date:20250515

[llvm-branch-commits] [mlir] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread Tom Eccles via llvm-branch-commits


https://github.com/tblah created 
https://github.com/llvm/llvm-project/pull/140090

Part of a series to fix https://github.com/llvm/llvm-project/issues/136357



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-mlir-llvm

Author: Tom Eccles (tblah)


Changes

Part of a series to fix https://github.com/llvm/llvm-project/issues/136357

---
Full diff: https://github.com/llvm/llvm-project/pull/140090.diff


2 Files Affected:

- (modified) 
mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp (+25-12) 
- (modified) mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir (+3-1) 


``diff
diff --git 
a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp 
b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index 9f7b5605556e6..65d496ad8b774 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -1512,10 +1512,11 @@ allocatePrivateVars(llvm::IRBuilderBase &builder,
 }
 
 static LogicalResult copyFirstPrivateVars(
-llvm::IRBuilderBase &builder, LLVM::ModuleTranslation &moduleTranslation,
+mlir::Operation *op, llvm::IRBuilderBase &builder,
+LLVM::ModuleTranslation &moduleTranslation,
 SmallVectorImpl &mlirPrivateVars,
 ArrayRef llvmPrivateVars,
-SmallVectorImpl &privateDecls,
+SmallVectorImpl &privateDecls, bool insertBarrier,
 llvm::DenseMap *mappedPrivateVars = nullptr) {
   // Apply copy region for firstprivate.
   bool needsFirstprivate =
@@ -1563,6 +1564,14 @@ static LogicalResult copyFirstPrivateVars(
 moduleTranslation.forgetMapping(copyRegion);
   }
 
+  if (insertBarrier) {
+llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
+llvm::OpenMPIRBuilder::InsertPointOrErrorTy res =
+ompBuilder->createBarrier(builder.saveIP(), llvm::omp::OMPD_barrier);
+if (failed(handleError(res, *op)))
+  return failure();
+  }
+
   return success();
 }
 
@@ -2171,8 +2180,9 @@ convertOmpTaskOp(omp::TaskOp taskOp, llvm::IRBuilderBase 
&builder,
   // firstprivate copy region
   setInsertPointForPossiblyEmptyBlock(builder, copyBlock);
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers)))
+  taskOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers,
+  taskOp.getPrivateNeedsBarrier(
 return llvm::failure();
 
   // Set up for call to createTask()
@@ -2392,8 +2402,9 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase 
&builder,
 return failure();
 
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+  wsloopOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+  wsloopOp.getPrivateNeedsBarrier(
 return failure();
 
   assert(afterAllocas.get()->getSinglePredecessor());
@@ -2512,8 +2523,9 @@ convertOmpParallel(omp::ParallelOp opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
-privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+opInst, builder, moduleTranslation, privateVarsInfo.mlirVars,
+privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+opInst.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 if (failed(
@@ -4461,8 +4473,9 @@ convertOmpDistribute(Operation &opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privVarsInfo.mlirVars,
-privVarsInfo.llvmVars, privVarsInfo.privatizers)))
+distributeOp, builder, moduleTranslation, privVarsInfo.mlirVars,
+privVarsInfo.llvmVars, privVarsInfo.privatizers,
+distributeOp.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
@@ -5222,9 +5235,9 @@ convertOmpTarget(Operation &opInst, llvm::IRBuilderBase 
&builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
+targetOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
 privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
-&mappedPrivateVars)))
+targetOp.getPrivateNeedsBarrier(), &mappedPrivateVars)))
   return llvm::make_error();
 
 SmallVector privateCleanupRegions;
diff --git a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir 
b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
index 23a0ae5713aa2..0b1f45ad7ce1c 100644
--- a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
@@

[llvm-branch-commits] [mlir] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-mlir

Author: Tom Eccles (tblah)


Changes

Part of a series to fix https://github.com/llvm/llvm-project/issues/136357

---
Full diff: https://github.com/llvm/llvm-project/pull/140090.diff


2 Files Affected:

- (modified) 
mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp (+25-12) 
- (modified) mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir (+3-1) 


``diff
diff --git 
a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp 
b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index 9f7b5605556e6..65d496ad8b774 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -1512,10 +1512,11 @@ allocatePrivateVars(llvm::IRBuilderBase &builder,
 }
 
 static LogicalResult copyFirstPrivateVars(
-llvm::IRBuilderBase &builder, LLVM::ModuleTranslation &moduleTranslation,
+mlir::Operation *op, llvm::IRBuilderBase &builder,
+LLVM::ModuleTranslation &moduleTranslation,
 SmallVectorImpl &mlirPrivateVars,
 ArrayRef llvmPrivateVars,
-SmallVectorImpl &privateDecls,
+SmallVectorImpl &privateDecls, bool insertBarrier,
 llvm::DenseMap *mappedPrivateVars = nullptr) {
   // Apply copy region for firstprivate.
   bool needsFirstprivate =
@@ -1563,6 +1564,14 @@ static LogicalResult copyFirstPrivateVars(
 moduleTranslation.forgetMapping(copyRegion);
   }
 
+  if (insertBarrier) {
+llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
+llvm::OpenMPIRBuilder::InsertPointOrErrorTy res =
+ompBuilder->createBarrier(builder.saveIP(), llvm::omp::OMPD_barrier);
+if (failed(handleError(res, *op)))
+  return failure();
+  }
+
   return success();
 }
 
@@ -2171,8 +2180,9 @@ convertOmpTaskOp(omp::TaskOp taskOp, llvm::IRBuilderBase 
&builder,
   // firstprivate copy region
   setInsertPointForPossiblyEmptyBlock(builder, copyBlock);
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers)))
+  taskOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers,
+  taskOp.getPrivateNeedsBarrier(
 return llvm::failure();
 
   // Set up for call to createTask()
@@ -2392,8 +2402,9 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase 
&builder,
 return failure();
 
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+  wsloopOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+  wsloopOp.getPrivateNeedsBarrier(
 return failure();
 
   assert(afterAllocas.get()->getSinglePredecessor());
@@ -2512,8 +2523,9 @@ convertOmpParallel(omp::ParallelOp opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
-privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+opInst, builder, moduleTranslation, privateVarsInfo.mlirVars,
+privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+opInst.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 if (failed(
@@ -4461,8 +4473,9 @@ convertOmpDistribute(Operation &opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privVarsInfo.mlirVars,
-privVarsInfo.llvmVars, privVarsInfo.privatizers)))
+distributeOp, builder, moduleTranslation, privVarsInfo.mlirVars,
+privVarsInfo.llvmVars, privVarsInfo.privatizers,
+distributeOp.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
@@ -5222,9 +5235,9 @@ convertOmpTarget(Operation &opInst, llvm::IRBuilderBase 
&builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
+targetOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
 privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
-&mappedPrivateVars)))
+targetOp.getPrivateNeedsBarrier(), &mappedPrivateVars)))
   return llvm::make_error();
 
 SmallVector privateCleanupRegions;
diff --git a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir 
b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
index 23a0ae5713aa2..0b1f45ad7ce1c 100644
--- a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
@@ -37,7

[llvm-branch-commits] [mlir] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-flang-openmp

Author: Tom Eccles (tblah)


Changes

Part of a series to fix https://github.com/llvm/llvm-project/issues/136357

---
Full diff: https://github.com/llvm/llvm-project/pull/140090.diff


2 Files Affected:

- (modified) 
mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp (+25-12) 
- (modified) mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir (+3-1) 


``diff
diff --git 
a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp 
b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index 9f7b5605556e6..65d496ad8b774 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -1512,10 +1512,11 @@ allocatePrivateVars(llvm::IRBuilderBase &builder,
 }
 
 static LogicalResult copyFirstPrivateVars(
-llvm::IRBuilderBase &builder, LLVM::ModuleTranslation &moduleTranslation,
+mlir::Operation *op, llvm::IRBuilderBase &builder,
+LLVM::ModuleTranslation &moduleTranslation,
 SmallVectorImpl &mlirPrivateVars,
 ArrayRef llvmPrivateVars,
-SmallVectorImpl &privateDecls,
+SmallVectorImpl &privateDecls, bool insertBarrier,
 llvm::DenseMap *mappedPrivateVars = nullptr) {
   // Apply copy region for firstprivate.
   bool needsFirstprivate =
@@ -1563,6 +1564,14 @@ static LogicalResult copyFirstPrivateVars(
 moduleTranslation.forgetMapping(copyRegion);
   }
 
