[llvm-branch-commits] [openmp] release/18.x: [OpenMP][OMPT] Fix hsa include when building amdgpu/src/rtl.cpp (PR #95484)
jhuber6 wrote: Made https://github.com/llvm/llvm-project/pull/95769, don't know if the window for the 18.x window is still open for a back port. https://github.com/llvm/llvm-project/pull/95484 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [mlir] 46a67e7 - Revert "[mlir][Target] Improve ROCDL gpu serialization API (#95456)"
Author: Fabian Mora
Date: 2024-06-17T09:08:42-05:00
New Revision: 46a67e7de24fa997a97a0b7df7391727dfc3aa5e
URL:
https://github.com/llvm/llvm-project/commit/46a67e7de24fa997a97a0b7df7391727dfc3aa5e
DIFF:
https://github.com/llvm/llvm-project/commit/46a67e7de24fa997a97a0b7df7391727dfc3aa5e.diff
LOG: Revert "[mlir][Target] Improve ROCDL gpu serialization API (#95456)"
This reverts commit 954cb5f9a220eb18afe2d0025722067a8299c41c.
Added:
Modified:
mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
mlir/lib/Dialect/GPU/CMakeLists.txt
mlir/lib/Target/LLVM/CMakeLists.txt
mlir/lib/Target/LLVM/ROCDL/Target.cpp
Removed:
diff --git a/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
b/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
index 44c9ded317fa5..374fa65bd02e3 100644
--- a/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
+++ b/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
@@ -27,19 +27,6 @@ namespace ROCDL {
/// 5. Returns an empty string.
StringRef getROCMPath();
-/// Helper enum for specifying the AMD GCN device libraries required for
-/// compilation.
-enum class AMDGCNLibraries : uint32_t {
- None = 0,
- Ockl = 1,
- Ocml = 2,
- OpenCL = 4,
- Hip = 8,
- LastLib = Hip,
- LLVM_MARK_AS_BITMASK_ENUM(LastLib),
- All = (LastLib << 1) - 1
-};
-
/// Base class for all ROCDL serializations from GPU modules into binary
/// strings. By default this class serializes into LLVM bitcode.
class SerializeGPUModuleBase : public LLVM::ModuleToObject {
@@ -62,8 +49,8 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
/// Returns the bitcode files to be loaded.
ArrayRef getFileList() const;
- /// Appends standard ROCm device libraries to `fileList`.
- LogicalResult appendStandardLibs(AMDGCNLibraries libs);
+ /// Appends standard ROCm device libraries like `ocml.bc`, `ockl.bc`, etc.
+ LogicalResult appendStandardLibs();
/// Loads the bitcode files in `fileList`.
virtual std::optional>>
@@ -76,20 +63,15 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
LogicalResult handleBitcodeFile(llvm::Module &module) override;
protected:
- /// Adds `oclc` control variables to the LLVM module.
- void addControlVariables(llvm::Module &module, AMDGCNLibraries libs,
- bool wave64, bool daz, bool finiteOnly,
- bool unsafeMath, bool fastMath, bool correctSqrt,
- StringRef abiVer);
+ /// Appends the paths of common ROCm device libraries to `libs`.
+ LogicalResult getCommonBitcodeLibs(llvm::SmallVector &libs,
+ SmallVector &libPath,
+ StringRef isaVersion);
- /// Compiles assembly to a binary.
- virtual std::optional>
- compileToBinary(const std::string &serializedISA);
-
- /// Default implementation of `ModuleToObject::moduleToObject`.
- std::optional>
- moduleToObjectImpl(const gpu::TargetOptions &targetOptions,
- llvm::Module &llvmModule);
+ /// Adds `oclc` control variables to the LLVM module.
+ void addControlVariables(llvm::Module &module, bool wave64, bool daz,
+ bool finiteOnly, bool unsafeMath, bool fastMath,
+ bool correctSqrt, StringRef abiVer);
/// Returns the assembled ISA.
std::optional> assembleIsa(StringRef isa);
@@ -102,9 +84,6 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
/// List of LLVM bitcode files to link to.
SmallVector fileList;
-
- /// AMD GCN libraries to use when linking, the default is using none.
- AMDGCNLibraries deviceLibs = AMDGCNLibraries::None;
};
} // namespace ROCDL
} // namespace mlir
diff --git a/mlir/lib/Dialect/GPU/CMakeLists.txt
b/mlir/lib/Dialect/GPU/CMakeLists.txt
index 08c8aea36fac9..61ab298ebfb98 100644
--- a/mlir/lib/Dialect/GPU/CMakeLists.txt
+++ b/mlir/lib/Dialect/GPU/CMakeLists.txt
@@ -106,7 +106,7 @@ if(MLIR_ENABLE_ROCM_CONVERSIONS)
"Building mlir with ROCm support requires the AMDGPU backend")
endif()
- set(DEFAULT_ROCM_PATH "" CACHE PATH "Fallback path to search for ROCm
installs")
+ set(DEFAULT_ROCM_PATH "/opt/rocm" CACHE PATH "Fallback path to search for
ROCm installs")
target_compile_definitions(obj.MLIRGPUTransforms
PRIVATE
__DEFAULT_ROCM_PATH__="${DEFAULT_ROCM_PATH}"
diff --git a/mlir/lib/Target/LLVM/CMakeLists.txt
b/mlir/lib/Target/LLVM/CMakeLists.txt
index 4393ff1775ef9..5a3fa160850b4 100644
--- a/mlir/lib/Target/LLVM/CMakeLists.txt
+++ b/mlir/lib/Target/LLVM/CMakeLists.txt
@@ -123,12 +123,17 @@ add_mlir_dialect_library(MLIRROCDLTarget
)
if(MLIR_ENABLE_ROCM_CONVERSIONS)
+ if (NOT ("AMDGPU" IN_LIST LLVM_TARGETS_TO_BUILD))
+message(SEND_ERROR
+ "Building mlir with ROCm support requires the AMDGPU backend")
+ endif()
+
if (DEFINED ROCM_PATH)
set(DEFAULT_ROCM_PATH "${ROCM_PATH}" CACHE PATH "Fallback path
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,28 @@
+// RUN: %clang_cc1 -triple arm64-apple-ios -fptrauth-calls
-fptrauth-intrinsics -emit-llvm %s -o - | FileCheck -check-prefix=CHECK %s
+
+void test_call();
+
+// CHECK-LABEL: define void @test_direct_call()
+void test_direct_call() {
kovdan01 wrote:
Are indirect calls tested? I see
clang/test/CodeGen/ptrauth-function-lvalue-cast.c testing indirect call via
`char *ptr` casted to desired function type, but I don't see a test when we
just have a signed function pointer (for example, passed by an argument to a
function) and call the function that way. Please let me know if I miss smth.
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,13 @@ +// RUN: %clang_cc1 -triple arm64-apple-ios-emit-llvm %s -o - | FileCheck %s --check-prefixes=ALL,OFF kovdan01 wrote: All the tests introduced in the PR seem not Apple-specific - so, it would be nice if you also add RUN lines for a triple like `aarch64-elf`. Manually specifying `-fptrauth-calls` will be needed since we now don't have an equivalent of Apple's arm64e which enabled desired functionality automatically. This applies to all the tests. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,10 @@ +// RUN: %clang_cc1 -triple arm64-apple-ios -fptrauth-calls -fptrauth-intrinsics -emit-llvm %s -o - | FileCheck %s kovdan01 wrote: Can we move this test file contents to clang/test/CodeGen/ptrauth-function.c? I'm happy with both variants, but now as for me we have many small test files, and combining some of them would not harm readability IMHO since they'll still be small even after concatenating. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -128,12 +136,12 @@ class CGCallee {
static CGCallee forDirect(llvm::Constant *functionPtr,
const CGCalleeInfo &abstractInfo = CGCalleeInfo())
{
-return CGCallee(abstractInfo, functionPtr);
+return CGCallee(abstractInfo, functionPtr, CGPointerAuthInfo());
kovdan01 wrote:
Even though the parameter is going to be mandatory later (as said in
https://github.com/llvm/llvm-project/pull/93906/files#r1636843492), now it's
not, so right now there is probably no reason for this change and a similar one
below.
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -61,3 +79,28 @@ CodeGen::getConstantSignedPointer(CodeGenModule &CGM,
llvm::Constant *Pointer,
return CGM.getConstantSignedPointer(Pointer, Key, StorageAddress,
OtherDiscriminator);
}
+
+/// If applicable, sign a given constant function pointer with the ABI rules
for
+/// functionType.
+llvm::Constant *CodeGenModule::getFunctionPointer(llvm::Constant *Pointer,
+ QualType FunctionType,
+ GlobalDecl GD) {
kovdan01 wrote:
Is `GD` used in the function?
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 requested changes to this pull request. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,9 @@ +// RUN: %clang -target arm64-apple-macosx -fptrauth-calls -c %s -### 2>&1 | FileCheck %s --check-prefix PTRAUTH_CALLS kovdan01 wrote: We already have these tests in aarch64-ptrauth.c. If ptrauth.c is a better name for them - consider just renaming the old test file (probably in a separate PR since this one is clang codegen-related) instead of introducing a new one with duplicating checks. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -61,3 +79,28 @@ CodeGen::getConstantSignedPointer(CodeGenModule &CGM, llvm::Constant *Pointer, return CGM.getConstantSignedPointer(Pointer, Key, StorageAddress, OtherDiscriminator); } + +/// If applicable, sign a given constant function pointer with the ABI rules for kovdan01 wrote: Do we need this comments? In CodeGenModule.h, we already have more detailed description. It looks like that other member functions of `CodeGenModule` only contain description in the header, but not in the sources, and there is probably no reason for such duplication when adding new member functions. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -104,10 +109,13 @@ class CGCallee {
/// Construct a callee. Call this constructor directly when this
/// isn't a direct call.
- CGCallee(const CGCalleeInfo &abstractInfo, llvm::Value *functionPtr)
+ CGCallee(
+ const CGCalleeInfo &abstractInfo, llvm::Value *functionPtr,
+ const CGPointerAuthInfo &pointerAuthInfo = /*FIXME*/ CGPointerAuthInfo())
kovdan01 wrote:
Thanks for explanation. Do I get it correct that you'll make the parameter
mandatory in a subsequent PR since this would require small changes in many
places and you don't want to clutter this PR with them?
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -3041,3 +3046,20 @@ llvm::Value
*CodeGenFunction::emitBoolVecConversion(llvm::Value *SrcVec,
return Builder.CreateShuffleVector(SrcVec, ShuffleMask, Name);
}
+
+void CodeGenFunction::EmitPointerAuthOperandBundle(
+const CGPointerAuthInfo &PointerAuth,
+SmallVectorImpl &Bundles) {
+ if (!PointerAuth.isSigned())
+return;
+
+ auto Key = Builder.getInt32(PointerAuth.getKey());
+
+ llvm::Value *Discriminator = PointerAuth.getDiscriminator();
+ if (!Discriminator) {
kovdan01 wrote:
nit: omit braces for simple if statements
https://llvm.org/docs/CodingStandards.html#don-t-use-braces-on-simple-single-statement-bodies-of-if-else-loop-statements
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -938,6 +939,32 @@ class CodeGenModule : public CodeGenTypeCache {
// Return the function body address of the given function.
llvm::Constant *GetFunctionStart(const ValueDecl *Decl);
+ /// Return a function pointer for a reference to the given function.
+ /// This correctly handles weak references, but does not apply a
+ /// pointer signature.
+ llvm::Constant *getRawFunctionPointer(GlobalDecl GD,
+llvm::Type *Ty = nullptr);
+
+ /// Return the ABI-correct function pointer value for a reference
+ /// to the given function. This will apply a pointer signature if
+ /// necessary, caching the result for the given function.
+ llvm::Constant *getFunctionPointer(GlobalDecl GD, llvm::Type *Ty = nullptr);
+
+ /// Return the ABI-correct function pointer value for a reference
+ /// to the given function. This will apply a pointer signature if
+ /// necessary, but will only cache the result if \p FD is passed.
kovdan01 wrote:
> but will only cache the result if \p FD is passed.
You probably misspelled GD?
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -351,6 +351,9 @@ def err_drv_omp_host_ir_file_not_found : Error< "target regions but cannot be found">; def err_drv_omp_host_target_not_supported : Error< "target '%0' is not a supported OpenMP host target">; +def err_drv_ptrauth_not_supported : Error< + "target '%0' does not support native pointer authentication">; kovdan01 wrote: This was marked as resolved, but still looks untested. Do I miss smth? I suppose that on driver side, it's not a problem to have a test case for unsupported target (while I agree that for backend testing the same requires cluttering tests unrelated target). https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -61,3 +79,28 @@ CodeGen::getConstantSignedPointer(CodeGenModule &CGM,
llvm::Constant *Pointer,
return CGM.getConstantSignedPointer(Pointer, Key, StorageAddress,
OtherDiscriminator);
}
+
+/// If applicable, sign a given constant function pointer with the ABI rules
for
+/// functionType.
