Author: Piotr Zegar
Date: 2024-07-23T08:46:14+02:00
New Revision: a96af6c18ea45269c4d2e1fd762a14d763be0358
URL:
https://github.com/llvm/llvm-project/commit/a96af6c18ea45269c4d2e1fd762a14d763be0358
DIFF:
https://github.com/llvm/llvm-project/commit/a96af6c18ea45269c4d2e1fd762a14d763be0358.diff
L
@@ -266,6 +287,47 @@ class StaleMatcher {
}
return BestBlock;
}
+ // Uses pseudo probe information to attach the profile to the appropriate
+ // block.
+ const FlowBlock *matchWithPseudoProbes(
+ const std::vector &PseudoProbes) const {
+// Searches for th
@@ -241,6 +242,10 @@ class BinaryContext {
/// Function fragments to skip.
std::unordered_set FragmentsToSkip;
+ /// Fragment equivalence classes to query belonging to the same "family" in
+ /// presence of multiple fragments/multiple parents.
+ EquivalenceClasses Fragm
h-vetinari wrote:
This now closes #100018
https://github.com/llvm/llvm-project/pull/99375
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/aaupov updated
https://github.com/llvm/llvm-project/pull/99988
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/aaupov updated
https://github.com/llvm/llvm-project/pull/99988
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/aaupov updated
https://github.com/llvm/llvm-project/pull/99979
>From f6478e36a962843329c519ba35ad2a132ffd8c9e Mon Sep 17 00:00:00 2001
From: Amir Ayupov
Date: Mon, 22 Jul 2024 16:34:02 -0700
Subject: [PATCH] fix getOrCreateJumpTable
Created using spr 1.3.4
---
bolt/lib/Core
https://github.com/aaupov created
https://github.com/llvm/llvm-project/pull/99988
Multi-way splitting can cause multiple fragments to access the same jump
table. Relax the assumption that a jump table can only have up to two
parents.
Test Plan: added bolt/test/X86/three-way-split-jt.s
__
https://github.com/aaupov created
https://github.com/llvm/llvm-project/pull/99979
Three-way splitting can create references between split fragments (warm
to cold or vice versa) that are not handled by
`isChildOf/isParentOf/isChildOrParentOf`. Generalize fragment
relationships to allow checking i
Author: RoseZhang03
Date: 2024-07-22T15:25:52-07:00
New Revision: 052f002d33e247b9e00994a100bf32531fefe615
URL:
https://github.com/llvm/llvm-project/commit/052f002d33e247b9e00994a100bf32531fefe615
DIFF:
https://github.com/llvm/llvm-project/commit/052f002d33e247b9e00994a100bf32531fefe615.diff
L
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/99891
>From 0274f697376264c2d77816190f9a434f64e79089 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 22 Jul 2024 11:56:23 -0700
Subject: [PATCH 1/2] Changed assignment of profiles with pseudo probe index
Creat
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/99891
>From 0274f697376264c2d77816190f9a434f64e79089 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 22 Jul 2024 11:56:23 -0700
Subject: [PATCH] Changed assignment of profiles with pseudo probe index
Created u
https://github.com/PiotrZSL commented:
Except pointed out issues, looks fine for me.
https://github.com/llvm/llvm-project/pull/99681
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo
@@ -1455,6 +1462,64 @@ struct SwitchCoroutineSplitter {
setCoroInfo(F, Shape, Clones);
}
+ static Function *createNoAllocVariant(Function &F, coro::Shape &Shape,
+SmallVectorImpl &Clones) {
+auto *OrigFnTy = F.getFunctionType(
@@ -968,8 +969,8 @@ PassBuilder::buildInlinerPipeline(OptimizationLevel Level,
// it's been modified since.