+  if (insertBarrier) {
+llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
+llvm::OpenMPIRBuilder::InsertPointOrErrorTy res =
+ompBuilder->createBarrier(builder.saveIP(), llvm::omp::OMPD_barrier);
+if (failed(handleError(res, *op)))
+  return failure();
+  }
+
   return success();
 }
 
@@ -2171,8 +2180,9 @@ convertOmpTaskOp(omp::TaskOp taskOp, llvm::IRBuilderBase 
&builder,
   // firstprivate copy region
   setInsertPointForPossiblyEmptyBlock(builder, copyBlock);
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers)))
+  taskOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers,
+  taskOp.getPrivateNeedsBarrier(
 return llvm::failure();
 
   // Set up for call to createTask()
@@ -2392,8 +2402,9 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase 
&builder,
 return failure();
 
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+  wsloopOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+  wsloopOp.getPrivateNeedsBarrier(
 return failure();
 
   assert(afterAllocas.get()->getSinglePredecessor());
@@ -2512,8 +2523,9 @@ convertOmpParallel(omp::ParallelOp opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
-privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+opInst, builder, moduleTranslation, privateVarsInfo.mlirVars,
+privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+opInst.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 if (failed(
@@ -4461,8 +4473,9 @@ convertOmpDistribute(Operation &opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privVarsInfo.mlirVars,
-privVarsInfo.llvmVars, privVarsInfo.privatizers)))
+distributeOp, builder, moduleTranslation, privVarsInfo.mlirVars,
+privVarsInfo.llvmVars, privVarsInfo.privatizers,
+distributeOp.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
@@ -5222,9 +5235,9 @@ convertOmpTarget(Operation &opInst, llvm::IRBuilderBase 
&builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
+targetOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
 privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
-&mappedPrivateVars)))
+targetOp.getPrivateNeedsBarrier(), &mappedPrivateVars)))
   return llvm::make_error();
 
 SmallVector privateCleanupRegions;
diff --git a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir 
b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
index 23a0ae5713aa2..0b1f45ad7ce1c 100644
--- a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir

[llvm-branch-commits] [mlir] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-mlir-openmp

Author: Tom Eccles (tblah)


Changes

Part of a series to fix https://github.com/llvm/llvm-project/issues/136357

---
Full diff: https://github.com/llvm/llvm-project/pull/140090.diff


2 Files Affected:

- (modified) 
mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp (+25-12) 
- (modified) mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir (+3-1) 


``diff
diff --git 
a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp 
b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index 9f7b5605556e6..65d496ad8b774 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -1512,10 +1512,11 @@ allocatePrivateVars(llvm::IRBuilderBase &builder,
 }
 
 static LogicalResult copyFirstPrivateVars(
-llvm::IRBuilderBase &builder, LLVM::ModuleTranslation &moduleTranslation,
+mlir::Operation *op, llvm::IRBuilderBase &builder,
+LLVM::ModuleTranslation &moduleTranslation,
 SmallVectorImpl &mlirPrivateVars,
 ArrayRef llvmPrivateVars,
-SmallVectorImpl &privateDecls,
+SmallVectorImpl &privateDecls, bool insertBarrier,
 llvm::DenseMap *mappedPrivateVars = nullptr) {
   // Apply copy region for firstprivate.
   bool needsFirstprivate =
@@ -1563,6 +1564,14 @@ static LogicalResult copyFirstPrivateVars(
 moduleTranslation.forgetMapping(copyRegion);
   }
 
+  if (insertBarrier) {
+llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
+llvm::OpenMPIRBuilder::InsertPointOrErrorTy res =
+ompBuilder->createBarrier(builder.saveIP(), llvm::omp::OMPD_barrier);
+if (failed(handleError(res, *op)))
+  return failure();
+  }
+
   return success();
 }
 
@@ -2171,8 +2180,9 @@ convertOmpTaskOp(omp::TaskOp taskOp, llvm::IRBuilderBase 
&builder,
   // firstprivate copy region
   setInsertPointForPossiblyEmptyBlock(builder, copyBlock);
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers)))
+  taskOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  taskStructMgr.getLLVMPrivateVarGEPs(), privateVarsInfo.privatizers,
+  taskOp.getPrivateNeedsBarrier(
 return llvm::failure();
 
   // Set up for call to createTask()
@@ -2392,8 +2402,9 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase 
&builder,
 return failure();
 
   if (failed(copyFirstPrivateVars(
-  builder, moduleTranslation, privateVarsInfo.mlirVars,
-  privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+  wsloopOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
+  privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+  wsloopOp.getPrivateNeedsBarrier(
 return failure();
 
   assert(afterAllocas.get()->getSinglePredecessor());
@@ -2512,8 +2523,9 @@ convertOmpParallel(omp::ParallelOp opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
-privateVarsInfo.llvmVars, privateVarsInfo.privatizers)))
+opInst, builder, moduleTranslation, privateVarsInfo.mlirVars,
+privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
+opInst.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 if (failed(
@@ -4461,8 +4473,9 @@ convertOmpDistribute(Operation &opInst, 
llvm::IRBuilderBase &builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privVarsInfo.mlirVars,
-privVarsInfo.llvmVars, privVarsInfo.privatizers)))
+distributeOp, builder, moduleTranslation, privVarsInfo.mlirVars,
+privVarsInfo.llvmVars, privVarsInfo.privatizers,
+distributeOp.getPrivateNeedsBarrier(
   return llvm::make_error();
 
 llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
@@ -5222,9 +5235,9 @@ convertOmpTarget(Operation &opInst, llvm::IRBuilderBase 
&builder,
   return llvm::make_error();
 
 if (failed(copyFirstPrivateVars(
-builder, moduleTranslation, privateVarsInfo.mlirVars,
+targetOp, builder, moduleTranslation, privateVarsInfo.mlirVars,
 privateVarsInfo.llvmVars, privateVarsInfo.privatizers,
-&mappedPrivateVars)))
+targetOp.getPrivateNeedsBarrier(), &mappedPrivateVars)))
   return llvm::make_error();
 
 SmallVector privateCleanupRegions;
diff --git a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir 
b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
index 23a0ae5713aa2..0b1f45ad7ce1c 100644
--- a/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-wsloop-private.mlir
@

[llvm-branch-commits] [flang] [flang][OpenMP] use attribute for delayed privatization barrier (PR #140092)

2025-05-15 Thread Tom Eccles via llvm-branch-commits


https://github.com/tblah created 
https://github.com/llvm/llvm-project/pull/140092

Fixes #136357

The barrier needs to go between the copying into firstprivate variables and the 
initialization call for the OpenMP construct (e.g. wsloop). There is no way of 
expressing this in MLIR because for delayed privatization that is all implicit 
(added in MLIR->LLVMIR conversion).

The previous approach put the barrier immediately before the wsloop (or 
similar). For delayed privatization, the firstprivate copy code would then be 
inserted after that, opening the possibility for the race observed in the bug 
report.

This patch solves the issue by instead setting an attribute on the mlir 
operation, which will instruct openmp dialect to llvm ir conversion to insert a 
barrier in the correct place.



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Paul Kirth via llvm-branch-commits


ilovepi wrote:

Well, with my patch the program in initial bug report from #70703 is passing w/ 
this patch, and the rest of the test suite seems OK ... maybe its fine to clone 
now? 


https://github.com/llvm/llvm-project/pull/13
___
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] [StaticDataLayout][PGO]Implement reader and writer change for data access profiles (PR #139997)

2025-05-15 Thread Mingming Liu via llvm-branch-commits


https://github.com/mingmingl-llvm updated 
https://github.com/llvm/llvm-project/pull/139997

>From d15ae3e73589b50edeb2ab823976734eaa3806e5 Mon Sep 17 00:00:00 2001
From: mingmingl 
Date: Thu, 15 May 2025 09:56:52 -0700
Subject: [PATCH] Support reading and writing data access profiles in memprof
 v4.

