[AMDGPU][UnifyDivergentExitNodes][StructurizeCFG] Add support for callbr instruction with inline-asm #152161
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…lbr instruction with basic inline-asm Finishes adding basic inline-asm callbr support for AMDGPU, started by #149308.
ro-i
requested review from
arsenm,
nickdesaulniers,
Meinersbur,
shiltian and
ssahasra
|
@llvm/pr-subscribers-llvm-transforms @llvm/pr-subscribers-backend-amdgpu Author: Robert Imschweiler (ro-i) ChangesFinishes adding basic inline-asm callbr support for AMDGPU, started by #149308. Patch is 42.29 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/152161.diff 10 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp b/llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp
index 733c5d520fb23..2df6bbc74f6da 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUUnifyDivergentExitNodes.cpp
@@ -181,14 +181,52 @@ BasicBlock *AMDGPUUnifyDivergentExitNodesImpl::unifyReturnBlockSet(
return NewRetBlock;
}
+static BasicBlock *
+createDummyReturnBlock(Function &F,
+ SmallVector<BasicBlock *, 4> &ReturningBlocks) {
+ BasicBlock *DummyReturnBB =
+ BasicBlock::Create(F.getContext(), "DummyReturnBlock", &F);
+ Type *RetTy = F.getReturnType();
+ Value *RetVal = RetTy->isVoidTy() ? nullptr : PoisonValue::get(RetTy);
+ ReturnInst::Create(F.getContext(), RetVal, DummyReturnBB);
+ ReturningBlocks.push_back(DummyReturnBB);
+ return DummyReturnBB;
+}
+
+/// Handle conditional branch instructions (-> 2 targets) and callbr
+/// instructions with N targets.
+static void handleNBranch(Function &F, BasicBlock *BB, Instruction *BI,
+ BasicBlock *DummyReturnBB,
+ std::vector<DominatorTree::UpdateType> &Updates) {
+ SmallVector<BasicBlock *, 2> Successors(successors(BB));
+
+ // Create a new transition block to hold the conditional branch.
+ BasicBlock *TransitionBB = BB->splitBasicBlock(BI, "TransitionBlock");
+
+ Updates.reserve(Updates.size() + 2 * Successors.size() + 2);
+
+ // 'Successors' become successors of TransitionBB instead of BB,
+ // and TransitionBB becomes a single successor of BB.
+ Updates.emplace_back(DominatorTree::Insert, BB, TransitionBB);
+ for (BasicBlock *Successor : Successors) {
+ Updates.emplace_back(DominatorTree::Insert, TransitionBB, Successor);
+ Updates.emplace_back(DominatorTree::Delete, BB, Successor);
+ }
+
+ // Create a branch that will always branch to the transition block and
+ // references DummyReturnBB.
+ BB->getTerminator()->eraseFromParent();
+ BranchInst::Create(TransitionBB, DummyReturnBB,
+ ConstantInt::getTrue(F.getContext()), BB);
+ Updates.emplace_back(DominatorTree::Insert, BB, DummyReturnBB);
+}
+
bool AMDGPUUnifyDivergentExitNodesImpl::run(Function &F, DominatorTree *DT,
const PostDominatorTree &PDT,
const UniformityInfo &UA) {
- assert(hasOnlySimpleTerminator(F) && "Unsupported block terminator.");
-
if (PDT.root_size() == 0 ||
(PDT.root_size() == 1 &&
- !isa<BranchInst>(PDT.getRoot()->getTerminator())))
+ !isa<BranchInst, CallBrInst>(PDT.getRoot()->getTerminator())))
return false;
// Loop over all of the blocks in a function, tracking all of the blocks that
@@ -222,46 +260,27 @@ bool AMDGPUUnifyDivergentExitNodesImpl::run(Function &F, DominatorTree *DT,
if (HasDivergentExitBlock)
UnreachableBlocks.push_back(BB);
} else if (BranchInst *BI = dyn_cast<BranchInst>(BB->getTerminator())) {
-
- ConstantInt *BoolTrue = ConstantInt::getTrue(F.getContext());
- if (DummyReturnBB == nullptr) {
- DummyReturnBB = BasicBlock::Create(F.getContext(),
- "DummyReturnBlock", &F);
- Type *RetTy = F.getReturnType();
- Value *RetVal = RetTy->isVoidTy() ? nullptr : PoisonValue::get(RetTy);
- ReturnInst::Create(F.getContext(), RetVal, DummyReturnBB);
- ReturningBlocks.push_back(DummyReturnBB);
- }
+ if (DummyReturnBB == nullptr)
+ DummyReturnBB = createDummyReturnBlock(F, ReturningBlocks);
if (BI->isUnconditional()) {
BasicBlock *LoopHeaderBB = BI->getSuccessor(0);
BI->eraseFromParent(); // Delete the unconditional branch.
