT,SW
VBCSTNEBF16PS ymmreg,mem16 [rm: vex.256.f3.0f38.w0 b1 /r] AVXNECONVERT,SW
VBCSTNEBF162PS xmmreg,mem16 [rm: vex.128.f3.0f38.w0 b1 /r] AVXNECONVERT,SW
VBCSTNEBF162PS ymmreg,mem16 [rm: vex.256.f3.0f38.w0 b1 /r] AVXNECONVERT,SW
VBCSTNESH2PS xmmreg,mem16 [rm: vex.128.66.0f38.w0 b1 /r] AVXNECONVERT,SW
VBCSTNESH2PS ymmreg,mem16 [rm: vex.256.66.0f38.w0 b1 /r] AVXNECONVERT,SW
VCVTNEEBF162PS xmmreg,mem128 [rm: vex.128.f3.0f38.w0 b0 /r] AVXNECONVERT,SO
VCVTNEEBF162PS ymmreg,mem256 [rm: vex.256.f3.0f38.w0 b0 /r] AVXNECONVERT,SY
VCVTNEEPH2PS xmmreg,mem128 [rm: vex.128.66.0f38.w0 b0 /r] AVXNECONVERT,SO
VCVTNEEPH2PS ymmreg,mem256 [rm: vex.256.66.0f38.w0 b0 /r] AVXNECONVERT,SY
VCVTNEOBF162PS xmmreg,mem128 [rm: vex.128.f2.0f38.w0 b0 /r] AVXNECONVERT,SO
VCVTNEOBF162PS ymmreg,mem256 [rm: vex.256.f2.0f38.w0 b0 /r] AVXNECONVERT,SY
VCVTNEOPH2PS xmmreg,mem128 [rm: vex.128.np.0f38.w0 b0 /r] AVXNECONVERT,SO
VCVTNEOPH2PS ymmreg,mem256 [rm: vex.256.np.0f38.w0 b0 /r] AVXNECONVERT,SY
VCVTNEPS2BF16 xmmreg,xmmrm128 [rm: vex.128.f3.0f38.w0 72 /r] AVXNECONVERT,LATEVEX,SO
VCVTNEPS2BF16 ymmreg,ymmrm256 [rm: vex.256.f3.0f38.w0 72 /r] AVXNECONVERT,LATEVEX,SY
VCVTNEPS2BF16 xmmreg,ymmrm256 [rm: vex.256.f3.0f38.w0 72 /r] AVXNECONVERT,LATEVEX,SY

;# AVX Vector Neural Network Instructions
; Must precede AVX-512 versions
Expand Down