+llvm::Constant *CodeGenModule::getFunctionPointer(llvm::Constant *Pointer,
+ QualType FunctionType,
+ GlobalDecl GD) {
+ assert(FunctionType->isFunctionType() ||
+ FunctionType->isFunctionReferenceType() ||
+ FunctionType->isFunctionPointerType());
+
+ if (auto PointerAuth = getFunctionPointerAuthInfo(FunctionType)) {
+return getConstantSignedPointer(
+ Pointer, PointerAuth.getKey(), nullptr,
kovdan01 wrote:
nit
```suggestion
Pointer, PointerAuth.getKey(), /*StorageAddress=*/nullptr,
```
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,5 @@
+// RUN: %clang_cc1 -triple arm64e-apple-ios -fptrauth-calls %s -verify
-emit-llvm -o -
+
+void f(void);
+
+int *pf = (int *)&f + 1; // expected-error{{cannot compile this static
initializer yet}}
kovdan01 wrote:
Feel free to ignore: it's probably worth having a more specific error message
here. It's not actually obvious why we are erroring out - is this because
signing function pointers with non-zero offsets is not supported? Even if we
leave this error message as is, it's probably worth at least having a comment
here (and in clang/test/CodeGen/ptrauth-function-init.c as wel)
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -0,0 +1,13 @@ +// RUN: %clang_cc1 -triple arm64-apple-ios-emit-llvm %s -o - | FileCheck %s --check-prefixes=ALL,OFF +// RUN: %clang_cc1 -triple arm64e-apple-ios -emit-llvm %s -o - | FileCheck %s --check-prefixes=ALL,OFF + +// RUN: %clang_cc1 -triple arm64-apple-ios -fptrauth-calls -emit-llvm %s -o - | FileCheck %s --check-prefixes=ALL,CALLS kovdan01 wrote: I'm OK with any Apple-specific RUN lines if you really want to have all of them, but this one and the next one seem to duplicate each other. I suppose there should be a separate test which checks that arm64e target enabled desired ptrauth flags, and then you can just use either arm64e or arm64 with explicitly specified flags in RUN lines in codegen tests. Feel free to ignore - everything Apple-related is not what I'm going to debate :) https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 requested changes to this pull request. https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -938,6 +939,32 @@ class CodeGenModule : public CodeGenTypeCache {
// Return the function body address of the given function.
llvm::Constant *GetFunctionStart(const ValueDecl *Decl);
+ /// Return a function pointer for a reference to the given function.
+ /// This correctly handles weak references, but does not apply a
+ /// pointer signature.
+ llvm::Constant *getRawFunctionPointer(GlobalDecl GD,
+llvm::Type *Ty = nullptr);
+
+ /// Return the ABI-correct function pointer value for a reference
+ /// to the given function. This will apply a pointer signature if
+ /// necessary, caching the result for the given function.
+ llvm::Constant *getFunctionPointer(GlobalDecl GD, llvm::Type *Ty = nullptr);
+
+ /// Return the ABI-correct function pointer value for a reference
+ /// to the given function. This will apply a pointer signature if
+ /// necessary, but will only cache the result if \p FD is passed.
+ llvm::Constant *getFunctionPointer(llvm::Constant *Pointer,
+ QualType FunctionType,
+ GlobalDecl GD = GlobalDecl());
+
+ CGPointerAuthInfo getFunctionPointerAuthInfo(QualType T);
+
+ llvm::Constant *getConstantSignedPointer(llvm::Constant *Pointer,
kovdan01 wrote:
This member function declaration seems to be unused and not corresponding to
any definition, so it can probably be removed.
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
@@ -3041,3 +3046,20 @@ llvm::Value
*CodeGenFunction::emitBoolVecConversion(llvm::Value *SrcVec,
return Builder.CreateShuffleVector(SrcVec, ShuffleMask, Name);
}
+
+void CodeGenFunction::EmitPointerAuthOperandBundle(
+const CGPointerAuthInfo &PointerAuth,
+SmallVectorImpl &Bundles) {
+ if (!PointerAuth.isSigned())
+return;
+
+ auto Key = Builder.getInt32(PointerAuth.getKey());
kovdan01 wrote:
```suggestion
auto *Key = Builder.getInt32(PointerAuth.getKey());
```
https://github.com/llvm/llvm-project/pull/93906
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
@@ -394,7 +400,7 @@ createFlowFunction(const
BinaryFunction::BasicBlockOrderType &BlockOrder) {
void matchWeightsByHashes(BinaryContext &BC,
const BinaryFunction::BasicBlockOrderType
&BlockOrder,
const yaml::bolt::BinaryFunctionProfile &YamlBF,
- FlowFunction &Func) {
+ FlowFunction &Func, uint64_t &MatchedBlocksCount) {
shawbyoung wrote:
Sounds good
https://github.com/llvm/llvm-project/pull/95156
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95156
>From aa441dc0163d3d0f63de1e4dd1fa359180f82f1f Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Tue, 11 Jun 2024 11:43:13 -0700
Subject: [PATCH 01/13] Summary: Functions with little exact matching
Created using spr 1.3.4
---
bolt/docs/CommandLineArgumentReference.md | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/bolt/docs/CommandLineArgumentReference.md
b/bolt/docs/CommandLineArgumentReference.md
index 8887d1f5d5bd4..bdc1d9dfd735c 100644
--- a/bolt/docs/CommandLineArgumentReference.md
+++ b/bolt/docs/CommandLineArgumentReference.md
@@ -614,6 +614,17 @@
- `--lite-threshold-pct=`
+ Threshold (in percent) of matched profile at which stale profile inference is
+ applied to functions. Argument corresponds to the sum of matched execution
+ counts of function blocks divided by the sum of execution counts of function
+ blocks. E.g if the sum of a function blocks' execution counts is 100, the sum
+ of the function blocks' matched execution counts is 10, and the argument is
15
+ (15%), profile inference will not be applied to that function. A higher
+ threshold will correlate with fewer functions to process in cases of stale
+ profile. Default set to %5.
+
+- `--matched-profile-threshold=`
+
Threshold (in percent) for selecting functions to process in lite mode.
Higher
threshold means fewer functions to process. E.g threshold of 90 means only
top
10 percent of functions with profile will be processed.
@@ -1161,4 +1172,4 @@
- `--print-options`
- Print non-default options after command line parsing
\ No newline at end of file
+ Print non-default options after command line parsing
>From 46fa37a054a129ca36e7b6ae126273e40fddea98 Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:32:40 -0700
Subject: [PATCH 02/13] Update SampleProfileInference.h
---
llvm/include/llvm/Transforms/Utils/SampleProfileInference.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
index c654715c0ae9f..9ccbd0fa88f3d 100644
--- a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
+++ b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
@@ -58,6 +58,7 @@ struct FlowFunction {
std::vector Jumps;
/// The index of the entry block.
uint64_t Entry{0};
+ uint64_t Sink{UINT64_MAX};
// Matched execution count for the function.
uint64_t MatchedExecCount{0};
};
>From d532514257feb5e86232e76c437c99a41d5f2cea Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:39:28 -0700
Subject: [PATCH 03/13] Update StaleProfileMatching.cpp
---
bolt/lib/Profile/StaleProfileMatching.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 41afa6b4bbb19..47335163263a4 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -604,8 +604,8 @@ bool canApplyInference(const FlowFunction &Func,
if (Func.Blocks.size() > opts::StaleMatchingMaxFuncSize)
return false;
- if (Func.MatchedExecCount / YamlBF.ExecCount >=
- opts::MatchedProfileThreshold / 100)
+ if ((double)Func.MatchedExecCount / YamlBF.ExecCount >=
+ opts::MatchedProfileThreshold / 100.0)
return false;
bool HasExitBlocks = llvm::any_of(
>From 3fc6d72d866333d8ce964fdfaa748791d4f8d2b4 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Fri, 14 Jun 2024 08:38:19 -0700
Subject: [PATCH 04/13] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/StaleProfileMatching.cpp | 37 +++
.../Transforms/Utils/SampleProfileInference.h | 3 --
2 files changed, 29 insertions(+), 11 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 47335163263a4..cb356afdd2948 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -53,9 +53,9 @@ cl::opt
cl::opt MatchedProfileThreshold(
"matched-profile-threshold",
-cl::desc("Percentage threshold of matched execution counts at which stale "
+cl::desc("Percentage threshold of matched basic blocks at which stale "
"profile inference is executed."),
-cl::init(5), cl::Hidden, cl::cat(BoltOptCategory));
+cl::init(0), cl::Hidden, cl::cat(BoltOptCategory));
cl::opt StaleMatchingMaxFuncSize(
"stale-matching-max-func-size",
@@ -186,6 +186,17 @@ struct BlendedBlockHash {
uint8_t SuccHash{0};
};
+/// A data object containing function matching information.
+struct FunctionMatchingData {
+public:
+ /// The number of blocks matched exactly.
+ uint64_t MatchedExactBlocks{0};
+ /// The number of b
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
steakhal wrote: Ping. @NagyDonat @Xazax-hun @haoNoQ https://github.com/llvm/llvm-project/pull/95129 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
@@ -44,21 +47,43 @@ class Z3CrosscheckVisitor final : public BugReporterVisitor
{
/// Holds the constraints in a given path.
ConstraintMap Constraints;
Z3Result &Result;
+ const AnalyzerOptions &Opts;
};
/// The oracle will decide if a report should be accepted or rejected based on
-/// the results of the Z3 solver.
+/// the results of the Z3 solver and the statistics of the queries of a report
+/// equivalenece class.
class Z3CrosscheckOracle {
public:
+ explicit Z3CrosscheckOracle(const AnalyzerOptions &Opts) : Opts(Opts) {}
+
enum Z3Decision {
-AcceptReport, // The report was SAT.
-RejectReport, // The report was UNSAT or UNDEF.
+AcceptReport, // The report was SAT.
+RejectReport, // The report was UNSAT or UNDEF.
+RejectEQClass, // The heuristic suggests to skip the current eqclass.
};
- /// Makes a decision for accepting or rejecting the report based on the
- /// result of the corresponding Z3 query.
- static Z3Decision
- interpretQueryResult(const Z3CrosscheckVisitor::Z3Result &Query);
+ /// Updates the internal state with the new Z3Result and makes a decision how
+ /// to proceed:
+ /// - Accept the report if the Z3Result was SAT.
+ /// - Suggest dropping the report equvalence class based on the accumulated
+ /// statistics.
+ /// - Otherwise, reject the report if the Z3Result was UNSAT or UNDEF.
+ ///
+ /// Conditions for dropping the equivalence class:
+ /// - Accumulative time spent in Z3 checks is more than 700ms in the eqclass.
+ /// - Hit the 300ms query timeout in the report eqclass.
+ /// - Hit the 400'000 rlimit in the report eqclass.
+ ///
+ /// Refer to
+ /// https://discourse.llvm.org/t/analyzer-rfc-taming-z3-query-times/79520 to
+ /// see why this heuristic was chosen.
+ Z3Decision interpretQueryResult(const Z3CrosscheckVisitor::Z3Result &Meta);
+
+private:
+ const AnalyzerOptions &Opts;
+ unsigned NumZ3QueriesDoneInEqClass = 0;
steakhal wrote:
Hmm, I dont think I use this anymore.
https://github.com/llvm/llvm-project/pull/95129
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/aaupov approved this pull request. LGTM with one nit https://github.com/llvm/llvm-project/pull/95156 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
@@ -391,10 +397,9 @@ createFlowFunction(const
BinaryFunction::BasicBlockOrderType &BlockOrder) {
/// of the basic blocks in the binary, the count is "matched" to the block.
/// Similarly, if both the source and the target of a count in the profile are
/// matched to a jump in the binary, the count is recorded in CFG.
-void matchWeightsByHashes(BinaryContext &BC,
- const BinaryFunction::BasicBlockOrderType
&BlockOrder,
- const yaml::bolt::BinaryFunctionProfile &YamlBF,
- FlowFunction &Func) {
+uint64_t matchWeightsByHashes(
aaupov wrote:
```suggestion
size_t matchWeightsByHashes(
```
https://github.com/llvm/llvm-project/pull/95156
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/aaupov edited https://github.com/llvm/llvm-project/pull/95156 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95156
>From aa441dc0163d3d0f63de1e4dd1fa359180f82f1f Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Tue, 11 Jun 2024 11:43:13 -0700
Subject: [PATCH 01/14] Summary: Functions with little exact matching
Created using spr 1.3.4
---
bolt/docs/CommandLineArgumentReference.md | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/bolt/docs/CommandLineArgumentReference.md
b/bolt/docs/CommandLineArgumentReference.md
index 8887d1f5d5bd4..bdc1d9dfd735c 100644
--- a/bolt/docs/CommandLineArgumentReference.md
+++ b/bolt/docs/CommandLineArgumentReference.md
@@ -614,6 +614,17 @@
- `--lite-threshold-pct=`
+ Threshold (in percent) of matched profile at which stale profile inference is
+ applied to functions. Argument corresponds to the sum of matched execution
+ counts of function blocks divided by the sum of execution counts of function
+ blocks. E.g if the sum of a function blocks' execution counts is 100, the sum
+ of the function blocks' matched execution counts is 10, and the argument is
15
+ (15%), profile inference will not be applied to that function. A higher
+ threshold will correlate with fewer functions to process in cases of stale
+ profile. Default set to %5.