MainCGPipeline.addPass(createCGSCCToFunctionPassAdaptor(
RequireAnalysisPass()));
-
MainCGPipeline.addPass(CoroSplitPass(Level != OptimizationLevel::O0));
+
@@ -1,4 +1,6 @@
-// RUN: %check_clang_tidy %s llvmlibc-implementation-in-namespace %t
+// RUN: %check_clang_tidy %s llvmlibc-implementation-in-namespace %t -fix
ilovepi wrote:
ack. I saw those running, but I saw `-fix` on a lot of other tests outside of
llvmlibc
@@ -41,8 +50,26 @@ void ImplementationInNamespaceCheck::check(
// Enforce that the namespace is the result of macro expansion
if (Result.SourceManager->isMacroBodyExpansion(NS->getLocation()) == false) {
-diag(NS->getLocation(), "the outermost namespace should be the '
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/97050
>From 2541faa229c0099ff05eb8e451772d9410005463 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Thu, 27 Jun 2024 16:32:48 +0200
Subject: [PATCH] AMDGPU: Remove flat/global atomic fadd v2bf16 intrinsics
These a
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/96876
>From 06b6158450a2af1d4b9f8cb585a9bd8597c54ed8 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Wed, 26 Jun 2024 23:18:32 +0200
Subject: [PATCH] clang/AMDGPU: Emit atomicrmw for flat/global atomic min/max
f64
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/96875
>From 415dfe8521d5065f103727449adb7bc30d4193fe Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Wed, 26 Jun 2024 19:34:43 +0200
Subject: [PATCH] clang/AMDGPU: Emit atomicrmw for global/flat fadd v2bf16
builtin
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/96874
>From 9b0e5ee3db46e3a39f2ec5f4d5b9d5c99bbe2feb Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Wed, 26 Jun 2024 19:15:26 +0200
Subject: [PATCH] clang/AMDGPU: Emit atomicrmw from flat_atomic_{f32|f64}
builtins
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/96873
>From 48fd0a62922171fd563f9c5fdba640153075578e Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Wed, 26 Jun 2024 19:12:59 +0200
Subject: [PATCH] clang/AMDGPU: Emit atomicrmw from
{global|flat}_atomic_fadd_v2f1
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/96872
>From e1da39eaef8785570f687dd2a00ae89b918252b1 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Tue, 11 Jun 2024 10:58:44 +0200
Subject: [PATCH 1/2] clang/AMDGPU: Emit atomicrmw for
__builtin_amdgcn_global_ato
@@ -3,7 +3,7 @@
; RUN: llvm-profdata merge %S/Inputs/misexpect-branch-correct.proftext -o
%t.profdata
-; RUN: opt < %s -passes="function(lower-expect),pgo-instr-use"
-pgo-test-profile-file=%t.profdata -pgo-missing-annotations
-pass-remarks=missing-annotation -S 2>&1 | File
https://github.com/mordante updated
https://github.com/llvm/llvm-project/pull/99375
>From be0f12219752be987ca6674fe384281e890520a2 Mon Sep 17 00:00:00 2001
From: Mark de Wever
Date: Mon, 15 Jul 2024 07:45:02 +0200
Subject: [PATCH] [libc++][spaceship] Marks P1614 as complete.
Implements parts o
@@ -369,9 +369,21 @@ static bool lowerExpectIntrinsic(Function &F) {
if (BranchInst *BI = dyn_cast(BB.getTerminator())) {
if (handleBranchExpect(*BI))
ExpectIntrinsicsHandled++;
+ else {
+SmallVector Weights;
+if (extractBranchWeights(*BI,
@@ -48,7 +48,7 @@ void test_local_add_2f16_noret(__local half2 *addr, half2 x) {
}
// CHECK-LABEL: test_flat_add_2f16
-// CHECK: [[RMW:%.+]] = atomicrmw fadd ptr %{{.+}}, <2 x half> %{{.+}}
syncscope("agent") seq_cst, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+// C
https://github.com/shawbyoung created
https://github.com/llvm/llvm-project/pull/99891
Implemented pseudo probe block matching. When matched functions have
equal pseudo probe checksums, the indices of block pseudo probes are
used to match blocks following opcode and call hash block matching.
Tes
https://github.com/arsenm approved this pull request.
https://github.com/llvm/llvm-project/pull/96163
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -48,7 +48,7 @@ void test_local_add_2f16_noret(__local half2 *addr, half2 x) {
}
// CHECK-LABEL: test_flat_add_2f16
-// CHECK: [[RMW:%.+]] = atomicrmw fadd ptr %{{.+}}, <2 x half> %{{.+}}
syncscope("agent") seq_cst, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+// C
@@ -369,9 +369,21 @@ static bool lowerExpectIntrinsic(Function &F) {
if (BranchInst *BI = dyn_cast(BB.getTerminator())) {
if (handleBranchExpect(*BI))
ExpectIntrinsicsHandled++;
+ else {
+SmallVector Weights;
+if (extractBranchWeights(*BI,
@@ -3,7 +3,7 @@
; RUN: llvm-profdata merge %S/Inputs/misexpect-branch-correct.proftext -o
%t.profdata
-; RUN: opt < %s -passes="function(lower-expect),pgo-instr-use"
-pgo-test-profile-file=%t.profdata -pgo-missing-annotations
-pass-remarks=missing-annotation -S 2>&1 | File
https://github.com/shiltian approved this pull request.
