[llvm] [AMDGPU] Reset WavesPerEU in AAAMDWavesPerEU::initialize (PR #114162)

via llvm-commits llvm-commits at lists.llvm.org
Tue Oct 29 17:54:08 PDT 2024


llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-amdgpu

Author: choikwa (choikwa)

<details>
<summary>Changes</summary>

Since we run AMDGPUAttributorPass twice for LTO run, WavesPerEU field was re-using stale, conservative values from the 1st pass and propagating it despite there being a new CallGraph. Resetting this in initialize makes it reflect the new CallGraph generated by LTO. Update testcases.

---

Patch is 149.51 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/114162.diff


27 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp (+3-1) 
- (modified) llvm/test/CodeGen/AMDGPU/aa-as-infer.ll (+6-6) 
- (modified) llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll (+1-2) 
- (modified) llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll (+16-19) 
- (modified) llvm/test/CodeGen/AMDGPU/annotate-existing-abi-attributes.ll (+10-10) 
- (modified) llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll (+48-50) 
- (modified) llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll (+1-2) 
- (modified) llvm/test/CodeGen/AMDGPU/attributor-loop-issue-58639.ll (+2-1) 
- (modified) llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll (+3-2) 
- (modified) llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll (+1-1) 
- (modified) llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll (+75-59) 
- (modified) llvm/test/CodeGen/AMDGPU/indirect-call-set-from-other-function.ll (+2-2) 
- (modified) llvm/test/CodeGen/AMDGPU/inline-attr.ll (+3-3) 
- (modified) llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll (+13-11) 
- (modified) llvm/test/CodeGen/AMDGPU/propagate-waves-per-eu.ll (+38-42) 
- (modified) llvm/test/CodeGen/AMDGPU/recursive_global_initializer.ll (+2-2) 
- (modified) llvm/test/CodeGen/AMDGPU/remove-no-kernel-id-attribute.ll (+14-13) 
- (modified) llvm/test/CodeGen/AMDGPU/simple-indirect-call-2.ll (+3-3) 
- (modified) llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll (+1-1) 
- (modified) llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll (+3-3) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll (+3-4) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll (+5-4) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll (+2-2) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll (+3-4) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll (+3-3) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll (+2-2) 
- (modified) llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll (+3-4) 


``````````diff
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
index 687a7339da379d..0e83221b468be7 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
@@ -843,7 +843,9 @@ struct AAAMDWavesPerEU : public AAAMDSizeRangeAttribute {
                AssumedGroupSize->getAssumed().getUpper().getZExtValue() - 1});
 
       ConstantRange Range(APInt(32, Min), APInt(32, Max + 1));
-      intersectKnown(Range);
+      
+      // Reset in case this is subsequent run and CallGraph is updated
+      *this &= Range;
     }
 
     if (AMDGPU::isEntryFunctionCC(F->getCallingConv()))
diff --git a/llvm/test/CodeGen/AMDGPU/aa-as-infer.ll b/llvm/test/CodeGen/AMDGPU/aa-as-infer.ll
index d1a6414fe49ae1..cbf5eebf3fd78b 100644
--- a/llvm/test/CodeGen/AMDGPU/aa-as-infer.ll
+++ b/llvm/test/CodeGen/AMDGPU/aa-as-infer.ll
@@ -22,7 +22,7 @@ define internal void @volatile_load_store_as_0(ptr %p) {
 
 define void @call_volatile_load_store_as_0(ptr %p1, ptr %p2) {
 ; CHECK-LABEL: define void @call_volatile_load_store_as_0(
-; CHECK-SAME: ptr [[P1:%.*]], ptr [[P2:%.*]]) #[[ATTR0]] {