+
+- `--matched-profile-threshold=`
+
Threshold (in percent) for selecting functions to process in lite mode.
Higher
threshold means fewer functions to process. E.g threshold of 90 means only
top
10 percent of functions with profile will be processed.
@@ -1161,4 +1172,4 @@
- `--print-options`
- Print non-default options after command line parsing
\ No newline at end of file
+ Print non-default options after command line parsing
>From 46fa37a054a129ca36e7b6ae126273e40fddea98 Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:32:40 -0700
Subject: [PATCH 02/14] Update SampleProfileInference.h
---
llvm/include/llvm/Transforms/Utils/SampleProfileInference.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
index c654715c0ae9f..9ccbd0fa88f3d 100644
--- a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
+++ b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
@@ -58,6 +58,7 @@ struct FlowFunction {
std::vector Jumps;
/// The index of the entry block.
uint64_t Entry{0};
+ uint64_t Sink{UINT64_MAX};
// Matched execution count for the function.
uint64_t MatchedExecCount{0};
};
>From d532514257feb5e86232e76c437c99a41d5f2cea Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:39:28 -0700
Subject: [PATCH 03/14] Update StaleProfileMatching.cpp
---
bolt/lib/Profile/StaleProfileMatching.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 41afa6b4bbb19..47335163263a4 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -604,8 +604,8 @@ bool canApplyInference(const FlowFunction &Func,
if (Func.Blocks.size() > opts::StaleMatchingMaxFuncSize)
return false;
- if (Func.MatchedExecCount / YamlBF.ExecCount >=
- opts::MatchedProfileThreshold / 100)
+ if ((double)Func.MatchedExecCount / YamlBF.ExecCount >=
+ opts::MatchedProfileThreshold / 100.0)
return false;
bool HasExitBlocks = llvm::any_of(
>From 3fc6d72d866333d8ce964fdfaa748791d4f8d2b4 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Fri, 14 Jun 2024 08:38:19 -0700
Subject: [PATCH 04/14] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/StaleProfileMatching.cpp | 37 +++
.../Transforms/Utils/SampleProfileInference.h | 3 --
2 files changed, 29 insertions(+), 11 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 47335163263a4..cb356afdd2948 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -53,9 +53,9 @@ cl::opt
cl::opt MatchedProfileThreshold(
"matched-profile-threshold",
-cl::desc("Percentage threshold of matched execution counts at which stale "
+cl::desc("Percentage threshold of matched basic blocks at which stale "
"profile inference is executed."),
-cl::init(5), cl::Hidden, cl::cat(BoltOptCategory));
+cl::init(0), cl::Hidden, cl::cat(BoltOptCategory));
cl::opt StaleMatchingMaxFuncSize(
"stale-matching-max-func-size",
@@ -186,6 +186,17 @@ struct BlendedBlockHash {
uint8_t SuccHash{0};
};
+/// A data object containing function matching information.
+struct FunctionMatchingData {
+public:
+ /// The number of blocks matched exactly.
+ uint64_t MatchedExactBlocks{0};
+ /// The number of b
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95156
>From aa441dc0163d3d0f63de1e4dd1fa359180f82f1f Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Tue, 11 Jun 2024 11:43:13 -0700
Subject: [PATCH 01/15] Summary: Functions with little exact matching
Created using spr 1.3.4
---
bolt/docs/CommandLineArgumentReference.md | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/bolt/docs/CommandLineArgumentReference.md
b/bolt/docs/CommandLineArgumentReference.md
index 8887d1f5d5bd4..bdc1d9dfd735c 100644
--- a/bolt/docs/CommandLineArgumentReference.md
+++ b/bolt/docs/CommandLineArgumentReference.md
@@ -614,6 +614,17 @@
- `--lite-threshold-pct=`
+ Threshold (in percent) of matched profile at which stale profile inference is
+ applied to functions. Argument corresponds to the sum of matched execution
+ counts of function blocks divided by the sum of execution counts of function
+ blocks. E.g if the sum of a function blocks' execution counts is 100, the sum
+ of the function blocks' matched execution counts is 10, and the argument is
15
+ (15%), profile inference will not be applied to that function. A higher
+ threshold will correlate with fewer functions to process in cases of stale
+ profile. Default set to %5.
+
+- `--matched-profile-threshold=`
+
Threshold (in percent) for selecting functions to process in lite mode.
Higher
threshold means fewer functions to process. E.g threshold of 90 means only
top
10 percent of functions with profile will be processed.
@@ -1161,4 +1172,4 @@
- `--print-options`
- Print non-default options after command line parsing
\ No newline at end of file
+ Print non-default options after command line parsing
>From 46fa37a054a129ca36e7b6ae126273e40fddea98 Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:32:40 -0700
Subject: [PATCH 02/15] Update SampleProfileInference.h
---
llvm/include/llvm/Transforms/Utils/SampleProfileInference.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
index c654715c0ae9f..9ccbd0fa88f3d 100644
--- a/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
+++ b/llvm/include/llvm/Transforms/Utils/SampleProfileInference.h
@@ -58,6 +58,7 @@ struct FlowFunction {
std::vector Jumps;
/// The index of the entry block.
uint64_t Entry{0};
+ uint64_t Sink{UINT64_MAX};
// Matched execution count for the function.
uint64_t MatchedExecCount{0};
};
>From d532514257feb5e86232e76c437c99a41d5f2cea Mon Sep 17 00:00:00 2001
From: shaw young <58664393+shawbyo...@users.noreply.github.com>
Date: Tue, 11 Jun 2024 14:39:28 -0700
Subject: [PATCH 03/15] Update StaleProfileMatching.cpp
---
bolt/lib/Profile/StaleProfileMatching.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 41afa6b4bbb19..47335163263a4 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -604,8 +604,8 @@ bool canApplyInference(const FlowFunction &Func,
if (Func.Blocks.size() > opts::StaleMatchingMaxFuncSize)
return false;
- if (Func.MatchedExecCount / YamlBF.ExecCount >=
- opts::MatchedProfileThreshold / 100)
+ if ((double)Func.MatchedExecCount / YamlBF.ExecCount >=
+ opts::MatchedProfileThreshold / 100.0)
return false;
bool HasExitBlocks = llvm::any_of(
>From 3fc6d72d866333d8ce964fdfaa748791d4f8d2b4 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Fri, 14 Jun 2024 08:38:19 -0700
Subject: [PATCH 04/15] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/StaleProfileMatching.cpp | 37 +++
.../Transforms/Utils/SampleProfileInference.h | 3 --
2 files changed, 29 insertions(+), 11 deletions(-)
diff --git a/bolt/lib/Profile/StaleProfileMatching.cpp
b/bolt/lib/Profile/StaleProfileMatching.cpp
index 47335163263a4..cb356afdd2948 100644
--- a/bolt/lib/Profile/StaleProfileMatching.cpp
+++ b/bolt/lib/Profile/StaleProfileMatching.cpp
@@ -53,9 +53,9 @@ cl::opt
cl::opt MatchedProfileThreshold(
"matched-profile-threshold",
-cl::desc("Percentage threshold of matched execution counts at which stale "
+cl::desc("Percentage threshold of matched basic blocks at which stale "
"profile inference is executed."),
-cl::init(5), cl::Hidden, cl::cat(BoltOptCategory));
+cl::init(0), cl::Hidden, cl::cat(BoltOptCategory));
cl::opt StaleMatchingMaxFuncSize(
"stale-matching-max-func-size",
@@ -186,6 +186,17 @@ struct BlendedBlockHash {
uint8_t SuccHash{0};
};
+/// A data object containing function matching information.
+struct FunctionMatchingData {
+public:
+ /// The number of blocks matched exactly.
+ uint64_t MatchedExactBlocks{0};
+ /// The number of b
[llvm-branch-commits] [llvm] AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (PR #95591)
@@ -1608,14 +1598,14 @@ defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMAX",
"int_amdgcn_flat_atomic_fmax
}
let OtherPredicates = [isGFX10Only] in {
-defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_FMIN_X2",
"atomic_load_fmin_global", f64>;
-defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_FMAX_X2",
"atomic_load_fmax_global", f64>;
-defm : GlobalFLATAtomicIntrPats <"GLOBAL_ATOMIC_FMIN_X2",
"int_amdgcn_global_atomic_fmin", f64>;
-defm : GlobalFLATAtomicIntrPats <"GLOBAL_ATOMIC_FMAX_X2",
"int_amdgcn_global_atomic_fmax", f64>;
-defm : FlatSignedAtomicPat <"FLAT_ATOMIC_FMIN_X2", "atomic_load_fmin_flat",
f64>;
-defm : FlatSignedAtomicPat <"FLAT_ATOMIC_FMAX_X2", "atomic_load_fmax_flat",
f64>;
-defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMIN_X2",
"int_amdgcn_flat_atomic_fmin", f64>;
-defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMAX_X2",
"int_amdgcn_flat_atomic_fmax", f64>;
+defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_MIN_F64",
"atomic_load_fmin_global", f64>;
arsenm wrote:
Yes, there's a lot of predicate junk here that needs to be cleaned up. Most of
these atomics are imprecisely guarded with isGFX10/11/12 checks that are
available on a subset of gfx9
https://github.com/llvm/llvm-project/pull/95591
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
https://github.com/NagyDonat edited https://github.com/llvm/llvm-project/pull/95129 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
https://github.com/NagyDonat approved this pull request. Oops, I vaguely recalled that I already reviewed this, but it turns out I didn't. Anyway, the change LGTM, I only added a few minor remarks. (We already discussed a high-level overview of this on Discourse.) https://github.com/llvm/llvm-project/pull/95129 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
@@ -101,18 +123,37 @@ void Z3CrosscheckVisitor::Profile(llvm::FoldingSetNodeID
&ID) const {
Z3CrosscheckOracle::Z3Decision Z3CrosscheckOracle::interpretQueryResult(
const Z3CrosscheckVisitor::Z3Result &Query) {
++NumZ3QueriesDone;
+ ++NumZ3QueriesDoneInEqClass;
+ AccumulatedZ3QueryTimeInEqClass += Query.Z3QueryTimeMilliseconds;
- if (!Query.IsSAT.has_value()) {
-// For backward compatibility, let's accept the first timeout.
+ if (Query.IsSAT && Query.IsSAT.value()) {
+++NumTimesZ3QueryAcceptsReport;
+return AcceptReport; // sat
NagyDonat wrote:
```suggestion
return AcceptReport;
```
This old comment doesn't add anything + the "unset" on the other branch was
slightly incorrect.
https://github.com/llvm/llvm-project/pull/95129
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
@@ -101,18 +123,37 @@ void Z3CrosscheckVisitor::Profile(llvm::FoldingSetNodeID
&ID) const {
Z3CrosscheckOracle::Z3Decision Z3CrosscheckOracle::interpretQueryResult(
const Z3CrosscheckVisitor::Z3Result &Query) {
++NumZ3QueriesDone;
+ ++NumZ3QueriesDoneInEqClass;
+ AccumulatedZ3QueryTimeInEqClass += Query.Z3QueryTimeMilliseconds;
- if (!Query.IsSAT.has_value()) {
-// For backward compatibility, let's accept the first timeout.
+ if (Query.IsSAT && Query.IsSAT.value()) {
+++NumTimesZ3QueryAcceptsReport;
+return AcceptReport; // sat
+ }
+
+ // Suggest cutting the EQClass if certain heuristics trigger.
+ if (Opts.Z3CrosscheckTimeoutThreshold &&
+ Query.Z3QueryTimeMilliseconds >= Opts.Z3CrosscheckTimeoutThreshold) {
++NumTimesZ3TimedOut;
-return AcceptReport;
+++NumTimesZ3QueryRejectEQClass;
+return RejectEQClass;
}
- if (Query.IsSAT.value()) {
-++NumTimesZ3QueryAcceptsReport;
-return AcceptReport; // sat
+ if (Opts.Z3CrosscheckRLimitThreshold &&
+ Query.UsedRLimit >= Opts.Z3CrosscheckRLimitThreshold) {
+++NumTimesZ3ExhaustedRLimit;
+++NumTimesZ3QueryRejectEQClass;
+return RejectEQClass;
+ }
+
+ if (AccumulatedZ3QueryTimeInEqClass > 700 /*ms*/) {
NagyDonat wrote:
It's counter-intuitive that this limit is hardcoded, while the others are not
-- e.g. this will silently override timeout thresholds above 700 ms.
