[llvm-branch-commits] [clang] [llvm] [AMDGPU][Attributor] Rework update of `AAAMDWavesPerEU` (PR #123995)
Shilei Tian via llvm-branch-commits
llvm-branch-commits at lists.llvm.org
Thu May 1 11:15:58 PDT 2025
https://github.com/shiltian updated https://github.com/llvm/llvm-project/pull/123995
>From 3c09c1dd14d75f1326b396336885059ff99ab59a Mon Sep 17 00:00:00 2001
From: Shilei Tian <i at tianshilei.me>
Date: Fri, 21 Mar 2025 14:56:24 -0400
Subject: [PATCH 1/3] [AMDGPU][Attributor] Rework calculation of waves per eu
---
builtin.diff | 84 +++++++++++
.../builtins-amdgcn-gfx13-load-mcast.cl | 14 ++
clang/test/CodeGenOpenCL/builtins-amdgcn.cl | 3 +-
llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp | 130 +++++++++++-------
.../AMDGPU/amdgpu-attributor-no-agpr.ll | 16 +--
.../annotate-existing-abi-attributes.ll | 20 +--
...ttr-amdgpu-max-num-workgroups-propagate.ll | 52 ++++---
.../AMDGPU/attributor-flatscratchinit.ll | 24 ++--
.../AMDGPU/attributor-loop-issue-58639.ll | 5 +-
.../CodeGen/AMDGPU/direct-indirect-call.ll | 7 +-
.../AMDGPU/duplicate-attribute-indirect.ll | 4 +-
.../AMDGPU/implicitarg-offset-attributes.ll | 32 ++---
.../indirect-call-set-from-other-function.ll | 4 +-
...120256-annotate-constexpr-addrspacecast.ll | 4 +-
.../AMDGPU/pal-simple-indirect-call.ll | 4 +-
.../AMDGPU/propagate-flat-work-group-size.ll | 6 +-
.../CodeGen/AMDGPU/propagate-waves-per-eu.ll | 83 ++++++-----
.../AMDGPU/recursive_global_initializer.ll | 2 +-
.../AMDGPU/remove-no-kernel-id-attribute.ll | 15 +-
.../CodeGen/AMDGPU/simple-indirect-call-2.ll | 12 +-
.../CodeGen/AMDGPU/simple-indirect-call.ll | 4 +-
llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll | 2 +-
.../uniform-work-group-attribute-missing.ll | 2 +-
.../AMDGPU/uniform-work-group-multistep.ll | 11 +-
...niform-work-group-nested-function-calls.ll | 4 +-
...ork-group-prevent-attribute-propagation.ll | 4 +-
.../uniform-work-group-propagate-attribute.ll | 8 +-
.../uniform-work-group-recursion-test.ll | 4 +-
.../CodeGen/AMDGPU/uniform-work-group-test.ll | 2 +-
29 files changed, 347 insertions(+), 215 deletions(-)
create mode 100644 builtin.diff
create mode 100644 clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
diff --git a/builtin.diff b/builtin.diff
new file mode 100644
index 0000000000000..0021b62e655ea
--- /dev/null
+++ b/builtin.diff
@@ -0,0 +1,84 @@
+diff --git a/clang/include/clang/Basic/BuiltinsAMDGPU.def b/clang/include/clang/Basic/BuiltinsAMDGPU.def
+index 39fef9e4601f..55befac12b46 100644
+--- a/clang/include/clang/Basic/BuiltinsAMDGPU.def
++++ b/clang/include/clang/Basic/BuiltinsAMDGPU.def
+@@ -635,5 +635,7 @@ TARGET_BUILTIN(__builtin_amdgcn_bitop3_b16, "ssssIUi", "nc", "bitop3-insts")
+ TARGET_BUILTIN(__builtin_amdgcn_cvt_sr_bf16_f32, "V2yV2yfUiIb", "nc", "f32-to-f16bf16-cvt-sr-insts")
+ TARGET_BUILTIN(__builtin_amdgcn_cvt_sr_f16_f32, "V2hV2hfUiIb", "nc", "f32-to-f16bf16-cvt-sr-insts")
+
++TARGET_BUILTIN(__builtin_amdgcn_load_mcast_b32, "vi*3i*Iii", "", "gfx950-insts")
++
+ #undef BUILTIN
+ #undef TARGET_BUILTIN
+diff --git a/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp b/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
+index ad012d98635f..601319b2c225 100644
+--- a/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
++++ b/clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
+@@ -1083,6 +1083,20 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
+
+ return Builder.CreateBitCast(RMW, OrigTy);
+ }
++ case AMDGPU::BI__builtin_amdgcn_load_mcast_b32: {
++ Intrinsic::ID IID;
++ switch (BuiltinID) {
++ case AMDGPU::BI__builtin_amdgcn_load_mcast_b32:
++ IID = Intrinsic::amdgcn_load_mcast_b32;
++ break;
++ }
++ SmallVector<Value *, 3> Args;
++ auto *SrcType = ConvertType(E->getArg(1)->getType());
++ for (int i = 0, e = E->getNumArgs(); i != e; ++i)
++ Args.push_back(EmitScalarExpr(E->getArg(i)));
++ llvm::Function *F = CGM.getIntrinsic(IID, {SrcType});
++ return Builder.CreateCall(F, {Args});
++ }
+ case AMDGPU::BI__builtin_amdgcn_s_sendmsg_rtn:
+ case AMDGPU::BI__builtin_amdgcn_s_sendmsg_rtnl: {
+ llvm::Value *Arg = EmitScalarExpr(E->getArg(0));
+diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
+new file mode 100644
+index 000000000000..f6b058cb3fb5
+--- /dev/null
++++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
+@@ -0,0 +1,14 @@
++// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
++// REQUIRES: amdgpu-registered-target
++// RUN: %clang_cc1 -cl-std=CL2.0 -triple amdgcn-unknown-unknown -target-cpu gfx950 -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX13
++
++void test_amdgcn_load_mcast_b32(local void* outptr, global void* inptr, int mask)
++{
++ __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
++}
++
++//void test_amdgcn_load_mcast_local_b32(local void* outptr, local void* inptr, int mask)
++//{
++// __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
++//}
++
+diff --git a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
+index a57eb4a6dba4..a765ef8c3a42 100644
+--- a/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
++++ b/llvm/include/llvm/IR/IntrinsicsAMDGPU.td
+@@ -3499,6 +3499,22 @@ def int_amdgcn_fdiv_fast : DefaultAttrsIntrinsic<
+ [IntrNoMem, IntrSpeculatable]
+ >;
+
++class AMDGPULoadMCast:
++ Intrinsic<
++ [],
++ [local_ptr_ty, // laneshared destination pointer
++ llvm_anyptr_ty, // Global, LDS, or DDS pointer to load from
++ llvm_i32_ty, // gfx12+ cachepolicy:
++ // bits [0-2] = th
++ // bits [3-4] = scope
++ llvm_i32_ty], // workgroup broadcast mask (in M0)
++ [IntrArgMemOnly, ReadOnly<ArgIndex<1>>, IntrConvergent, NoCapture<ArgIndex<0>>,
++ NoCapture<ArgIndex<1>>, ImmArg<ArgIndex<2>>],
++ "", [SDNPMemOperand]
++ >;
++
++def int_amdgcn_load_mcast_b32 : AMDGPULoadMCast;
++
+ /// Emit an addrspacecast without null pointer checking.
+ /// Should only be inserted by a pass based on analysis of an addrspacecast's src.
+ def int_amdgcn_addrspacecast_nonnull : DefaultAttrsIntrinsic<
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
new file mode 100644
index 0000000000000..4e8b6744ba7a9
--- /dev/null
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
@@ -0,0 +1,14 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// REQUIRES: amdgpu-registered-target
+// RUN: %clang_cc1 -cl-std=CL2.0 -triple amdgcn-unknown-unknown -target-cpu gfx950 -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX13
+
+void test_amdgcn_load_mcast_b32(local int* outptr, global int* inptr, int mask)
+{
+ __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
+}
+
+//void test_amdgcn_load_mcast_local_b32(local void* outptr, local void* inptr, int mask)
+//{
+// __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
+//}
+
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn.cl
index bf022bc6eb446..d97fa62ab7249 100644
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn.cl
+++ b/clang/test/CodeGenOpenCL/builtins-amdgcn.cl
@@ -903,4 +903,5 @@ void test_set_fpenv(unsigned long env) {
// CHECK-DAG: [[$GRID_RANGE]] = !{i32 1, i32 0}
// CHECK-DAG: [[$WS_RANGE]] = !{i16 1, i16 1025}
-// CHECK-DAG: attributes #[[$NOUNWIND_READONLY]] = { convergent mustprogress nocallback nofree nounwind willreturn memory(none) }
+// CHECK-SPIRV-DAG: attributes #[[$NOUNWIND_READONLY]] = { convergent mustprogress nocallback nofree nounwind willreturn memory(none) }
+// CHECK-AMDGCN-DAG: attributes #[[$NOUNWIND_READONLY]] = { convergent mustprogress nocallback nofree nounwind willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
index 78e75f888c99c..a49e41a79bab7 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
@@ -1117,47 +1117,25 @@ struct AAAMDWavesPerEU : public AAAMDSizeRangeAttribute {
Function *F = getAssociatedFunction();
auto &InfoCache = static_cast<AMDGPUInformationCache &>(A.getInfoCache());
- auto TakeRange = [&](std::pair<unsigned, unsigned> R) {
- auto [Min, Max] = R;
- ConstantRange Range(APInt(32, Min), APInt(32, Max + 1));
- IntegerRangeState RangeState(Range);
- clampStateAndIndicateChange(this->getState(), RangeState);
- indicateOptimisticFixpoint();
- };
-
- std::pair<unsigned, unsigned> MaxWavesPerEURange{
- 1U, InfoCache.getMaxWavesPerEU(*F)};
-
// If the attribute exists, we will honor it if it is not the default.
if (auto Attr = InfoCache.getWavesPerEUAttr(*F)) {
+ std::pair<unsigned, unsigned> MaxWavesPerEURange{
+ 1U, InfoCache.getMaxWavesPerEU(*F)};
if (*Attr != MaxWavesPerEURange) {
- TakeRange(*Attr);
+ auto [Min, Max] = *Attr;
+ ConstantRange Range(APInt(32, Min), APInt(32, Max + 1));
+ IntegerRangeState RangeState(Range);
+ this->getState() = RangeState;
+ indicateOptimisticFixpoint();
return;
}
}
- // Unlike AAAMDFlatWorkGroupSize, it's getting trickier here. Since the
- // calculation of waves per EU involves flat work group size, we can't
- // simply use an assumed flat work group size as a start point, because the
- // update of flat work group size is in an inverse direction of waves per
- // EU. However, we can still do something if it is an entry function. Since
- // an entry function is a terminal node, and flat work group size either
- // from attribute or default will be used anyway, we can take that value and
- // calculate the waves per EU based on it. This result can't be updated by
- // no means, but that could still allow us to propagate it.
- if (AMDGPU::isEntryFunctionCC(F->getCallingConv())) {
- std::pair<unsigned, unsigned> FlatWorkGroupSize;
- if (auto Attr = InfoCache.getFlatWorkGroupSizeAttr(*F))
- FlatWorkGroupSize = *Attr;
- else
- FlatWorkGroupSize = InfoCache.getDefaultFlatWorkGroupSize(*F);
- TakeRange(InfoCache.getEffectiveWavesPerEU(*F, MaxWavesPerEURange,
- FlatWorkGroupSize));
- }
+ if (AMDGPU::isEntryFunctionCC(F->getCallingConv()))
+ indicatePessimisticFixpoint();
}
ChangeStatus updateImpl(Attributor &A) override {
- auto &InfoCache = static_cast<AMDGPUInformationCache &>(A.getInfoCache());
ChangeStatus Change = ChangeStatus::UNCHANGED;
auto CheckCallSite = [&](AbstractCallSite CS) {
@@ -1166,24 +1144,21 @@ struct AAAMDWavesPerEU : public AAAMDSizeRangeAttribute {
LLVM_DEBUG(dbgs() << '[' << getName() << "] Call " << Caller->getName()
<< "->" << Func->getName() << '\n');
- const auto *CallerInfo = A.getAAFor<AAAMDWavesPerEU>(
+ const auto *CallerAA = A.getAAFor<AAAMDWavesPerEU>(
*this, IRPosition::function(*Caller), DepClassTy::REQUIRED);
- const auto *AssumedGroupSize = A.getAAFor<AAAMDFlatWorkGroupSize>(
- *this, IRPosition::function(*Func), DepClassTy::REQUIRED);
- if (!CallerInfo || !AssumedGroupSize || !CallerInfo->isValidState() ||
- !AssumedGroupSize->isValidState())
+ if (!CallerAA || !CallerAA->isValidState())
return false;
- unsigned Min, Max;
- std::tie(Min, Max) = InfoCache.getEffectiveWavesPerEU(
- *Caller,
- {CallerInfo->getAssumed().getLower().getZExtValue(),
- CallerInfo->getAssumed().getUpper().getZExtValue() - 1},
- {AssumedGroupSize->getAssumed().getLower().getZExtValue(),
- AssumedGroupSize->getAssumed().getUpper().getZExtValue() - 1});
- ConstantRange CallerRange(APInt(32, Min), APInt(32, Max + 1));
- IntegerRangeState CallerRangeState(CallerRange);
- Change |= clampStateAndIndicateChange(this->getState(), CallerRangeState);
+ auto Assumed = this->getAssumed();
+ unsigned Min = std::max(Assumed.getLower().getZExtValue(),
+ CallerAA->getAssumed().getLower().getZExtValue());
+ unsigned Max = std::max(Assumed.getUpper().getZExtValue(),
+ CallerAA->getAssumed().getUpper().getZExtValue());
+ ConstantRange Range(APInt(32, Min), APInt(32, Max));
+ IntegerRangeState RangeState(Range);
+ this->getState() = RangeState;
+ Change |= this->getState() == Assumed ? ChangeStatus::UNCHANGED
+ : ChangeStatus::CHANGED;
return true;
};
@@ -1342,6 +1317,59 @@ static void addPreloadKernArgHint(Function &F, TargetMachine &TM) {
}
}
+static void checkWavesPerEU(Module &M, TargetMachine &TM) {
+ for (Function &F : M) {
+ const GCNSubtarget &ST = TM.getSubtarget<GCNSubtarget>(F);
+
+ auto FlatWgrpSizeAttr =
+ AMDGPU::getIntegerPairAttribute(F, "amdgpu-flat-work-group-size");
+ auto WavesPerEUAttr = AMDGPU::getIntegerPairAttribute(
+ F, "amdgpu-waves-per-eu", /*OnlyFirstRequired=*/true);
+
+ unsigned MinWavesPerEU = ST.getMinWavesPerEU();
+ unsigned MaxWavesPerEU = ST.getMaxWavesPerEU();
+
+ unsigned MinFlatWgrpSize = 1U;
+ unsigned MaxFlatWgrpSize = 1024U;
+ if (FlatWgrpSizeAttr.has_value()) {
+ MinFlatWgrpSize = FlatWgrpSizeAttr->first;
+ MaxFlatWgrpSize = *(FlatWgrpSizeAttr->second);
+ }
+
+ // Start with the max range.