+; CHECK-SAME: ptr [[P1:%.*]], ptr [[P2:%.*]]) #[[ATTR1:[0-9]+]] {
 ; CHECK-NEXT:    call void @volatile_load_store_as_0(ptr [[P1]])
 ; CHECK-NEXT:    call void @volatile_load_store_as_0(ptr [[P2]])
 ; CHECK-NEXT:    ret void
@@ -50,7 +50,7 @@ define internal void @volatile_load_store_as_1(ptr %p) {
 
 define void @call_volatile_load_store_as_1(ptr addrspace(1) %p1, ptr addrspace(1) %p2) {
 ; CHECK-LABEL: define void @call_volatile_load_store_as_1(
-; CHECK-SAME: ptr addrspace(1) [[P1:%.*]], ptr addrspace(1) [[P2:%.*]]) #[[ATTR0]] {
+; CHECK-SAME: ptr addrspace(1) [[P1:%.*]], ptr addrspace(1) [[P2:%.*]]) #[[ATTR1]] {
 ; CHECK-NEXT:    [[P1_CAST:%.*]] = addrspacecast ptr addrspace(1) [[P1]] to ptr
 ; CHECK-NEXT:    [[P2_CAST:%.*]] = addrspacecast ptr addrspace(1) [[P2]] to ptr
 ; CHECK-NEXT:    call void @volatile_load_store_as_1(ptr [[P1_CAST]])
@@ -74,7 +74,7 @@ define internal void @volatile_load_store_as_4(ptr %p) {
 
 define void @call_volatile_load_store_as_4(ptr addrspace(4) %p1, ptr addrspace(4) %p2) {
 ; CHECK-LABEL: define void @call_volatile_load_store_as_4(
-; CHECK-SAME: ptr addrspace(4) [[P1:%.*]], ptr addrspace(4) [[P2:%.*]]) #[[ATTR0]] {
+; CHECK-SAME: ptr addrspace(4) [[P1:%.*]], ptr addrspace(4) [[P2:%.*]]) #[[ATTR1]] {
 ; CHECK-NEXT:    [[P1_CAST:%.*]] = addrspacecast ptr addrspace(4) [[P1]] to ptr
 ; CHECK-NEXT:    [[P2_CAST:%.*]] = addrspacecast ptr addrspace(4) [[P2]] to ptr
 ; CHECK-NEXT:    call void @volatile_load_store_as_1(ptr [[P1_CAST]])
@@ -213,7 +213,7 @@ define internal void @can_not_infer_atomicrmw(ptr %word) {
 
 define void @foo(ptr addrspace(3) %val) {
 ; CHECK-LABEL: define void @foo(
-; CHECK-SAME: ptr addrspace(3) [[VAL:%.*]]) #[[ATTR1:[0-9]+]] {
+; CHECK-SAME: ptr addrspace(3) [[VAL:%.*]]) #[[ATTR2:[0-9]+]] {
 ; CHECK-NEXT:    [[VAL_CAST:%.*]] = addrspacecast ptr addrspace(3) [[VAL]] to ptr
 ; CHECK-NEXT:    call void @can_infer_cmpxchg(ptr addrspacecast (ptr addrspace(1) @g1 to ptr))
 ; CHECK-NEXT:    call void @can_infer_cmpxchg(ptr addrspacecast (ptr addrspace(1) @g2 to ptr))
@@ -245,7 +245,7 @@ define void @foo(ptr addrspace(3) %val) {
 
 define void @kernel_argument_promotion_pattern_intra_procedure(ptr %p, i32 %val) {
 ; CHECK-LABEL: define void @kernel_argument_promotion_pattern_intra_procedure(
-; CHECK-SAME: ptr [[P:%.*]], i32 [[VAL:%.*]]) #[[ATTR0]] {
+; CHECK-SAME: ptr [[P:%.*]], i32 [[VAL:%.*]]) #[[ATTR1]] {
 ; CHECK-NEXT:    [[P_CAST_0:%.*]] = addrspacecast ptr [[P]] to ptr addrspace(1)
 ; CHECK-NEXT:    store i32 [[VAL]], ptr addrspace(1) [[P_CAST_0]], align 4
 ; CHECK-NEXT:    ret void
@@ -269,7 +269,7 @@ define internal void @use_argument_after_promotion(ptr %p, i32 %val) {
 
 define void @kernel_argument_promotion_pattern_inter_procedure(ptr %p, i32 %val) {
 ; CHECK-LABEL: define void @kernel_argument_promotion_pattern_inter_procedure(
-; CHECK-SAME: ptr [[P:%.*]], i32 [[VAL:%.*]]) #[[ATTR0]] {
+; CHECK-SAME: ptr [[P:%.*]], i32 [[VAL:%.*]]) #[[ATTR1]] {
 ; CHECK-NEXT:    call void @use_argument_after_promotion(ptr [[P]], i32 [[VAL]])
 ; CHECK-NEXT:    ret void
 ;
diff --git a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
index cff9ce05066793..37a46bfcae8290 100644
--- a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
@@ -217,7 +217,7 @@ define ptr addrspace(3) @ret_constant_cast_group_gv_gep_to_flat_to_group() #1 {
 ; AKF_HSA-NEXT:    ret ptr addrspace(3) addrspacecast (ptr addrspace(4) getelementptr ([256 x i32], ptr addrspace(4) addrspacecast (ptr addrspace(3) @lds.arr to ptr addrspace(4)), i64 0, i64 8) to ptr addrspace(3))
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@ret_constant_cast_group_gv_gep_to_flat_to_group