Obviously this is bikeshedding, because practically nobody will use these
config options, but if we want to make this configurable, then we should make
the config clear and user-friendly. Either introduce this limit as a third
configurable value, or e.g. say that it's always twice the individual timeout
(which would yield 600 ms instead of 700 ms).
https://github.com/llvm/llvm-project/pull/95129
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
@@ -17,43 +18,126 @@ using Z3Decision = Z3CrosscheckOracle::Z3Decision;
static constexpr Z3Decision AcceptReport = Z3Decision::AcceptReport;
static constexpr Z3Decision RejectReport = Z3Decision::RejectReport;
+static constexpr Z3Decision RejectEQClass = Z3Decision::RejectEQClass;
static constexpr std::optional SAT = true;
static constexpr std::optional UNSAT = false;
static constexpr std::optional UNDEF = std::nullopt;
+static unsigned operator""_ms(unsigned long long ms) { return ms; }
+static unsigned operator""_step(unsigned long long rlimit) { return rlimit; }
+
+static const AnalyzerOptions DefaultOpts = [] {
+ AnalyzerOptions Config;
+#define ANALYZER_OPTION_DEPENDS_ON_USER_MODE(TYPE, NAME, CMDFLAG, DESC,
\
+ SHALLOW_VAL, DEEP_VAL)
\
+ ANALYZER_OPTION(TYPE, NAME, CMDFLAG, DESC, DEEP_VAL)
+#define ANALYZER_OPTION(TYPE, NAME, CMDFLAG, DESC, DEFAULT_VAL)
\
+ Config.NAME = DEFAULT_VAL;
+#include "clang/StaticAnalyzer/Core/AnalyzerOptions.def"
+
+ // Remember to update the tests in this file when these values change.
+ // Also update the doc comment of `interpretQueryResult`.
+ assert(Config.Z3CrosscheckRLimitThreshold == 400'000);
+ assert(Config.Z3CrosscheckTimeoutThreshold == 300 /*ms*/);
+ // Usually, when the timeout/rlimit threshold is reached, Z3 only slightly
+ // overshoots until it realizes that it overshoot and needs to back off.
+ // Consequently, the measured timeout should be fairly close to the
threshold.
+ // Same reasoning applies to the rlimit too.
+ return Config;
+}();
+
namespace {
-struct Z3CrosscheckOracleTest : public testing::Test {
- Z3Decision interpretQueryResult(const Z3Result &Result) const {
-return Z3CrosscheckOracle::interpretQueryResult(Result);
+class Z3CrosscheckOracleTest : public testing::Test {
+public:
+ Z3Decision interpretQueryResult(const Z3Result &Result) {
+return Oracle.interpretQueryResult(Result);
}
+
+private:
+ Z3CrosscheckOracle Oracle = Z3CrosscheckOracle(DefaultOpts);
};
TEST_F(Z3CrosscheckOracleTest, AcceptsFirstSAT) {
- ASSERT_EQ(AcceptReport, interpretQueryResult({SAT}));
+ ASSERT_EQ(AcceptReport, interpretQueryResult({SAT, 25_ms, 1000_step}));
}
TEST_F(Z3CrosscheckOracleTest, AcceptsSAT) {
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(AcceptReport, interpretQueryResult({SAT}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(AcceptReport, interpretQueryResult({SAT, 25_ms, 1000_step}));
+}
+
+TEST_F(Z3CrosscheckOracleTest, SATWhenItGoesOverTime) {
+ // Even if it times out, if it is SAT, we should accept it.
+ ASSERT_EQ(AcceptReport, interpretQueryResult({SAT, 310_ms, 1000_step}));
}
-TEST_F(Z3CrosscheckOracleTest, AcceptsFirstTimeout) {
- ASSERT_EQ(AcceptReport, interpretQueryResult({UNDEF}));
+TEST_F(Z3CrosscheckOracleTest, UNSATWhenItGoesOverTime) {
+ ASSERT_EQ(RejectEQClass, interpretQueryResult({UNSAT, 310_ms, 1000_step}));
}
-TEST_F(Z3CrosscheckOracleTest, AcceptsTimeout) {
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(AcceptReport, interpretQueryResult({UNDEF}));
+TEST_F(Z3CrosscheckOracleTest, RejectsTimeout) {
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(RejectEQClass, interpretQueryResult({UNDEF, 310_ms, 1000_step}));
}
TEST_F(Z3CrosscheckOracleTest, RejectsUNSATs) {
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
- ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 25_ms, 1000_step}));
+}
+
+// Testing cut heuristics:
+// ===
+
+TEST_F(Z3CrosscheckOracleTest, RejectEQClassIfSpendsTooMuchTotalTime) {
+ // Simulate long queries, that barely doesn't trigger the timeout.
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 290_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 290_ms, 1000_step}));
+ ASSERT_EQ(RejectEQClass, interpretQueryResult({UNSAT, 290_ms, 1000_step}));
+}
+
+TEST_F(Z3CrosscheckOracleTest, SATWhenItSpendsTooMuchTotalTime) {
+ // Simulate long queries, that barely doesn't trigger the timeout.
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 290_ms, 1000_step}));
+ ASSERT_EQ(RejectReport, interpretQueryResult({UNSAT, 290_ms, 1000_step}));
+ ASSERT_EQ(AcceptReport, interpretQueryResult({SAT, 290_ms, 1000_step}));
+}
+
+TEST
[llvm-branch-commits] [analyzer] Harden safeguards for Z3 query times (PR #95129)
@@ -101,18 +123,37 @@ void Z3CrosscheckVisitor::Profile(llvm::FoldingSetNodeID
&ID) const {
Z3CrosscheckOracle::Z3Decision Z3CrosscheckOracle::interpretQueryResult(
const Z3CrosscheckVisitor::Z3Result &Query) {
++NumZ3QueriesDone;
+ ++NumZ3QueriesDoneInEqClass;
+ AccumulatedZ3QueryTimeInEqClass += Query.Z3QueryTimeMilliseconds;
- if (!Query.IsSAT.has_value()) {
-// For backward compatibility, let's accept the first timeout.
+ if (Query.IsSAT && Query.IsSAT.value()) {
+++NumTimesZ3QueryAcceptsReport;
+return AcceptReport; // sat
+ }
+
+ // Suggest cutting the EQClass if certain heuristics trigger.
+ if (Opts.Z3CrosscheckTimeoutThreshold &&
+ Query.Z3QueryTimeMilliseconds >= Opts.Z3CrosscheckTimeoutThreshold) {
++NumTimesZ3TimedOut;
-return AcceptReport;
+++NumTimesZ3QueryRejectEQClass;
+return RejectEQClass;
}
- if (Query.IsSAT.value()) {
-++NumTimesZ3QueryAcceptsReport;
-return AcceptReport; // sat
+ if (Opts.Z3CrosscheckRLimitThreshold &&
+ Query.UsedRLimit >= Opts.Z3CrosscheckRLimitThreshold) {
+++NumTimesZ3ExhaustedRLimit;
+++NumTimesZ3QueryRejectEQClass;
+return RejectEQClass;
+ }
+
+ if (AccumulatedZ3QueryTimeInEqClass > 700 /*ms*/) {
+++NumTimesZ3SpendsTooMuchTimeOnASingleEQClass;
+++NumTimesZ3QueryRejectEQClass;
+return RejectEQClass;
}
+ // If no cutoff heuristics trigger, and the report is "unsat" or "undef",
+ // then reject the report.
++NumTimesZ3QueryRejectReport;
return RejectReport; // unsat
NagyDonat wrote:
```suggestion
return RejectReport;
```
This doesn't add anything and is technically incorrect.
https://github.com/llvm/llvm-project/pull/95129
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (PR #95591)
@@ -1608,14 +1598,14 @@ defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMAX",
"int_amdgcn_flat_atomic_fmax
}
let OtherPredicates = [isGFX10Only] in {
-defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_FMIN_X2",
"atomic_load_fmin_global", f64>;
-defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_FMAX_X2",
"atomic_load_fmax_global", f64>;
-defm : GlobalFLATAtomicIntrPats <"GLOBAL_ATOMIC_FMIN_X2",
"int_amdgcn_global_atomic_fmin", f64>;
-defm : GlobalFLATAtomicIntrPats <"GLOBAL_ATOMIC_FMAX_X2",
"int_amdgcn_global_atomic_fmax", f64>;
-defm : FlatSignedAtomicPat <"FLAT_ATOMIC_FMIN_X2", "atomic_load_fmin_flat",
f64>;
-defm : FlatSignedAtomicPat <"FLAT_ATOMIC_FMAX_X2", "atomic_load_fmax_flat",
f64>;
-defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMIN_X2",
"int_amdgcn_flat_atomic_fmin", f64>;
-defm : FlatSignedAtomicIntrPat <"FLAT_ATOMIC_FMAX_X2",
"int_amdgcn_flat_atomic_fmax", f64>;
+defm : GlobalFLATAtomicPats <"GLOBAL_ATOMIC_MIN_F64",
"atomic_load_fmin_global", f64>;
arsenm wrote:
This specific cleanup requires the next patch in the series to start
decomposing the features to the granularity we need
https://github.com/llvm/llvm-project/pull/95591
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (PR #95591)
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/95591
>From 71287fe3e9b31cf09dfc4a0a47e6bc02e40a1445 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Thu, 13 Jun 2024 18:33:11 +0200
Subject: [PATCH] AMDGPU: Create pseudo to real mapping for flat/buffer atomic
fmin/fmax
The global/flat/buffer atomic fmin/fmax situation is a mess. These
instructions have been renamed 3 times. We currently have
separate pseudos defined for the same opcodes with the different names
(e.g. GLOBAL_ATOMIC_MIN_F64 from gfx90a and GLOBAL_ATOMIC_FMIN_X2 from gfx10).
Use the _FMIN versions as the canonical name for the f32 versions. Use the
_MIN_F64 style as the canonical name for the f64 case. This is because
gfx90a has the most sensible names, but does not have the f32 versions.t sho
Wire through the pseudo to use for the instruction properties vs. the assembly
name like in other cases. This will simplify handling of direct atomicrmw
selection.
This will simplify directly selecting these from atomicrmw.
---
llvm/lib/Target/AMDGPU/AMDGPU.td | 4 +-
llvm/lib/Target/AMDGPU/BUFInstructions.td | 103 +
llvm/lib/Target/AMDGPU/FLATInstructions.td| 107 +-
.../AMDGPU/fp-atomic-to-s_denormmode.mir | 40 +++
4 files changed, 129 insertions(+), 125 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPU.td b/llvm/lib/Target/AMDGPU/AMDGPU.td
index d0d7a9dc17247..0a1550ccb53c4 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPU.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPU.td
@@ -1864,7 +1864,9 @@ def HasFlatAddressSpace :
Predicate<"Subtarget->hasFlatAddressSpace()">,
def HasBufferFlatGlobalAtomicsF64 :
Predicate<"Subtarget->hasBufferFlatGlobalAtomicsF64()">,
- AssemblerPredicate<(any_of FeatureGFX90AInsts)>;
+ // FIXME: This is too coarse, and working around using pseudo's predicates
on real instruction.
+ AssemblerPredicate<(any_of FeatureGFX90AInsts, FeatureGFX10Insts,
FeatureSouthernIslands, FeatureSeaIslands)>;
+
def HasLdsAtomicAddF64 :
Predicate<"Subtarget->hasLdsAtomicAddF64()">,
AssemblerPredicate<(any_of FeatureGFX90AInsts)>;
diff --git a/llvm/lib/Target/AMDGPU/BUFInstructions.td
b/llvm/lib/Target/AMDGPU/BUFInstructions.td
index 43e5434ea2700..9d21f93a957cc 100644
--- a/llvm/lib/Target/AMDGPU/BUFInstructions.td
+++ b/llvm/lib/Target/AMDGPU/BUFInstructions.td
@@ -1163,12 +1163,6 @@ let SubtargetPredicate = isGFX6GFX7GFX10 in {
defm BUFFER_ATOMIC_FCMPSWAP_X2 : MUBUF_Pseudo_Atomics <
"buffer_atomic_fcmpswap_x2", VReg_128, v2f64, null_frag
>;
-defm BUFFER_ATOMIC_FMIN_X2 : MUBUF_Pseudo_Atomics <
- "buffer_atomic_fmin_x2", VReg_64, f64, null_frag
->;
-defm BUFFER_ATOMIC_FMAX_X2 : MUBUF_Pseudo_Atomics <
- "buffer_atomic_fmax_x2", VReg_64, f64, null_frag
->;
}
@@ -1318,6 +1312,9 @@ let SubtargetPredicate = isGFX90APlus in {
let SubtargetPredicate = HasBufferFlatGlobalAtomicsF64 in {
defm BUFFER_ATOMIC_ADD_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_add_f64",
VReg_64, f64>;
+
+ // Note the names can be buffer_atomic_fmin_x2/buffer_atomic_fmax_x2
+ // depending on some subtargets.
defm BUFFER_ATOMIC_MIN_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_min_f64",
VReg_64, f64>;
defm BUFFER_ATOMIC_MAX_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_max_f64",
VReg_64, f64>;
} // End SubtargetPredicate = HasBufferFlatGlobalAtomicsF64
@@ -1763,8 +1760,8 @@ let OtherPredicates = [isGFX6GFX7GFX10Plus] in {
defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f32, "BUFFER_ATOMIC_FMAX">;
}
let SubtargetPredicate = isGFX6GFX7GFX10 in {
- defm : SIBufferAtomicPat<"SIbuffer_atomic_fmin", f64,
"BUFFER_ATOMIC_FMIN_X2">;
- defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f64,
"BUFFER_ATOMIC_FMAX_X2">;
+ defm : SIBufferAtomicPat<"SIbuffer_atomic_fmin", f64,
"BUFFER_ATOMIC_MIN_F64">;
+ defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f64,
"BUFFER_ATOMIC_MAX_F64">;
}
class NoUseBufferAtomic : PatFrag <
@@ -2315,6 +2312,12 @@ let OtherPredicates = [HasPackedD16VMem] in {
// Target-specific instruction encodings.