+ unsigned Min = MinWavesPerEU;
+ unsigned Max = MaxWavesPerEU;
+
+ // If the attribute exists, set them to the value from the attribute.
+ if (WavesPerEUAttr.has_value()) {
+ Min = WavesPerEUAttr->first;
+ if (WavesPerEUAttr->second.has_value())
+ Max = *(WavesPerEUAttr->second);
+ }
+
+ // Compute the range from flat workgroup size.
+ auto [MinFromFlatWgrpSize, MaxFromFlatWgrpSize] =
+ ST.getWavesPerEU(F, std::make_pair(MinFlatWgrpSize, MaxFlatWgrpSize));
+
+ // For the lower bound, we have to "tighten" it.
+ Min = std::max(Min, MinFromFlatWgrpSize);
+ // For the upper bound, we have to "extend" it.
+ Max = std::max(Max, MaxFromFlatWgrpSize);
+
+ // Clamp the range to the max range.
+ Min = std::max(Min, MinWavesPerEU);
+ Max = std::min(Max, MaxWavesPerEU);
+
+ // Update the attribute if it is not the max.
+ if (Min != MinWavesPerEU || Max != MaxWavesPerEU) {
+ SmallString<10> Buffer;
+ raw_svector_ostream OS(Buffer);
+ OS << Min << ',' << Max;
+ F.addFnAttr("amdgpu-waves-per-eu", OS.str());
+ }
+ }
+}
+
static bool runImpl(Module &M, AnalysisGetter &AG, TargetMachine &TM,
AMDGPUAttributorOptions Options,
ThinOrFullLTOPhase LTOPhase) {
@@ -1417,8 +1445,14 @@ static bool runImpl(Module &M, AnalysisGetter &AG, TargetMachine &TM,
}
}
- ChangeStatus Change = A.run();
- return Change == ChangeStatus::CHANGED;
+ bool Changed = A.run() == ChangeStatus::CHANGED;
+
+ if (Changed && (LTOPhase == ThinOrFullLTOPhase::None ||
+ LTOPhase == ThinOrFullLTOPhase::FullLTOPostLink ||
+ LTOPhase == ThinOrFullLTOPhase::ThinLTOPostLink))
+ checkWavesPerEU(M, TM);
+
+ return Changed;
}
class AMDGPUAttributorLegacy : public ModulePass {
diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
index 7e9cb7adf4fc2..b2c5801ad1e52 100644
--- a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
@@ -252,13 +252,13 @@ define amdgpu_kernel void @indirect_calls_none_agpr(i1 %cond) {
}
-attributes #0 = { "amdgpu-agpr-alloc"="0" }
+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-flat-scratch-init" "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-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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]] = { "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-agpr-alloc"="0" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR2]] = { "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3:[0-9]+]] = { convergent nocallback nofree nosync nounwind willreturn memory(none) "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR5:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) "amdgpu-waves-per-eu"="4,8" "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 7e0208cd1f45a..28722021e0448 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" "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" }
+; 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" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/attr-amdgpu-max-num-workgroups-propagate.ll b/llvm/test/CodeGen/AMDGPU/attr-amdgpu-max-num-workgroups-propagate.ll
index 678c3a0158ec1..366432e0fc6cb 100644
--- a/llvm/test/CodeGen/AMDGPU/attr-amdgpu-max-num-workgroups-propagate.ll
+++ b/llvm/test/CodeGen/AMDGPU/attr-amdgpu-max-num-workgroups-propagate.ll
@@ -27,7 +27,7 @@ define internal void @callee_1_2_3() {
define amdgpu_kernel void @kernel_1_2_3() #0 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_1_2_3(
-; CHECK-SAME: ) #[[ATTR2:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR1]] {
; CHECK-NEXT: call void @callee_1_2_3()
; CHECK-NEXT: call void @extern_callee()
; CHECK-NEXT: call void @dummy()
@@ -44,7 +44,7 @@ attributes #0 = {"amdgpu-max-num-workgroups"="1,2,3"}
; -> 100,10,99
define internal void @callee_merge_100_8_32__16_10_99() {
; CHECK-LABEL: define internal void @callee_merge_100_8_32__16_10_99(
-; CHECK-SAME: ) #[[ATTR3:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR2:[0-9]+]] {
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
;
@@ -54,7 +54,7 @@ define internal void @callee_merge_100_8_32__16_10_99() {
define amdgpu_kernel void @kernel_100_8_32() #1 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_100_8_32(
-; CHECK-SAME: ) #[[ATTR4:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR3:[0-9]+]] {
; CHECK-NEXT: call void @callee_merge_100_8_32__16_10_99()
; CHECK-NEXT: ret void
;
@@ -64,7 +64,7 @@ define amdgpu_kernel void @kernel_100_8_32() #1 {
define amdgpu_cs void @amdgpu_cs_100_8_32() #1 {
; CHECK-LABEL: define amdgpu_cs void @amdgpu_cs_100_8_32(
-; CHECK-SAME: ) #[[ATTR4]] {
+; CHECK-SAME: ) #[[ATTR3]] {
; CHECK-NEXT: call void @callee_merge_100_8_32__16_10_99()
; CHECK-NEXT: ret void
;
@@ -76,7 +76,7 @@ attributes #1 = {"amdgpu-max-num-workgroups"="100,8,32"}
define amdgpu_kernel void @kernel_16_10_99() #2 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_16_10_99(
-; CHECK-SAME: ) #[[ATTR5:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR4:[0-9]+]] {
; CHECK-NEXT: call void @callee_merge_100_8_32__16_10_99()
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
@@ -110,7 +110,7 @@ define internal void @callee_x_worst_case() {
define amdgpu_kernel void @kernel_x_maximum() #3 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_x_maximum(
-; CHECK-SAME: ) #[[ATTR6:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR5:[0-9]+]] {
; CHECK-NEXT: call void @merge_to_worst_case()
; CHECK-NEXT: call void @callee_x_worst_case()
; CHECK-NEXT: call void @dummy()
@@ -126,7 +126,7 @@ attributes #3 = {"amdgpu-max-num-workgroups"="4294967295,1,1"}
define amdgpu_kernel void @kernel_y_maximum() #4 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_y_maximum(
-; CHECK-SAME: ) #[[ATTR7:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR6:[0-9]+]] {
; CHECK-NEXT: call void @merge_to_worst_case()
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
@@ -140,7 +140,7 @@ attributes #4 = {"amdgpu-max-num-workgroups"="1,4294967295,1"}
define amdgpu_kernel void @kernel_z_maximum() #5 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_z_maximum(
-; CHECK-SAME: ) #[[ATTR8:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR7:[0-9]+]] {
; CHECK-NEXT: call void @merge_to_worst_case()
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
@@ -155,7 +155,7 @@ attributes #5 = {"amdgpu-max-num-workgroups"="1,1,4294967295"}
; Make sure the attribute isn't lost from the callee.
define internal void @annotated_callee_from_unannotated_kernel() #6 {
; CHECK-LABEL: define internal void @annotated_callee_from_unannotated_kernel(
-; CHECK-SAME: ) #[[ATTR9:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR8:[0-9]+]] {
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
;
@@ -167,7 +167,7 @@ attributes #6 = {"amdgpu-max-num-workgroups"="42,99,123"}
define amdgpu_kernel void @unannotated_kernel_calls_annotated_callee() {
; CHECK-LABEL: define amdgpu_kernel void @unannotated_kernel_calls_annotated_callee(
-; CHECK-SAME: ) #[[ATTR10:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR0]] {
; CHECK-NEXT: call void @annotated_callee_from_unannotated_kernel()
; CHECK-NEXT: ret void
;
@@ -178,7 +178,7 @@ define amdgpu_kernel void @unannotated_kernel_calls_annotated_callee() {
define internal void @annotated_callee_merge_caller() #7 {
; CHECK-LABEL: define internal void @annotated_callee_merge_caller(
-; CHECK-SAME: ) #[[ATTR11:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR9:[0-9]+]] {
; CHECK-NEXT: call void @dummy()
; CHECK-NEXT: ret void
;
@@ -190,7 +190,7 @@ attributes #7 = {"amdgpu-max-num-workgroups"="512,256,1024"}
define amdgpu_kernel void @call_annotated_callee_merge_caller() #8 {
; CHECK-LABEL: define amdgpu_kernel void @call_annotated_callee_merge_caller(
-; CHECK-SAME: ) #[[ATTR12:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR10:[0-9]+]] {
; CHECK-NEXT: call void @annotated_callee_merge_caller()
; CHECK-NEXT: ret void
;
@@ -212,7 +212,7 @@ define internal void @called_by_explicit_worst_case() {
define amdgpu_kernel void @kernel_explicit_worst_case() #9 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_explicit_worst_case(
-; CHECK-SAME: ) #[[ATTR13:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR11:[0-9]+]] {
; CHECK-NEXT: call void @called_by_explicit_worst_case()
; CHECK-NEXT: ret void
;
@@ -223,18 +223,16 @@ define amdgpu_kernel void @kernel_explicit_worst_case() #9 {
attributes #9 = {"amdgpu-max-num-workgroups"="4294967295,4294967295,4294967295"}
;.
-; CHECK: attributes #[[ATTR0]] = { "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-max-num-workgroups"="1,2,3" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-max-num-workgroups"="1,2,3" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3]] = { "amdgpu-max-num-workgroups"="100,10,99" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { "amdgpu-max-num-workgroups"="100,8,32" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR5]] = { "amdgpu-max-num-workgroups"="16,10,99" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR6]] = { "amdgpu-max-num-workgroups"="4294967295,1,1" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR7]] = { "amdgpu-max-num-workgroups"="1,4294967295,1" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR8]] = { "amdgpu-max-num-workgroups"="1,1,4294967295" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR9]] = { "amdgpu-max-num-workgroups"="42,99,123" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR10]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR11]] = { "amdgpu-max-num-workgroups"="256,128,1024" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR12]] = { "amdgpu-max-num-workgroups"="256,128,2048" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR13]] = { "amdgpu-max-num-workgroups"="4294967295,4294967295,4294967295" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-max-num-workgroups"="1,2,3" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { "amdgpu-max-num-workgroups"="100,10,99" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-max-num-workgroups"="100,8,32" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR4]] = { "amdgpu-max-num-workgroups"="16,10,99" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR5]] = { "amdgpu-max-num-workgroups"="4294967295,1,1" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR6]] = { "amdgpu-max-num-workgroups"="1,4294967295,1" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR7]] = { "amdgpu-max-num-workgroups"="1,1,4294967295" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR8]] = { "amdgpu-max-num-workgroups"="42,99,123" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR9]] = { "amdgpu-max-num-workgroups"="256,128,1024" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR10]] = { "amdgpu-max-num-workgroups"="256,128,2048" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR11]] = { "amdgpu-max-num-workgroups"="4294967295,4294967295,4294967295" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
index a3d6c3b729523..3286d78b758fd 100644
--- a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
+++ b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
@@ -865,17 +865,17 @@ define amdgpu_kernel void @with_inline_asm() {
}
;.
-; GFX9: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx900" "uniform-work-group-size"="false" }
-; GFX9: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="gfx900" "uniform-work-group-size"="false" }
-; GFX9: attributes #[[ATTR2]] = { "target-cpu"="gfx900" "uniform-work-group-size"="false" }
-; GFX9: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx900" "uniform-work-group-size"="false" }
-; GFX9: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx900" }
-; GFX9: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR2]] = { "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" }
+; GFX9: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
;.
-; GFX10: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx1010" "uniform-work-group-size"="false" }
-; GFX10: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="gfx1010" "uniform-work-group-size"="false" }
-; GFX10: attributes #[[ATTR2]] = { "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
-; GFX10: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx1010" "uniform-work-group-size"="false" }
-; GFX10: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx1010" }
-; GFX10: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR2]] = { "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" }
+; GFX10: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/attributor-loop-issue-58639.ll b/llvm/test/CodeGen/AMDGPU/attributor-loop-issue-58639.ll
index 8481cea4d7c35..2b9f579e6a183 100644
--- a/llvm/test/CodeGen/AMDGPU/attributor-loop-issue-58639.ll
+++ b/llvm/test/CodeGen/AMDGPU/attributor-loop-issue-58639.ll
@@ -51,7 +51,7 @@ bb5: ; preds = %bb5, %bb3
define amdgpu_kernel void @entry() {
; CHECK-LABEL: define {{[^@]+}}@entry
-; CHECK-SAME: () #[[ATTR1:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR0]] {
; CHECK-NEXT: [[ALLOCA:%.*]] = alloca [[TMP0:%.*]], align 8, addrspace(5)
; CHECK-NEXT: [[CAST:%.*]] = addrspacecast ptr addrspace(5) [[ALLOCA]] to ptr
; CHECK-NEXT: [[ARST:%.*]] = call double @baz(ptr [[CAST]])
@@ -63,6 +63,5 @@ define amdgpu_kernel void @entry() {
ret void
}
;.
-; CHECK: attributes #[[ATTR0]] = { "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll b/llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll
index b8d4fcb7c99a7..20f48dac61d5c 100644
--- a/llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll
@@ -27,7 +27,7 @@ define internal void @direct() {
define amdgpu_kernel void @test_direct_indirect_call() {
; CHECK-LABEL: define {{[^@]+}}@test_direct_indirect_call
-; CHECK-SAME: () #[[ATTR2:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR1]] {
; CHECK-NEXT: call void @direct()
; CHECK-NEXT: ret void
;
@@ -35,7 +35,6 @@ define amdgpu_kernel void @test_direct_indirect_call() {
ret void
}
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll b/llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll
index f964170ccdda5..38d4998a098e5 100644
--- a/llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll
+++ b/llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll
@@ -29,6 +29,6 @@ attributes #0 = { "amdgpu-no-dispatch-id" }
;.
;.
-; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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_GCN: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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_GCN: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_GCN: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll b/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
index 2a693e1001cd9..e35057aec2b13 100644
--- a/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
+++ b/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
@@ -275,24 +275,24 @@ attributes #0 = { nocallback nofree nosync nounwind speculatable willreturn memo
;.