// Add a new conditional branch with a dummy edge to the return block.
- BranchInst::Create(LoopHeaderBB, DummyReturnBB, BoolTrue, BB);
- Updates.emplace_back(DominatorTree::Insert, BB, DummyReturnBB);
- } else { // Conditional branch.
- SmallVector<BasicBlock *, 2> Successors(successors(BB));
-
- // Create a new transition block to hold the conditional branch.
- BasicBlock *TransitionBB = BB->splitBasicBlock(BI, "TransitionBlock");
-
- Updates.reserve(Updates.size() + 2 * Successors.size() + 2);
-
- // 'Successors' become successors of TransitionBB instead of BB,
- // and TransitionBB becomes a single successor of BB.
- Updates.emplace_back(DominatorTree::Insert, BB, TransitionBB);
- for (BasicBlock *Successor : Successors) {
- Updates.emplace_back(DominatorTree::Insert, TransitionBB, Successor);
- Updates.emplace_back(DominatorTree::Delete, BB, Successor);
- }
-
- // Create a branch that will always branch to the transition block and
- // references DummyReturnBB.
- BB->getTerminator()->eraseFromParent();
- BranchInst::Create(TransitionBB, DummyReturnBB, BoolTrue, BB);
+ BranchInst::Create(LoopHeaderBB, DummyReturnBB,
+ ConstantInt::getTrue(F.getContext()), BB);
Updates.emplace_back(DominatorTree::Insert, BB, DummyReturnBB);
+ } else {
+ handleNBranch(F, BB, BI, DummyReturnBB, Updates);
}
Changed = true;
+ } else if (CallBrInst *CBI = dyn_cast<CallBrInst>(BB->getTerminator())) {
+ if (DummyReturnBB == nullptr)
+ DummyReturnBB = createDummyReturnBlock(F, ReturningBlocks);
+
+ handleNBranch(F, BB, CBI, DummyReturnBB, Updates);
+ } else {
+ llvm_unreachable("unsupported block terminator");
}
}
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index dfe6f65d240e6..5f09cfbc6d817 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -16476,12 +16476,12 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_,
const TargetRegisterClass *RC = nullptr;
if (Constraint.size() == 1) {
- const unsigned BitWidth = VT.getSizeInBits();
switch (Constraint[0]) {
default:
return TargetLowering::getRegForInlineAsmConstraint(TRI, Constraint, VT);
case 's':
- case 'r':
+ case 'r': {
+ const unsigned BitWidth = VT.getSizeInBits();
switch (BitWidth) {
case 16:
RC = &AMDGPU::SReg_32RegClass;
@@ -16496,7 +16496,9 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_,
break;
}
break;
- case 'v':
+ }
+ case 'v': {
+ const unsigned BitWidth = VT.getSizeInBits();
switch (BitWidth) {
case 16:
RC = Subtarget->useRealTrue16Insts() ? &AMDGPU::VGPR_16RegClass
@@ -16509,9 +16511,11 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_,
break;
}
break;
+ }
case 'a':
if (!Subtarget->hasMAIInsts())
break;
+ const unsigned BitWidth = VT.getSizeInBits();
switch (BitWidth) {
case 16:
RC = &AMDGPU::AGPR_32RegClass;
diff --git a/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp b/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp
index a69d64956d6d9..d43ed8a9364b8 100644