//===--===//
+// Shortcut to default Mnemonic from BUF_Pseudo. Hides the cast to the
+// specific pseudo (bothen in this case) since any of them will work.
+class get_BUF_ps {
+ string Mnemonic = !cast(name # "_OFFSET").Mnemonic;
+}
+
//===--===//
// Base ENC_MUBUF for GFX6, GFX7, GFX10, GFX11.
//===--===//
@@ -2346,8 +2349,8 @@ multiclass MUBUF_Real_gfx11 op, string real_name
= !cast(N
}
}
-class Base_MUBUF_Real_gfx6_gfx7_gfx10 op, MUBUF_Pseudo ps, int ef> :
- Base_MUBUF_Real_gfx6_gfx7_gfx10_gfx11 {
+class Base_MUBUF_Real_gfx6_gfx7_gfx10 op, MUBUF_Pseudo ps, int ef,
string asmName> :
+ Base_MUBUF_Real_gfx6_gfx7_gfx10_gfx11 {
let Inst{12}= ps.offen;
let Inst{13}= ps.idxen;
let Inst
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Implement function pointer signing and authenticated function calls (PR #93906)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93906 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Fix buffer load/store of pointers (PR #95379)
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/95379
>From cad588c3cb951121a7a2ee4024839b0f40ecd147 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Wed, 12 Jun 2024 10:10:20 +0200
Subject: [PATCH] AMDGPU: Fix buffer load/store of pointers
Make sure we test all the address spaces since this support isn't
free in gisel.
---
llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 31 +-
.../AMDGPU/llvm.amdgcn.raw.ptr.buffer.load.ll | 596 ++
.../llvm.amdgcn.raw.ptr.buffer.store.ll | 456 ++
3 files changed, 1071 insertions(+), 12 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index d9a163ded6bab..1faccd7e061ce 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -1112,29 +1112,33 @@ unsigned
SITargetLowering::getVectorTypeBreakdownForCallingConv(
Context, CC, VT, IntermediateVT, NumIntermediates, RegisterVT);
}
-static EVT memVTFromLoadIntrData(Type *Ty, unsigned MaxNumLanes) {
+static EVT memVTFromLoadIntrData(const SITargetLowering &TLI,
+ const DataLayout &DL, Type *Ty,
+ unsigned MaxNumLanes) {
assert(MaxNumLanes != 0);
+ LLVMContext &Ctx = Ty->getContext();
if (auto *VT = dyn_cast(Ty)) {
unsigned NumElts = std::min(MaxNumLanes, VT->getNumElements());
-return EVT::getVectorVT(Ty->getContext(),
-EVT::getEVT(VT->getElementType()),
+return EVT::getVectorVT(Ctx, TLI.getValueType(DL, VT->getElementType()),
NumElts);
}
- return EVT::getEVT(Ty);
+ return TLI.getValueType(DL, Ty);
}
// Peek through TFE struct returns to only use the data size.
-static EVT memVTFromLoadIntrReturn(Type *Ty, unsigned MaxNumLanes) {
+static EVT memVTFromLoadIntrReturn(const SITargetLowering &TLI,
+ const DataLayout &DL, Type *Ty,
+ unsigned MaxNumLanes) {
auto *ST = dyn_cast(Ty);
if (!ST)
-return memVTFromLoadIntrData(Ty, MaxNumLanes);
+return memVTFromLoadIntrData(TLI, DL, Ty, MaxNumLanes);
// TFE intrinsics return an aggregate type.
assert(ST->getNumContainedTypes() == 2 &&
ST->getContainedType(1)->isIntegerTy(32));
- return memVTFromLoadIntrData(ST->getContainedType(0), MaxNumLanes);
+ return memVTFromLoadIntrData(TLI, DL, ST->getContainedType(0), MaxNumLanes);
}
/// Map address space 7 to MVT::v5i32 because that's its in-memory
@@ -1219,10 +1223,12 @@ bool SITargetLowering::getTgtMemIntrinsic(IntrinsicInfo
&Info,
MaxNumLanes = DMask == 0 ? 1 : llvm::popcount(DMask);
}
-Info.memVT = memVTFromLoadIntrReturn(CI.getType(), MaxNumLanes);
+Info.memVT = memVTFromLoadIntrReturn(*this, MF.getDataLayout(),
+ CI.getType(), MaxNumLanes);
} else {
-Info.memVT = memVTFromLoadIntrReturn(
-CI.getType(), std::numeric_limits::max());
+Info.memVT =
+memVTFromLoadIntrReturn(*this, MF.getDataLayout(), CI.getType(),
+std::numeric_limits::max());
}
// FIXME: What does alignment mean for an image?
@@ -1235,9 +1241,10 @@ bool SITargetLowering::getTgtMemIntrinsic(IntrinsicInfo
&Info,
if (RsrcIntr->IsImage) {
unsigned DMask =
cast(CI.getArgOperand(1))->getZExtValue();
unsigned DMaskLanes = DMask == 0 ? 1 : llvm::popcount(DMask);
-Info.memVT = memVTFromLoadIntrData(DataTy, DMaskLanes);
+Info.memVT = memVTFromLoadIntrData(*this, MF.getDataLayout(), DataTy,
+ DMaskLanes);
} else
-Info.memVT = EVT::getEVT(DataTy);
+Info.memVT = getValueType(MF.getDataLayout(), DataTy);
Info.flags |= MachineMemOperand::MOStore;
} else {
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.raw.ptr.buffer.load.ll
b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.raw.ptr.buffer.load.ll
index 3e3371091ef72..4d557c76dc4d0 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.raw.ptr.buffer.load.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.raw.ptr.buffer.load.ll
@@ -1280,6 +1280,602 @@ define <2 x i64> @buffer_load_v2i64__voffset_add(ptr
addrspace(8) inreg %rsrc, i
ret <2 x i64> %data
}
+define ptr @buffer_load_p0__voffset_add(ptr addrspace(8) inreg %rsrc, i32
%voffset) {
+; PREGFX10-LABEL: buffer_load_p0__voffset_add:
+; PREGFX10: ; %bb.0:
+; PREGFX10-NEXT:s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; PREGFX10-NEXT:buffer_load_dwordx2 v[0:1], v0, s[4:7], 0 offen offset:60
+; PREGFX10-NEXT:s_waitcnt vmcnt(0)
+; PREGFX10-NEXT:s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: buffer_load_p0__voffset_add:
+; GFX10: ; %bb.0:
+; GFX10-NEXT:s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT:buffer_load_dwordx2 v[0:1], v0, s[4:7
[llvm-branch-commits] [clang] [clang] Define ptrauth_sign_constant builtin. (PR #93904)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93904 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [clang] [clang] Define ptrauth_sign_constant builtin. (PR #93904)
https://github.com/kovdan01 edited https://github.com/llvm/llvm-project/pull/93904 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Create pseudo to real mapping for flat/buffer atomic fmin/fmax (PR #95591)
https://github.com/arsenm updated
https://github.com/llvm/llvm-project/pull/95591
>From a5b973d9b4816fb5097ba005929601639347c657 Mon Sep 17 00:00:00 2001
From: Matt Arsenault
Date: Thu, 13 Jun 2024 18:33:11 +0200
Subject: [PATCH] AMDGPU: Create pseudo to real mapping for flat/buffer atomic
fmin/fmax
The global/flat/buffer atomic fmin/fmax situation is a mess. These
instructions have been renamed 3 times. We currently have
separate pseudos defined for the same opcodes with the different names
(e.g. GLOBAL_ATOMIC_MIN_F64 from gfx90a and GLOBAL_ATOMIC_FMIN_X2 from gfx10).
Use the _FMIN versions as the canonical name for the f32 versions. Use the
_MIN_F64 style as the canonical name for the f64 case. This is because
gfx90a has the most sensible names, but does not have the f32 versions.t sho
Wire through the pseudo to use for the instruction properties vs. the assembly
name like in other cases. This will simplify handling of direct atomicrmw
selection.
This will simplify directly selecting these from atomicrmw.
---
llvm/lib/Target/AMDGPU/AMDGPU.td | 4 +-
llvm/lib/Target/AMDGPU/BUFInstructions.td | 103 +
llvm/lib/Target/AMDGPU/FLATInstructions.td| 107 +-
.../AMDGPU/fp-atomic-to-s_denormmode.mir | 40 +++
4 files changed, 129 insertions(+), 125 deletions(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPU.td b/llvm/lib/Target/AMDGPU/AMDGPU.td
index 65c4abef2bf8a..cb5ceb9959325 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPU.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPU.td
@@ -1864,7 +1864,9 @@ def HasFlatAddressSpace :
Predicate<"Subtarget->hasFlatAddressSpace()">,
def HasBufferFlatGlobalAtomicsF64 :
Predicate<"Subtarget->hasBufferFlatGlobalAtomicsF64()">,
- AssemblerPredicate<(any_of FeatureGFX90AInsts)>;
+ // FIXME: This is too coarse, and working around using pseudo's predicates
on real instruction.
+ AssemblerPredicate<(any_of FeatureGFX90AInsts, FeatureGFX10Insts,
FeatureSouthernIslands, FeatureSeaIslands)>;
+
def HasLdsAtomicAddF64 :
Predicate<"Subtarget->hasLdsAtomicAddF64()">,
AssemblerPredicate<(any_of FeatureGFX90AInsts)>;
diff --git a/llvm/lib/Target/AMDGPU/BUFInstructions.td
b/llvm/lib/Target/AMDGPU/BUFInstructions.td
index 43e5434ea2700..9d21f93a957cc 100644
--- a/llvm/lib/Target/AMDGPU/BUFInstructions.td
+++ b/llvm/lib/Target/AMDGPU/BUFInstructions.td
@@ -1163,12 +1163,6 @@ let SubtargetPredicate = isGFX6GFX7GFX10 in {
defm BUFFER_ATOMIC_FCMPSWAP_X2 : MUBUF_Pseudo_Atomics <
"buffer_atomic_fcmpswap_x2", VReg_128, v2f64, null_frag
>;
-defm BUFFER_ATOMIC_FMIN_X2 : MUBUF_Pseudo_Atomics <
- "buffer_atomic_fmin_x2", VReg_64, f64, null_frag
->;
-defm BUFFER_ATOMIC_FMAX_X2 : MUBUF_Pseudo_Atomics <
- "buffer_atomic_fmax_x2", VReg_64, f64, null_frag
->;
}
@@ -1318,6 +1312,9 @@ let SubtargetPredicate = isGFX90APlus in {
let SubtargetPredicate = HasBufferFlatGlobalAtomicsF64 in {
defm BUFFER_ATOMIC_ADD_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_add_f64",
VReg_64, f64>;
+
+ // Note the names can be buffer_atomic_fmin_x2/buffer_atomic_fmax_x2
+ // depending on some subtargets.
defm BUFFER_ATOMIC_MIN_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_min_f64",
VReg_64, f64>;
defm BUFFER_ATOMIC_MAX_F64 : MUBUF_Pseudo_Atomics<"buffer_atomic_max_f64",
VReg_64, f64>;
} // End SubtargetPredicate = HasBufferFlatGlobalAtomicsF64
@@ -1763,8 +1760,8 @@ let OtherPredicates = [isGFX6GFX7GFX10Plus] in {
defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f32, "BUFFER_ATOMIC_FMAX">;
}
let SubtargetPredicate = isGFX6GFX7GFX10 in {
- defm : SIBufferAtomicPat<"SIbuffer_atomic_fmin", f64,
"BUFFER_ATOMIC_FMIN_X2">;
- defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f64,
"BUFFER_ATOMIC_FMAX_X2">;
+ defm : SIBufferAtomicPat<"SIbuffer_atomic_fmin", f64,
"BUFFER_ATOMIC_MIN_F64">;
+ defm : SIBufferAtomicPat<"SIbuffer_atomic_fmax", f64,
"BUFFER_ATOMIC_MAX_F64">;
}
class NoUseBufferAtomic : PatFrag <
@@ -2315,6 +2312,12 @@ let OtherPredicates = [HasPackedD16VMem] in {
// Target-specific instruction encodings.
//===--===//
+// Shortcut to default Mnemonic from BUF_Pseudo. Hides the cast to the
+// specific pseudo (bothen in this case) since any of them will work.
+class get_BUF_ps {
+ string Mnemonic = !cast(name # "_OFFSET").Mnemonic;
+}
+
//===--===//
// Base ENC_MUBUF for GFX6, GFX7, GFX10, GFX11.