-; V4: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; V4: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V4: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V4: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V4: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V4: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
+; V4: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V4: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V4: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V4: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V4: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V4: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
-; V5: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; V5: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V5: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V5: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V5: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
+; V5: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V5: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V5: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V5: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V5: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
-; V6: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; V6: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V6: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V6: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
-; V6: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "uniform-work-group-size"="false" }
+; V6: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V6: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V6: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V6: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; V6: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
; V4: [[META0:![0-9]+]] = !{i32 1, !"amdhsa_code_object_version", i32 400}
;.
diff --git a/llvm/test/CodeGen/AMDGPU/indirect-call-set-from-other-function.ll b/llvm/test/CodeGen/AMDGPU/indirect-call-set-from-other-function.ll
index d09b4fda1b697..9d98c70196dff 100644
--- a/llvm/test/CodeGen/AMDGPU/indirect-call-set-from-other-function.ll
+++ b/llvm/test/CodeGen/AMDGPU/indirect-call-set-from-other-function.ll
@@ -68,6 +68,6 @@ if.end:
ret void
}
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/issue120256-annotate-constexpr-addrspacecast.ll b/llvm/test/CodeGen/AMDGPU/issue120256-annotate-constexpr-addrspacecast.ll
index 8e87256c24ce5..7af6d3a0c640e 100644
--- a/llvm/test/CodeGen/AMDGPU/issue120256-annotate-constexpr-addrspacecast.ll
+++ b/llvm/test/CodeGen/AMDGPU/issue120256-annotate-constexpr-addrspacecast.ll
@@ -55,8 +55,8 @@ define amdgpu_kernel void @issue120256_private(ptr addrspace(1) %out) {
; FIXME: Inference of amdgpu-no-queue-ptr should not depend on code object version.
!0 = !{i32 1, !"amdhsa_code_object_version", i32 400}
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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"="gfx803" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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-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"="gfx803" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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" "target-cpu"="gfx803" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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-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" "target-cpu"="gfx803" "uniform-work-group-size"="false" }
;.
; CHECK: [[META0:![0-9]+]] = !{i32 1, !"amdhsa_code_object_version", i32 400}
;.
diff --git a/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll b/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
index 971b0dbc730c5..31f8e0ea1f8fa 100644
--- a/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
@@ -59,6 +59,6 @@ declare i64 @llvm.amdgcn.s.getpc() #0
attributes #0 = { nounwind readnone speculatable willreturn }
;.
;.
-; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "uniform-work-group-size"="false" }
-; ATTRIBUTOR_GCN: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
+; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_GCN: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll b/llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll
index 74d908f43b5ab..d6bf3670a9145 100644
--- a/llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll
+++ b/llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll
@@ -170,10 +170,10 @@ define amdgpu_kernel void @kernel_64_256() #7 {
define internal void @default_captured_address() {
; CHECK-LABEL: define {{[^@]+}}@default_captured_address
; CHECK-SAME: () #[[ATTR8:[0-9]+]] {
-; CHECK-NEXT: store volatile ptr @default_captured_address, ptr poison, align 8
+; CHECK-NEXT: store volatile ptr @default_captured_address, ptr undef, align 8
; CHECK-NEXT: ret void
;
- store volatile ptr @default_captured_address, ptr poison, align 8
+ store volatile ptr @default_captured_address, ptr undef, align 8
ret void
}
@@ -211,5 +211,5 @@ attributes #7 = { "amdgpu-flat-work-group-size"="64,256" }
; CHECK: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="512,1024" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR6]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="512,512" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,10" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR7]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="64,256" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR8]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR8]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/propagate-waves-per-eu.ll b/llvm/test/CodeGen/AMDGPU/propagate-waves-per-eu.ll
index ae114f3213d8f..5c33ba510668b 100644
--- a/llvm/test/CodeGen/AMDGPU/propagate-waves-per-eu.ll
+++ b/llvm/test/CodeGen/AMDGPU/propagate-waves-per-eu.ll
@@ -14,7 +14,7 @@ define internal void @default_to_1_8_a() {
define amdgpu_kernel void @kernel_1_8() #0 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_1_8
-; CHECK-SAME: () #[[ATTR0]] {
+; CHECK-SAME: () #[[ATTR1:[0-9]+]] {
; CHECK-NEXT: call void @default_to_1_8_a()
; CHECK-NEXT: ret void
;
@@ -25,7 +25,7 @@ define amdgpu_kernel void @kernel_1_8() #0 {
; Called from a single kernel with 1,2
define internal void @default_to_1_2() {
; CHECK-LABEL: define internal void @default_to_1_2
-; CHECK-SAME: () #[[ATTR1:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR2:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -33,7 +33,7 @@ define internal void @default_to_1_2() {
define amdgpu_kernel void @kernel_1_2() #1 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_1_2
-; CHECK-SAME: () #[[ATTR1]] {
+; CHECK-SAME: () #[[ATTR2]] {
; CHECK-NEXT: call void @default_to_1_2()
; CHECK-NEXT: call void @flat_group_1_1()
; CHECK-NEXT: call void @default_to_1_8_b()
@@ -50,7 +50,7 @@ define amdgpu_kernel void @kernel_1_2() #1 {
; Called from a single kernel with 1,4
define internal void @default_to_1_4() {
; CHECK-LABEL: define internal void @default_to_1_4
-; CHECK-SAME: () #[[ATTR2:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR3:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -58,7 +58,7 @@ define internal void @default_to_1_4() {
define amdgpu_kernel void @kernel_1_4() #2 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_1_4
-; CHECK-SAME: () #[[ATTR2]] {
+; CHECK-SAME: () #[[ATTR3]] {
; CHECK-NEXT: call void @default_to_1_4()
; CHECK-NEXT: ret void
;
@@ -69,7 +69,7 @@ define amdgpu_kernel void @kernel_1_4() #2 {
; Called from kernels with 2,9 and 9,9
define internal void @default_to_2_9() {
; CHECK-LABEL: define internal void @default_to_2_9
-; CHECK-SAME: () #[[ATTR3:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR4:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -79,7 +79,7 @@ define internal void @default_to_2_9() {
; bounds, and should not be changed.
define internal void @flat_group_1_1() #3 {
; CHECK-LABEL: define internal void @flat_group_1_1
-; CHECK-SAME: () #[[ATTR4:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR5:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -88,7 +88,7 @@ define internal void @flat_group_1_1() #3 {
; 2,8 -> 2,2
define internal void @flat_group_2_8() #4 {
; CHECK-LABEL: define internal void @flat_group_2_8
-; CHECK-SAME: () #[[ATTR5:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR0]] {
; CHECK-NEXT: ret void
;
ret void
@@ -105,7 +105,7 @@ define internal void @flat_group_9_10() #5 {
define amdgpu_kernel void @kernel_2_9() #6 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_2_9
-; CHECK-SAME: () #[[ATTR3]] {
+; CHECK-SAME: () #[[ATTR7:[0-9]+]] {
; CHECK-NEXT: call void @default_to_2_9()
; CHECK-NEXT: call void @flat_group_1_1()
; CHECK-NEXT: ret void
@@ -117,7 +117,7 @@ define amdgpu_kernel void @kernel_2_9() #6 {
define amdgpu_kernel void @kernel_9_9() #7 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_9_9
-; CHECK-SAME: () #[[ATTR7:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR4]] {
; CHECK-NEXT: call void @default_to_2_9()
; CHECK-NEXT: call void @flat_group_9_10()
; CHECK-NEXT: ret void
@@ -140,7 +140,7 @@ define internal void @default_to_1_8_b() {
; this should probably be illegal.
define amdgpu_kernel void @kernel_2_8() #4 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_2_8
-; CHECK-SAME: () #[[ATTR5]] {
+; CHECK-SAME: () #[[ATTR0]] {
; CHECK-NEXT: call void @default_to_1_8_a()
; CHECK-NEXT: call void @default_to_1_8_b()
; CHECK-NEXT: ret void
@@ -153,7 +153,7 @@ define amdgpu_kernel void @kernel_2_8() #4 {
; 1,2 -> 2,2
define internal void @merge_cycle_0() #1 {
; CHECK-LABEL: define internal void @merge_cycle_0
-; CHECK-SAME: () #[[ATTR1]] {
+; CHECK-SAME: () #[[ATTR2]] {
; CHECK-NEXT: call void @merge_cycle_1()
; CHECK-NEXT: ret void
;
@@ -165,7 +165,7 @@ define internal void @merge_cycle_0() #1 {
; 2,8 -> 2,8
define internal void @merge_cycle_1() #4 {
; CHECK-LABEL: define internal void @merge_cycle_1
-; CHECK-SAME: () #[[ATTR5]] {
+; CHECK-SAME: () #[[ATTR0]] {
; CHECK-NEXT: call void @merge_cycle_0()
; CHECK-NEXT: ret void
;
@@ -192,10 +192,10 @@ define amdgpu_kernel void @kernel_3_8() #8 {
define internal void @default_captured_address() {
; CHECK-LABEL: define internal void @default_captured_address
; CHECK-SAME: () #[[ATTR9:[0-9]+]] {
-; CHECK-NEXT: store volatile ptr @default_captured_address, ptr poison, align 8
+; CHECK-NEXT: store volatile ptr @default_captured_address, ptr undef, align 8
; CHECK-NEXT: ret void
;
- store volatile ptr @default_captured_address, ptr poison, align 8
+ store volatile ptr @default_captured_address, ptr undef, align 8
ret void
}
@@ -218,7 +218,7 @@ define internal i32 @bitcasted_function() {
define internal void @called_from_invalid_bounds_0() {
; CHECK-LABEL: define internal void @called_from_invalid_bounds_0
-; CHECK-SAME: () #[[ATTR10:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR1]] {
; CHECK-NEXT: ret void
;
ret void
@@ -226,7 +226,7 @@ define internal void @called_from_invalid_bounds_0() {
define internal void @called_from_invalid_bounds_1() {
; CHECK-LABEL: define internal void @called_from_invalid_bounds_1
-; CHECK-SAME: () #[[ATTR10]] {
+; CHECK-SAME: () #[[ATTR10:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -235,7 +235,7 @@ define internal void @called_from_invalid_bounds_1() {
; Invalid range for amdgpu-waves-per-eu
define amdgpu_kernel void @kernel_invalid_bounds_0_8() #9 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_invalid_bounds_0_8
-; CHECK-SAME: () #[[ATTR0]] {
+; CHECK-SAME: () #[[ATTR1]] {
; CHECK-NEXT: call void @called_from_invalid_bounds_0()
; CHECK-NEXT: ret void
;
@@ -279,7 +279,7 @@ define amdgpu_kernel void @kernel_3_6() #12 {
; 3,6 -> 6,9
define internal void @refine_upper_func_3_6() #13 {
; CHECK-LABEL: define internal void @refine_upper_func_3_6
-; CHECK-SAME: () #[[ATTR14:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR9]] {
; CHECK-NEXT: ret void
;
ret void
@@ -288,7 +288,7 @@ define internal void @refine_upper_func_3_6() #13 {
; 4,8 -> 6,8
define internal void @refine_lower_func_4_8() #14 {
; CHECK-LABEL: define internal void @refine_lower_func_4_8
-; CHECK-SAME: () #[[ATTR15:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR14:[0-9]+]] {
; CHECK-NEXT: call void @refine_upper_func_3_6()
; CHECK-NEXT: ret void
;
@@ -298,7 +298,7 @@ define internal void @refine_lower_func_4_8() #14 {
define amdgpu_kernel void @kernel_foo_6_8() #15 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_foo_6_8
-; CHECK-SAME: () #[[ATTR16:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR15:[0-9]+]] {
; CHECK-NEXT: call void @refine_upper_func_3_6()
; CHECK-NEXT: call void @refine_lower_func_4_8()
; CHECK-NEXT: call void @func_9_10_a()
@@ -313,7 +313,7 @@ define amdgpu_kernel void @kernel_foo_6_8() #15 {
; 5,5 -> 5,5
define internal void @func_5_5() #16 {
; CHECK-LABEL: define internal void @func_5_5
-; CHECK-SAME: () #[[ATTR17:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR16:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -322,7 +322,7 @@ define internal void @func_5_5() #16 {
; 5,8 -> 8,8
define internal void @func_5_8() #17 {
; CHECK-LABEL: define internal void @func_5_8
-; CHECK-SAME: () #[[ATTR18:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR17:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -331,7 +331,7 @@ define internal void @func_5_8() #17 {
; 9,10 -> 9,10
define internal void @func_9_10_a() #18 {
; CHECK-LABEL: define internal void @func_9_10_a
-; CHECK-SAME: () #[[ATTR19:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR18:[0-9]+]] {
; CHECK-NEXT: ret void
;
ret void
@@ -340,7 +340,7 @@ define internal void @func_9_10_a() #18 {
; 9,10 -> 9,9
define internal void @func_9_10_b() #18 {
; CHECK-LABEL: define internal void @func_9_10_b
-; CHECK-SAME: () #[[ATTR19]] {
+; CHECK-SAME: () #[[ATTR18]] {
; CHECK-NEXT: ret void
;
ret void
@@ -348,7 +348,7 @@ define internal void @func_9_10_b() #18 {
define amdgpu_kernel void @kernel_bar_8_9() #19 {
; CHECK-LABEL: define amdgpu_kernel void @kernel_bar_8_9
-; CHECK-SAME: () #[[ATTR20:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR19:[0-9]+]] {
; CHECK-NEXT: call void @refine_upper_func_3_6()
; CHECK-NEXT: call void @func_5_5()
; CHECK-NEXT: call void @func_9_10_b()
@@ -399,25 +399,24 @@ attributes #17 = { "amdgpu-waves-per-eu"="5,8" }
attributes #18 = { "amdgpu-waves-per-eu"="9,10" }
attributes #19 = { "amdgpu-waves-per-eu"="8,9" }
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,8" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,2" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,4" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,9" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,1" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,8" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,8" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,8" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,2" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,4" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="9,9" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,1" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR6]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="9,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR7]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="9,9" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR7]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,9" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR8]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="3,8" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR9]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR9]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR10]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
; CHECK: attributes #[[ATTR11]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,64" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="1,123" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR12]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,512" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR12]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,512" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="2,10" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR13]] = { "amdgpu-agpr-alloc"="0" "amdgpu-flat-work-group-size"="1,512" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="3,6" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR14]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="3,6" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR15]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR16]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="6,8" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR17]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="5,5" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR18]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="5,8" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR19]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="9,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR20]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,9" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR14]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR15]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="6,8" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR16]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="5,5" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR17]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="5,8" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR18]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="9,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR19]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,9" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/recursive_global_initializer.ll b/llvm/test/CodeGen/AMDGPU/recursive_global_initializer.ll
index 2b07fc716e8b9..6e9e4e47ffa0f 100644
--- a/llvm/test/CodeGen/AMDGPU/recursive_global_initializer.ll
+++ b/llvm/test/CodeGen/AMDGPU/recursive_global_initializer.ll
@@ -19,5 +19,5 @@ define void @hoge() {
ret void
}
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/remove-no-kernel-id-attribute.ll b/llvm/test/CodeGen/AMDGPU/remove-no-kernel-id-attribute.ll
index 56409999f1b95..ed4e691fbf154 100644
--- a/llvm/test/CodeGen/AMDGPU/remove-no-kernel-id-attribute.ll
+++ b/llvm/test/CodeGen/AMDGPU/remove-no-kernel-id-attribute.ll
@@ -179,7 +179,7 @@ define internal void @mutual_recursion_1(i16 %arg) {
define amdgpu_kernel void @kernel_lds_recursion() {
; CHECK-LABEL: define amdgpu_kernel void @kernel_lds_recursion(
; CHECK-SAME: ) #[[ATTR5:[0-9]+]] !llvm.amdgcn.lds.kernel.id [[META9:![0-9]+]] {
-; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(ptr addrspace(3) @llvm.amdgcn.kernel.kernel_lds_recursion.lds) ]
+; CHECK-NEXT: call void @llvm.donothing() [ "ExplicitUse"(ptr addrspace(3) @llvm.amdgcn.kernel.kernel_lds_recursion.lds) ], !alias.scope [[META10:![0-9]+]], !noalias [[META13:![0-9]+]]
; CHECK-NEXT: call void @mutual_recursion_0(i16 0)
; CHECK-NEXT: ret void
;
@@ -191,11 +191,11 @@ define amdgpu_kernel void @kernel_lds_recursion() {
!1 = !{i32 1, !"amdhsa_code_object_version", i32 400}
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "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" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "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" }
-; CHECK: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-lds-size"="2" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "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" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "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" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "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" }
+; CHECK: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-lds-size"="2" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "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" }
; CHECK: attributes #[[ATTR3]] = { "amdgpu-lds-size"="4" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-lds-size"="2" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-lds-size"="2" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
; CHECK: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-lds-size"="4" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "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" }
; CHECK: attributes #[[ATTR6:[0-9]+]] = { nocallback nofree nosync nounwind willreturn memory(none) }
; CHECK: attributes #[[ATTR7:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
@@ -210,6 +210,11 @@ define amdgpu_kernel void @kernel_lds_recursion() {
; CHECK: [[META7]] = !{[[META8:![0-9]+]]}
; CHECK: [[META8]] = distinct !{[[META8]], [[META6]]}
; CHECK: [[META9]] = !{i32 2}
+; CHECK: [[META10]] = !{[[META11:![0-9]+]]}
+; CHECK: [[META11]] = distinct !{[[META11]], [[META12:![0-9]+]]}
+; CHECK: [[META12]] = distinct !{[[META12]]}
+; CHECK: [[META13]] = !{[[META14:![0-9]+]]}
+; CHECK: [[META14]] = distinct !{[[META14]], [[META12]]}
;.