--- a/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp
+++ b/llvm/lib/Transforms/Scalar/StructurizeCFG.cpp
@@ -480,11 +480,10 @@ void StructurizeCFG::analyzeLoops(RegionNode *N) {
} else {
// Test for successors as back edge
BasicBlock *BB = N->getNodeAs<BasicBlock>();
- BranchInst *Term = cast<BranchInst>(BB->getTerminator());
-
- for (BasicBlock *Succ : Term->successors())
- if (Visited.count(Succ))
- Loops[Succ] = BB;
+ if (BranchInst *Term = dyn_cast<BranchInst>(BB->getTerminator()))
+ for (BasicBlock *Succ : Term->successors())
+ if (Visited.count(Succ))
+ Loops[Succ] = BB;
}
}
@@ -516,7 +515,7 @@ void StructurizeCFG::gatherPredicates(RegionNode *N) {
for (BasicBlock *P : predecessors(BB)) {
// Ignore it if it's a branch from outside into our region entry
- if (!ParentRegion->contains(P))
+ if (!ParentRegion->contains(P) || !dyn_cast<BranchInst>(P->getTerminator()))
continue;
Region *R = RI->getRegionFor(P);
@@ -1284,13 +1283,13 @@ bool StructurizeCFG::makeUniformRegion(Region *R, UniformityInfo &UA) {
/// Run the transformation for each region found
bool StructurizeCFG::run(Region *R, DominatorTree *DT) {
- if (R->isTopLevelRegion())
+ // CallBr and its corresponding blocks must not be modified by this pass.
+ if (R->isTopLevelRegion() || isa<CallBrInst>(R->getEntry()->getTerminator()))
return false;
this->DT = DT;
Func = R->getEntry()->getParent();
- assert(hasOnlySimpleTerminator(*Func) && "Unsupported block terminator.");
ParentRegion = R;
diff --git a/llvm/test/CodeGen/AMDGPU/callbr.ll b/llvm/test/CodeGen/AMDGPU/callbr.ll
new file mode 100644
index 0000000000000..253a6ec100eae
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/callbr.ll
@@ -0,0 +1,54 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=amdgcn -mcpu=gfx90a < %s | FileCheck %s
+
+define void @callbr_inline_asm(ptr %src, ptr %dst1, ptr %dst2, i32 %c) {
+; CHECK-LABEL: callbr_inline_asm:
+; CHECK: ; %bb.0:
+; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT: flat_load_dword v0, v[0:1]
+; CHECK-NEXT: ;;#ASMSTART
+; CHECK-NEXT: v_cmp_gt_i32 vcc v6, 42; s_cbranch_vccnz .LBB0_2
+; CHECK-NEXT: ;;#ASMEND
+; CHECK-NEXT: ; %bb.1: ; %fallthrough
+; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
+; CHECK-NEXT: flat_store_dword v[2:3], v0
+; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
+; CHECK-NEXT: s_setpc_b64 s[30:31]
+; CHECK-NEXT: .LBB0_2: ; Inline asm indirect target
+; CHECK-NEXT: ; %indirect
+; CHECK-NEXT: ; Label of block must be emitted
+; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
+; CHECK-NEXT: flat_store_dword v[4:5], v0
+; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
+; CHECK-NEXT: s_setpc_b64 s[30:31]
+ %a = load i32, ptr %src, align 4
+ callbr void asm "v_cmp_gt_i32 vcc $0, 42; s_cbranch_vccnz ${1:l}", "r,!i"(i32 %c) to label %fallthrough [label %indirect]
+fallthrough:
+ store i32 %a, ptr %dst1, align 4
+ br label %ret
+indirect:
+ store i32 %a, ptr %dst2, align 4