LGTM with one question regarding the memory order
https://github.com/llvm/llvm-project/pull/96875
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi
@@ -48,7 +48,7 @@ void test_local_add_2f16_noret(__local half2 *addr, half2 x) {
}
// CHECK-LABEL: test_flat_add_2f16
-// CHECK: [[RMW:%.+]] = atomicrmw fadd ptr %{{.+}}, <2 x half> %{{.+}}
syncscope("agent") seq_cst, align 4, !amdgpu.no.fine.grained.memory !{{[0-9]+$}}
+// C
cdevadas wrote:
The latest patch optimizes the PatFrag and the patterns written further by
using OtherPredicates. The lit test changes in the latest patch are a missed
optimization I incorrectly introduced earlier in this PR for GFX7. It is now
fixed and matches the default behavior with the c
https://github.com/jayfoad approved this pull request.
LGTM.
https://github.com/llvm/llvm-project/pull/96163
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -34,18 +34,17 @@ entry:
}
define amdgpu_kernel void @test_llvm_amdgcn_fdot2_bf16_bf16_dpp(
-; SDAG-GFX11-LABEL: test_llvm_amdgcn_fdot2_bf16_bf16_dpp:
-; SDAG-GFX11: ; %bb.0: ; %entry
-; SDAG-GFX11-NEXT:s_load_b128 s[0:3], s[0:1], 0x24
-; SDAG-GFX11-NEXT:s_wait
https://github.com/jayfoad edited
https://github.com/llvm/llvm-project/pull/96163
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
arsenm wrote:
ping
https://github.com/llvm/llvm-project/pull/96875
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
https://github.com/jayfoad approved this pull request.
https://github.com/llvm/llvm-project/pull/96162
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -6,7 +6,7 @@ declare i32 @llvm.amdgcn.global.atomic.csub(ptr addrspace(1),
i32)
; GCN-LABEL: {{^}}global_atomic_csub_rtn:
; PREGFX12: global_atomic_csub v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9:]+}},
s{{\[[0-9]+:[0-9]+\]}} glc
-; GFX12PLUS: global_atomic_sub_clamp_u32 v0, v0, v1
arsenm wrote:
> > I still think it is terribly surprising all of the test diff shows up in
> > this commit, and not the selection case
>
> Because the selection support is done in the next PR of the review stack,
> #96162. This patch takes care of choosing the right opcode while merging the
>
Author: Lukacma
Date: 2024-07-22T12:32:23+01:00
New Revision: 9fb427c61f9444e90bad89b14723f1c369f67aad
URL:
https://github.com/llvm/llvm-project/commit/9fb427c61f9444e90bad89b14723f1c369f67aad
DIFF:
https://github.com/llvm/llvm-project/commit/9fb427c61f9444e90bad89b14723f1c369f67aad.diff
LOG:
@@ -49,7 +49,7 @@ void test_s_wait_event_export_ready() {
}
// CHECK-LABEL: @test_global_add_f32
-// CHECK: {{.*}}call{{.*}} float
@llvm.amdgcn.global.atomic.fadd.f32.p1.f32(ptr addrspace(1) %{{.*}}, float
%{{.*}})
+// CHECK: = atomicrmw fadd ptr addrspace(1) %addr, float %x
https://github.com/tblah edited https://github.com/llvm/llvm-project/pull/99507
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
@@ -432,9 +439,10 @@ def SimdOp : OpenMP_Op<"simd", traits = [
AttrSizedOperandSegments, DeclareOpInterfaceMethods,
RecursiveMemoryEffects, SingleBlock
], clauses = [
-// TODO: Complete clause list (linear, private, reduction).
+// TODO: Complete clause list (
@@ -238,9 +237,9 @@ def SectionOp : OpenMP_Op<"section",
[HasParent<"SectionsOp">],
def SectionsOp : OpenMP_Op<"sections", traits = [
AttrSizedOperandSegments
], clauses = [
-// TODO: Complete clause list (private).
// TODO: Sort clauses alphabetically.
https://github.com/tblah commented:
I think it would be best to add errors for these unsupported clauses to the
OpenMPToLLVMIR translation, so that clauses in MLIR are not silently ignored.
https://github.com/llvm/llvm-project/pull/99507
___
llvm-bran
Author: Max Beck-Jones
Date: 2024-07-22T10:24:54+01:00
New Revision: ebf0fc9ae845af15baed663d79a5e4e88542f1e4
URL:
https://github.com/llvm/llvm-project/commit/ebf0fc9ae845af15baed663d79a5e4e88542f1e4
DIFF:
https://github.com/llvm/llvm-project/commit/ebf0fc9ae845af15baed663d79a5e4e88542f1e4.diff
49 matches
Mail list logo