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR3:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR2]] {
 ; ATTRIBUTOR_HSA-NEXT:    ret ptr addrspace(3) addrspacecast (ptr addrspace(4) getelementptr ([256 x i32], ptr addrspace(4) addrspacecast (ptr addrspace(3) @lds.arr to ptr addrspace(4)), i64 0, i64 8) to ptr addrspace(3))
 ;
   ret ptr addrspace(3) addrspacecast (ptr addrspace(4) getelementptr ([256 x i32], ptr addrspace(4) addrspacecast (ptr addrspace(3) @lds.arr to ptr addrspace(4)), i64 0, i64 8) to ptr addrspace(3))
@@ -235,7 +235,6 @@ attributes #1 = { nounwind }
 ; ATTRIBUTOR_HSA: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) }
 ; ATTRIBUTOR_HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
 ; ATTRIBUTOR_HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR3]] = { nounwind "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
 ;.
 ; AKF_HSA: [[META0:![0-9]+]] = !{i32 1, !"amdhsa_code_object_version", i32 500}
 ;.
diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
index e5d440b96349f6..67fcc5c1b4d09d 100644
--- a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
@@ -73,7 +73,7 @@ define amdgpu_kernel void @kernel_uses_asm_physreg_tuple() {
 
 define void @func_uses_asm_virtreg_agpr() {
 ; CHECK-LABEL: define void @func_uses_asm_virtreg_agpr(
-; CHECK-SAME: ) #[[ATTR2:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR0]] {
 ; CHECK-NEXT:    call void asm sideeffect "
 ; CHECK-NEXT:    ret void
 ;
@@ -83,7 +83,7 @@ define void @func_uses_asm_virtreg_agpr() {
 
 define void @func_uses_asm_physreg_agpr() {
 ; CHECK-LABEL: define void @func_uses_asm_physreg_agpr(
-; CHECK-SAME: ) #[[ATTR2]] {
+; CHECK-SAME: ) #[[ATTR0]] {
 ; CHECK-NEXT:    call void asm sideeffect "
 ; CHECK-NEXT:    ret void
 ;
@@ -93,7 +93,7 @@ define void @func_uses_asm_physreg_agpr() {
 
 define void @func_uses_asm_physreg_agpr_tuple() {
 ; CHECK-LABEL: define void @func_uses_asm_physreg_agpr_tuple(
-; CHECK-SAME: ) #[[ATTR2]] {
+; CHECK-SAME: ) #[[ATTR0]] {
 ; CHECK-NEXT:    call void asm sideeffect "
 ; CHECK-NEXT:    ret void
 ;
@@ -105,7 +105,7 @@ declare void @unknown()
 
 define amdgpu_kernel void @kernel_calls_extern() {
 ; CHECK-LABEL: define amdgpu_kernel void @kernel_calls_extern(
-; CHECK-SAME: ) #[[ATTR4:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR2:[0-9]+]] {
 ; CHECK-NEXT:    call void @unknown()
 ; CHECK-NEXT:    ret void
 ;
@@ -115,8 +115,8 @@ define amdgpu_kernel void @kernel_calls_extern() {
 
 define amdgpu_kernel void @kernel_calls_extern_marked_callsite() {
 ; CHECK-LABEL: define amdgpu_kernel void @kernel_calls_extern_marked_callsite(
-; CHECK-SAME: ) #[[ATTR4]] {
-; CHECK-NEXT:    call void @unknown() #[[ATTR9:[0-9]+]]
+; CHECK-SAME: ) #[[ATTR2]] {
+; CHECK-NEXT:    call void @unknown() #[[ATTR6:[0-9]+]]
 ; CHECK-NEXT:    ret void
 ;
   call void @unknown() #0
@@ -125,7 +125,7 @@ define amdgpu_kernel void @kernel_calls_extern_marked_callsite() {
 
 define amdgpu_kernel void @kernel_calls_indirect(ptr %indirect) {
 ; CHECK-LABEL: define amdgpu_kernel void @kernel_calls_indirect(
-; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR4]] {
+; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR2]] {
 ; CHECK-NEXT:    call void [[INDIRECT]]()
 ; CHECK-NEXT:    ret void
 ;
@@ -135,8 +135,8 @@ define amdgpu_kernel void @kernel_calls_indirect(ptr %indirect) {
 
 define amdgpu_kernel void @kernel_calls_indirect_marked_callsite(ptr %indirect) {
 ; CHECK-LABEL: define amdgpu_kernel void @kernel_calls_indirect_marked_callsite(
-; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR4]] {
-; CHECK-NEXT:    call void [[INDIRECT]]() #[[ATTR9]]
+; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR2]] {
+; CHECK-NEXT:    call void [[INDIRECT]]() #[[ATTR6]]