---
 .../include/llvm/ProfileData/DataAccessProf.h | 12 +++-
 .../llvm/ProfileData/IndexedMemProfData.h | 12 +++-
 .../llvm/ProfileData/InstrProfReader.h|  6 +-
 .../llvm/ProfileData/InstrProfWriter.h|  6 ++
 llvm/include/llvm/ProfileData/MemProfReader.h | 15 +
 llvm/include/llvm/ProfileData/MemProfYAML.h   | 65 +++
 llvm/lib/ProfileData/DataAccessProf.cpp   |  6 +-
 llvm/lib/ProfileData/IndexedMemProfData.cpp   | 61 +
 llvm/lib/ProfileData/InstrProfReader.cpp  | 14 
 llvm/lib/ProfileData/InstrProfWriter.cpp  | 20 --
 llvm/lib/ProfileData/MemProfReader.cpp| 30 +
 .../tools/llvm-profdata/memprof-yaml.test | 11 
 llvm/tools/llvm-profdata/llvm-profdata.cpp|  4 ++
 .../ProfileData/DataAccessProfTest.cpp| 11 ++--
 14 files changed, 242 insertions(+), 31 deletions(-)

diff --git a/llvm/include/llvm/ProfileData/DataAccessProf.h 
b/llvm/include/llvm/ProfileData/DataAccessProf.h
index e8504102238d1..f5f6abf0a2817 100644
--- a/llvm/include/llvm/ProfileData/DataAccessProf.h
+++ b/llvm/include/llvm/ProfileData/DataAccessProf.h
@@ -41,6 +41,8 @@ namespace data_access_prof {
 struct SourceLocation {
   SourceLocation(StringRef FileNameRef, uint32_t Line)
   : FileName(FileNameRef.str()), Line(Line) {}
+
+  SourceLocation() {}
   /// The filename where the data is located.
   std::string FileName;
   /// The line number in the source code.
@@ -53,6 +55,8 @@ namespace internal {
 // which strings are owned by `DataAccessProfData`. Used by 
`DataAccessProfData`
 // to represent data locations internally.
 struct SourceLocationRef {
+  SourceLocationRef(StringRef FileNameRef, uint32_t Line)
+  : FileName(FileNameRef), Line(Line) {}
   // The filename where the data is located.
   StringRef FileName;
   // The line number in the source code.
@@ -100,8 +104,9 @@ using SymbolHandle = std::variant;
 /// The data access profiles for a symbol.
 struct DataAccessProfRecord {
 public:
-  DataAccessProfRecord(SymbolHandleRef SymHandleRef,
-   ArrayRef LocRefs) {
+  DataAccessProfRecord(SymbolHandleRef SymHandleRef, uint64_t AccessCount,
+   ArrayRef LocRefs)
+  : AccessCount(AccessCount) {
 if (std::holds_alternative(SymHandleRef)) {
   SymHandle = std::get(SymHandleRef).str();
 } else
@@ -110,8 +115,9 @@ struct DataAccessProfRecord {
 for (auto Loc : LocRefs)
   Locations.push_back(SourceLocation(Loc.FileName, Loc.Line));
   }
+  DataAccessProfRecord() {}
   SymbolHandle SymHandle;
-
+  uint64_t AccessCount;
   // The locations of data in the source code. Optional.
   SmallVector Locations;
 };
diff --git a/llvm/include/llvm/ProfileData/IndexedMemProfData.h 
b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
index 3c6c329d1c49d..66fa38472059b 100644
--- a/llvm/include/llvm/ProfileData/IndexedMemProfData.h
+++ b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
@@ -10,14 +10,20 @@
 //
 
//===--===//
 
+#include "llvm/ProfileData/DataAccessProf.h"
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/MemProf.h"
 
+#include 
+#include 
+
 namespace llvm {
 
 // Write the MemProf data to OS.
-Error writeMemProf(ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
-   memprof::IndexedVersion MemProfVersionRequested,
-   bool MemProfFullSchema);
+Error writeMemProf(
+ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
+memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema,
+std::optional>
+DataAccessProfileData);
 
 } // namespace llvm
diff --git a/llvm/include/llvm/ProfileData/InstrProfReader.h 
b/llvm/include/llvm/ProfileData/InstrProfReader.h
index c250a9ede39bc..210df6be46f04 100644
--- a/llvm/include/llvm/ProfileData/InstrProfReader.h
+++ b/llvm/include/llvm/ProfileData/InstrProfReader.h
@@ -18,6 +18,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/IR/ProfileSummary.h"
 #include "llvm/Object/BuildID.h"
+#include "llvm/ProfileData/DataAccessProf.h"
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/InstrProfCorrelator.h"
 #include "llvm/ProfileData/MemProf.h"
@@ -704,9 +705,12 @@ class IndexedMemProfReader {
   // The number of elements in the radix tree array.
   unsigned RadixTreeSize = 0;
 
+  std::unique_ptr DataAccessProfileData;
+
   Error deserializeV2(const unsigned char *Start, const unsigned char *Ptr);
   Error deserializeRadixTreeBased(const unsigned char *Start,
-  const unsigned char *Ptr);
+

[llvm-branch-commits] [llvm] [StaticDataLayout][PGO]Implement reader and writer change for data access profiles (PR #139997)

2025-05-15 Thread Mingming Liu via llvm-branch-commits


https://github.com/mingmingl-llvm updated 
https://github.com/llvm/llvm-project/pull/139997

>From 4045c943966609cc9a92693752af0e29a19e1ef9 Mon Sep 17 00:00:00 2001
From: mingmingl 
Date: Thu, 15 May 2025 09:56:52 -0700
Subject: [PATCH] Support reading and writing data access profiles in memprof
 v4.

---
 .../include/llvm/ProfileData/DataAccessProf.h | 12 +++-
 .../llvm/ProfileData/IndexedMemProfData.h | 12 +++-
 .../llvm/ProfileData/InstrProfReader.h|  6 +-
 .../llvm/ProfileData/InstrProfWriter.h|  6 ++
 llvm/include/llvm/ProfileData/MemProfReader.h | 15 +
 llvm/include/llvm/ProfileData/MemProfYAML.h   | 58 ++
 llvm/lib/ProfileData/DataAccessProf.cpp   |  6 +-
 llvm/lib/ProfileData/IndexedMemProfData.cpp   | 61 +++
 llvm/lib/ProfileData/InstrProfReader.cpp  | 14 +
 llvm/lib/ProfileData/InstrProfWriter.cpp  | 20 --
 llvm/lib/ProfileData/MemProfReader.cpp| 30 +
 .../tools/llvm-profdata/memprof-yaml.test | 11 
 llvm/tools/llvm-profdata/llvm-profdata.cpp|  4 ++
 .../ProfileData/DataAccessProfTest.cpp| 11 ++--
 14 files changed, 235 insertions(+), 31 deletions(-)

diff --git a/llvm/include/llvm/ProfileData/DataAccessProf.h 
b/llvm/include/llvm/ProfileData/DataAccessProf.h
index e8504102238d1..f5f6abf0a2817 100644
--- a/llvm/include/llvm/ProfileData/DataAccessProf.h
+++ b/llvm/include/llvm/ProfileData/DataAccessProf.h
@@ -41,6 +41,8 @@ namespace data_access_prof {
 struct SourceLocation {
   SourceLocation(StringRef FileNameRef, uint32_t Line)
   : FileName(FileNameRef.str()), Line(Line) {}
+
+  SourceLocation() {}
   /// The filename where the data is located.
   std::string FileName;
   /// The line number in the source code.
@@ -53,6 +55,8 @@ namespace internal {
 // which strings are owned by `DataAccessProfData`. Used by 
`DataAccessProfData`
 // to represent data locations internally.
 struct SourceLocationRef {
+  SourceLocationRef(StringRef FileNameRef, uint32_t Line)
+  : FileName(FileNameRef), Line(Line) {}
   // The filename where the data is located.
   StringRef FileName;
   // The line number in the source code.
@@ -100,8 +104,9 @@ using SymbolHandle = std::variant;
 /// The data access profiles for a symbol.
 struct DataAccessProfRecord {
 public:
-  DataAccessProfRecord(SymbolHandleRef SymHandleRef,
-   ArrayRef LocRefs) {
+  DataAccessProfRecord(SymbolHandleRef SymHandleRef, uint64_t AccessCount,
+   ArrayRef LocRefs)
+  : AccessCount(AccessCount) {
 if (std::holds_alternative(SymHandleRef)) {
   SymHandle = std::get(SymHandleRef).str();
 } else
@@ -110,8 +115,9 @@ struct DataAccessProfRecord {
 for (auto Loc : LocRefs)
   Locations.push_back(SourceLocation(Loc.FileName, Loc.Line));
   }
+  DataAccessProfRecord() {}
   SymbolHandle SymHandle;
-
+  uint64_t AccessCount;
   // The locations of data in the source code. Optional.
   SmallVector Locations;
 };
diff --git a/llvm/include/llvm/ProfileData/IndexedMemProfData.h 
b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
index 3c6c329d1c49d..66fa38472059b 100644
--- a/llvm/include/llvm/ProfileData/IndexedMemProfData.h
+++ b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
@@ -10,14 +10,20 @@
 //
 