//===--===//
@@ -2346,8 +2349,8 @@ multiclass MUBUF_Real_gfx11 op, string real_name
= !cast(N
}
}
-class Base_MUBUF_Real_gfx6_gfx7_gfx10 op, MUBUF_Pseudo ps, int ef> :
- Base_MUBUF_Real_gfx6_gfx7_gfx10_gfx11 {
+class Base_MUBUF_Real_gfx6_gfx7_gfx10 op, MUBUF_Pseudo ps, int ef,
string asmName> :
+ Base_MUBUF_Real_gfx6_gfx7_gfx10_gfx11 {
let Inst{12}= ps.offen;
let Inst{13}= ps.idxen;
let Inst
[llvm-branch-commits] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung created https://github.com/llvm/llvm-project/pull/95821 Using the hashes of binary and profiled functions to recover functions with changed names. Test Plan: tbd ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [BOLT] Hash-based function matching (PR #95821)
llvmbot wrote:
@llvm/pr-subscribers-bolt
Author: shaw young (shawbyoung)
Changes
Using the hashes of binary and profiled functions
to recover functions with changed names.
Test Plan: tbd
---
Full diff: https://github.com/llvm/llvm-project/pull/95821.diff
1 Files Affected:
- (modified) bolt/lib/Profile/YAMLProfileReader.cpp (+17)
``diff
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f25f59201f1cd..f0fcb1c130002 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -415,6 +415,23 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
if (!YamlBF.Used && BF && !ProfiledFunctions.count(BF))
matchProfileToFunction(YamlBF, *BF);
+ // Uses the strict hash of profiled and binary functions to match functions
+ // that are not matched by name or common name.
+ std::unordered_map StrictBinaryFunctionHashes;
+ StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
+
+ for (auto& [_, BF] : BC.getBinaryFunctions()) {
+StrictBinaryFunctionHashes[BF.getHash()] = &BF;
+ }
+
+ for (auto YamlBF : YamlBP.Functions) {
+auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
+if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
+ auto *BF = It->second;
+ matchProfileToFunction(YamlBF, *BF);
+}
+ }
+
for (yaml::bolt::BinaryFunctionProfile &YamlBF : YamlBP.Functions)
if (!YamlBF.Used && opts::Verbosity >= 1)
errs() << "BOLT-WARNING: profile ignored for function " << YamlBF.Name
``
https://github.com/llvm/llvm-project/pull/95821
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [mlir] 6b51c7a - Revert "[mlir][sparse] implement lowering rules for IterateOp. (#95286)"
Author: Peiming Liu
Date: 2024-06-17T11:35:01-07:00
New Revision: 6b51c7a96be4d260d74d9992745f92c1aadc78ff
URL:
https://github.com/llvm/llvm-project/commit/6b51c7a96be4d260d74d9992745f92c1aadc78ff
DIFF:
https://github.com/llvm/llvm-project/commit/6b51c7a96be4d260d74d9992745f92c1aadc78ff.diff
LOG: Revert "[mlir][sparse] implement lowering rules for IterateOp. (#95286)"
This reverts commit 3a2e4428938476a4822f4da7193e866e9dd585d9.
Added:
Modified:
mlir/lib/Dialect/SparseTensor/Transforms/SparseIterationToScf.cpp
mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.cpp
mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h
mlir/test/Dialect/SparseTensor/sparse_iteration_to_scf.mlir
Removed:
diff --git a/mlir/lib/Dialect/SparseTensor/Transforms/SparseIterationToScf.cpp
b/mlir/lib/Dialect/SparseTensor/Transforms/SparseIterationToScf.cpp
index f57be49f21b8c..62887c75c872b 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/SparseIterationToScf.cpp
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/SparseIterationToScf.cpp
@@ -34,20 +34,6 @@ convertIterSpaceType(IterSpaceType itSp,
SmallVectorImpl &fields) {
return success();
}
-static std::optional
-convertIteratorType(IteratorType itTp, SmallVectorImpl &fields) {
- // The actually Iterator Values (that are updated every iteration).
- auto idxTp = IndexType::get(itTp.getContext());
- // TODO: handle batch dimension.
- assert(itTp.getEncoding().getBatchLvlRank() == 0);
- if (!itTp.isUnique()) {
-// Segment high for non-unique iterator.
-fields.push_back(idxTp);
- }
- fields.push_back(idxTp);
- return success();
-}
-
namespace {
/// Sparse codegen rule for number of entries operator.
@@ -71,114 +57,10 @@ class ExtractIterSpaceConverter
}
};
-class SparseIterateOpConverter : public OneToNOpConversionPattern {
-public:
- using OneToNOpConversionPattern::OneToNOpConversionPattern;
- LogicalResult
- matchAndRewrite(IterateOp op, OpAdaptor adaptor,
- OneToNPatternRewriter &rewriter) const override {
-if (!op.getCrdUsedLvls().empty())
- return rewriter.notifyMatchFailure(
- op, "non-empty coordinates list not implemented.");
-
-Location loc = op.getLoc();
-
-auto iterSpace = SparseIterationSpace::fromValues(
-op.getIterSpace().getType(), adaptor.getIterSpace(), 0);
-
-std::unique_ptr it =
-iterSpace.extractIterator(rewriter, loc);
-
-if (it->iteratableByFor()) {
- auto [lo, hi] = it->genForCond(rewriter, loc);
- Value step = constantIndex(rewriter, loc, 1);
- SmallVector ivs;
- for (ValueRange inits : adaptor.getInitArgs())
-llvm::append_range(ivs, inits);
- scf::ForOp forOp = rewriter.create(loc, lo, hi, step, ivs);
-
- Block *loopBody = op.getBody();
- OneToNTypeMapping bodyTypeMapping(loopBody->getArgumentTypes());
- if (failed(typeConverter->convertSignatureArgs(
- loopBody->getArgumentTypes(), bodyTypeMapping)))
-return failure();
- rewriter.applySignatureConversion(loopBody, bodyTypeMapping);
-
- forOp.getBody()->erase();
- Region &dstRegion = forOp.getRegion();
- rewriter.inlineRegionBefore(op.getRegion(), dstRegion, dstRegion.end());
-
- auto yieldOp =
- llvm::cast(forOp.getBody()->getTerminator());
-
- rewriter.setInsertionPointToEnd(forOp.getBody());
- // replace sparse_tensor.yield with scf.yield.
- rewriter.create(loc, yieldOp.getResults());
- yieldOp.erase();
-
- const OneToNTypeMapping &resultMapping = adaptor.getResultMapping();
- rewriter.replaceOp(op, forOp.getResults(), resultMapping);
-} else {
- SmallVector ivs;
- llvm::append_range(ivs, it->getCursor());
- for (ValueRange inits : adaptor.getInitArgs())
-llvm::append_range(ivs, inits);
-
- assert(llvm::all_of(ivs, [](Value v) { return v != nullptr; }));
-
- TypeRange types = ValueRange(ivs).getTypes();
- auto whileOp = rewriter.create(loc, types, ivs);
- SmallVector l(types.size(), op.getIterator().getLoc());
-
- // Generates loop conditions.
- Block *before = rewriter.createBlock(&whileOp.getBefore(), {}, types, l);
- rewriter.setInsertionPointToStart(before);
- ValueRange bArgs = before->getArguments();
- auto [whileCond, remArgs] = it->genWhileCond(rewriter, loc, bArgs);
- assert(remArgs.size() == adaptor.getInitArgs().size());
- rewriter.create(loc, whileCond,
before->getArguments());
-
- // Generates loop body.
- Block *loopBody = op.getBody();
- OneToNTypeMapping bodyTypeMapping(loopBody->getArgumentTypes());
- if (failed(typeConverter->convertSignatureArgs(
- loopBody->getArgumentTypes(), bodyTypeMapping)))
-return failure();
- rewriter.applySignatureConversion(loopBody, bodyTypeMapping)
[llvm-branch-commits] [clang] 676be4f - Revert "[Clang] Add fix-it hint to insert `*` when a modifiable lvalue is req…"
Author: Sirraide
Date: 2024-06-17T21:43:34+02:00
New Revision: 676be4fb82172ad1ea701ce219e9706b91570ea3
URL:
https://github.com/llvm/llvm-project/commit/676be4fb82172ad1ea701ce219e9706b91570ea3
DIFF:
https://github.com/llvm/llvm-project/commit/676be4fb82172ad1ea701ce219e9706b91570ea3.diff
LOG: Revert "[Clang] Add fix-it hint to insert `*` when a modifiable lvalue is
req…"
This reverts commit 9d7299fb84bcaafc6b7368dbe5ee6de29e71e655.
Added:
Modified:
clang/include/clang/Basic/DiagnosticSemaKinds.td
clang/lib/Sema/SemaExpr.cpp
clang/test/C/drs/dr1xx.c
clang/test/Sema/exprs.c
clang/test/Sema/va_arg_x86_32.c
clang/test/SemaObjCXX/sel-address.mm
Removed:
clang/test/Sema/debug-93066.cpp
diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td
b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index 9b8f5b7e80e7e..8f85371df3b85 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -8778,9 +8778,6 @@ def err_typecheck_incomplete_type_not_modifiable_lvalue :
Error<
def err_typecheck_lvalue_casts_not_supported : Error<
"assignment to cast is illegal, lvalue casts are not supported">;
-def note_typecheck_add_deref_star_not_modifiable_lvalue : Note<
- "add '*' to dereference it">;
-
def err_typecheck_duplicate_vector_components_not_mlvalue : Error<
"vector is not assignable (contains duplicate components)">;
def err_block_decl_ref_not_modifiable_lvalue : Error<
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index 4db8b4130c3c7..453af8f28f946 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -13288,23 +13288,6 @@ enum {
ConstUnknown, // Keep as last element
};
-static void MaybeSuggestDerefFixIt(Sema &S, const Expr *E, SourceLocation Loc)
{
- ExprResult Deref;
- Expr *TE = const_cast(E);
- {
-Sema::TentativeAnalysisScope Trap(S);
-Deref = S.ActOnUnaryOp(S.getCurScope(), Loc, tok::star, TE);
- }
- if (Deref.isUsable() &&
- Deref.get()->isModifiableLvalue(S.Context, &Loc) == Expr::MLV_Valid &&
- !E->getType()->isObjCObjectPointerType()) {
-S.Diag(E->getBeginLoc(),
- diag::note_typecheck_add_deref_star_not_modifiable_lvalue)
-<< E->getSourceRange()
-<< FixItHint::CreateInsertion(E->getBeginLoc(), "*");
- }
-}
-
/// Emit the "read-only variable not assignable" error and print notes to give
/// more information about why the variable is not assignable, such as pointing
/// to the declaration of a const variable, showing that a method is const, or
@@ -13399,7 +13382,6 @@ static void DiagnoseConstAssignment(Sema &S, const Expr
*E,
if (!DiagnosticEmitted) {
S.Diag(Loc, diag::err_typecheck_assign_const)
<< ExprRange << ConstVariable << VD << VD->getType();
- MaybeSuggestDerefFixIt(S, E, Loc);
DiagnosticEmitted = true;
}
S.Diag(VD->getLocation(), diag::note_typecheck_assign_const)
@@ -13620,12 +13602,10 @@ static bool CheckForModifiableLvalue(Expr *E,
SourceLocation Loc, Sema &S) {
SourceRange Assign;
if (Loc != OrigLoc)
Assign = SourceRange(OrigLoc, OrigLoc);
- if (NeedType) {
+ if (NeedType)
S.Diag(Loc, DiagID) << E->getType() << E->getSourceRange() << Assign;
- } else {
+ else
S.Diag(Loc, DiagID) << E->getSourceRange() << Assign;
-MaybeSuggestDerefFixIt(S, E, Loc);
- }
return true;
}
diff --git a/clang/test/C/drs/dr1xx.c b/clang/test/C/drs/dr1xx.c
index 20e953b2c20ac..47538e44428c3 100644
--- a/clang/test/C/drs/dr1xx.c
+++ b/clang/test/C/drs/dr1xx.c
@@ -296,7 +296,6 @@ void dr126(void) {
*/
*object = 12; /* ok */
++object; /* expected-error {{cannot assign to variable 'object' with
const-qualified type 'const IP' (aka 'int *const')}} */
- /* expected-note@-1 {{add '*' to dereference it}} */
}
/* WG14 DR128: yes
diff --git a/clang/test/Sema/debug-93066.cpp b/clang/test/Sema/debug-93066.cpp
deleted file mode 100644
index e64d378e6f90d..0
--- a/clang/test/Sema/debug-93066.cpp
+++ /dev/null
@@ -1,49 +0,0 @@
-// RUN: %clang_cc1 -std=c++23 -fsyntax-only -verify %s
-
-struct S {
- void f() {
-++this; // expected-error {{expression is not assignable}}
-// expected-note@-1 {{add '*' to dereference it}}
- }
-
- void g() const {
-++this; // expected-error {{expression is not assignable}}
- }
-};
-
-void f(int* a, int* const b, const int* const c, __UINTPTR_TYPE__ d) {
- // expected-note@-1 {{variable 'b' declared const here}}
- // expected-note@-2 {{variable 'c' declared const here}}
- (int*)d = 4; // expected-error {{assignment to cast is illegal, lvalue casts
are not supported}}
- // expected-note@-1 {{add '*' to dereference it}}
-
- ++a;
- ++b; // expected-error {{cannot assign to variable 'b' with const-qualified
type 'int *const'}}
-
[llvm-branch-commits] [openmp] release/18.x: [OpenMP][OMPT] Fix hsa include when building amdgpu/src/rtl.cpp (PR #95484)
jhuber6 wrote: I merged my patch, I think the window to backport to 18.x is passed so you'll need to use the main branch for now. Thanks for bringing this to my attention. https://github.com/llvm/llvm-project/pull/95484 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [openmp] release/18.x: [OpenMP][OMPT] Fix hsa include when building amdgpu/src/rtl.cpp (PR #95484)
https://github.com/jhuber6 closed https://github.com/llvm/llvm-project/pull/95484 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Drop high discrepancy profiles in matching (PR #95156)
https://github.com/shawbyoung closed https://github.com/llvm/llvm-project/pull/95156 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [mlir] 768bda0 - Revert "Reland [mlir][Target] Improve ROCDL gpu serialization API (#95813)"
Author: Fabian Mora
Date: 2024-06-17T15:49:52-05:00
New Revision: 768bda06368a96d186ed292e15d979f7ee695819
URL:
https://github.com/llvm/llvm-project/commit/768bda06368a96d186ed292e15d979f7ee695819
DIFF:
https://github.com/llvm/llvm-project/commit/768bda06368a96d186ed292e15d979f7ee695819.diff
LOG: Revert "Reland [mlir][Target] Improve ROCDL gpu serialization API (#95813)"
This reverts commit dcb6c0d71c8dbb6bb17391c968c3716cfafd3765.