+ br label %ret
+ret:
+ ret void
+}
+
+define void @callbr_self_loop(i1 %c) {
+; CHECK-LABEL: callbr_self_loop:
+; CHECK: ; %bb.0:
+; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT: .LBB1_1: ; %callbr
+; CHECK-NEXT: ; =>This Inner Loop Header: Depth=1
+; CHECK-NEXT: ;;#ASMSTART
+; CHECK-NEXT: ;;#ASMEND
+; CHECK-NEXT: s_branch .LBB1_1
+; CHECK-NEXT: .LBB1_2: ; Inline asm indirect target
+; CHECK-NEXT: ; %callbr.target.ret
+; CHECK-NEXT: ; Label of block must be emitted
+; CHECK-NEXT: s_setpc_b64 s[30:31]
+ br label %callbr
+callbr:
+ callbr void asm "", "!i"() to label %callbr [label %ret]
+ret:
+ ret void
+}
diff --git a/llvm/test/CodeGen/AMDGPU/do-not-unify-divergent-exit-nodes-with-musttail.ll b/llvm/test/CodeGen/AMDGPU/do-not-unify-divergent-exit-nodes-with-musttail.ll
index 007e3f0a6bdbc..076a99ff8588f 100644
--- a/llvm/test/CodeGen/AMDGPU/do-not-unify-divergent-exit-nodes-with-musttail.ll
+++ b/llvm/test/CodeGen/AMDGPU/do-not-unify-divergent-exit-nodes-with-musttail.ll
@@ -3,6 +3,7 @@
declare void @foo(ptr)
declare i1 @bar(ptr)
+declare i32 @bar32(ptr)
define void @musttail_call_without_return_value(ptr %p) {
; CHECK-LABEL: define void @musttail_call_without_return_value(
@@ -28,6 +29,31 @@ bb.1:
ret void
}
+define void @musttail_call_without_return_value_callbr(ptr %p) {
+; CHECK-LABEL: define void @musttail_call_without_return_value_callbr(
+; CHECK-SAME: ptr [[P:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT: [[ENTRY:.*:]]
+; CHECK-NEXT: [[LOAD:%.*]] = load i32, ptr [[P]], align 1
+; CHECK-NEXT: callbr void asm "", "r,!i"(i32 [[LOAD]])
+; CHECK-NEXT: to label %[[BB_0:.*]] [label %bb.1]
+; CHECK: [[BB_0]]:
+; CHECK-NEXT: musttail call void @foo(ptr [[P]])
+; CHECK-NEXT: ret void
+; CHECK: [[BB_1:.*:]]
+; CHECK-NEXT: ret void
+;
+entry:
+ %load = load i32, ptr %p, align 1
+ callbr void asm "", "r,!i"(i32 %load) to label %bb.0 [label %bb.1]
+
+bb.0:
+ musttail call void @foo(ptr %p)
+ ret void
+
+bb.1:
+ ret void
+}
+
define i1 @musttail_call_with_return_value(ptr %p) {
; CHECK-LABEL: define i1 @musttail_call_with_return_value(
; CHECK-SAME: ptr [[P:%.*]]) #[[ATTR0]] {
@@ -51,3 +77,28 @@ bb.0:
bb.1:
ret i1 %load
}
+
+define i32 @musttail_call_with_return_value_callbr(ptr %p) {
+; CHECK-LABEL: define i32 @musttail_call_with_return_value_callbr(
+; CHECK-SAME: ptr [[P:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT: [[ENTRY:.*:]]
+; CHECK-NEXT: [[LOAD:%.*]] = load i32, ptr [[P]], align 1
+; CHECK-NEXT: callbr void asm "", "r,!i"(i32 [[LOAD]])
+; CHECK-NEXT: to label %[[BB_0:.*]] [label %bb.1]
+; CHECK: [[BB_0]]:
+; CHECK-NEXT: [[RET:%.*]] = musttail call i32 @bar32(ptr [[P]])
+; CHECK-NEXT: ret i32 [[RET]]
+; CHECK: [[BB_1:.*:]]
+; CHECK-NEXT: ret i32 [[LOAD]]
+;
+entry:
+ %load = load i32, ptr %p, align 1
+ callbr void asm "", "r,!i"(i32 %load) to label %bb.0 [label %bb.1]
+
+bb.0:
+ %ret = musttail call i32 @bar32(ptr %p)