//===--===//
 
+#include "llvm/ProfileData/DataAccessProf.h"
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/MemProf.h"
 
+#include 
+#include 
+
 namespace llvm {
 
 // Write the MemProf data to OS.
-Error writeMemProf(ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
-   memprof::IndexedVersion MemProfVersionRequested,
-   bool MemProfFullSchema);
+Error writeMemProf(
+ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
+memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema,
+std::optional>
+DataAccessProfileData);
 
 } // namespace llvm
diff --git a/llvm/include/llvm/ProfileData/InstrProfReader.h 
b/llvm/include/llvm/ProfileData/InstrProfReader.h
index c250a9ede39bc..a3436e1dfe711 100644
--- a/llvm/include/llvm/ProfileData/InstrProfReader.h
+++ b/llvm/include/llvm/ProfileData/InstrProfReader.h
@@ -18,6 +18,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/IR/ProfileSummary.h"
 #include "llvm/Object/BuildID.h"
+#include "llvm/ProfileData/DataAccessProf.h"
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/InstrProfCorrelator.h"
 #include "llvm/ProfileData/MemProf.h"
@@ -703,10 +704,13 @@ class IndexedMemProfReader {
   const unsigned char *CallStackBase = nullptr;
   // The number of elements in the radix tree array.
   unsigned RadixTreeSize = 0;
+  /// The data access profiles, deserialized from binary data.
+  std::unique_ptr DataAccessProfileData;
 
   Error deserializeV2(const unsigned char *Start, const unsigned char *Ptr);
   Error deserializeRadixTreeBased(const u

[llvm-branch-commits] Add pointer field protection feature. (PR #133538)

2025-05-15 Thread Peter Collingbourne via llvm-branch-commits



@@ -362,6 +362,17 @@ class LangOptionsBase {
 BKey
   };
 
+  enum class PointerFieldProtectionKind {

pcc wrote:

I think that allowing this level of customization should be implemented as part 
of the separate opt-in solution (e.g. it may be a property of the qualifier). 
It is generally a reasonable assumption that operator new is the same for all 
types. In the unlikely case that it isn't, we aren't really doing anything 
wrong by using the malloc-derived decision here. It just means that we lose 
some entropy from the type or the pointer tag.

https://github.com/llvm/llvm-project/pull/133538
___
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] Add pointer field protection feature. (PR #133538)

2025-05-15 Thread Oliver Hunt via llvm-branch-commits

ojhunt wrote:

> Hi Oliver, thanks for your comments! I'll address them below.
> 
> > Thoughts:
> > This should be opt-in on a field or struct granularity, not just a global 
> > behavior.
> 
> This would certainly be easier if it were an opt-in behavior, as it would 
> allow avoiding a substantial amount of complexity that you point out below. 
> The problem is that would not make the solution very effective as a UAF 
> mitigation, as it would only help with code that was opted in, which is 
> likely a tiny fraction of an entire codebase. That fraction is unlikely to be 
> the fraction with UAF bugs, not only because of the size of the fraction but 
> also because if developers aren't thinking about memory safety issues, they 
> certainly won't be manually opting into this. With PFP, the problem that I 
> set out to solve was to find a way to automatically protect as many pointers 
> stored in memory as possible in standards-conforming code.
> 

There are a significant number of standard layout types that contain 
meaningfully powerful data if they are compromised, which leads to saying the 
standard layout types should also be covered.

That said, there's a significant cost to applying pointer protection 
universally though. The cost of existing mitigations rely on them not being 
global properties as they can saturate the required compute resources.

I would be interested in a performance comparison of enumerating struct fields 
as you increase the number of fields being accessed just to see if there's a 
point that perf starts to collapse - I suspect there would be, but if the 
number of concurrent (from the cpu pov, not threading) fields accesses to get 
to that point is high enough it may not matter. Similar tests for increasing 
levels of dependent loads are also probably worth while. The intent is to make 
this behavior at least partially universal (all pointers in objects that are 
not standard layout), which means that extrapolating from individual loads, or 
loads from the same addresses, are not necessarily representative.

In the RFC you talk about the time and/or cycles required to perform 
operations, but that's the best case, where the required units are not 
saturated. This gate on performance isn't unique to pointer protections: every 
operation consumes some amount of cpu resources, once those resources are 
saturated the pipeline stalls, so it's necessary to compare performance when 
many of these operations are happening in succession.

Now there's some caching (saying the tag remains constant, etc) you might be 
able to do, but that's dependent on being willing to say "this is a software 
mitigation, so we're ok if we don't catch everything" - these are things that a 
cpu can mitigate in hardware via direct knowledge of what is happening on a 
given core, and peeking on the memory bus. (There are also performance costs 
when you're accessing the same memory on different cores, but that goes so bad 
so quickly even in normal cases I'm not concerned about that yet).

> > In the RFC I think you mentioned not applying PFP to C types, but I'm 
> > unsure how you're deciding what is a C type?
> 
> This is based on whether the type has standard layout. C does not permit 
> declaring a type that is non-standard layout.

The issue here is that there are many cases where standard layout structs 
contain important data, which is why I'm not 100% sold on excluding C types 
from this.

> > There are a lot of special cases being added in places that should not need 
> > to be aware of PFP - the core type queries should be returning correct 
> > values for types containing PFP fields.
> 
> Agreed on type queries. The current adjustment to how trivially-relocatable 
> is defined is a workaround that will be removed once P2786 is fully 
> implemented in libc++, and at that point we shouldn't need 
> `__has_non_relocatable_fields` either. The intent is that turning on PFP 
> should be invisible to the developer (in other words, the modification to the 
> storage format of pointer fields in certain structs is permitted by the as-if 
> rule). As a result, most of the implementation is in CodeGen and below.

My point was that this should not have a separate attempt to make trivial 
relocation work, it should make sure the type queries report the correct 
information, and the implementation of trivially_relocate agrees with that 
definition. Also trivial relocation isn't meaningfully supported or used 
anywhere yet - it's a new feature - so I don't think you should be prioritizing 
supporting that yet: just make sure that the (internal) trait queries return 
correct/consistent results for what can be handled, and worry about optimizing 
that later.

> 
> > A lot of the "this read/write/copy/init special work" exactly aligns with 
> > the constraints of pointer auth, and I think the overall implementation 
> > could be made better by introducing the concept of a 
> > `HardenedPointerQualifier` or similar,

[llvm-branch-commits] [clang] [llvm] [KeyInstr][Clang] Add ApplyAtomGroup (PR #134632)

2025-05-15 Thread Orlando Cazalet-Hyams via llvm-branch-commits

OCHyams wrote:

> This all seems fine -- I guess the plumbing here has to get in without a 
> test, before then later real changes come in and can be tested.
> 
> The "Override" vs new-source-atom distinction seems a little clunky, although 
> I haven't read how it's used to get the full context. IMO it's worth putting 
> thought into a better name: can we pick an abstract name describing the 
> purpose ("New key operation"?) where the override thing is just an 
> implementation feature?

Hmm you're right it's not very clear. The "override" situation is needed 
because of how returns are handled in Clang (ret atoms review - #134652), which 
is that multiple returns are emitted as branches to a single return-block. 
Those branches get the source location info for the return, and the actual ret 
is associated with the closing brace. However, if there's only one pred to the 
return block it's folded into it, and the return takes the source location of 
the branch. I can't remember exactly why the atom application had to be 
structured in this slightly convoluted way (should've written better 
comments!). I'll see if it can be simplified a bit, and if not I'll rename the 
functions and add better comments.

https://github.com/llvm/llvm-project/pull/134632
___
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] [KeyInstr][Clang] Add ApplyAtomGroup (PR #134632)

2025-05-15 Thread Orlando Cazalet-Hyams via llvm-branch-commits


https://github.com/OCHyams updated 
https://github.com/llvm/llvm-project/pull/134632

>From 0544aed12dd76bd64c42ce186c586702d3e01589 Mon Sep 17 00:00:00 2001
From: Orlando Cazalet-Hyams 
Date: Tue, 1 Apr 2025 11:59:24 +0100
Subject: [PATCH] [KeyInstr][Clang] Add ApplyAtomGroup

This is a scoped helper similar to ApplyDebugLocation that creates a new source
atom group which instructions can be added to.

A source atom is a source construct that is "interesting" for debug stepping
purposes. We use an atom group number to track the instruction(s) that implement
the functionality for the atom, plus backup instructions/source locations.

---

This patch is part of a stack that teaches Clang to generate Key Instructions
metadata for C and C++.