;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
; TABLE: {{.*}}
diff --git a/llvm/test/CodeGen/AMDGPU/simple-indirect-call-2.ll b/llvm/test/CodeGen/AMDGPU/simple-indirect-call-2.ll
index 35108d8fcfb1e..6ffc8ca407431 100644
--- a/llvm/test/CodeGen/AMDGPU/simple-indirect-call-2.ll
+++ b/llvm/test/CodeGen/AMDGPU/simple-indirect-call-2.ll
@@ -98,14 +98,14 @@ entry:
}
;.
-; NO: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; NO: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; NO: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; NO: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
-; OW: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; OW: attributes #[[ATTR1]] = { "uniform-work-group-size"="false" }
+; OW: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; OW: attributes #[[ATTR1]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
-; CW: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CW: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CW: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CW: attributes #[[ATTR1]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
; NO: [[META0]] = !{ptr @bar1, ptr @bar2}
;.
diff --git a/llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll b/llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll
index 80e8d7485a8f3..71544c82d0aee 100644
--- a/llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll
@@ -63,6 +63,6 @@ define amdgpu_kernel void @test_simple_indirect_call() {
;.
;.
-; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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_GCN: attributes #[[ATTR1]] = { "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_GCN: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_GCN: attributes #[[ATTR1]] = { "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" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll b/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
index 6a88be6e55859..cb0a7a146104a 100644
--- a/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
+++ b/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
@@ -826,5 +826,5 @@ entry:
; GCN-PRELINK: declare float @_Z4cbrtf(float) local_unnamed_addr #[[$NOUNWIND_READONLY:[0-9]+]]
; GCN-PRELINK-DAG: attributes #[[$NOUNWIND]] = { nounwind }
-; GCN-PRELINK-DAG: attributes #[[$NOUNWIND_READONLY]] = { nounwind memory(read) "uniform-work-group-size"="false" }
+; GCN-PRELINK-DAG: attributes #[[$NOUNWIND_READONLY]] = { nounwind memory(read) "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll
index 473d996bf721d..0fc655af37fa1 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll
@@ -31,5 +31,5 @@ define amdgpu_kernel void @kernel1() #1 {
attributes #0 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll
index d7641f976968b..b5c14c56b0aae 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll
@@ -87,7 +87,7 @@ define internal void @internal2() {
define amdgpu_kernel void @kernel2() #0 {
; CHECK-LABEL: define {{[^@]+}}@kernel2
-; CHECK-SAME: () #[[ATTR4:[0-9]+]] {
+; CHECK-SAME: () #[[ATTR3]] {
; CHECK-NEXT: call void @internal2()
; CHECK-NEXT: ret void
;
@@ -97,9 +97,8 @@ define amdgpu_kernel void @kernel2() #0 {
attributes #0 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR2]] = { "uniform-work-group-size"="true" }
-; CHECK: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="true" }
-; CHECK: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="true" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll
index 308f8b595eb06..6464ff3a50dff 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll
@@ -41,6 +41,6 @@ define amdgpu_kernel void @kernel3() #2 {
attributes #2 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="true" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll
index 7e2b085f5a879..ea3c0a33c2da0 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll
@@ -41,6 +41,6 @@ define amdgpu_kernel void @kernel2() #2 {
attributes #1 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="true" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll
index 3d6454cc9f99b..3823d17658fd5 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll
@@ -52,8 +52,8 @@ attributes #0 = { nounwind }
attributes #1 = { "uniform-work-group-size"="false" }
attributes #2 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR2]] = { nounwind "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3]] = { "uniform-work-group-size"="true" }
+; CHECK: attributes #[[ATTR0]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { nounwind "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="true" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll
index 6cd622a582d18..e4783597b5d00 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll
@@ -101,7 +101,7 @@ define amdgpu_kernel void @kernel(ptr addrspace(1) %m) #1 {
attributes #0 = { nounwind readnone }
attributes #1 = { "uniform-work-group-size"="true" }
;.
-; CHECK: attributes #[[ATTR0]] = { nounwind memory(none) "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR1]] = { nounwind memory(none) "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="true" }
+; CHECK: attributes #[[ATTR0]] = { nounwind memory(none) "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { nounwind memory(none) "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
; CHECK: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="true" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll b/llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll
index e315e0454f424..1c054fb97ee19 100644
--- a/llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll
+++ b/llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll
@@ -61,5 +61,5 @@ define amdgpu_kernel void @kernel3() #0 {
attributes #0 = { "uniform-work-group-size"="false" }
;.
-; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; CHECK: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
;.
>From 5b7ec580cbe0df6330fc547f6251744a2b500579 Mon Sep 17 00:00:00 2001
From: Shilei Tian <i at tianshilei.me>
Date: Mon, 24 Mar 2025 13:54:48 -0400
Subject: [PATCH 2/3] resolve review comments
---
llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp | 29 +++++++++------------
1 file changed, 13 insertions(+), 16 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
index a49e41a79bab7..448b5322d6294 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
@@ -1149,7 +1149,7 @@ struct AAAMDWavesPerEU : public AAAMDSizeRangeAttribute {
if (!CallerAA || !CallerAA->isValidState())
return false;
- auto Assumed = this->getAssumed();
+ ConstantRange Assumed = this->getAssumed();
unsigned Min = std::max(Assumed.getLower().getZExtValue(),
CallerAA->getAssumed().getLower().getZExtValue());
unsigned Max = std::max(Assumed.getUpper().getZExtValue(),
@@ -1317,20 +1317,23 @@ static void addPreloadKernArgHint(Function &F, TargetMachine &TM) {
}
}
-static void checkWavesPerEU(Module &M, TargetMachine &TM) {
+/// The final check and update of the attribute 'amdgpu-waves-per-eu' based on
+/// the determined 'amdgpu-flat-work-group-size' attribute. We can't do this
+/// during attributor run because the two attributes grow in opposite direction,
+/// we should not use any intermediate value to calculate waves per eu until we
+/// have a determined flat workgroup size.
+static void updateWavesPerEU(Module &M, TargetMachine &TM) {
for (Function &F : M) {
const GCNSubtarget &ST = TM.getSubtarget<GCNSubtarget>(F);
auto FlatWgrpSizeAttr =
AMDGPU::getIntegerPairAttribute(F, "amdgpu-flat-work-group-size");
- auto WavesPerEUAttr = AMDGPU::getIntegerPairAttribute(
- F, "amdgpu-waves-per-eu", /*OnlyFirstRequired=*/true);
unsigned MinWavesPerEU = ST.getMinWavesPerEU();
unsigned MaxWavesPerEU = ST.getMaxWavesPerEU();
- unsigned MinFlatWgrpSize = 1U;
- unsigned MaxFlatWgrpSize = 1024U;
+ unsigned MinFlatWgrpSize = ST.getMinFlatWorkGroupSize();
+ unsigned MaxFlatWgrpSize = ST.getMaxFlatWorkGroupSize();
if (FlatWgrpSizeAttr.has_value()) {
MinFlatWgrpSize = FlatWgrpSizeAttr->first;
MaxFlatWgrpSize = *(FlatWgrpSizeAttr->second);
@@ -1338,16 +1341,10 @@ static void checkWavesPerEU(Module &M, TargetMachine &TM) {
// Start with the max range.
unsigned Min = MinWavesPerEU;
- unsigned Max = MaxWavesPerEU;
+ unsigned Max = MinWavesPerEU;
- // If the attribute exists, set them to the value from the attribute.
- if (WavesPerEUAttr.has_value()) {
- Min = WavesPerEUAttr->first;
- if (WavesPerEUAttr->second.has_value())
- Max = *(WavesPerEUAttr->second);
- }
-
- // Compute the range from flat workgroup size.
+ // Compute the range from flat workgroup size. `getWavesPerEU` will also
+ // account for the 'amdgpu-waves-er-eu' attribute.
