Skip to content

[AMDGPU] Fix a crash when a bool variable is used in inline asm #171004

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
shiltian merged 4 commits into from
Dec 8, 2025
+28 −0
Merged

[AMDGPU] Fix a crash when a bool variable is used in inline asm #171004

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
4044fc5
[AMDGPU] Fix a crash when a bool variable is used in inline asm
shiltian Dec 7, 2025
88a9db4
move test
shiltian Dec 8, 2025
4846cd2
review comments
shiltian Dec 8, 2025
e65d0e7
add more tests
shiltian Dec 8, 2025
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
4 changes: 4 additions & 0 deletions llvm/lib/Target/AMDGPU/SIISelLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17684,6 +17684,8 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_,
break;
case 'v':
switch (BitWidth) {
case 1:
return std::pair(0U, nullptr);
case 16:
RC = Subtarget->useRealTrue16Insts() ? &AMDGPU::VGPR_16RegClass
: &AMDGPU::VGPR_32_Lo256RegClass;
Expand All @@ -17701,6 +17703,8 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_,
if (!Subtarget->hasMAIInsts())
break;
switch (BitWidth) {
case 1:
return std::pair(0U, nullptr);
case 16:
RC = &AMDGPU::AGPR_32RegClass;
break;
Expand Down
24 changes: 24 additions & 0 deletions llvm/test/CodeGen/AMDGPU/inlineasm-mismatched-size-error.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,3 +153,27 @@ define <4 x i32> @misaligned_sgpr_4xi32_out_2() {
%asm = call <4 x i32> asm sideeffect "; def $0", "={s[2:5]}"()
ret <4 x i32> %asm
}

; ERR: error: couldn't allocate input reg for constraint 'v'
define void @i1_used_as_vgpr_operand(ptr %p, i1 %b) {
tail call void asm sideeffect "global_store_byte $0, $1, off glc slc", "v,v"(ptr %p, i1 %b)
ret void
}
Copy link
Contributor

@arsenm arsenm Dec 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Check the def case for good measure?

Copy link
Contributor

@arsenm arsenm Dec 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also AGPR + SGPR cases, those are probably equally broken

Copy link
Contributor Author

@shiltian shiltian Dec 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Check the def case for good measure?

What do you mean by "def case"?

Copy link
Contributor

@arsenm arsenm Dec 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

call i1 asm "; def $0", "=v"()


; ERR: error: couldn't allocate input reg for constraint 'a'
define void @i1_used_as_agpr_operand(ptr %p, i1 %b) {
tail call void asm sideeffect "global_store_byte $0, $1, off glc slc", "v,a"(ptr %p, i1 %b)
ret void
}

; ERR: error: couldn't allocate input reg for constraint 's'
define void @i1_used_as_sgpr_operand_s(ptr %p, i1 %b) {
tail call void asm sideeffect "global_store_byte $0, $1, off glc slc", "v,s"(ptr %p, i1 %b)
ret void
}

; ERR: error: couldn't allocate input reg for constraint 'r'
define void @i1_used_as_sgpr_operand_r(ptr %p, i1 %b) {
tail call void asm sideeffect "global_store_byte $0, $1, off glc slc", "v,r"(ptr %p, i1 %b)
ret void
}
Loading