+ ret i32 %ret
+
+bb.1:
+ ret i32 %load
+}
diff --git a/llvm/test/CodeGen/AMDGPU/infinite-loop.ll b/llvm/test/CodeGen/AMDGPU/infinite-loop.ll
index bea532bd52955..2181c77f0e6ef 100644
--- a/llvm/test/CodeGen/AMDGPU/infinite-loop.ll
+++ b/llvm/test/CodeGen/AMDGPU/infinite-loop.ll
@@ -36,26 +36,60 @@ loop:
br label %loop
}
+define amdgpu_kernel void @infinite_loop_callbr(ptr addrspace(1) %out) {
+; SI-LABEL: infinite_loop_callbr:
+; SI: ; %bb.0: ; %entry
+; SI-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9
+; SI-NEXT: ;;#ASMSTART
+; SI-NEXT: ;;#ASMEND
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_mov_b32_e32 v0, 0x3e7
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_waitcnt vmcnt(0)
+; SI-NEXT: s_endpgm
+; IR-LABEL: @infinite_loop_callbr(
+; IR-NEXT: entry:
+; IR-NEXT: callbr void asm "", ""()
+; IR-NEXT: to label [[LOOP:%.*]] []
+; IR: loop:
+; IR-NEXT: store volatile i32 999, ptr addrspace(1) [[OUT:%.*]], align 4
+; IR-NEXT: br i1 true, label [[TRANSITIONBLOCK:%.*]], label [[DUMMYRETURNBLOCK:%.*]]
+; IR: TransitionBlock:
+; IR-NEXT: callbr void asm "", ""()
+; IR-NEXT: to label [[LOOP]] []
+; IR: DummyReturnBlock:
+; IR-NEXT: ret void
+;
+entry:
+ callbr void asm "", ""() to label %loop []
+
+loop:
+ store volatile i32 999, ptr addrspace(1) %out, align 4
+ callbr void asm "", ""() to label %loop []
+}
+
define amdgpu_kernel void @infinite_loop_ret(ptr addrspace(1) %out) {
; SI-LABEL: infinite_loop_ret:
; SI: ; %bb.0: ; %entry
; SI-NEXT: v_cmp_eq_u32_e32 vcc, 1, v0
; SI-NEXT: s_and_saveexec_b64 s[0:1], vcc
-; SI-NEXT: s_cbranch_execz .LBB1_3
+; SI-NEXT: s_cbranch_execz .LBB2_3
; SI-NEXT: ; %bb.1: ; %loop.preheader
; SI-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT: s_mov_b32 s3, 0xf000
; SI-NEXT: s_mov_b32 s2, -1
; SI-NEXT: v_mov_b32_e32 v0, 0x3e7
; SI-NEXT: s_and_b64 vcc, exec, -1
-; SI-NEXT: .LBB1_2: ; %loop
+; SI-NEXT: .LBB2_2: ; %loop
; SI-NEXT: ; =>This Inner Loop Header: Depth=1
; SI-NEXT: s_waitcnt lgkmcnt(0)
; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; SI-NEXT: s_waitcnt vmcnt(0)
; SI-NEXT: s_mov_b64 vcc, vcc
-; SI-NEXT: s_cbranch_vccnz .LBB1_2
-; SI-NEXT: .LBB1_3: ; %UnifiedReturnBlock
+; SI-NEXT: s_cbranch_vccnz .LBB2_2
+; SI-NEXT: .LBB2_3: ; %UnifiedReturnBlock
; SI-NEXT: s_endpgm
; IR-LABEL: @infinite_loop_ret(
; IR-NEXT: entry:
@@ -81,44 +115,93 @@ return:
ret void
}
+define amdgpu_kernel void @infinite_loop_ret_callbr(ptr addrspace(1) %out) {
+; SI-LABEL: infinite_loop_ret_callbr:
+; SI: ; %bb.0: ; %entry
+; SI-NEXT: v_cmp_eq_u32_e32 vcc, 1, v0
+; SI-NEXT: v_cndmask_b32_e64 v0, 0, 1, vcc
+; SI-NEXT: ;;#ASMSTART
+; SI-NEXT: ;;#ASMEND
+; SI-NEXT: ; %bb.1: ; %loop.preheader
+; SI-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_mov_b32_e32 v0, 0x3e7
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_waitcnt vmcnt(0)
+; SI-NEXT: .LBB3_2: ; Inline asm indirect target
+; SI-NEXT: ; %UnifiedReturnBlock
