Skip to content

[AMDGPU] Fix stack size metadata for functions with direct and indirect calls #110828

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 2 commits into from
Oct 2, 2024
Merged

[AMDGPU] Fix stack size metadata for functions with direct and indirect calls #110828

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ec3398a
[AMDGPU] Fix stack size resource usage metadata for functions with bo…
JanekvO Oct 2, 2024
17aa1b9
Add case with repeated calls
JanekvO Oct 2, 2024
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
8 changes: 6 additions & 2 deletions llvm/lib/Target/AMDGPU/AMDGPUMCResourceInfo.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -157,8 +157,12 @@ void MCResourceInfo::gatherResourceInfo(
ArgExprs.push_back(
MCConstantExpr::create(FRI.CalleeSegmentSize, OutContext));

if (!FRI.HasIndirectCall) {
for (const Function *Callee : FRI.Callees) {
SmallPtrSet<const Function *, 8> Seen;
Seen.insert(&MF.getFunction());
for (const Function *Callee : FRI.Callees) {
if (!Seen.insert(Callee).second)
continue;
if (!Callee->isDeclaration()) {
MCSymbol *calleeValSym =
getSymbol(Callee->getName(), RIK_PrivateSegSize, OutContext);
ArgExprs.push_back(MCSymbolRefExpr::create(calleeValSym, OutContext));
Expand Down
27 changes: 25 additions & 2 deletions llvm/test/CodeGen/AMDGPU/function-resource-usage.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -358,22 +358,45 @@ declare void @external() #0
; GCN: .set multi_call_with_external.num_vgpr, max(41, amdgpu.max_num_vgpr)
; GCN: .set multi_call_with_external.num_agpr, max(0, amdgpu.max_num_agpr)
; GCN: .set multi_call_with_external.numbered_sgpr, max(42, amdgpu.max_num_sgpr)
; GCN: .set multi_call_with_external.private_seg_size, 0
; GCN: .set multi_call_with_external.private_seg_size, 0+(max(use_stack0.private_seg_size, use_stack1.private_seg_size))
; GCN: .set multi_call_with_external.uses_vcc, 1
; GCN: .set multi_call_with_external.uses_flat_scratch, 1
; GCN: .set multi_call_with_external.has_dyn_sized_stack, 1
; GCN: .set multi_call_with_external.has_recursion, 0
; GCN: .set multi_call_with_external.has_indirect_call, 1
; GCN: TotalNumSgprs: multi_call_with_external.numbered_sgpr+6
; GCN: NumVgprs: multi_call_with_external.num_vgpr
; GCN: ScratchSize: 0
; GCN: ScratchSize: 2052
define amdgpu_kernel void @multi_call_with_external() #0 {
call void @use_stack0()
call void @use_stack1()
call void @external()
ret void
}

; GCN-LABEL: {{^}}multi_call_with_external_and_duplicates:
; GCN: .set multi_call_with_external_and_duplicates.num_vgpr, max(41, amdgpu.max_num_vgpr)
; GCN: .set multi_call_with_external_and_duplicates.num_agpr, max(0, amdgpu.max_num_agpr)
; GCN: .set multi_call_with_external_and_duplicates.numbered_sgpr, max(44, amdgpu.max_num_sgpr)
; GCN: .set multi_call_with_external_and_duplicates.private_seg_size, 0+(max(use_stack0.private_seg_size, use_stack1.private_seg_size))
; GCN: .set multi_call_with_external_and_duplicates.uses_vcc, 1
; GCN: .set multi_call_with_external_and_duplicates.uses_flat_scratch, 1
; GCN: .set multi_call_with_external_and_duplicates.has_dyn_sized_stack, 1
; GCN: .set multi_call_with_external_and_duplicates.has_recursion, 0
; GCN: .set multi_call_with_external_and_duplicates.has_indirect_call, 1
; GCN: TotalNumSgprs: multi_call_with_external_and_duplicates.numbered_sgpr+6
; GCN: NumVgprs: multi_call_with_external_and_duplicates.num_vgpr
; GCN: ScratchSize: 2052
define amdgpu_kernel void @multi_call_with_external_and_duplicates() #0 {
call void @use_stack0()
call void @use_stack0()
call void @use_stack1()
call void @use_stack1()
call void @external()
call void @external()
ret void
}

; GCN-LABEL: {{^}}usage_external:
; GCN: .set usage_external.num_vgpr, max(32, amdgpu.max_num_vgpr)
; GCN: .set usage_external.num_agpr, max(0, amdgpu.max_num_agpr)
Expand Down
Loading