The Key Instructions project is introduced, including a "quick summary" section
at the top which adds context for this PR, here:
https://discourse.llvm.org/t/rfc-improving-is-stmt-placement-for-better-interactive-debugging/82668

The feature is only functional in LLVM if LLVM is built with CMake flag
LLVM_EXPERIMENTAL_KEY_INSTRUCTIONs. Eventually that flag will be removed.

The Clang-side work is demoed here:
https://github.com/llvm/llvm-project/pull/130943
---
 clang/lib/CodeGen/CGDebugInfo.cpp   | 119 +++-
 clang/lib/CodeGen/CGDebugInfo.h |  50 
 clang/lib/CodeGen/CodeGenFunction.h |  14 
 3 files changed, 182 insertions(+), 1 deletion(-)

diff --git a/clang/lib/CodeGen/CGDebugInfo.cpp 
b/clang/lib/CodeGen/CGDebugInfo.cpp
index 2a11eebf1b682..f4c5c57a38b3e 100644
--- a/clang/lib/CodeGen/CGDebugInfo.cpp
+++ b/clang/lib/CodeGen/CGDebugInfo.cpp
@@ -43,6 +43,7 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/DerivedTypes.h"
+#include "llvm/IR/Instruction.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/Intrinsics.h"
 #include "llvm/IR/Metadata.h"
@@ -52,6 +53,7 @@
 #include "llvm/Support/SHA1.h"
 #include "llvm/Support/SHA256.h"
 #include "llvm/Support/TimeProfiler.h"
+#include 
 #include 
 using namespace clang;
 using namespace clang::CodeGen;
@@ -119,6 +121,114 @@ CGDebugInfo::~CGDebugInfo() {
  "Region stack mismatch, stack not empty!");
 }
 
+void CGDebugInfo::addInstSourceAtomMetadata(llvm::Instruction *I,
+uint64_t Group, uint8_t Rank) {
+  if (!I->getDebugLoc() || Group == 0 || !I->getDebugLoc()->getLine())
+return;
+
+  // Saturate the 3-bit rank.
+  Rank = std::min(Rank, 7);
+
+  const llvm::DebugLoc &DL = I->getDebugLoc();
+
+  // Each instruction can only be attributed to one source atom (a limitation 
of
+  // the implementation). If this instruction is already part of a source atom,
+  // pick the group in which it has highest precedence (lowest rank).
+  if (DL.get()->getAtomGroup() && DL.get()->getAtomRank() &&
+  DL.get()->getAtomRank() < Rank) {
+Group = DL.get()->getAtomGroup();
+Rank = DL.get()->getAtomRank();
+  }
+
+  // Update the function-local watermark so we don't reuse this number for
+  // another atom.
+  KeyInstructionsInfo.HighestEmittedAtom =
+  std::max(Group, KeyInstructionsInfo.HighestEmittedAtom);
+
+  // Apply the new DILocation to the instruction.
+  llvm::DILocation *NewDL = llvm::DILocation::get(
+  I->getContext(), DL.getLine(), DL.getCol(), DL.getScope(),
+  DL.getInlinedAt(), DL.isImplicitCode(), Group, Rank);
+  I->setDebugLoc(NewDL);
+};
+
+void CGDebugInfo::addInstToCurrentSourceAtom(llvm::Instruction *KeyInstruction,
+ llvm::Value *Backup) {
+  if (!CGM.getCodeGenOpts().DebugKeyInstructions)
+return;
+
+  uint64_t Group = KeyInstructionsInfo.CurrentAtom;
+  if (!Group)
+return;
+
+  addInstSourceAtomMetadata(KeyInstruction, Group, /*Rank=*/1);
+
+  llvm::Instruction *BackupI =
+  llvm::dyn_cast_or_null(Backup);
+  if (!BackupI)
+return;
+
+  // Add the backup instruction to the group.
+  addInstSourceAtomMetadata(BackupI, Group, /*Rank=*/2);
+
+  // Look through chains of casts too, as they're probably going to evaporate.
+  // FIXME: And other nops like zero length geps?
+  // FIXME: Should use Cast->isNoopCast()?
+  uint8_t Rank = 3;
+  while (auto *Cast = dyn_cast(BackupI)) {
+BackupI = dyn_cast(Cast->getOperand(0));
+if (!BackupI)
+  break;
+addInstSourceAtomMetadata(BackupI, Group, Rank++);
+  }
+}
+
+void CGDebugInfo::addRetToOverrideOrNewSourceAtom(llvm::ReturnInst *Ret,
+  llvm::Value *Backup) {
+  if (KeyInstructionsInfo.RetAtomOverride) {
+uint64_t CurrentAtom = KeyInstructionsInfo.CurrentAtom;
+KeyInstructionsInfo.CurrentAtom = KeyInstructionsInfo.RetAtomOverride;
+addInstToCurrentSourceAtom(Ret, Backup);
+KeyInstructionsInfo.CurrentAtom = CurrentAtom;
+KeyInstructionsInfo.RetAtomOverride = 0;
+  } else {
+auto Grp = ApplyAtomGroup(this);
+addInstToCurrentSourceAtom(Ret, Backup);
+  }
+}
+
+void CGDe

[llvm-branch-commits] [mlir] [MLIR][OpenMP] Assert on map translation functions, NFC (PR #137199)

2025-05-15 Thread Sergio Afonso via llvm-branch-commits


https://github.com/skatrak updated 
https://github.com/llvm/llvm-project/pull/137199

>From c6954b3120a87eef7d9cf86f18d4ef342b2e7b25 Mon Sep 17 00:00:00 2001
From: Sergio Afonso 
Date: Fri, 11 Apr 2025 13:40:14 +0100
Subject: [PATCH] [MLIR][OpenMP] Assert on map translation functions, NFC

This patch adds assertions to map-related MLIR to LLVM IR translation functions
and utils to explicitly document whether they are intended for host or device
compilation only.

Over time, map-related handling has increased in complexity. This is compounded
by the fact that some handling is device-specific and some is host-specific. By
explicitly asserting on these functions on the expected compilation pass, the
flow should become slighlty easier to follow.
---
 .../OpenMP/OpenMPToLLVMIRTranslation.cpp  | 24 +--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git 
a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp 
b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index 9f7b5605556e6..010c46358f7df 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -3720,6 +3720,9 @@ static llvm::omp::OpenMPOffloadMappingFlags 
mapParentWithMembers(
 LLVM::ModuleTranslation &moduleTranslation, llvm::IRBuilderBase &builder,
 llvm::OpenMPIRBuilder &ompBuilder, DataLayout &dl, MapInfosTy 
&combinedInfo,
 MapInfoData &mapData, uint64_t mapDataIndex, bool isTargetParams) {
+  assert(!ompBuilder.Config.isTargetDevice() &&
+ "function only supported for host device codegen");
+
   // Map the first segment of our structure
   combinedInfo.Types.emplace_back(
   isTargetParams
@@ -3828,6 +3831,8 @@ static void processMapMembersWithParent(
 llvm::OpenMPIRBuilder &ompBuilder, DataLayout &dl, MapInfosTy 
&combinedInfo,
 MapInfoData &mapData, uint64_t mapDataIndex,
 llvm::omp::OpenMPOffloadMappingFlags memberOfFlag) {
+  assert(!ompBuilder.Config.isTargetDevice() &&
+ "function only supported for host device codegen");
 
   auto parentClause =
   llvm::cast(mapData.MapClause[mapDataIndex]);
@@ -3941,6 +3946,9 @@ static void 
processMapWithMembersOf(LLVM::ModuleTranslation &moduleTranslation,
 DataLayout &dl, MapInfosTy &combinedInfo,
 MapInfoData &mapData, uint64_t 
mapDataIndex,
 bool isTargetParams) {
+  assert(!ompBuilder.Config.isTargetDevice() &&
+ "function only supported for host device codegen");
+
   auto parentClause =
   llvm::cast(mapData.MapClause[mapDataIndex]);
 