+; SI-NEXT: ; Label of block must be emitted
+; SI-NEXT: s_endpgm
+; IR-LABEL: @infinite_loop_ret_callbr(
+; IR-NEXT: entry:
+; IR-NEXT: [[TMP:%.*]] = tail call i32 @llvm.amdgcn.workitem.id.x()
+; IR-NEXT: [[COND:%.*]] = icmp eq i32 [[TMP]], 1
+; IR-NEXT: [[COND32:%.*]] = zext i1 [[COND]] to i32
+; IR-NEXT: callbr void asm "", "r,!i"(i32 [[COND32]])
+; IR-NEXT: to label [[LOOP:%.*]] [label %UnifiedReturnBlock]
+; IR: loop:
+; IR-NEXT: store volatile i32 999, ptr addrspace(1) [[OUT:%.*]], align 4
+; IR-NEXT: br i1 true, label [[TRANSITIONBLOCK:%.*]], label [[UNIFIEDRETURNBLOCK:%.*]]
+; IR: TransitionBlock:
+; IR-NEXT: callbr void asm "", ""()
+; IR-NEXT: to label [[LOOP]] []
+; IR: UnifiedReturnBlock:
+; IR-NEXT: ret void
+;
+entry:
+ %tmp = tail call i32 @llvm.amdgcn.workitem.id.x()
+ %cond = icmp eq i32 %tmp, 1
+ %cond32 = zext i1 %cond to i32
+ callbr void asm "", "r,!i"(i32 %cond32) to label %loop [label %return]
+
+loop:
+ store volatile i32 999, ptr addrspace(1) %out, align 4
+ callbr void asm "", ""() to label %loop []
+
+return:
+ ret void
+}
+
define amdgpu_kernel void @infinite_loops(ptr addrspace(1) %out) {
; SI-LABEL: infinite_loops:
; SI: ; %bb.0: ; %entry
; SI-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT: s_mov_b64 s[2:3], -1
-; SI-NEXT: s_cbranch_scc1 .LBB2_4
+; SI-NEXT: s_cbranch_scc1 .LBB4_4
; SI-NEXT: ; %bb.1:
; SI-NEXT: s_mov_b32 s3, 0xf000
; SI-NEXT: s_mov_b32 s2, -1
; SI-NEXT: v_mov_b32_e32 v0, 0x378
; SI-NEXT: s_and_b64 vcc, exec, -1
-; SI-NEXT: .LBB2_2: ; %loop2
+; SI-NEXT: .LBB4_2: ; %loop2
; SI-NEXT: ; =>This Inner Loop Header: Depth=1
; SI-NEXT: s_waitcnt lgkmcnt(0)
; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; SI-NEXT: s_waitcnt vmcnt(0)
; SI-NEXT: s_mov_b64 vcc, vcc
-; SI-NEXT: s_cbranch_vccnz .LBB2_2
+; SI-NEXT: s_cbranch_vccnz .LBB4_2
; SI-NEXT: ; %bb.3: ; %Flow
; SI-NEXT: s_mov_b64 s[2:3], 0
-; SI-NEXT: .LBB2_4: ; %Flow2
+; SI-NEXT: .LBB4_4: ; %Flow2
; SI-NEXT: s_and_b64 vcc, exec, s[2:3]
; SI-NEXT: s_waitcnt lgkmcnt(0)
; SI-NEXT: s_mov_b64 vcc, vcc
-; SI-NEXT: s_cbranch_vccz .LBB2_7
+; SI-NEXT: s_cbranch_vccz .LBB4_7
; SI-NEXT: ; %bb.5:
; SI-NEXT: s_mov_b32 s3, 0xf000
; SI-NEXT: s_mov_b32 s2, -1
; SI-NEXT: s_waitcnt expcnt(0)
; SI-NEXT: v_mov_b32_e32 v0, 0x3e7
; SI-NEXT: s_and_b64 vcc, exec, 0
-; SI-NEXT: .LBB2_6: ; %loop1
+; SI-NEXT: .LBB4_6: ; %loop1
; SI-NEXT: ; =>This Inner Loop Header: Depth=1
; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
; SI-NEXT: s_waitcnt vmcnt(0)
; SI-NEXT: s_mov_b64 vcc, vcc
-; SI-NEXT: s_cbranch_vccz .LBB2_6
-; SI-NEXT: .LBB2_7: ; %DummyReturnBlock
+; SI-NEXT: s_cbranch_vccz .LBB4_6
+; SI-NEXT: .LBB4_7: ; %DummyReturnBlock
; SI-NEXT: s_endpgm
; IR-LABEL: @infinite_loops(
; IR-NEXT: entry:
@@ -144,24 +227,78 @@ loop2:
br label %loop2
}
+define amdgpu_kernel void @infinite_loops_callbr(ptr addrspace(1) %out) {
+; SI-LABEL: infinite_loops_callbr:
+; SI: ; %bb.0: ; %entry