auto [MinFromFlatWgrpSize, MaxFromFlatWgrpSize] =
ST.getWavesPerEU(F, std::make_pair(MinFlatWgrpSize, MaxFlatWgrpSize));
@@ -1450,7 +1447,7 @@ static bool runImpl(Module &M, AnalysisGetter &AG, TargetMachine &TM,
if (Changed && (LTOPhase == ThinOrFullLTOPhase::None ||
LTOPhase == ThinOrFullLTOPhase::FullLTOPostLink ||
LTOPhase == ThinOrFullLTOPhase::ThinLTOPostLink))
- checkWavesPerEU(M, TM);
+ updateWavesPerEU(M, TM);
return Changed;
}
>From 63b735b47b341b5e191ccb6613df383c882b6653 Mon Sep 17 00:00:00 2001
From: Shilei Tian <i at tianshilei.me>
Date: Thu, 1 May 2025 13:40:40 -0400
Subject: [PATCH 3/3] rebase and resolve review comments
---
.../builtins-amdgcn-gfx13-load-mcast.cl | 14 -
clang/test/CodeGenOpenCL/builtins-amdgcn.cl | 907 ------------------
llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp | 12 +-
llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp | 9 +
llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h | 7 +
.../AMDGPU/addrspacecast-constantexpr.ll | 4 +-
.../AMDGPU/amdgpu-attributor-no-agpr.ll | 25 +-
.../annotate-kernel-features-hsa-call.ll | 70 +-
.../AMDGPU/annotate-kernel-features-hsa.ll | 26 +-
.../AMDGPU/annotate-kernel-features.ll | 18 +-
...utor-flatscratchinit-undefined-behavior.ll | 4 +-
.../AMDGPU/attributor-flatscratchinit.ll | 4 +-
.../AMDGPU/implicitarg-offset-attributes.ll | 6 +-
llvm/test/CodeGen/AMDGPU/inline-attr.ll | 19 +-
.../AMDGPU/pal-simple-indirect-call.ll | 3 +-
llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll | 2 +-
16 files changed, 116 insertions(+), 1014 deletions(-)
delete mode 100644 clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
delete mode 100644 clang/test/CodeGenOpenCL/builtins-amdgcn.cl
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
deleted file mode 100644
index 4e8b6744ba7a9..0000000000000
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn-gfx13-load-mcast.cl
+++ /dev/null
@@ -1,14 +0,0 @@
-// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// REQUIRES: amdgpu-registered-target
-// RUN: %clang_cc1 -cl-std=CL2.0 -triple amdgcn-unknown-unknown -target-cpu gfx950 -emit-llvm -o - %s | FileCheck %s --check-prefix=CHECK-GFX13
-
-void test_amdgcn_load_mcast_b32(local int* outptr, global int* inptr, int mask)
-{
- __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
-}
-
-//void test_amdgcn_load_mcast_local_b32(local void* outptr, local void* inptr, int mask)
-//{
-// __builtin_amdgcn_load_mcast_b32(outptr, inptr, 10, mask);
-//}
-
diff --git a/clang/test/CodeGenOpenCL/builtins-amdgcn.cl b/clang/test/CodeGenOpenCL/builtins-amdgcn.cl
deleted file mode 100644
index d97fa62ab7249..0000000000000
--- a/clang/test/CodeGenOpenCL/builtins-amdgcn.cl
+++ /dev/null
@@ -1,907 +0,0 @@
-// REQUIRES: amdgpu-registered-target
-// RUN: %clang_cc1 -cl-std=CL2.0 -triple amdgcn-unknown-unknown -target-cpu tahiti -emit-llvm -o - %s | FileCheck -enable-var-scope --check-prefixes=CHECK-AMDGCN,CHECK %s
-// RUN: %clang_cc1 -cl-std=CL2.0 -triple spirv64-amd-amdhsa -emit-llvm -o - %s | FileCheck -enable-var-scope --check-prefixes=CHECK,CHECK-SPIRV %s
-
-
-#pragma OPENCL EXTENSION cl_khr_fp64 : enable
-
-typedef unsigned long ulong;
-typedef unsigned int uint;
-typedef unsigned short ushort;
-typedef half __attribute__((ext_vector_type(2))) half2;
-typedef short __attribute__((ext_vector_type(2))) short2;
-typedef ushort __attribute__((ext_vector_type(2))) ushort2;
-typedef uint __attribute__((ext_vector_type(4))) uint4;
-
-// CHECK-LABEL: @test_div_scale_f64
-// CHECK: {{.*}}call{{.*}} { double, i1 } @llvm.amdgcn.div.scale.f64(double %a, double %b, i1 true)
-// CHECK-DAG: [[FLAG:%.+]] = extractvalue { double, i1 } %{{.+}}, 1
-// CHECK-DAG: [[VAL:%.+]] = extractvalue { double, i1 } %{{.+}}, 0
-// CHECK: [[FLAGEXT:%.+]] = zext i1 [[FLAG]] to i32
-// CHECK: store i32 [[FLAGEXT]]
-void test_div_scale_f64(global double* out, global int* flagout, double a, double b)
-{
- bool flag;
- *out = __builtin_amdgcn_div_scale(a, b, true, &flag);
- *flagout = flag;
-}
-
-// CHECK-LABEL: @test_div_scale_f32(
-// CHECK: {{.*}}call{{.*}} { float, i1 } @llvm.amdgcn.div.scale.f32(float %a, float %b, i1 true)
-// CHECK-DAG: [[FLAG:%.+]] = extractvalue { float, i1 } %{{.+}}, 1
-// CHECK-DAG: [[VAL:%.+]] = extractvalue { float, i1 } %{{.+}}, 0
-// CHECK: [[FLAGEXT:%.+]] = zext i1 [[FLAG]] to i8
-// CHECK: store i8 [[FLAGEXT]]
-void test_div_scale_f32(global float* out, global bool* flagout, float a, float b)
-{
- bool flag;
- *out = __builtin_amdgcn_div_scalef(a, b, true, &flag);
- *flagout = flag;
-}
-
-// CHECK-LABEL: @test_div_scale_f32_global_ptr(
-// CHECK: {{.*}}call{{.*}} { float, i1 } @llvm.amdgcn.div.scale.f32(float %a, float %b, i1 true)
-// CHECK-DAG: [[FLAG:%.+]] = extractvalue { float, i1 } %{{.+}}, 1
-// CHECK-DAG: [[VAL:%.+]] = extractvalue { float, i1 } %{{.+}}, 0
-// CHECK: [[FLAGEXT:%.+]] = zext i1 [[FLAG]] to i8
-// CHECK: store i8 [[FLAGEXT]]
-void test_div_scale_f32_global_ptr(global float* out, global int* flagout, float a, float b, global bool* flag)
-{
- *out = __builtin_amdgcn_div_scalef(a, b, true, flag);
-}
-
-// CHECK-LABEL: @test_div_scale_f32_generic_ptr(
-// CHECK: {{.*}}call{{.*}} { float, i1 } @llvm.amdgcn.div.scale.f32(float %a, float %b, i1 true)
-// CHECK-DAG: [[FLAG:%.+]] = extractvalue { float, i1 } %{{.+}}, 1
-// CHECK-DAG: [[VAL:%.+]] = extractvalue { float, i1 } %{{.+}}, 0
-// CHECK: [[FLAGEXT:%.+]] = zext i1 [[FLAG]] to i8
-// CHECK: store i8 [[FLAGEXT]]
-void test_div_scale_f32_generic_ptr(global float* out, global int* flagout, float a, float b, global bool* flag_arg)
-{
- generic bool* flag = flag_arg;
- *out = __builtin_amdgcn_div_scalef(a, b, true, flag);
-}
-
-// CHECK-LABEL: @test_div_fmas_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.div.fmas.f32
-void test_div_fmas_f32(global float* out, float a, float b, float c, int d)
-{
- *out = __builtin_amdgcn_div_fmasf(a, b, c, d);
-}
-
-// CHECK-LABEL: @test_div_fmas_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.div.fmas.f64
-void test_div_fmas_f64(global double* out, double a, double b, double c, int d)
-{
- *out = __builtin_amdgcn_div_fmas(a, b, c, d);
-}
-
-// CHECK-LABEL: @test_div_fixup_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.div.fixup.f32
-void test_div_fixup_f32(global float* out, float a, float b, float c)
-{
- *out = __builtin_amdgcn_div_fixupf(a, b, c);
-}
-
-// CHECK-LABEL: @test_div_fixup_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.div.fixup.f64
-void test_div_fixup_f64(global double* out, double a, double b, double c)
-{
- *out = __builtin_amdgcn_div_fixup(a, b, c);
-}
-
-// CHECK-LABEL: @test_trig_preop_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.trig.preop.f32
-void test_trig_preop_f32(global float* out, float a, int b)
-{
- *out = __builtin_amdgcn_trig_preopf(a, b);
-}
-
-// CHECK-LABEL: @test_trig_preop_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.trig.preop.f64
-void test_trig_preop_f64(global double* out, double a, int b)
-{
- *out = __builtin_amdgcn_trig_preop(a, b);
-}
-
-// CHECK-LABEL: @test_rcp_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.rcp.f32
-void test_rcp_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_rcpf(a);
-}
-
-// CHECK-LABEL: @test_rcp_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.rcp.f64
-void test_rcp_f64(global double* out, double a)
-{
- *out = __builtin_amdgcn_rcp(a);
-}
-
-// CHECK-LABEL: @test_sqrt_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.{{((amdgcn.){0,1})}}sqrt.f32
-void test_sqrt_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_sqrtf(a);
-}
-
-// CHECK-LABEL: @test_sqrt_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.sqrt.f64
-void test_sqrt_f64(global double* out, double a)
-{
- *out = __builtin_amdgcn_sqrt(a);
-}
-
-// CHECK-LABEL: @test_rsq_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.rsq.f32
-void test_rsq_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_rsqf(a);
-}
-
-// CHECK-LABEL: @test_rsq_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.rsq.f64
-void test_rsq_f64(global double* out, double a)
-{
- *out = __builtin_amdgcn_rsq(a);
-}
-
-// CHECK-LABEL: @test_rsq_clamp_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.rsq.clamp.f32
-void test_rsq_clamp_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_rsq_clampf(a);
-}
-
-// CHECK-LABEL: @test_rsq_clamp_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.rsq.clamp.f64
-void test_rsq_clamp_f64(global double* out, double a)
-{
- *out = __builtin_amdgcn_rsq_clamp(a);
-}
-
-// CHECK-LABEL: @test_sin_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.sin.f32
-void test_sin_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_sinf(a);
-}
-
-// CHECK-LABEL: @test_cos_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.cos.f32
-void test_cos_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_cosf(a);
-}
-
-// CHECK-LABEL: @test_log_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.log.f32
-void test_log_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_logf(a);
-}
-
-// CHECK-LABEL: @test_exp2_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.exp2.f32
-void test_exp2_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_exp2f(a);
-}
-
-// CHECK-LABEL: @test_log_clamp_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.log.clamp.f32
-void test_log_clamp_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_log_clampf(a);
-}
-
-// CHECK-LABEL: @test_ldexp_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.ldexp.f32.i32
-void test_ldexp_f32(global float* out, float a, int b)
-{
- *out = __builtin_amdgcn_ldexpf(a, b);
-}
-
-// CHECK-LABEL: @test_ldexp_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.ldexp.f64.i32
-void test_ldexp_f64(global double* out, double a, int b)
-{
- *out = __builtin_amdgcn_ldexp(a, b);
-}
-
-// CHECK-LABEL: @test_frexp_mant_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.frexp.mant.f32
-void test_frexp_mant_f32(global float* out, float a)
-{
- *out = __builtin_amdgcn_frexp_mantf(a);
-}
-
-// CHECK-LABEL: @test_frexp_mant_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.frexp.mant.f64
-void test_frexp_mant_f64(global double* out, double a)
-{
- *out = __builtin_amdgcn_frexp_mant(a);
-}
-
-// CHECK-LABEL: @test_frexp_exp_f32
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.frexp.exp.i32.f32
-void test_frexp_exp_f32(global int* out, float a)
-{
- *out = __builtin_amdgcn_frexp_expf(a);
-}
-
-// CHECK-LABEL: @test_frexp_exp_f64
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.frexp.exp.i32.f64
-void test_frexp_exp_f64(global int* out, double a)
-{
- *out = __builtin_amdgcn_frexp_exp(a);
-}
-
-// CHECK-LABEL: @test_fract_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.fract.f32
-void test_fract_f32(global int* out, float a)
-{
- *out = __builtin_amdgcn_fractf(a);
-}
-
-// CHECK-LABEL: @test_fract_f64
-// CHECK: {{.*}}call{{.*}} double @llvm.amdgcn.fract.f64
-void test_fract_f64(global int* out, double a)
-{
- *out = __builtin_amdgcn_fract(a);
-}
-
-// CHECK-LABEL: @test_lerp
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.lerp
-void test_lerp(global int* out, int a, int b, int c)
-{
- *out = __builtin_amdgcn_lerp(a, b, c);
-}
-
-// CHECK-LABEL: @test_sicmp_i32
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.icmp.i64.i32(i32 %a, i32 %b, i32 32)
-void test_sicmp_i32(global ulong* out, int a, int b)
-{
- *out = __builtin_amdgcn_sicmp(a, b, 32);
-}
-
-// CHECK-LABEL: @test_uicmp_i32
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.icmp.i64.i32(i32 %a, i32 %b, i32 32)
-void test_uicmp_i32(global ulong* out, uint a, uint b)
-{
- *out = __builtin_amdgcn_uicmp(a, b, 32);
-}
-
-// CHECK-LABEL: @test_sicmp_i64
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.icmp.i64.i64(i64 %a, i64 %b, i32 38)
-void test_sicmp_i64(global ulong* out, long a, long b)
-{
- *out = __builtin_amdgcn_sicmpl(a, b, 39-1);
-}
-
-// CHECK-LABEL: @test_uicmp_i64
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.icmp.i64.i64(i64 %a, i64 %b, i32 35)
-void test_uicmp_i64(global ulong* out, ulong a, ulong b)
-{
- *out = __builtin_amdgcn_uicmpl(a, b, 30+5);
-}
-
-// CHECK-LABEL: @test_ds_swizzle
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ds.swizzle(i32 %a, i32 32)
-void test_ds_swizzle(global int* out, int a)
-{
- *out = __builtin_amdgcn_ds_swizzle(a, 32);
-}
-
-// CHECK-LABEL: @test_ds_permute
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ds.permute(i32 %a, i32 %b)
-void test_ds_permute(global int* out, int a, int b)
-{
- out[0] = __builtin_amdgcn_ds_permute(a, b);
-}
-
-// CHECK-LABEL: @test_ds_bpermute
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ds.bpermute(i32 %a, i32 %b)
-void test_ds_bpermute(global int* out, int a, int b)
-{
- *out = __builtin_amdgcn_ds_bpermute(a, b);
-}
-
-// CHECK-LABEL: @test_readfirstlane
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.readfirstlane.i32(i32 %a)
-void test_readfirstlane(global int* out, int a)
-{
- *out = __builtin_amdgcn_readfirstlane(a);
-}
-
-// CHECK-LABEL: @test_readlane
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.readlane.i32(i32 %a, i32 %b)
-void test_readlane(global int* out, int a, int b)
-{
- *out = __builtin_amdgcn_readlane(a, b);
-}
-
-// CHECK-LABEL: @test_fcmp_f32
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.fcmp.i64.f32(float %a, float %b, i32 5)
-void test_fcmp_f32(global ulong* out, float a, float b)
-{
- *out = __builtin_amdgcn_fcmpf(a, b, 5);
-}
-
-// CHECK-LABEL: @test_fcmp_f64
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.fcmp.i64.f64(double %a, double %b, i32 6)