@@ -3982,6 +3990,8 @@ static void
 createAlteredByCaptureMap(MapInfoData &mapData,
   LLVM::ModuleTranslation &moduleTranslation,
   llvm::IRBuilderBase &builder) {
+  assert(!moduleTranslation.getOpenMPBuilder()->Config.isTargetDevice() &&
+ "function only supported for host device codegen");
   for (size_t i = 0; i < mapData.MapClause.size(); ++i) {
 // if it's declare target, skip it, it's handled separately.
 if (!mapData.IsDeclareTarget[i]) {
@@ -4046,6 +4056,9 @@ static void genMapInfos(llvm::IRBuilderBase &builder,
 LLVM::ModuleTranslation &moduleTranslation,
 DataLayout &dl, MapInfosTy &combinedInfo,
 MapInfoData &mapData, bool isTargetParams = false) {
+  assert(!moduleTranslation.getOpenMPBuilder()->Config.isTargetDevice() &&
+ "function only supported for host device codegen");
+
   // We wish to modify some of the methods in which arguments are
   // passed based on their capture type by the target region, this can
   // involve generating new loads and stores, which changes the
@@ -4057,8 +4070,7 @@ static void genMapInfos(llvm::IRBuilderBase &builder,
   // kernel arg structure. It primarily becomes relevant in cases like
   // bycopy, or byref range'd arrays. In the default case, we simply
   // pass thee pointer byref as both basePointer and pointer.
-  if (!moduleTranslation.getOpenMPBuilder()->Config.isTargetDevice())
-createAlteredByCaptureMap(mapData, moduleTranslation, builder);
+  createAlteredByCaptureMap(mapData, moduleTranslation, builder);
 
   llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
 
@@ -4092,6 +4104,8 @@ emitUserDefinedMapper(Operation *declMapperOp, 
llvm::IRBuilderBase &builder,
 static llvm::Expected
 getOrCreateUserDefinedMapperFunc(Operation *op, llvm::IRBuilderBase &builder,
  LLVM::ModuleTranslation &moduleTranslation) {
+  assert(!moduleTranslation.getOpenMPBuilder()->Config.isTargetDevice() &&
+ "function only supported for host device codegen");
   auto declMapperOp = cast(op);
   std::string mapperFuncName =
   moduleTranslation.getOpenMPBuilder()->createPlatformSpecificName(
@@

[llvm-branch-commits] [llvm] [ConstraintElim] Simplify `usub_with_overflow` when A uge B (PR #135785)

2025-05-15 Thread Iris Shi via llvm-branch-commits


https://github.com/el-ev updated 
https://github.com/llvm/llvm-project/pull/135785

>From b5cac9305a6af33bc0bc30d01965c5e7e333 Mon Sep 17 00:00:00 2001
From: Iris Shi <0...@owo.li>
Date: Tue, 15 Apr 2025 20:20:45 +0800
Subject: [PATCH] [ConstraintElim] Simplify `usub_with_overflow` when A uge B

---
 .../Scalar/ConstraintElimination.cpp  | 11 
 .../usub-with-overflow.ll | 54 +++
 2 files changed, 44 insertions(+), 21 deletions(-)

diff --git a/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp 
b/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
index 15aba43721e57..71afd7a2203ef 100644
--- a/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
+++ b/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
@@ -1124,6 +1124,7 @@ void State::addInfoFor(BasicBlock &BB) {
 // Enqueue overflow intrinsics for simplification.
 case Intrinsic::sadd_with_overflow:
 case Intrinsic::ssub_with_overflow:
+case Intrinsic::usub_with_overflow:
 case Intrinsic::ucmp:
 case Intrinsic::scmp:
   WorkList.push_back(
@@ -1765,6 +1766,16 @@ tryToSimplifyOverflowMath(IntrinsicInst *II, 
ConstraintInfo &Info,
 Changed = true;
 break;
   }
+  case Intrinsic::usub_with_overflow: {
+// usub overflows iff A < B
+// TODO: If the operation is guaranteed to overflow, we could
+// also apply some simplifications.
+if (DoesConditionHold(CmpInst::ICMP_UGE, A, B, Info)) {
+  replaceAddOrSubOverflowUses(II, A, B, ToRemove);
+  Changed = true;
+}
+break;
+  }
   }
 
   return Changed;
diff --git a/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll 
b/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
index 06bfd8269d97d..722116cc6ebd0 100644
--- a/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
+++ b/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
@@ -9,12 +9,14 @@ define i8 @usub_no_overflow_due_to_cmp_condition(i8 %a, i8 
%b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp uge i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[MATH:%.*]], label [[EXIT_FAIL:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
-; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[TMP2:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
+; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[TMP2]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 } [[TMP2]], 0
+; CHECK-NEXT:ret i8 [[RES1]]
 ; CHECK:   exit.fail:
 ; CHECK-NEXT:ret i8 0
 ;
@@ -41,12 +43,14 @@ define i8 @usub_no_overflow_due_to_cmp_condition2(i8 %a, i8 
%b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp ule i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[EXIT_FAIL:%.*]], label [[MATH:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
-; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[TMP2:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
+; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[TMP2]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 } [[TMP2]], 0
+; CHECK-NEXT:ret i8 [[RES1]]
 ; CHECK:   exit.fail:
 ; CHECK-NEXT:ret i8 0
 ;
@@ -75,13 +79,15 @@ define i8 
@sub_no_overflow_due_to_cmp_condition_result_used(i8 %a, i8 %b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp ule i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[EXIT_FAIL:%.*]], label [[MATH:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[OP:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
 ; CHECK-NEXT:call void @use_res({ i8, i1 } [[OP]])
 ; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 }

[llvm-branch-commits] [llvm] [ConstraintElim] Simplify `usub_with_overflow` when A uge B (PR #135785)

2025-05-15 Thread Iris Shi via llvm-branch-commits


https://github.com/el-ev updated 
https://github.com/llvm/llvm-project/pull/135785

>From b5cac9305a6af33bc0bc30d01965c5e7e333 Mon Sep 17 00:00:00 2001
From: Iris Shi <0...@owo.li>
Date: Tue, 15 Apr 2025 20:20:45 +0800
Subject: [PATCH] [ConstraintElim] Simplify `usub_with_overflow` when A uge B

---
 .../Scalar/ConstraintElimination.cpp  | 11 
 .../usub-with-overflow.ll | 54 +++
 2 files changed, 44 insertions(+), 21 deletions(-)

diff --git a/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp 
b/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
index 15aba43721e57..71afd7a2203ef 100644
--- a/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
+++ b/llvm/lib/Transforms/Scalar/ConstraintElimination.cpp
@@ -1124,6 +1124,7 @@ void State::addInfoFor(BasicBlock &BB) {
 // Enqueue overflow intrinsics for simplification.
 case Intrinsic::sadd_with_overflow:
 case Intrinsic::ssub_with_overflow:
+case Intrinsic::usub_with_overflow:
 case Intrinsic::ucmp:
 case Intrinsic::scmp:
   WorkList.push_back(
@@ -1765,6 +1766,16 @@ tryToSimplifyOverflowMath(IntrinsicInst *II, 
ConstraintInfo &Info,
 Changed = true;
 break;
   }
+  case Intrinsic::usub_with_overflow: {
+// usub overflows iff A < B
+// TODO: If the operation is guaranteed to overflow, we could
+// also apply some simplifications.
+if (DoesConditionHold(CmpInst::ICMP_UGE, A, B, Info)) {
+  replaceAddOrSubOverflowUses(II, A, B, ToRemove);
+  Changed = true;
+}
+break;
+  }
   }
 
   return Changed;
diff --git a/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll 
b/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
index 06bfd8269d97d..722116cc6ebd0 100644
--- a/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
+++ b/llvm/test/Transforms/ConstraintElimination/usub-with-overflow.ll
@@ -9,12 +9,14 @@ define i8 @usub_no_overflow_due_to_cmp_condition(i8 %a, i8 
%b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp uge i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[MATH:%.*]], label [[EXIT_FAIL:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
-; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[TMP2:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
+; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[TMP2]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 } [[TMP2]], 0
+; CHECK-NEXT:ret i8 [[RES1]]
 ; CHECK:   exit.fail:
 ; CHECK-NEXT:ret i8 0
 ;
@@ -41,12 +43,14 @@ define i8 @usub_no_overflow_due_to_cmp_condition2(i8 %a, i8 
%b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp ule i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[EXIT_FAIL:%.*]], label [[MATH:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
-; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[TMP2:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
+; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[TMP2]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 } [[TMP2]], 0
+; CHECK-NEXT:ret i8 [[RES1]]
 ; CHECK:   exit.fail:
 ; CHECK-NEXT:ret i8 0
 ;
@@ -75,13 +79,15 @@ define i8 
@sub_no_overflow_due_to_cmp_condition_result_used(i8 %a, i8 %b) {
 ; CHECK-NEXT:[[C_1:%.*]] = icmp ule i8 [[B:%.*]], [[A:%.*]]
 ; CHECK-NEXT:br i1 [[C_1]], label [[EXIT_FAIL:%.*]], label [[MATH:%.*]]
 ; CHECK:   math:
-; CHECK-NEXT:[[OP:%.*]] = tail call { i8, i1 } 
@llvm.usub.with.overflow.i8(i8 [[B]], i8 [[A]])
+; CHECK-NEXT:[[RES:%.*]] = sub i8 [[B]], [[A]]
+; CHECK-NEXT:[[TMP1:%.*]] = insertvalue { i8, i1 } poison, i8 [[RES]], 0
+; CHECK-NEXT:[[OP:%.*]] = insertvalue { i8, i1 } [[TMP1]], i1 false, 1
 ; CHECK-NEXT:call void @use_res({ i8, i1 } [[OP]])
 ; CHECK-NEXT:[[STATUS:%.*]] = extractvalue { i8, i1 } [[OP]], 1
 ; CHECK-NEXT:br i1 [[STATUS]], label [[EXIT_FAIL]], label [[EXIT_OK:%.*]]
 ; CHECK:   exit.ok:
-; CHECK-NEXT:[[RES:%.*]] = extractvalue { i8, i1 } [[OP]], 0
-; CHECK-NEXT:ret i8 [[RES]]
+; CHECK-NEXT:[[RES1:%.*]] = extractvalue { i8, i1 }