+; SI-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x9
+; SI-NEXT: s_waitcnt lgkmcnt(0)
+; SI-NEXT: ;;#ASMSTART
+; SI-NEXT: ;;#ASMEND
+; SI-NEXT: ; %bb.1: ; %loop1
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_mov_b32_e32 v0, 0x3e7
+; SI-NEXT: buffer_store_dword v0, off, s[0:3], 0
+; SI-NEXT: s_waitcnt vmcnt(0)
+; SI-NEXT: s_endpgm
+; SI-NEXT: .LBB5_2: ; Inline asm indirect target
+; SI-NEXT: ; %loop2.preheader
+; SI-NEXT: ; Label of block must be emitted
+; SI-NEXT: s_mov_b32 s3, 0xf000
+; SI-NEXT: s_mov_b32 s2, -1
+; SI-NEXT: v_mov...
[truncated]
|
|
StructurizeCFG is very sensitive to changes. Could you do a full CQE cycles for this? |
arsenm
reviewed
Aug 5, 2025
| @@ -16476,12 +16476,12 @@ SITargetLowering::getRegForInlineAsmConstraint(const TargetRegisterInfo *TRI_, | |||
|
|
|||
| const TargetRegisterClass *RC = nullptr; | |||
There was a problem hiding this comment.
Everything in this file looks like unrelated style changes
There was a problem hiding this comment.
No, this is actually important to support the asm label constraint "!i". This should fall through to the default case (imho) without causing VT.getSizeInBits() to crash.
There was a problem hiding this comment.
Can you do this separately? You should be able to refer to labels in regular inline asm
arsenm
reviewed
Aug 5, 2025
Will be done after the reviews before the merge 👍 |
Co-authored-by: Matt Arsenault <[email protected]>
|
A couple of comments about the commit description. These also apply to #149308.
|
ssahasra
reviewed
Aug 6, 2025
| @@ -1284,13 +1283,13 @@ bool StructurizeCFG::makeUniformRegion(Region *R, UniformityInfo &UA) { | |||
|
|
|||
| /// Run the transformation for each region found | |||
| bool StructurizeCFG::run(Region *R, DominatorTree *DT) { | |||
| if (R->isTopLevelRegion()) | |||
| // CallBr and its corresponding blocks must not be modified by this pass. | |||
There was a problem hiding this comment.
I think this is saying that if a region is headed by a CallBr, then anything inside it is separately structurized by visiting the nested regions. Is that correct? Are there any assertions we could put here about the current region R, to demonstrate that there is nothing to structurize?
ro-i
changed the title
I think they do. I developed and submitted the PR to support inline-asm for amdgpu as two PRs so that it's not too large and easier to review. So, if I would call #149308 "[AMDGPU] Support callbr with inline-asm" instead of "[AMDGPU][FixIrreducible][UnifyLoopExits] Support callbr with inline-asm" this would be a lie since this PR here is needed, too.
Thanks, I removed the "basic" since it's misleading. For me, "basic" = "inline-asm" and "complete" = "inline-asm + intrinsics", but that is in no way based on any commonly used definition. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Finishes adding inline-asm callbr support for AMDGPU, started by #149308.