-void test_fcmp_f64(global ulong* out, double a, double b)
-{
- *out = __builtin_amdgcn_fcmp(a, b, 3+3);
-}
-
-// CHECK-LABEL: @test_class_f32
-// CHECK: {{.*}}call{{.*}} i1 @llvm.amdgcn.class.f32
-void test_class_f32(global float* out, float a, int b)
-{
- *out = __builtin_amdgcn_classf(a, b);
-}
-
-// CHECK-LABEL: @test_class_f64
-// CHECK: {{.*}}call{{.*}} i1 @llvm.amdgcn.class.f64
-void test_class_f64(global double* out, double a, int b)
-{
- *out = __builtin_amdgcn_class(a, b);
-}
-
-// CHECK-LABEL: @test_buffer_wbinvl1
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.buffer.wbinvl1(
-void test_buffer_wbinvl1()
-{
- __builtin_amdgcn_buffer_wbinvl1();
-}
-
-// CHECK-LABEL: @test_s_dcache_inv
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.dcache.inv(
-void test_s_dcache_inv()
-{
- __builtin_amdgcn_s_dcache_inv();
-}
-
-// CHECK-LABEL: @test_s_waitcnt
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.waitcnt(
-void test_s_waitcnt()
-{
- __builtin_amdgcn_s_waitcnt(0);
-}
-
-// CHECK-LABEL: @test_s_sendmsg
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sendmsg(
-void test_s_sendmsg()
-{
- __builtin_amdgcn_s_sendmsg(1, 0);
-}
-
-// CHECK-LABEL: @test_s_sendmsg_var
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sendmsg(
-void test_s_sendmsg_var(int in)
-{
- __builtin_amdgcn_s_sendmsg(1, in);
-}
-
-// CHECK-LABEL: @test_s_sendmsghalt
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sendmsghalt(
-void test_s_sendmsghalt()
-{
- __builtin_amdgcn_s_sendmsghalt(1, 0);
-}
-
-// CHECK-LABEL: @test_s_sendmsghalt
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sendmsghalt(
-void test_s_sendmsghalt_var(int in)
-{
- __builtin_amdgcn_s_sendmsghalt(1, in);
-}
-
-// CHECK-LABEL: @test_s_barrier
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.barrier(
-void test_s_barrier()
-{
- __builtin_amdgcn_s_barrier();
-}
-
-// CHECK-LABEL: @test_wave_barrier
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.wave.barrier(
-void test_wave_barrier()
-{
- __builtin_amdgcn_wave_barrier();
-}
-
-// CHECK-LABEL: @test_sched_barrier
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.barrier(i32 0)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.barrier(i32 1)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.barrier(i32 4)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.barrier(i32 15)
-void test_sched_barrier()
-{
- __builtin_amdgcn_sched_barrier(0);
- __builtin_amdgcn_sched_barrier(1);
- __builtin_amdgcn_sched_barrier(4);
- __builtin_amdgcn_sched_barrier(15);
-}
-
-// CHECK-LABEL: @test_sched_group_barrier
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.group.barrier(i32 0, i32 1, i32 2)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.group.barrier(i32 1, i32 2, i32 4)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.group.barrier(i32 4, i32 8, i32 16)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.sched.group.barrier(i32 15, i32 10000, i32 -1)
-void test_sched_group_barrier()
-{
- __builtin_amdgcn_sched_group_barrier(0, 1, 2);
- __builtin_amdgcn_sched_group_barrier(1, 2, 4);
- __builtin_amdgcn_sched_group_barrier(4, 8, 16);
- __builtin_amdgcn_sched_group_barrier(15, 10000, -1);
-}
-
-// CHECK-LABEL: @test_iglp_opt
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.iglp.opt(i32 0)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.iglp.opt(i32 1)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.iglp.opt(i32 4)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.iglp.opt(i32 15)
-void test_iglp_opt()
-{
- __builtin_amdgcn_iglp_opt(0);
- __builtin_amdgcn_iglp_opt(1);
- __builtin_amdgcn_iglp_opt(4);
- __builtin_amdgcn_iglp_opt(15);
-}
-
-// CHECK-LABEL: @test_s_sleep
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sleep(i32 1)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.sleep(i32 15)
-void test_s_sleep()
-{
- __builtin_amdgcn_s_sleep(1);
- __builtin_amdgcn_s_sleep(15);
-}
-
-// CHECK-LABEL: @test_s_incperflevel
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.incperflevel(i32 1)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.incperflevel(i32 15)
-void test_s_incperflevel()
-{
- __builtin_amdgcn_s_incperflevel(1);
- __builtin_amdgcn_s_incperflevel(15);
-}
-
-// CHECK-LABEL: @test_s_decperflevel
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.decperflevel(i32 1)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.decperflevel(i32 15)
-void test_s_decperflevel()
-{
- __builtin_amdgcn_s_decperflevel(1);
- __builtin_amdgcn_s_decperflevel(15);
-}
-
-// CHECK-LABEL: @test_s_setprio
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.setprio(i16 0)
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.setprio(i16 3)
-void test_s_setprio()
-{
- __builtin_amdgcn_s_setprio(0);
- __builtin_amdgcn_s_setprio(3);
-}
-
-// CHECK-LABEL: @test_cubeid(
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.cubeid(float %a, float %b, float %c)
-void test_cubeid(global float* out, float a, float b, float c) {
- *out = __builtin_amdgcn_cubeid(a, b, c);
-}
-
-// CHECK-LABEL: @test_cubesc(
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.cubesc(float %a, float %b, float %c)
-void test_cubesc(global float* out, float a, float b, float c) {
- *out = __builtin_amdgcn_cubesc(a, b, c);
-}
-
-// CHECK-LABEL: @test_cubetc(
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.cubetc(float %a, float %b, float %c)
-void test_cubetc(global float* out, float a, float b, float c) {
- *out = __builtin_amdgcn_cubetc(a, b, c);
-}
-
-// CHECK-LABEL: @test_cubema(
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.cubema(float %a, float %b, float %c)
-void test_cubema(global float* out, float a, float b, float c) {
- *out = __builtin_amdgcn_cubema(a, b, c);
-}
-
-// CHECK-LABEL: @test_read_exec(
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.ballot.i64(i1 true)
-void test_read_exec(global ulong* out) {
- *out = __builtin_amdgcn_read_exec();
-}
-
-// CHECK: declare i64 @llvm.amdgcn.ballot.i64(i1){{.*}} #[[$NOUNWIND_READONLY:[0-9]+]]
-
-// CHECK-LABEL: @test_read_exec_lo(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ballot.i32(i1 true)
-void test_read_exec_lo(global uint* out) {
- *out = __builtin_amdgcn_read_exec_lo();
-}
-
-// CHECK: declare i32 @llvm.amdgcn.ballot.i32(i1){{.*}} #[[$NOUNWIND_READONLY:[0-9]+]]
-
-// CHECK-LABEL: @test_read_exec_hi(
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.ballot.i64(i1 true)
-// CHECK: lshr i64 [[A:%.*]], 32
-// CHECK: trunc nuw i64 [[B:%.*]] to i32
-void test_read_exec_hi(global uint* out) {
- *out = __builtin_amdgcn_read_exec_hi();
-}
-
-// CHECK-LABEL: @test_dispatch_ptr
-// CHECK: {{.*}}call align 4 dereferenceable(64){{.*}} ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
-#if !defined(__SPIRV__)
-void test_dispatch_ptr(__constant unsigned char ** out)
-#else
-void test_dispatch_ptr(__attribute__((address_space(4))) unsigned char ** out)
-#endif
-{
- *out = __builtin_amdgcn_dispatch_ptr();
-}
-
-// CHECK-LABEL: @test_queue_ptr
-// CHECK: {{.*}}call{{.*}} ptr addrspace(4) @llvm.amdgcn.queue.ptr()
-#if !defined(__SPIRV__)
-void test_queue_ptr(__constant unsigned char ** out)
-#else
-void test_queue_ptr(__attribute__((address_space(4))) unsigned char ** out)
-#endif
-{
- *out = __builtin_amdgcn_queue_ptr();
-}
-
-// CHECK-LABEL: @test_kernarg_segment_ptr
-// CHECK: {{.*}}call{{.*}} ptr addrspace(4) @llvm.amdgcn.kernarg.segment.ptr()
-#if !defined(__SPIRV__)
-void test_kernarg_segment_ptr(__constant unsigned char ** out)
-#else
-void test_kernarg_segment_ptr(__attribute__((address_space(4))) unsigned char ** out)
-#endif
-{
- *out = __builtin_amdgcn_kernarg_segment_ptr();
-}
-
-// CHECK-LABEL: @test_implicitarg_ptr
-// CHECK: {{.*}}call{{.*}} ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr()
-#if !defined(__SPIRV__)
-void test_implicitarg_ptr(__constant unsigned char ** out)
-#else
-void test_implicitarg_ptr(__attribute__((address_space(4))) unsigned char ** out)
-#endif
-{
- *out = __builtin_amdgcn_implicitarg_ptr();
-}
-
-// CHECK-LABEL: @test_get_group_id(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workgroup.id.x()
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workgroup.id.y()
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workgroup.id.z()
-void test_get_group_id(int d, global int *out)
-{
- switch (d) {
- case 0: *out = __builtin_amdgcn_workgroup_id_x(); break;
- case 1: *out = __builtin_amdgcn_workgroup_id_y(); break;
- case 2: *out = __builtin_amdgcn_workgroup_id_z(); break;
- default: *out = 0;
- }
-}
-
-// CHECK-LABEL: @test_s_getreg(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.s.getreg(i32 0)
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.s.getreg(i32 1)
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.s.getreg(i32 65535)
-void test_s_getreg(volatile global uint *out)
-{
- *out = __builtin_amdgcn_s_getreg(0);
- *out = __builtin_amdgcn_s_getreg(1);
- *out = __builtin_amdgcn_s_getreg(65535);
-}
-
-// CHECK-LABEL: @test_get_local_id(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workitem.id.x()
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workitem.id.y()
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.workitem.id.z()
-void test_get_local_id(int d, global int *out)
-{
- switch (d) {
- case 0: *out = __builtin_amdgcn_workitem_id_x(); break;
- case 1: *out = __builtin_amdgcn_workitem_id_y(); break;
- case 2: *out = __builtin_amdgcn_workitem_id_z(); break;
- default: *out = 0;
- }
-}
-
-// CHECK: declare noundef range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.x()
-// CHECK: declare noundef range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.y()
-// CHECK: declare noundef range(i32 0, 1024) i32 @llvm.amdgcn.workitem.id.z()
-
-// CHECK-LABEL: @test_get_workgroup_size(
-// CHECK: {{.*}}call align 8 dereferenceable(256){{.*}} ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr()
-// CHECK: getelementptr inbounds nuw i8, ptr addrspace(4) %{{.*}}, i64 12
-// CHECK: load i16, ptr addrspace(4) %{{.*}}, align 4, !range [[$WS_RANGE:![0-9]*]], !invariant.load{{.*}}, !noundef
-// CHECK: getelementptr inbounds nuw i8, ptr addrspace(4) %{{.*}}, i64 14
-// CHECK: load i16, ptr addrspace(4) %{{.*}}, align 2, !range [[$WS_RANGE:![0-9]*]], !invariant.load{{.*}}, !noundef
-// CHECK: getelementptr inbounds nuw i8, ptr addrspace(4) %{{.*}}, i64 16
-// CHECK: load i16, ptr addrspace(4) %{{.*}}, align 8, !range [[$WS_RANGE:![0-9]*]], !invariant.load{{.*}}, !noundef
-void test_get_workgroup_size(int d, global int *out)
-{
- switch (d) {
- case 0: *out = __builtin_amdgcn_workgroup_size_x() + 1; break;
- case 1: *out = __builtin_amdgcn_workgroup_size_y(); break;
- case 2: *out = __builtin_amdgcn_workgroup_size_z(); break;
- default: *out = 0;
- }
-}
-
-// CHECK-LABEL: @test_get_grid_size(
-// CHECK: {{.*}}call align 4 dereferenceable(64){{.*}} ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
-// CHECK: getelementptr inbounds nuw i8, ptr addrspace(4) %{{.*}}, i64 %{{.+}}
-// CHECK: load i32, ptr addrspace(4) %{{.*}}, align 4, !range [[$GRID_RANGE:![0-9]+]], !invariant.load
-void test_get_grid_size(int d, global int *out)
-{
- switch (d) {
- case 0: *out = __builtin_amdgcn_grid_size_x(); break;
- case 1: *out = __builtin_amdgcn_grid_size_y(); break;
- case 2: *out = __builtin_amdgcn_grid_size_z(); break;
- default: *out = 0;
- }
-}
-
-// CHECK-LABEL: @test_fmed3_f32
-// CHECK: {{.*}}call{{.*}} float @llvm.amdgcn.fmed3.f32(
-void test_fmed3_f32(global float* out, float a, float b, float c)
-{
- *out = __builtin_amdgcn_fmed3f(a, b, c);
-}
-
-// CHECK-LABEL: @test_s_getpc
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.s.getpc()
-void test_s_getpc(global ulong* out)
-{
- *out = __builtin_amdgcn_s_getpc();
-}
-
-// CHECK-LABEL: @test_ds_append_lds(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ds.append.p3(ptr addrspace(3) %{{.+}}, i1 false)
-#if !defined(__SPIRV__)
-kernel void test_ds_append_lds(global int* out, local int* ptr) {
-#else
-kernel void test_ds_append_lds(__attribute__((address_space(1))) int* out, __attribute__((address_space(3))) int* ptr) {
-#endif
- *out = __builtin_amdgcn_ds_append(ptr);
-}
-
-// CHECK-LABEL: @test_ds_consume_lds(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.ds.consume.p3(ptr addrspace(3) %{{.+}}, i1 false)
-
-#if !defined(__SPIRV__)
-kernel void test_ds_consume_lds(global int* out, local int* ptr) {
-#else
-kernel void test_ds_consume_lds(__attribute__((address_space(1))) int* out, __attribute__((address_space(3))) int* ptr) {
-#endif
- *out = __builtin_amdgcn_ds_consume(ptr);
-}
-
-// CHECK-LABEL: @test_gws_init(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.ds.gws.init(i32 %value, i32 %id)
-kernel void test_gws_init(uint value, uint id) {
- __builtin_amdgcn_ds_gws_init(value, id);
-}
-
-// CHECK-LABEL: @test_gws_barrier(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.ds.gws.barrier(i32 %value, i32 %id)
-kernel void test_gws_barrier(uint value, uint id) {
- __builtin_amdgcn_ds_gws_barrier(value, id);
-}
-
-// CHECK-LABEL: @test_gws_sema_v(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.ds.gws.sema.v(i32 %id)
-kernel void test_gws_sema_v(uint id) {
- __builtin_amdgcn_ds_gws_sema_v(id);
-}
-
-// CHECK-LABEL: @test_gws_sema_br(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.ds.gws.sema.br(i32 %value, i32 %id)
-kernel void test_gws_sema_br(uint value, uint id) {
- __builtin_amdgcn_ds_gws_sema_br(value, id);
-}
-
-// CHECK-LABEL: @test_gws_sema_p(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.ds.gws.sema.p(i32 %id)