 ; CHECK-NEXT:    ret void
 ;
   call void %indirect() #0
@@ -155,7 +155,7 @@ define amdgpu_kernel void @kernel_transitively_uses_agpr_asm() {
 
 define void @empty() {
 ; CHECK-LABEL: define void @empty(
-; CHECK-SAME: ) #[[ATTR5:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR1]] {
 ; CHECK-NEXT:    ret void
 ;
   ret void
@@ -163,7 +163,7 @@ define void @empty() {
 
 define void @also_empty() {
 ; CHECK-LABEL: define void @also_empty(
-; CHECK-SAME: ) #[[ATTR5]] {
+; CHECK-SAME: ) #[[ATTR1]] {
 ; CHECK-NEXT:    ret void
 ;
   ret void
@@ -256,12 +256,9 @@ attributes #0 = { "amdgpu-no-agpr" }
 ;.
 ; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
 ; CHECK: attributes #[[ATTR1]] = { "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3:[0-9]+]] = { "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR5]] = { "amdgpu-no-agpr" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR6:[0-9]+]] = { convergent nocallback nofree nosync nounwind willreturn memory(none) "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR7:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR8:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR9]] = { "amdgpu-no-agpr" }
+; CHECK: attributes #[[ATTR2]] = { "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3:[0-9]+]] = { convergent nocallback nofree nosync nounwind willreturn memory(none) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR5:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR6]] = { "amdgpu-no-agpr" }
 ;.
diff --git a/llvm/test/CodeGen/AMDGPU/annotate-existing-abi-attributes.ll b/llvm/test/CodeGen/AMDGPU/annotate-existing-abi-attributes.ll
index 28722021e0448f..7e0208cd1f45aa 100644
--- a/llvm/test/CodeGen/AMDGPU/annotate-existing-abi-attributes.ll
+++ b/llvm/test/CodeGen/AMDGPU/annotate-existing-abi-attributes.ll
@@ -117,14 +117,14 @@ define void @call_no_dispatch_id() {
   ret void
 }
 ;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-workitem-id-x" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-no-workitem-id-y" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3]] = { "amdgpu-no-workgroup-id-x" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { "amdgpu-no-workgroup-id-y" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR5]] = { "amdgpu-no-workgroup-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR6]] = { "amdgpu-no-dispatch-ptr" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR7]] = { "amdgpu-no-queue-ptr" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR8]] = { "amdgpu-no-implicitarg-ptr" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR9]] = { "amdgpu-no-dispatch-id" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-workitem-id-x" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-no-workitem-id-y" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-no-workgroup-id-x" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR4]] = { "amdgpu-no-workgroup-id-y" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR5]] = { "amdgpu-no-workgroup-id-z" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR6]] = { "amdgpu-no-dispatch-ptr" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR7]] = { "amdgpu-no-queue-ptr" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR8]] = { "amdgpu-no-implicitarg-ptr" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR9]] = { "amdgpu-no-dispatch-id" "uniform-work-group-size"="false" }
 ;.
diff --git a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll
index 3d4ae84d9c698e..5c5520ec32bb84 100644
--- a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll
@@ -593,7 +593,7 @@ define amdgpu_kernel void @kern_use_implicitarg_ptr() #1 {
 ; AKF_HSA-NEXT:    ret void
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@kern_use_implicitarg_ptr