Added:
Modified:
mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
mlir/lib/Dialect/GPU/CMakeLists.txt
mlir/lib/Target/LLVM/CMakeLists.txt
mlir/lib/Target/LLVM/ROCDL/Target.cpp
Removed:
diff --git a/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
b/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
index 44c9ded317fa5..374fa65bd02e3 100644
--- a/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
+++ b/mlir/include/mlir/Target/LLVM/ROCDL/Utils.h
@@ -27,19 +27,6 @@ namespace ROCDL {
/// 5. Returns an empty string.
StringRef getROCMPath();
-/// Helper enum for specifying the AMD GCN device libraries required for
-/// compilation.
-enum class AMDGCNLibraries : uint32_t {
- None = 0,
- Ockl = 1,
- Ocml = 2,
- OpenCL = 4,
- Hip = 8,
- LastLib = Hip,
- LLVM_MARK_AS_BITMASK_ENUM(LastLib),
- All = (LastLib << 1) - 1
-};
-
/// Base class for all ROCDL serializations from GPU modules into binary
/// strings. By default this class serializes into LLVM bitcode.
class SerializeGPUModuleBase : public LLVM::ModuleToObject {
@@ -62,8 +49,8 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
/// Returns the bitcode files to be loaded.
ArrayRef getFileList() const;
- /// Appends standard ROCm device libraries to `fileList`.
- LogicalResult appendStandardLibs(AMDGCNLibraries libs);
+ /// Appends standard ROCm device libraries like `ocml.bc`, `ockl.bc`, etc.
+ LogicalResult appendStandardLibs();
/// Loads the bitcode files in `fileList`.
virtual std::optional>>
@@ -76,20 +63,15 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
LogicalResult handleBitcodeFile(llvm::Module &module) override;
protected:
- /// Adds `oclc` control variables to the LLVM module.
- void addControlVariables(llvm::Module &module, AMDGCNLibraries libs,
- bool wave64, bool daz, bool finiteOnly,
- bool unsafeMath, bool fastMath, bool correctSqrt,
- StringRef abiVer);
+ /// Appends the paths of common ROCm device libraries to `libs`.
+ LogicalResult getCommonBitcodeLibs(llvm::SmallVector &libs,
+ SmallVector &libPath,
+ StringRef isaVersion);
- /// Compiles assembly to a binary.
- virtual std::optional>
- compileToBinary(const std::string &serializedISA);
-
- /// Default implementation of `ModuleToObject::moduleToObject`.
- std::optional>
- moduleToObjectImpl(const gpu::TargetOptions &targetOptions,
- llvm::Module &llvmModule);
+ /// Adds `oclc` control variables to the LLVM module.
+ void addControlVariables(llvm::Module &module, bool wave64, bool daz,
+ bool finiteOnly, bool unsafeMath, bool fastMath,
+ bool correctSqrt, StringRef abiVer);
/// Returns the assembled ISA.
std::optional> assembleIsa(StringRef isa);
@@ -102,9 +84,6 @@ class SerializeGPUModuleBase : public LLVM::ModuleToObject {
/// List of LLVM bitcode files to link to.
SmallVector fileList;
-
- /// AMD GCN libraries to use when linking, the default is using none.
- AMDGCNLibraries deviceLibs = AMDGCNLibraries::None;
};
} // namespace ROCDL
} // namespace mlir
diff --git a/mlir/lib/Dialect/GPU/CMakeLists.txt
b/mlir/lib/Dialect/GPU/CMakeLists.txt
index 08c8aea36fac9..61ab298ebfb98 100644
--- a/mlir/lib/Dialect/GPU/CMakeLists.txt
+++ b/mlir/lib/Dialect/GPU/CMakeLists.txt
@@ -106,7 +106,7 @@ if(MLIR_ENABLE_ROCM_CONVERSIONS)
"Building mlir with ROCm support requires the AMDGPU backend")
endif()
- set(DEFAULT_ROCM_PATH "" CACHE PATH "Fallback path to search for ROCm
installs")
+ set(DEFAULT_ROCM_PATH "/opt/rocm" CACHE PATH "Fallback path to search for
ROCm installs")
target_compile_definitions(obj.MLIRGPUTransforms
PRIVATE
__DEFAULT_ROCM_PATH__="${DEFAULT_ROCM_PATH}"
diff --git a/mlir/lib/Target/LLVM/CMakeLists.txt
b/mlir/lib/Target/LLVM/CMakeLists.txt
index 6e146710d67af..5a3fa160850b4 100644
--- a/mlir/lib/Target/LLVM/CMakeLists.txt
+++ b/mlir/lib/Target/LLVM/CMakeLists.txt
@@ -123,18 +123,17 @@ add_mlir_dialect_library(MLIRROCDLTarget
)
if(MLIR_ENABLE_ROCM_CONVERSIONS)
+ if (NOT ("AMDGPU" IN_LIST LLVM_TARGETS_TO_BUILD))
+message(SEND_ERROR
+ "Building mlir with ROCm support requires the AMDGPU backend")
+ endif()
+
if (DEFINED ROCM_PATH)
set(DEFAULT_ROCM_PATH "${ROCM_PATH}" CACHE PATH "Fallbac
[llvm-branch-commits] [compiler-rt] 43ca021 - Revert "[HWASan] [compiler-rt] support non-4k pages on Android (#95069)"
Author: Florian Mayer
Date: 2024-06-17T14:37:55-07:00
New Revision: 43ca021ef4765f212be166940f0305c05845b713
URL:
https://github.com/llvm/llvm-project/commit/43ca021ef4765f212be166940f0305c05845b713
DIFF:
https://github.com/llvm/llvm-project/commit/43ca021ef4765f212be166940f0305c05845b713.diff
LOG: Revert "[HWASan] [compiler-rt] support non-4k pages on Android (#95069)"
This reverts commit 5b04b6fe3fabba8f76d730da3c0d528e1dd0c184.
Added:
Modified:
compiler-rt/lib/asan/asan_linux.cpp
compiler-rt/lib/asan/asan_mac.cpp
compiler-rt/lib/asan/asan_premap_shadow.cpp
compiler-rt/lib/hwasan/hwasan_dynamic_shadow.cpp
compiler-rt/lib/memprof/memprof_linux.cpp
compiler-rt/lib/sanitizer_common/sanitizer_common.h
compiler-rt/lib/sanitizer_common/sanitizer_linux.cpp
compiler-rt/lib/sanitizer_common/sanitizer_linux_libcdep.cpp
compiler-rt/lib/sanitizer_common/sanitizer_mac.cpp
compiler-rt/lib/sanitizer_common/sanitizer_win.cpp
Removed:
diff --git a/compiler-rt/lib/asan/asan_linux.cpp
b/compiler-rt/lib/asan/asan_linux.cpp
index 0b470db86748f..a517de5af00dc 100644
--- a/compiler-rt/lib/asan/asan_linux.cpp
+++ b/compiler-rt/lib/asan/asan_linux.cpp
@@ -93,8 +93,7 @@ uptr FindDynamicShadowStart() {
# endif
return MapDynamicShadow(shadow_size_bytes, ASAN_SHADOW_SCALE,
- /*min_shadow_base_alignment*/ 0, kHighMemEnd,
- GetMmapGranularity());
+ /*min_shadow_base_alignment*/ 0, kHighMemEnd);
}
void AsanApplyToGlobals(globals_op_fptr op, const void *needle) {
diff --git a/compiler-rt/lib/asan/asan_mac.cpp
b/compiler-rt/lib/asan/asan_mac.cpp
index bfc349223258b..b250f796e165f 100644
--- a/compiler-rt/lib/asan/asan_mac.cpp
+++ b/compiler-rt/lib/asan/asan_mac.cpp
@@ -51,8 +51,7 @@ bool IsSystemHeapAddress (uptr addr) { return false; }
uptr FindDynamicShadowStart() {
return MapDynamicShadow(MemToShadowSize(kHighMemEnd), ASAN_SHADOW_SCALE,
- /*min_shadow_base_alignment*/ 0, kHighMemEnd,
- GetMmapGranularity());
+ /*min_shadow_base_alignment*/ 0, kHighMemEnd);
}
// No-op. Mac does not support static linkage anyway.
diff --git a/compiler-rt/lib/asan/asan_premap_shadow.cpp
b/compiler-rt/lib/asan/asan_premap_shadow.cpp
index 6e08b8f966507..bed2f62a22511 100644
--- a/compiler-rt/lib/asan/asan_premap_shadow.cpp
+++ b/compiler-rt/lib/asan/asan_premap_shadow.cpp
@@ -33,8 +33,7 @@ uptr PremapShadowSize() {
// PremapShadowSize() bytes on the right of it are mapped r/o.
uptr PremapShadow() {
return MapDynamicShadow(PremapShadowSize(), /*mmap_alignment_scale*/ 3,
- /*min_shadow_base_alignment*/ 0, kHighMemEnd,
- GetMmapGranularity());
+ /*min_shadow_base_alignment*/ 0, kHighMemEnd);
}
bool PremapShadowFailed() {
diff --git a/compiler-rt/lib/hwasan/hwasan_dynamic_shadow.cpp
b/compiler-rt/lib/hwasan/hwasan_dynamic_shadow.cpp
index 48bc3b631ac07..7642ba6c0bf08 100644
--- a/compiler-rt/lib/hwasan/hwasan_dynamic_shadow.cpp
+++ b/compiler-rt/lib/hwasan/hwasan_dynamic_shadow.cpp
@@ -36,20 +36,15 @@ decltype(__hwasan_shadow)* __hwasan_premap_shadow();
namespace __hwasan {
-// We cannot call anything in libc here (see comment above), so we need to
-// assume the biggest allowed page size.
-// Android max page size is defined as 16k here:
-//
https://android.googlesource.com/platform/bionic/+/main/libc/platform/bionic/page.h#41
-static constexpr uptr kMaxGranularity = 16384;
-
// Conservative upper limit.
static uptr PremapShadowSize() {
- return RoundUpTo(GetMaxVirtualAddress() >> kShadowScale, kMaxGranularity);
+ return RoundUpTo(GetMaxVirtualAddress() >> kShadowScale,
+ GetMmapGranularity());
}
static uptr PremapShadow() {
return MapDynamicShadow(PremapShadowSize(), kShadowScale,
- kShadowBaseAlignment, kHighMemEnd, kMaxGranularity);
+ kShadowBaseAlignment, kHighMemEnd);
}
static bool IsPremapShadowAvailable() {
@@ -61,7 +56,7 @@ static bool IsPremapShadowAvailable() {
}
static uptr FindPremappedShadowStart(uptr shadow_size_bytes) {
- const uptr granularity = kMaxGranularity;
+ const uptr granularity = GetMmapGranularity();
const uptr shadow_start = reinterpret_cast(&__hwasan_shadow);
const uptr premap_shadow_size = PremapShadowSize();
const uptr shadow_size = RoundUpTo(shadow_size_bytes, granularity);
@@ -114,7 +109,7 @@ uptr FindDynamicShadowStart(uptr shadow_size_bytes) {
if (IsPremapShadowAvailable())
return FindPremappedShadowStart(shadow_size_bytes);
return MapDynamicShadow(shadow_size_bytes, kShadowScale,
kShadowBaseAlignment,
- kHighMemEnd, kMaxGranularity);
+ kHighMemEnd);
}
}
[llvm-branch-commits] [llvm] release/18.x: [SystemZ] Bugfix in getDemandedSrcElements(). (#88623) (PR #95463)
tstellar wrote: Is this a critical bug? I'm trying to decide if we should add this to 18.1.8. https://github.com/llvm/llvm-project/pull/95463 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95821
>From 92212c96ea169d26ac10bf8d750539bc5dd72c49 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:02 -0700
Subject: [PATCH] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f0fcb1c130002..2bca83c9d11ec 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,6 +421,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto& [_, BF] : BC.getBinaryFunctions()) {
+if (!ProfiledFunctions.count(&BF))
+ continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95821
>From 92212c96ea169d26ac10bf8d750539bc5dd72c49 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:02 -0700
Subject: [PATCH 1/2] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f0fcb1c130002..2bca83c9d11ec 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,6 +421,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto& [_, BF] : BC.getBinaryFunctions()) {
+if (!ProfiledFunctions.count(&BF))
+ continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
>From 2497922ccc46e3189870563b1fe819b67172778d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:39 -0700
Subject: [PATCH 2/2] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 2bca83c9d11ec..56474a67307ed 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -417,10 +417,10 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
// Uses the strict hash of profiled and binary functions to match functions
// that are not matched by name or common name.