[llvm-branch-commits] [llvm] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Nikita Popov via llvm-branch-commits


https://github.com/nikic commented:

Didn't we switch away from cloning because it breaks blockaddress somehow?

https://github.com/llvm/llvm-project/pull/13
___
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] [ConstraintElim] Simplify `usub_with_overflow` when A uge B (PR #135785)

2025-05-15 Thread Florian Hahn via llvm-branch-commits


https://github.com/fhahn commented:

Could you add an Alive2 proof showing this transform is valid given the 
precondition for completeness?

https://github.com/llvm/llvm-project/pull/135785
___
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][OpenMP] Improve handling of non-C/C++ directives (PR #139961)

2025-05-15 Thread Aaron Ballman via llvm-branch-commits


https://github.com/AaronBallman approved this pull request.

LGTM, thank you for the general fix!

https://github.com/llvm/llvm-project/pull/139961
___
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] [ConstraintElim] Simplify `usub_with_overflow` when A uge B (PR #135785)

2025-05-15 Thread Iris Shi via llvm-branch-commits


el-ev wrote:

https://alive2.llvm.org/ce/z/ZRmQES

https://github.com/llvm/llvm-project/pull/135785
___
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] [ConstraintElim] Simplify `usub_with_overflow` when A uge B (PR #135785)

2025-05-15 Thread Iris Shi via llvm-branch-commits


https://github.com/el-ev edited https://github.com/llvm/llvm-project/pull/135785
___
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] [flang] [mlir] [MLIR][OpenMP] Simplify OpenMP device codegen (PR #137201)

2025-05-15 Thread Sergio Afonso via llvm-branch-commits


https://github.com/skatrak updated 
https://github.com/llvm/llvm-project/pull/137201



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] release/20.x: [Driver] Fix _XOPEN_SOURCE definition on Solaris (#137141) (PR #140044)

2025-05-15 Thread via llvm-branch-commits


https://github.com/llvmbot created 
https://github.com/llvm/llvm-project/pull/140044

Backport e71c8ea3cc73c8f7b0382468f355a254166d3a72

Requested by: @rorth

>From 0b9ca6aebb3ca5c0c4925403172731753bac1c6a Mon Sep 17 00:00:00 2001
From: Rainer Orth 
Date: Sat, 26 Apr 2025 17:06:04 +0200
Subject: [PATCH] [Driver] Fix _XOPEN_SOURCE definition on Solaris (#137141)

Since commit 613a077b05b8352a48695be295037306f5fca151, `flang` doesn't
build any longer on Solaris/amd64:
```
flang/lib/Evaluate/intrinsics-library.cpp:225:26:
error: address of overloaded function 'acos' does not match required type 
'__float128 (__float128)'
  225 |   FolderFactory::Create("acos"),
  |  ^
```
That patch led to the version of `quadmath.h` deep inside `/usr/gcc/`
to be found, thus `HAS_QUADMATHLIB` is defined. However, the `struct
HostRuntimeLibrary<__float128, LibraryVersion::Libm>` template is
guarded by `_POSIX_C_SOURCE >= 200112L || _XOPEN_SOURCE >= 600`, while
`clang` only predefines `_XOPEN_SOURCE=500`.

This code dates back to commit 0c1941cb055fcf008e17faa6605969673211bea3
back in 2012. Currently, this is long obsolete and `gcc` prefefines
`_XOPEN_SOURCE=600` instead since GCC 4.6 back in 2011.

This patch follows that.

Tested on `amd64-pc-solaris2.11` and `sparcv9-sun-solaris2.11`.

(cherry picked from commit e71c8ea3cc73c8f7b0382468f355a254166d3a72)
---
 clang/lib/Basic/Targets/OSTargets.h | 9 +
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/clang/lib/Basic/Targets/OSTargets.h 
b/clang/lib/Basic/Targets/OSTargets.h
index 991efd2bde01f..4cf4230273d38 100644
--- a/clang/lib/Basic/Targets/OSTargets.h
+++ b/clang/lib/Basic/Targets/OSTargets.h
@@ -618,14 +618,7 @@ class LLVM_LIBRARY_VISIBILITY SolarisTargetInfo : public 
OSTargetInfo {
 DefineStd(Builder, "unix", Opts);
 Builder.defineMacro("__svr4__");
 Builder.defineMacro("__SVR4");
-// Solaris headers require _XOPEN_SOURCE to be set to 600 for C99 and
-// newer, but to 500 for everything else.  feature_test.h has a check to
-// ensure that you are not using C99 with an old version of X/Open or C89
-// with a new version.
-if (Opts.C99)
-  Builder.defineMacro("_XOPEN_SOURCE", "600");
-else
-  Builder.defineMacro("_XOPEN_SOURCE", "500");
+Builder.defineMacro("_XOPEN_SOURCE", "600");
 if (Opts.CPlusPlus) {
   Builder.defineMacro("__C99FEATURES__");
   Builder.defineMacro("_FILE_OFFSET_BITS", "64");

___
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] release/20.x: [Driver] Fix _XOPEN_SOURCE definition on Solaris (#137141) (PR #140044)

2025-05-15 Thread via llvm-branch-commits


https://github.com/llvmbot milestoned 
https://github.com/llvm/llvm-project/pull/140044
___
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] release/20.x: [Driver] Fix _XOPEN_SOURCE definition on Solaris (#137141) (PR #140044)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:

@MaskRay What do you think about merging this PR to the release branch?

https://github.com/llvm/llvm-project/pull/140044
___
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] release/20.x: [Driver] Fix _XOPEN_SOURCE definition on Solaris (#137141) (PR #140044)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-clang

Author: None (llvmbot)


Changes

Backport e71c8ea3cc73c8f7b0382468f355a254166d3a72

Requested by: @rorth

---
Full diff: https://github.com/llvm/llvm-project/pull/140044.diff


1 Files Affected:

- (modified) clang/lib/Basic/Targets/OSTargets.h (+1-8) 


``diff
diff --git a/clang/lib/Basic/Targets/OSTargets.h 
b/clang/lib/Basic/Targets/OSTargets.h
index 991efd2bde01f..4cf4230273d38 100644
--- a/clang/lib/Basic/Targets/OSTargets.h
+++ b/clang/lib/Basic/Targets/OSTargets.h
@@ -618,14 +618,7 @@ class LLVM_LIBRARY_VISIBILITY SolarisTargetInfo : public 
OSTargetInfo {
 DefineStd(Builder, "unix", Opts);
 Builder.defineMacro("__svr4__");
 Builder.defineMacro("__SVR4");
-// Solaris headers require _XOPEN_SOURCE to be set to 600 for C99 and
-// newer, but to 500 for everything else.  feature_test.h has a check to
-// ensure that you are not using C99 with an old version of X/Open or C89
-// with a new version.
-if (Opts.C99)
-  Builder.defineMacro("_XOPEN_SOURCE", "600");
-else
-  Builder.defineMacro("_XOPEN_SOURCE", "500");
+Builder.defineMacro("_XOPEN_SOURCE", "600");
 if (Opts.CPlusPlus) {
   Builder.defineMacro("__C99FEATURES__");
   Builder.defineMacro("_FILE_OFFSET_BITS", "64");

``




https://github.com/llvm/llvm-project/pull/140044
___
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] [MLIR][ArmSVE] Add an ArmSVE dialect operation which maps to svusmmla (PR #135634)

2025-05-15 Thread Momchil Velikov via llvm-branch-commits


https://github.com/momchil-velikov updated 
https://github.com/llvm/llvm-project/pull/135634