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR15:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR12]] {
 ; ATTRIBUTOR_HSA-NEXT:    [[IMPLICITARG_PTR:%.*]] = call ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr()
 ; ATTRIBUTOR_HSA-NEXT:    store volatile ptr addrspace(4) [[IMPLICITARG_PTR]], ptr addrspace(1) undef, align 8
 ; ATTRIBUTOR_HSA-NEXT:    ret void
@@ -645,7 +645,7 @@ define internal void @defined.func() #3 {
 ; AKF_HSA-NEXT:    ret void
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@defined.func
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR17:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR16:[0-9]+]] {
 ; ATTRIBUTOR_HSA-NEXT:    ret void
 ;
   ret void
@@ -658,7 +658,7 @@ define void @func_call_external() #3 {
 ; AKF_HSA-NEXT:    ret void
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_call_external
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR16:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR15:[0-9]+]] {
 ; ATTRIBUTOR_HSA-NEXT:    call void @external.func()
 ; ATTRIBUTOR_HSA-NEXT:    ret void
 ;
@@ -673,7 +673,7 @@ define void @func_call_defined() #3 {
 ; AKF_HSA-NEXT:    ret void
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_call_defined
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR17]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR17:[0-9]+]] {
 ; ATTRIBUTOR_HSA-NEXT:    call void @defined.func()
 ; ATTRIBUTOR_HSA-NEXT:    ret void
 ;
@@ -688,7 +688,7 @@ define void @func_call_asm() #3 {
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_call_asm
 ; ATTRIBUTOR_HSA-SAME: () #[[ATTR17]] {
-; ATTRIBUTOR_HSA-NEXT:    call void asm sideeffect "", ""() #[[ATTR28:[0-9]+]]
+; ATTRIBUTOR_HSA-NEXT:    call void asm sideeffect "", ""() #[[ATTR26:[0-9]+]]
 ; ATTRIBUTOR_HSA-NEXT:    ret void
 ;
   call void asm sideeffect "", ""() #3
@@ -702,7 +702,7 @@ define amdgpu_kernel void @kern_call_external() #3 {
 ; AKF_HSA-NEXT:    ret void
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@kern_call_external
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR18:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR15]] {
 ; ATTRIBUTOR_HSA-NEXT:    call void @external.func()
 ; ATTRIBUTOR_HSA-NEXT:    ret void
 ;
@@ -769,7 +769,7 @@ define float @func_indirect_call(ptr %fptr) #3 {
 ; AKF_HSA-NEXT:    ret float [[FADD]]
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_indirect_call
-; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR16]] {
+; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR15]] {
 ; ATTRIBUTOR_HSA-NEXT:    [[F:%.*]] = call float [[FPTR]]()
 ; ATTRIBUTOR_HSA-NEXT:    [[FADD:%.*]] = fadd float [[F]], 1.000000e+00
 ; ATTRIBUTOR_HSA-NEXT:    ret float [[FADD]]
@@ -788,7 +788,7 @@ define float @func_extern_call() #3 {
 ; AKF_HSA-NEXT:    ret float [[FADD]]
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_extern_call
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR16]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR15]] {
 ; ATTRIBUTOR_HSA-NEXT:    [[F:%.*]] = call float @extern()
 ; ATTRIBUTOR_HSA-NEXT:    [[FADD:%.*]] = fadd float [[F]], 1.000000e+00
 ; ATTRIBUTOR_HSA-NEXT:    ret float [[FADD]]
@@ -806,7 +806,7 @@ define float @func_null_call(ptr %fptr) #3 {
 ; AKF_HSA-NEXT:    ret float [[FADD]]
 ;
 ; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_null_call
-; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR16]] {
+; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR15]] {
 ; ATTRIBUTOR_HSA-NEXT:    [[F:%.*]] = call float null()
 ; ATTRIBUT...
[truncated]

``````````

</details>


https://github.com/llvm/llvm-project/pull/114162


More information about the llvm-commits mailing list