-kernel void test_gws_sema_p(uint id) {
- __builtin_amdgcn_ds_gws_sema_p(id);
-}
-
-// CHECK-LABEL: @test_mbcnt_lo(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.mbcnt.lo(i32 %src0, i32 %src1)
-kernel void test_mbcnt_lo(global uint* out, uint src0, uint src1) {
- *out = __builtin_amdgcn_mbcnt_lo(src0, src1);
-}
-
-// CHECK-LABEL: @test_mbcnt_hi(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.mbcnt.hi(i32 %src0, i32 %src1)
-kernel void test_mbcnt_hi(global uint* out, uint src0, uint src1) {
- *out = __builtin_amdgcn_mbcnt_hi(src0, src1);
-}
-
-// CHECK-LABEL: @test_alignbit(
-// CHECK: tail call{{.*}} i32 @llvm.fshr.i32(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_alignbit(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_alignbit(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_alignbyte(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.alignbyte(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_alignbyte(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_alignbyte(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_ubfe(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.ubfe.i32(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_ubfe(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_ubfe(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_sbfe(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.sbfe.i32(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_sbfe(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_sbfe(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_cvt_pkrtz(
-// CHECK: tail call{{.*}} <2 x half> @llvm.amdgcn.cvt.pkrtz(float %src0, float %src1)
-kernel void test_cvt_pkrtz(global half2* out, float src0, float src1) {
- *out = __builtin_amdgcn_cvt_pkrtz(src0, src1);
-}
-
-// CHECK-LABEL: @test_cvt_pknorm_i16(
-// CHECK: tail call{{.*}} <2 x i16> @llvm.amdgcn.cvt.pknorm.i16(float %src0, float %src1)
-kernel void test_cvt_pknorm_i16(global short2* out, float src0, float src1) {
- *out = __builtin_amdgcn_cvt_pknorm_i16(src0, src1);
-}
-
-// CHECK-LABEL: @test_cvt_pknorm_u16(
-// CHECK: tail call{{.*}} <2 x i16> @llvm.amdgcn.cvt.pknorm.u16(float %src0, float %src1)
-kernel void test_cvt_pknorm_u16(global ushort2* out, float src0, float src1) {
- *out = __builtin_amdgcn_cvt_pknorm_u16(src0, src1);
-}
-
-// CHECK-LABEL: @test_cvt_pk_i16(
-// CHECK: tail call{{.*}} <2 x i16> @llvm.amdgcn.cvt.pk.i16(i32 %src0, i32 %src1)
-kernel void test_cvt_pk_i16(global short2* out, int src0, int src1) {
- *out = __builtin_amdgcn_cvt_pk_i16(src0, src1);
-}
-
-// CHECK-LABEL: @test_cvt_pk_u16(
-// CHECK: tail call{{.*}} <2 x i16> @llvm.amdgcn.cvt.pk.u16(i32 %src0, i32 %src1)
-kernel void test_cvt_pk_u16(global ushort2* out, uint src0, uint src1) {
- *out = __builtin_amdgcn_cvt_pk_u16(src0, src1);
-}
-
-// CHECK-LABEL: @test_cvt_pk_u8_f32
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src0, i32 %src1, i32 %src2)
-kernel void test_cvt_pk_u8_f32(global uint* out, float src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_cvt_pk_u8_f32(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_sad_u8(
-// CHECK: tail call{{.*}} i32 @llvm.amdgcn.sad.u8(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_sad_u8(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_sad_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: test_msad_u8(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.msad.u8(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_msad_u8(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_msad_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: test_sad_hi_u8(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.sad.hi.u8(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_sad_hi_u8(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_sad_hi_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_sad_u16(
-// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.sad.u16(i32 %src0, i32 %src1, i32 %src2)
-kernel void test_sad_u16(global uint* out, uint src0, uint src1, uint src2) {
- *out = __builtin_amdgcn_sad_u16(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_qsad_pk_u16_u8(
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.qsad.pk.u16.u8(i64 %src0, i32 %src1, i64 %src2)
-kernel void test_qsad_pk_u16_u8(global ulong* out, ulong src0, uint src1, ulong src2) {
- *out = __builtin_amdgcn_qsad_pk_u16_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: @test_mqsad_pk_u16_u8(
-// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.mqsad.pk.u16.u8(i64 %src0, i32 %src1, i64 %src2)
-kernel void test_mqsad_pk_u16_u8(global ulong* out, ulong src0, uint src1, ulong src2) {
- *out = __builtin_amdgcn_mqsad_pk_u16_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: test_mqsad_u32_u8(
-// CHECK: {{.*}}call{{.*}} <4 x i32> @llvm.amdgcn.mqsad.u32.u8(i64 %src0, i32 %src1, <4 x i32> %src2)
-kernel void test_mqsad_u32_u8(global uint4* out, ulong src0, uint src1, uint4 src2) {
- *out = __builtin_amdgcn_mqsad_u32_u8(src0, src1, src2);
-}
-
-// CHECK-LABEL: test_s_setreg(
-// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.setreg(i32 8193, i32 %val)
-kernel void test_s_setreg(uint val) {
- __builtin_amdgcn_s_setreg(8193, val);
-}
-
-// CHECK-LABEL test_atomic_inc_dec(
-#if !defined(__SPIRV__)
-void test_atomic_inc_dec(local uint *lptr, global uint *gptr, uint val) {
-#else
-void test_atomic_inc_dec(__attribute__((address_space(3))) uint *lptr, __attribute__((address_space(1))) uint *gptr, uint val) {
-#endif
- uint res;
-
- // CHECK: atomicrmw uinc_wrap ptr addrspace(3) %lptr, i32 %val syncscope("workgroup") seq_cst, align 4
- res = __builtin_amdgcn_atomic_inc32(lptr, val, __ATOMIC_SEQ_CST, "workgroup");
-
- // CHECK: atomicrmw udec_wrap ptr addrspace(3) %lptr, i32 %val syncscope("workgroup") seq_cst, align 4
- res = __builtin_amdgcn_atomic_dec32(lptr, val, __ATOMIC_SEQ_CST, "workgroup");
-
- // CHECK: atomicrmw uinc_wrap ptr addrspace(1) %gptr, i32 %val syncscope("agent") seq_cst, align 4
- res = __builtin_amdgcn_atomic_inc32(gptr, val, __ATOMIC_SEQ_CST, "agent");
-
- // CHECK: atomicrmw udec_wrap ptr addrspace(1) %gptr, i32 %val seq_cst, align 4
- res = __builtin_amdgcn_atomic_dec32(gptr, val, __ATOMIC_SEQ_CST, "");
-
- // CHECK: atomicrmw volatile udec_wrap ptr addrspace(1) %gptr, i32 %val seq_cst, align 4
- #if !defined(__SPIRV__)
- res = __builtin_amdgcn_atomic_dec32((volatile global uint*)gptr, val, __ATOMIC_SEQ_CST, "");
- #else
- res = __builtin_amdgcn_atomic_dec32((volatile __attribute__((address_space(1))) uint*)gptr, val, __ATOMIC_SEQ_CST, "");
- #endif
-}
-
-// CHECK-LABEL test_wavefrontsize(
-unsigned test_wavefrontsize() {
-
- // CHECK-AMDGCN: ret i32 {{[0-9]+}}
- // CHECK-SPIRV: {{.*}}call{{.*}} i32 @llvm.amdgcn.wavefrontsize()
- return __builtin_amdgcn_wavefrontsize();
-}
-
-// CHECK-LABEL test_flt_rounds(
-unsigned test_flt_rounds() {
-
- // CHECK: {{.*}}call{{.*}} i32 @llvm.get.rounding()
- unsigned mode = __builtin_flt_rounds();
-
-#if !defined(__SPIRV__)
- // CHECK-AMDGCN: call void @llvm.set.rounding(i32 %0)
- __builtin_set_flt_rounds(mode);
-#endif
-
- return mode;
-}
-
-// CHECK-LABEL test_get_fpenv(
-unsigned long test_get_fpenv() {
- // CHECK: {{.*}}call{{.*}} i64 @llvm.get.fpenv.i64()
- return __builtin_amdgcn_get_fpenv();
-}
-
-// CHECK-LABEL test_set_fpenv(
-void test_set_fpenv(unsigned long env) {
- // CHECK: {{.*}}call{{.*}} void @llvm.set.fpenv.i64(i64 %[[ENV:.+]])
- __builtin_amdgcn_set_fpenv(env);
-}
-
-// CHECK-DAG: [[$GRID_RANGE]] = !{i32 1, i32 0}
-// CHECK-DAG: [[$WS_RANGE]] = !{i16 1, i16 1025}
-// CHECK-SPIRV-DAG: attributes #[[$NOUNWIND_READONLY]] = { convergent mustprogress nocallback nofree nounwind willreturn memory(none) }
-// CHECK-AMDGCN-DAG: attributes #[[$NOUNWIND_READONLY]] = { convergent mustprogress nocallback nofree nounwind willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
index 448b5322d6294..5c1bf1a92da11 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp
@@ -1324,10 +1324,14 @@ static void addPreloadKernArgHint(Function &F, TargetMachine &TM) {
/// have a determined flat workgroup size.
static void updateWavesPerEU(Module &M, TargetMachine &TM) {
for (Function &F : M) {
+ if (F.isDeclaration())
+ continue;
+
const GCNSubtarget &ST = TM.getSubtarget<GCNSubtarget>(F);
- auto FlatWgrpSizeAttr =
- AMDGPU::getIntegerPairAttribute(F, "amdgpu-flat-work-group-size");
+ std::optional<std::pair<unsigned, std::optional<unsigned>>>
+ FlatWgrpSizeAttr =
+ AMDGPU::getIntegerPairAttribute(F, "amdgpu-flat-work-group-size");
unsigned MinWavesPerEU = ST.getMinWavesPerEU();
unsigned MaxWavesPerEU = ST.getMaxWavesPerEU();
@@ -1346,7 +1350,7 @@ static void updateWavesPerEU(Module &M, TargetMachine &TM) {
// Compute the range from flat workgroup size. `getWavesPerEU` will also
// account for the 'amdgpu-waves-er-eu' attribute.
auto [MinFromFlatWgrpSize, MaxFromFlatWgrpSize] =
- ST.getWavesPerEU(F, std::make_pair(MinFlatWgrpSize, MaxFlatWgrpSize));
+ ST.getWavesPerEU(F, {MinFlatWgrpSize, MaxFlatWgrpSize});
// For the lower bound, we have to "tighten" it.
Min = std::max(Min, MinFromFlatWgrpSize);
@@ -1444,6 +1448,8 @@ static bool runImpl(Module &M, AnalysisGetter &AG, TargetMachine &TM,
bool Changed = A.run() == ChangeStatus::CHANGED;
+ // We only update the waves-per-eu attribute at the final stage to avoid
+ // setting it with intermediate values.
if (Changed && (LTOPhase == ThinOrFullLTOPhase::None ||
LTOPhase == ThinOrFullLTOPhase::FullLTOPostLink ||
LTOPhase == ThinOrFullLTOPhase::ThinLTOPostLink))
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
index 563605f964cc6..776cc6258dbcd 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
@@ -216,6 +216,15 @@ AMDGPUSubtarget::getWavesPerEU(const Function &F) const {
return getWavesPerEU(FlatWorkGroupSizes, LDSBytes, F);
}
+std::pair<unsigned, unsigned> AMDGPUSubtarget::getWavesPerEU(
+ const Function &F, std::pair<unsigned, unsigned> FlatWorkGroupSizes) const {
+ // Minimum number of bytes allocated in the LDS.
+ unsigned LDSBytes = AMDGPU::getIntegerPairAttribute(F, "amdgpu-lds-size",
+ {0, UINT32_MAX}, true)
+ .first;
+ return getWavesPerEU(FlatWorkGroupSizes, LDSBytes, F);
+}
+
std::pair<unsigned, unsigned>
AMDGPUSubtarget::getWavesPerEU(std::pair<unsigned, unsigned> FlatWorkGroupSizes,
unsigned LDSBytes, const Function &F) const {
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
index 91fe2a69bc0b7..744bef4924b79 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
@@ -108,6 +108,13 @@ class AMDGPUSubtarget {
/// size, register usage, and/or lds usage.
std::pair<unsigned, unsigned> getWavesPerEU(const Function &F) const;
+ /// Overload which uses the specified values for the flat work group sizes,
+ /// rather than querying the function itself. \p FlatWorkGroupSizes Should
+ /// correspond to the function's value for getFlatWorkGroupSizes.
+ std::pair<unsigned, unsigned>
+ getWavesPerEU(const Function &F,
+ std::pair<unsigned, unsigned> FlatWorkGroupSizes) const;
+
/// Overload which uses the specified values for the flat workgroup sizes and
/// LDS space rather than querying the function itself. \p FlatWorkGroupSizes
/// should correspond to the function's value for getFlatWorkGroupSizes and \p
diff --git a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
index 42c7b90da63d3..28f55511ebb6f 100644
--- a/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll
@@ -169,6 +169,6 @@ attributes #1 = { nounwind }
;.
; HSA: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) }
-; HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
index b2c5801ad1e52..181dab8d4ca79 100644
--- a/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdgpu-attributor-no-agpr.ll
@@ -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: ) #[[ATTR2:[0-9]+]] {
+; CHECK-SAME: ) #[[ATTR3:[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: ) #[[ATTR2]] {
-; CHECK-NEXT: call void @unknown() #[[ATTR6:[0-9]+]]
+; CHECK-SAME: ) #[[ATTR3]] {
+; CHECK-NEXT: call void @unknown() #[[ATTR7:[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:%.*]]) #[[ATTR2]] {
+; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR3]] {
; 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:%.*]]) #[[ATTR2]] {
-; CHECK-NEXT: call void [[INDIRECT]]() #[[ATTR6]]
+; CHECK-SAME: ptr [[INDIRECT:%.*]]) #[[ATTR3]] {
+; CHECK-NEXT: call void [[INDIRECT]]() #[[ATTR7]]
; CHECK-NEXT: ret void
;
call void %indirect() #0
@@ -252,13 +252,14 @@ define amdgpu_kernel void @indirect_calls_none_agpr(i1 %cond) {
}
-attributes #0 = { "amdgpu-no-agpr" }
+attributes #0 = { "amdgpu-agpr-alloc"="0" }
;.
; CHECK: attributes #[[ATTR0]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR2]] = { "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3:[0-9]+]] = { convergent nocallback nofree nosync nounwind willreturn memory(none) "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR5:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" }
-; CHECK: attributes #[[ATTR6]] = { "amdgpu-no-agpr" }
+; CHECK: attributes #[[ATTR2:[0-9]+]] = { "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR3]] = { "amdgpu-waves-per-eu"="4,8" "target-cpu"="gfx90a" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR4:[0-9]+]] = { convergent nocallback nofree nosync nounwind willreturn memory(none) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR5:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR6:[0-9]+]] = { nocallback nofree nounwind willreturn memory(argmem: readwrite) "target-cpu"="gfx90a" }
+; CHECK: attributes #[[ATTR7]] = { "amdgpu-agpr-alloc"="0" }
;.