- std::unordered_map StrictBinaryFunctionHashes;
+ std::unordered_map StrictBinaryFunctionHashes;
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
- for (auto& [_, BF] : BC.getBinaryFunctions()) {
+ for (auto &[_, BF] : BC.getBinaryFunctions()) {
if (!ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
@@ -428,7 +428,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
for (auto YamlBF : YamlBP.Functions) {
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
-if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
+if (It != StrictBinaryFunctionHashes.end() &&
+!ProfiledFunctions.count(It->second)) {
auto *BF = It->second;
matchProfileToFunction(YamlBF, *BF);
}
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95821
>From 92212c96ea169d26ac10bf8d750539bc5dd72c49 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:02 -0700
Subject: [PATCH 1/3] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f0fcb1c130002..2bca83c9d11ec 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,6 +421,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto& [_, BF] : BC.getBinaryFunctions()) {
+if (!ProfiledFunctions.count(&BF))
+ continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
>From 2497922ccc46e3189870563b1fe819b67172778d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:39 -0700
Subject: [PATCH 2/3] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 2bca83c9d11ec..56474a67307ed 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -417,10 +417,10 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
// Uses the strict hash of profiled and binary functions to match functions
// that are not matched by name or common name.
- std::unordered_map StrictBinaryFunctionHashes;
+ std::unordered_map StrictBinaryFunctionHashes;
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
- for (auto& [_, BF] : BC.getBinaryFunctions()) {
+ for (auto &[_, BF] : BC.getBinaryFunctions()) {
if (!ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
@@ -428,7 +428,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
for (auto YamlBF : YamlBP.Functions) {
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
-if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
+if (It != StrictBinaryFunctionHashes.end() &&
+!ProfiledFunctions.count(It->second)) {
auto *BF = It->second;
matchProfileToFunction(YamlBF, *BF);
}
>From 8e7b2229a69c3795e723404c56e0d4298eef412a Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:55:58 -0700
Subject: [PATCH 3/3] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 +-
bolt/test/X86/profile-passthrough-block.test | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 56474a67307ed..779d60bce3b66 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,7 +421,7 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto &[_, BF] : BC.getBinaryFunctions()) {
-if (!ProfiledFunctions.count(&BF))
+if (ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
diff --git a/bolt/test/X86/profile-passthrough-block.test
b/bolt/test/X86/profile-passthrough-block.test
index 1b875885260dc..ed2a8117ddfc4 100644
--- a/bolt/test/X86/profile-passthrough-block.test
+++ b/bolt/test/X86/profile-passthrough-block.test
@@ -57,7 +57,7 @@ header:
functions:
- name:main
fid: 0
-hash:0x
+hash:0x0001
exec:1
nblocks: 6
blocks:
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95821
>From 92212c96ea169d26ac10bf8d750539bc5dd72c49 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:02 -0700
Subject: [PATCH 1/4] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f0fcb1c130002..2bca83c9d11ec 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,6 +421,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto& [_, BF] : BC.getBinaryFunctions()) {
+if (!ProfiledFunctions.count(&BF))
+ continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
>From 2497922ccc46e3189870563b1fe819b67172778d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:39 -0700
Subject: [PATCH 2/4] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 2bca83c9d11ec..56474a67307ed 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -417,10 +417,10 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
// Uses the strict hash of profiled and binary functions to match functions
// that are not matched by name or common name.
- std::unordered_map StrictBinaryFunctionHashes;
+ std::unordered_map StrictBinaryFunctionHashes;
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
- for (auto& [_, BF] : BC.getBinaryFunctions()) {
+ for (auto &[_, BF] : BC.getBinaryFunctions()) {
if (!ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
@@ -428,7 +428,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
for (auto YamlBF : YamlBP.Functions) {
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
-if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
+if (It != StrictBinaryFunctionHashes.end() &&
+!ProfiledFunctions.count(It->second)) {
auto *BF = It->second;
matchProfileToFunction(YamlBF, *BF);
}
>From 8e7b2229a69c3795e723404c56e0d4298eef412a Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:55:58 -0700
Subject: [PATCH 3/4] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 +-
bolt/test/X86/profile-passthrough-block.test | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 56474a67307ed..779d60bce3b66 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,7 +421,7 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto &[_, BF] : BC.getBinaryFunctions()) {
-if (!ProfiledFunctions.count(&BF))
+if (ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
diff --git a/bolt/test/X86/profile-passthrough-block.test
b/bolt/test/X86/profile-passthrough-block.test
index 1b875885260dc..ed2a8117ddfc4 100644
--- a/bolt/test/X86/profile-passthrough-block.test
+++ b/bolt/test/X86/profile-passthrough-block.test
@@ -57,7 +57,7 @@ header:
functions:
- name:main
fid: 0
-hash:0x
+hash:0x0001
exec:1
nblocks: 6
blocks:
>From ef5f0dac9185dbb7a62345938d4f309c3379a85d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:58:22 -0700
Subject: [PATCH 4/4] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 779d60bce3b66..e3d30bfdb74e4 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -427,6 +427,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
}
for (auto YamlBF : YamlBP.Functions) {
+if (YamlBF.Used)
+ continue;
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] [BOLT] Hash-based function matching (PR #95821)
https://github.com/shawbyoung updated
https://github.com/llvm/llvm-project/pull/95821
>From 92212c96ea169d26ac10bf8d750539bc5dd72c49 Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:02 -0700
Subject: [PATCH 1/5] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f0fcb1c130002..2bca83c9d11ec 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,6 +421,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto& [_, BF] : BC.getBinaryFunctions()) {
+if (!ProfiledFunctions.count(&BF))
+ continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
>From 2497922ccc46e3189870563b1fe819b67172778d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:39:39 -0700
Subject: [PATCH 2/5] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 2bca83c9d11ec..56474a67307ed 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -417,10 +417,10 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
// Uses the strict hash of profiled and binary functions to match functions
// that are not matched by name or common name.
- std::unordered_map StrictBinaryFunctionHashes;
+ std::unordered_map StrictBinaryFunctionHashes;
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
- for (auto& [_, BF] : BC.getBinaryFunctions()) {
+ for (auto &[_, BF] : BC.getBinaryFunctions()) {
if (!ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
@@ -428,7 +428,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
for (auto YamlBF : YamlBP.Functions) {
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
-if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
+if (It != StrictBinaryFunctionHashes.end() &&
+!ProfiledFunctions.count(It->second)) {
auto *BF = It->second;
matchProfileToFunction(YamlBF, *BF);
}
>From 8e7b2229a69c3795e723404c56e0d4298eef412a Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:55:58 -0700
Subject: [PATCH 3/5] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 +-
bolt/test/X86/profile-passthrough-block.test | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 56474a67307ed..779d60bce3b66 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -421,7 +421,7 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
StrictBinaryFunctionHashes.reserve(BC.getBinaryFunctions().size());
for (auto &[_, BF] : BC.getBinaryFunctions()) {
-if (!ProfiledFunctions.count(&BF))
+if (ProfiledFunctions.count(&BF))
continue;
StrictBinaryFunctionHashes[BF.getHash()] = &BF;
}
diff --git a/bolt/test/X86/profile-passthrough-block.test
b/bolt/test/X86/profile-passthrough-block.test
index 1b875885260dc..ed2a8117ddfc4 100644
--- a/bolt/test/X86/profile-passthrough-block.test
+++ b/bolt/test/X86/profile-passthrough-block.test
@@ -57,7 +57,7 @@ header:
functions:
- name:main
fid: 0
-hash:0x
+hash:0x0001
exec:1
nblocks: 6
blocks:
>From ef5f0dac9185dbb7a62345938d4f309c3379a85d Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 15:58:22 -0700
Subject: [PATCH 4/5] spr amend
Created using spr 1.3.4
---
bolt/lib/Profile/YAMLProfileReader.cpp | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index 779d60bce3b66..e3d30bfdb74e4 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -427,6 +427,8 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
}
for (auto YamlBF : YamlBP.Functions) {
+if (YamlBF.Used)
+ continue;
auto It = StrictBinaryFunctionHashes.find(YamlBF.Hash);
if (It != StrictBinaryFunctionHashes.end() &&
!ProfiledFunctions.count(It->second)) {
>From 41ce2897a445e47dfe685da66b4af080824e78ed Mon Sep 17 00:00:00 2001
From: shawbyoung
Date: Mon, 17 Jun 2024 16:00:27 -0700
Subject: [PATCH 5/5] spr amend
Created using spr 1.3.4
---
bolt/test/X86/profile-passthrough-block.test | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
dif
[llvm-branch-commits] [clang] [serialization] No transitive type change (PR #92511)
ChuanqiXu9 wrote: @jansvoboda11 @ilya-biryukov would you like to take look? https://github.com/llvm/llvm-project/pull/92511 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [BOLT] Name similarity function matching (PR #95884)
https://github.com/shawbyoung created https://github.com/llvm/llvm-project/pull/95884 Matching functions based on edit distance. Test Plan: tbd ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [BOLT] Name similarity function matching (PR #95884)
llvmbot wrote:
@llvm/pr-subscribers-bolt
Author: shaw young (shawbyoung)
Changes
Matching functions based on edit
distance.
Test Plan: tbd
---
Full diff: https://github.com/llvm/llvm-project/pull/95884.diff
1 Files Affected:
- (modified) bolt/lib/Profile/YAMLProfileReader.cpp (+29)
``diff
diff --git a/bolt/lib/Profile/YAMLProfileReader.cpp
b/bolt/lib/Profile/YAMLProfileReader.cpp
index f25f59201f1cd..a4f8ba9a440b6 100644
--- a/bolt/lib/Profile/YAMLProfileReader.cpp
+++ b/bolt/lib/Profile/YAMLProfileReader.cpp
@@ -13,6 +13,7 @@
#include "bolt/Profile/ProfileYAMLMapping.h"
#include "bolt/Utils/Utils.h"
#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/edit_distance.h"
#include "llvm/Support/CommandLine.h"
using namespace llvm;
@@ -23,6 +24,10 @@ extern cl::opt Verbosity;
extern cl::OptionCategory BoltOptCategory;
extern cl::opt InferStaleProfile;
+cl::opt NameSimilarityFunctionMatchingThreshold(
+"name-similarity-function-matching-threshold", cl::desc("edit distance."),
+cl::init(0), cl::Hidden, cl::cat(BoltOptCategory));
+
static llvm::cl::opt
IgnoreHash("profile-ignore-hash",
cl::desc("ignore hash while reading function profile"),
@@ -415,6 +420,30 @@ Error YAMLProfileReader::readProfile(BinaryContext &BC) {
if (!YamlBF.Used && BF && !ProfiledFunctions.count(BF))
matchProfileToFunction(YamlBF, *BF);
+ // Uses name similarity to match functions that were not matched by name.
+ for (auto YamlBF : YamlBP.Functions) {
+if (YamlBF.Used)
+ continue;
+
+unsigned MinEditDistance = UINT_MAX;
+BinaryFunction *ClosestNameBF = nullptr;
+
+for (auto &[_, BF] : BC.getBinaryFunctions()) {
+ if (ProfiledFunctions.count(&BF))
+continue;
+
+ unsigned BFEditDistance = BF.getOneName().edit_distance(YamlBF.Name);
+ if (BFEditDistance < MinEditDistance) {
+MinEditDistance = BFEditDistance;
+ClosestNameBF = &BF;
+ }
+}
+
+if (ClosestNameBF &&
+ MinEditDistance < opts::NameSimilarityFunctionMatchingThreshold)
+ matchProfileToFunction(YamlBF, *ClosestNameBF);
+ }
+
for (yaml::bolt::BinaryFunctionProfile &YamlBF : YamlBP.Functions)
if (!YamlBF.Used && opts::Verbosity >= 1)
errs() << "BOLT-WARNING: profile ignored for function " << YamlBF.Name
``
https://github.com/llvm/llvm-project/pull/95884
___
llvm-branch-commits mailing list
llvm-branch-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Support local atomicrmw fmin/fmax for float/double (PR #95590)
https://github.com/arsenm edited https://github.com/llvm/llvm-project/pull/95590 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
[llvm-branch-commits] [llvm] AMDGPU: Start selecting flat/global atomicrmw fmin/fmax. (PR #95592)
https://github.com/arsenm edited https://github.com/llvm/llvm-project/pull/95592 ___ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