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 dfa645c8379d5..cdb9801e660bc 100644
--- a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll
@@ -442,7 +442,7 @@ define internal void @defined.func() #3 {
define void @func_call_external() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_call_external
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR15:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR17:[0-9]+]] {
; ATTRIBUTOR_HSA-NEXT: call void @external.func()
; ATTRIBUTOR_HSA-NEXT: ret void
;
@@ -462,7 +462,7 @@ define void @func_call_defined() #3 {
define void @func_call_asm() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_call_asm
; ATTRIBUTOR_HSA-SAME: () #[[ATTR16]] {
-; ATTRIBUTOR_HSA-NEXT: call void asm sideeffect "", ""() #[[ATTR26:[0-9]+]]
+; ATTRIBUTOR_HSA-NEXT: call void asm sideeffect "", ""() #[[ATTR25:[0-9]+]]
; ATTRIBUTOR_HSA-NEXT: ret void
;
call void asm sideeffect "", ""() #3
@@ -471,7 +471,7 @@ define void @func_call_asm() #3 {
define amdgpu_kernel void @kern_call_external() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@kern_call_external
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR15]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR17]] {
; ATTRIBUTOR_HSA-NEXT: call void @external.func()
; ATTRIBUTOR_HSA-NEXT: ret void
;
@@ -481,7 +481,7 @@ define amdgpu_kernel void @kern_call_external() #3 {
define amdgpu_kernel void @func_kern_defined() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_kern_defined
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR17:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR16]] {
; ATTRIBUTOR_HSA-NEXT: call void @defined.func()
; ATTRIBUTOR_HSA-NEXT: ret void
;
@@ -515,7 +515,7 @@ define float @func_indirect_use_dispatch_ptr_constexpr_cast_func() #1 {
define float @func_indirect_call(ptr %fptr) #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_indirect_call
-; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR15]] {
+; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR17]] {
; ATTRIBUTOR_HSA-NEXT: [[F:%.*]] = call float [[FPTR]]()
; ATTRIBUTOR_HSA-NEXT: [[FADD:%.*]] = fadd float [[F]], 1.000000e+00
; ATTRIBUTOR_HSA-NEXT: ret float [[FADD]]
@@ -528,7 +528,7 @@ define float @func_indirect_call(ptr %fptr) #3 {
declare float @extern() #3
define float @func_extern_call() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_extern_call
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR15]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR17]] {
; ATTRIBUTOR_HSA-NEXT: [[F:%.*]] = call float @extern()
; ATTRIBUTOR_HSA-NEXT: [[FADD:%.*]] = fadd float [[F]], 1.000000e+00
; ATTRIBUTOR_HSA-NEXT: ret float [[FADD]]
@@ -540,7 +540,7 @@ define float @func_extern_call() #3 {
define float @func_null_call(ptr %fptr) #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@func_null_call
-; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR15]] {
+; ATTRIBUTOR_HSA-SAME: (ptr [[FPTR:%.*]]) #[[ATTR17]] {
; ATTRIBUTOR_HSA-NEXT: [[F:%.*]] = call float null()
; ATTRIBUTOR_HSA-NEXT: [[FADD:%.*]] = fadd float [[F]], 1.000000e+00
; ATTRIBUTOR_HSA-NEXT: ret float [[FADD]]
@@ -615,7 +615,7 @@ declare void @extern_func_sanitize_address() #5
define amdgpu_kernel void @kern_decl_sanitize_address() #3 {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@kern_decl_sanitize_address
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR15]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR17]] {
; ATTRIBUTOR_HSA-NEXT: call void @extern_func_sanitize_address()
; ATTRIBUTOR_HSA-NEXT: ret void
;
@@ -655,7 +655,7 @@ define amdgpu_kernel void @kern_call_enqueued_block_def() {
define void @unused_enqueue_block() {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@unused_enqueue_block
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR25:[0-9]+]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR24]] {
; ATTRIBUTOR_HSA-NEXT: ret void
;
ret void
@@ -663,7 +663,7 @@ define void @unused_enqueue_block() {
define internal void @known_func() {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@known_func
-; ATTRIBUTOR_HSA-SAME: () #[[ATTR25]] {
+; ATTRIBUTOR_HSA-SAME: () #[[ATTR24]] {
; ATTRIBUTOR_HSA-NEXT: ret void
;
ret void
@@ -673,7 +673,7 @@ define internal void @known_func() {
define amdgpu_kernel void @kern_callsite_enqueue_block() {
; ATTRIBUTOR_HSA-LABEL: define {{[^@]+}}@kern_callsite_enqueue_block
; ATTRIBUTOR_HSA-SAME: () #[[ATTR24]] {
-; ATTRIBUTOR_HSA-NEXT: call void @known_func() #[[ATTR27:[0-9]+]]
+; ATTRIBUTOR_HSA-NEXT: call void @known_func() #[[ATTR26:[0-9]+]]
; ATTRIBUTOR_HSA-NEXT: ret void
;
call void @known_func() #6
@@ -689,34 +689,32 @@ attributes #5 = { nounwind sanitize_address "amdgpu-no-implicitarg-ptr" }
attributes #6 = { "enqueued-block" }
-;.
;.
; ATTRIBUTOR_HSA: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; ATTRIBUTOR_HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-flat-scratch-init" "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"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR10]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR11]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR12]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "target-cpu"="fiji" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR13]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx900" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR14]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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"="gfx900" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR15]] = { nounwind "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR16]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR17]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR18]] = { nounwind sanitize_address "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-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 #[[ATTR19]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-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 #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-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" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR10]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR11]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR12]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "target-cpu"="fiji" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR13]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR14]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR15:[0-9]+]] = { nounwind "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR16]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR17]] = { nounwind "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR18]] = { nounwind sanitize_address "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-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" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR19]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-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" }
; ATTRIBUTOR_HSA: attributes #[[ATTR20:[0-9]+]] = { nounwind sanitize_address "amdgpu-no-implicitarg-ptr" "uniform-work-group-size"="false" }
; ATTRIBUTOR_HSA: attributes #[[ATTR21:[0-9]+]] = { "enqueued-block" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR22]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "enqueued-block" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR23]] = { "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR22]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "enqueued-block" "uniform-work-group-size"="false" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR23]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; ATTRIBUTOR_HSA: attributes #[[ATTR24]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_HSA: attributes #[[ATTR25]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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 #[[ATTR26]] = { nounwind }
-; ATTRIBUTOR_HSA: attributes #[[ATTR27]] = { "enqueued-block" }
+; ATTRIBUTOR_HSA: attributes #[[ATTR25]] = { nounwind }
+; ATTRIBUTOR_HSA: attributes #[[ATTR26]] = { "enqueued-block" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll
index 2426ab9df437c..894ef4fa7c976 100644
--- a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll
+++ b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll
@@ -474,17 +474,17 @@ attributes #1 = { nounwind }
; AKF_HSA: [[META0:![0-9]+]] = !{i32 1, !"amdhsa_code_object_version", i32 500}
;.
; HSA: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "uniform-work-group-size"="false" }
-; HSA: attributes #[[ATTR10]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-flat-scratch-init" "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" }
-; HSA: attributes #[[ATTR11]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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" }
-; HSA: attributes #[[ATTR12]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; HSA: attributes #[[ATTR13]] = { nounwind "amdgpu-agpr-alloc"="0" "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" }
+; HSA: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-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" }
+; HSA: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR10]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; HSA: attributes #[[ATTR11]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-multigrid-sync-arg" "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" }
+; HSA: attributes #[[ATTR12]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
+; HSA: attributes #[[ATTR13]] = { nounwind "amdgpu-agpr-alloc"="0" "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" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features.ll b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features.ll
index 81ccf16c4e4bc..aeca1384f09a2 100644
--- a/llvm/test/CodeGen/AMDGPU/annotate-kernel-features.ll
+++ b/llvm/test/CodeGen/AMDGPU/annotate-kernel-features.ll
@@ -294,13 +294,13 @@ attributes #1 = { nounwind }
;.
; CHECK: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
-; CHECK: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" }
-; CHECK: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-z" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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" "uniform-work-group-size"="false" }
-; CHECK: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR1]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR2]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-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" }
+; CHECK: attributes #[[ATTR3]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR4]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR5]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR6]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR7]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR8]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; CHECK: attributes #[[ATTR9]] = { nounwind "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-workitem-id-x" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit-undefined-behavior.ll b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit-undefined-behavior.ll
index 9642d6cceb514..93ebeafe0dee5 100644
--- a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit-undefined-behavior.ll
+++ b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit-undefined-behavior.ll
@@ -147,9 +147,9 @@ define amdgpu_kernel void @call_calls_intrin_ascast_cc_kernel(ptr addrspace(3) %
attributes #0 = { "amdgpu-no-flat-scratch-init" }
;.
-; GFX9: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx900" "uniform-work-group-size"="false" }
+; GFX9: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
; GFX9: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx900" }
;.
-; GFX10: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="gfx1010" "uniform-work-group-size"="false" }
+; GFX10: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
; GFX10: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx1010" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
index 3286d78b758fd..7f450ed2b9aa4 100644
--- a/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
+++ b/llvm/test/CodeGen/AMDGPU/attributor-flatscratchinit.ll
@@ -869,13 +869,13 @@ define amdgpu_kernel void @with_inline_asm() {
; GFX9: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
; GFX9: attributes #[[ATTR2]] = { "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
; GFX9: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
-; GFX9: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" }
+; GFX9: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx900" }
; GFX9: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="4,10" "target-cpu"="gfx900" "uniform-work-group-size"="false" }
;.
; GFX10: attributes #[[ATTR0]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
; GFX10: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
; GFX10: attributes #[[ATTR2]] = { "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
; GFX10: attributes #[[ATTR3]] = { "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
-; GFX10: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" }
+; GFX10: attributes #[[ATTR4:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "target-cpu"="gfx1010" }
; GFX10: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-flat-scratch-init" "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-y" "amdgpu-no-workitem-id-z" "amdgpu-waves-per-eu"="8,20" "target-cpu"="gfx1010" "uniform-work-group-size"="false" }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll b/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
index e35057aec2b13..c48613b4bf3bf 100644
--- a/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
+++ b/llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll
@@ -275,20 +275,20 @@ attributes #0 = { nocallback nofree nosync nounwind speculatable willreturn memo
;.
-; V4: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V4: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
; V4: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V4: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V4: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V4: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V4: attributes #[[ATTR5]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
-; V5: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V5: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
; V5: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V5: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V5: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V5: attributes #[[ATTR4]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
;.
-; V6: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; V6: attributes #[[ATTR0:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
; V6: attributes #[[ATTR1]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V6: attributes #[[ATTR2]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-completion-action" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
; V6: attributes #[[ATTR3]] = { "amdgpu-agpr-alloc"="0" "amdgpu-no-default-queue" "amdgpu-no-flat-scratch-init" "amdgpu-no-heap-ptr" "amdgpu-no-hostcall-ptr" "amdgpu-no-multigrid-sync-arg" "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
diff --git a/llvm/test/CodeGen/AMDGPU/inline-attr.ll b/llvm/test/CodeGen/AMDGPU/inline-attr.ll
index 4ae0ba065adfc..1e56f6f8e2c41 100644
--- a/llvm/test/CodeGen/AMDGPU/inline-attr.ll
+++ b/llvm/test/CodeGen/AMDGPU/inline-attr.ll
@@ -7,7 +7,7 @@ declare void @extern() #0
define float @foo(float %x) #0 {
; GCN-LABEL: define float @foo(
-; GCN-SAME: float [[X:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
+; GCN-SAME: float [[X:%.*]]) local_unnamed_addr #[[ATTR1:[0-9]+]] {
; GCN-NEXT: [[ENTRY:.*:]]
; GCN-NEXT: tail call void @extern()
; GCN-NEXT: [[MUL:%.*]] = fmul float [[X]], 1.500000e+01
@@ -21,7 +21,7 @@ entry:
define amdgpu_kernel void @caller(ptr addrspace(1) %p) #1 {
; GCN-LABEL: define amdgpu_kernel void @caller(
-; GCN-SAME: ptr addrspace(1) captures(none) [[P:%.*]]) local_unnamed_addr #[[ATTR1:[0-9]+]] {
+; GCN-SAME: ptr addrspace(1) captures(none) [[P:%.*]]) local_unnamed_addr #[[ATTR2:[0-9]+]] {
; GCN-NEXT: [[ENTRY:.*:]]
; GCN-NEXT: [[LOAD:%.*]] = load float, ptr addrspace(1) [[P]], align 4, !amdgpu.noclobber [[META0:![0-9]+]]
; GCN-NEXT: tail call void @extern()
@@ -40,14 +40,17 @@ attributes #0 = { nounwind "uniform-work-group-size"="false" "unsafe-fp-math"="t
attributes #1 = { nounwind "less-precise-fpmad"="true" "no-infs-fp-math"="true" "no-nans-fp-math"="true" "unsafe-fp-math"="true" }
;.
-; UNSAFE: attributes #[[ATTR0]] = { nounwind "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
-; UNSAFE: attributes #[[ATTR1]] = { nounwind "less-precise-fpmad"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; UNSAFE: attributes #[[ATTR0:[0-9]+]] = { nounwind "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; UNSAFE: attributes #[[ATTR1]] = { nounwind "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; UNSAFE: attributes #[[ATTR2]] = { nounwind "amdgpu-waves-per-eu"="4,10" "less-precise-fpmad"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
;.
-; NONANS: attributes #[[ATTR0]] = { nounwind "no-nans-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
-; NONANS: attributes #[[ATTR1]] = { nounwind "less-precise-fpmad"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NONANS: attributes #[[ATTR0:[0-9]+]] = { nounwind "no-nans-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NONANS: attributes #[[ATTR1]] = { nounwind "amdgpu-waves-per-eu"="4,10" "no-nans-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NONANS: attributes #[[ATTR2]] = { nounwind "amdgpu-waves-per-eu"="4,10" "less-precise-fpmad"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
;.
-; NOINFS: attributes #[[ATTR0]] = { nounwind "no-infs-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
-; NOINFS: attributes #[[ATTR1]] = { nounwind "less-precise-fpmad"="false" "no-infs-fp-math"="true" "no-nans-fp-math"="false" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NOINFS: attributes #[[ATTR0:[0-9]+]] = { nounwind "no-infs-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NOINFS: attributes #[[ATTR1]] = { nounwind "amdgpu-waves-per-eu"="4,10" "no-infs-fp-math"="true" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
+; NOINFS: attributes #[[ATTR2]] = { nounwind "amdgpu-waves-per-eu"="4,10" "less-precise-fpmad"="false" "no-infs-fp-math"="true" "no-nans-fp-math"="false" "uniform-work-group-size"="false" "unsafe-fp-math"="true" }
;.
; UNSAFE: [[META0]] = !{}
;.
diff --git a/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll b/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
index 31f8e0ea1f8fa..fc360423599e3 100644
--- a/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
+++ b/llvm/test/CodeGen/AMDGPU/pal-simple-indirect-call.ll
@@ -58,7 +58,6 @@ declare i64 @llvm.amdgcn.s.getpc() #0
attributes #0 = { nounwind readnone speculatable willreturn }
;.
-;.
; ATTRIBUTOR_GCN: attributes #[[ATTR0]] = { "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
-; ATTRIBUTOR_GCN: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) "amdgpu-waves-per-eu"="4,10" }
+; ATTRIBUTOR_GCN: attributes #[[ATTR1:[0-9]+]] = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
;.
diff --git a/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll b/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
index cb0a7a146104a..6a88be6e55859 100644
--- a/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
+++ b/llvm/test/CodeGen/AMDGPU/simplify-libcalls.ll
@@ -826,5 +826,5 @@ entry:
; GCN-PRELINK: declare float @_Z4cbrtf(float) local_unnamed_addr #[[$NOUNWIND_READONLY:[0-9]+]]
; GCN-PRELINK-DAG: attributes #[[$NOUNWIND]] = { nounwind }
-; GCN-PRELINK-DAG: attributes #[[$NOUNWIND_READONLY]] = { nounwind memory(read) "amdgpu-waves-per-eu"="4,10" "uniform-work-group-size"="false" }
+; GCN-PRELINK-DAG: attributes #[[$NOUNWIND_READONLY]] = { nounwind memory(read) "uniform-work-group-size"="false" }
attributes #0 = { nounwind }
More information about the llvm-branch-commits
mailing list