>From e60ca5aadf1043a0cb59d50da5f3dbf68bd50c51 Mon Sep 17 00:00:00 2001
From: Momchil Velikov 
Date: Thu, 10 Apr 2025 14:38:27 +
Subject: [PATCH] [MLIR][ArmSVE] Add an ArmSVE dialect operation which maps to
 `svusmmla`

---
 mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td | 95 +++
 .../Transforms/LegalizeForLLVMExport.cpp  |  4 +
 .../Dialect/ArmSVE/legalize-for-llvm.mlir | 12 +++
 mlir/test/Dialect/ArmSVE/roundtrip.mlir   | 11 +++
 mlir/test/Target/LLVMIR/arm-sve.mlir  | 12 +++
 5 files changed, 96 insertions(+), 38 deletions(-)

diff --git a/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td 
b/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
index 3a990f8464ef8..7385bb73b449a 100644
--- a/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
+++ b/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
@@ -147,11 +147,9 @@ class ScalableMaskedIOp,
-   AllTypesMatch<["acc", "dst"]>,
- ]> {
+def SdotOp : ArmSVE_Op<"sdot", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Vector-vector dot product and accumulate op";
   let description = [{
 SDOT: Signed integer addition of dot product.
@@ -178,11 +176,9 @@ def SdotOp : ArmSVE_Op<"sdot",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def SmmlaOp : ArmSVE_Op<"smmla",
-[Pure,
-AllTypesMatch<["src1", "src2"]>,
-AllTypesMatch<["acc", "dst"]>,
-  ]> {
+def SmmlaOp : ArmSVE_Op<"smmla", [Pure,
+  AllTypesMatch<["src1", "src2"]>,
+  AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Matrix-matrix multiply and accumulate op";
   let description = [{
 SMMLA: Signed integer matrix multiply-accumulate.
@@ -210,11 +206,9 @@ def SmmlaOp : ArmSVE_Op<"smmla",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def UdotOp : ArmSVE_Op<"udot",
-   [Pure,
-   AllTypesMatch<["src1", "src2"]>,
-   AllTypesMatch<["acc", "dst"]>,
- ]> {
+def UdotOp : ArmSVE_Op<"udot", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Vector-vector dot product and accumulate op";
   let description = [{
 UDOT: Unsigned integer addition of dot product.
@@ -241,11 +235,9 @@ def UdotOp : ArmSVE_Op<"udot",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def UmmlaOp : ArmSVE_Op<"ummla",
-[Pure,
-AllTypesMatch<["src1", "src2"]>,
-AllTypesMatch<["acc", "dst"]>,
-  ]> {
+def UmmlaOp : ArmSVE_Op<"ummla", [Pure,
+  AllTypesMatch<["src1", "src2"]>,
+  AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Matrix-matrix multiply and accumulate op";
   let description = [{
 UMMLA: Unsigned integer matrix multiply-accumulate.
@@ -273,14 +265,42 @@ def UmmlaOp : ArmSVE_Op<"ummla",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
+def UsmmlaOp : ArmSVE_Op<"usmmla", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
+  let summary = "Matrix-matrix multiply and accumulate op";
+  let description = [{
+USMMLA: Unsigned by signed integer matrix multiply-accumulate.
+
+The unsigned by signed integer matrix multiply-accumulate operation
+multiplies the 2×8 matrix of unsigned 8-bit integer values held
+the first source vector by the 8×2 matrix of signed 8-bit integer
+values in the second source vector. The resulting 2×2 widened 32-bit
+integer matrix product is then added to the 32-bit integer matrix
+accumulator.
+
+Source:
+https://developer.arm.com/documentation/100987/
+  }];
+  // Supports (vector<16xi8>, vector<16xi8>) -> (vector<4xi32>)
+  let arguments = (ins
+  ScalableVectorOfLengthAndType<[4], [I32]>:$acc,
+  ScalableVectorOfLengthAndType<[16], [I8]>:$src1,
+  ScalableVectorOfLengthAndType<[16], [I8]>:$src2
+  );
+  let results = (outs ScalableVectorOfLengthAndType<[4], [I32]>:$dst);
+  let assemblyFormat =
+"$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
+}
+
 class SvboolTypeConstraint : TypesMatchWith<
   "expected corresponding svbool type widened to [16]xi1",
   lhsArg, rhsArg,
   
"VectorType(VectorType::Builder(::llvm::cast($_self)).setDim(::llvm::cast($_self).getRank()
 - 1, 16))">;
 
 def ConvertFromSvboolOp : ArmSVE_Op<"convert_from_svbool",
-[Pure, SvboolTypeConstraint<"result", "source">]>
-{
+

[llvm-branch-commits] [mlir] [MLIR][ArmSVE] Add initial lowering of vector.contract to SVE `*MMLA` instructions (PR #135636)

2025-05-15 Thread Momchil Velikov via llvm-branch-commits


https://github.com/momchil-velikov updated 
https://github.com/llvm/llvm-project/pull/135636

>From aa8a667f206874af3b26811ec04d58be12ad43de Mon Sep 17 00:00:00 2001
From: Momchil Velikov 
Date: Tue, 8 Apr 2025 14:43:54 +
Subject: [PATCH 1/3] [MLIR][ArmSVE] Add initial lowering of `vector.contract`
 to SVE `*MMLA` instructions

---
 mlir/include/mlir/Conversion/Passes.td|   4 +
 .../Dialect/ArmSVE/Transforms/Transforms.h|   3 +
 .../Conversion/VectorToLLVM/CMakeLists.txt|   1 +
 .../VectorToLLVM/ConvertVectorToLLVMPass.cpp  |   7 +
 .../LowerContractionToSMMLAPattern.cpp|   5 +-
 .../Dialect/ArmSVE/Transforms/CMakeLists.txt  |   1 +
 .../LowerContractionToSVEI8MMPattern.cpp  | 304 ++
 .../Vector/CPU/ArmSVE/vector-smmla.mlir   |  94 ++
 .../Vector/CPU/ArmSVE/vector-summla.mlir  |  85 +
 .../Vector/CPU/ArmSVE/vector-ummla.mlir   |  94 ++
 .../Vector/CPU/ArmSVE/vector-usmmla.mlir  |  95 ++
 .../CPU/ArmSVE/contraction-smmla-4x8x4.mlir   | 117 +++
 .../ArmSVE/contraction-smmla-8x8x8-vs2.mlir   | 159 +
 .../CPU/ArmSVE/contraction-summla-4x8x4.mlir  | 118 +++
 .../CPU/ArmSVE/contraction-ummla-4x8x4.mlir   | 119 +++
 .../CPU/ArmSVE/contraction-usmmla-4x8x4.mlir  | 117 +++
 16 files changed, 1322 insertions(+), 1 deletion(-)
 create mode 100644 
mlir/lib/Dialect/ArmSVE/Transforms/LowerContractionToSVEI8MMPattern.cpp
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-smmla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-summla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-ummla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-usmmla.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-smmla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-smmla-8x8x8-vs2.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-summla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-ummla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-usmmla-4x8x4.mlir

diff --git a/mlir/include/mlir/Conversion/Passes.td 
b/mlir/include/mlir/Conversion/Passes.td
index 10557658d5d7d..b496ee0114910 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -1431,6 +1431,10 @@ def ConvertVectorToLLVMPass : 
Pass<"convert-vector-to-llvm"> {
"bool", /*default=*/"false",
"Enables the use of ArmSVE dialect while lowering the vector "
"dialect.">,
+Option<"armI8MM", "enable-arm-i8mm",
+   "bool", /*default=*/"false",
+   "Enables the use of Arm FEAT_I8MM instructions while lowering "
+   "the vector dialect.">,
 Option<"x86Vector", "enable-x86vector",
"bool", /*default=*/"false",
"Enables the use of X86Vector dialect while lowering the vector "
diff --git a/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h 
b/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
index 8665c8224cc45..232e2be29e574 100644
--- a/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
+++ b/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
@@ -20,6 +20,9 @@ class RewritePatternSet;
 void populateArmSVELegalizeForLLVMExportPatterns(
 const LLVMTypeConverter &converter, RewritePatternSet &patterns);
 
+void populateLowerContractionToSVEI8MMPatternPatterns(
+RewritePatternSet &patterns);
+
 /// Configure the target to support lowering ArmSVE ops to ops that map to LLVM
 /// intrinsics.
 void configureArmSVELegalizeForExportTarget(LLVMConversionTarget &target);
diff --git a/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt 
b/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
index 330474a718e30..8e2620029c354 100644
--- a/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
+++ b/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
@@ -35,6 +35,7 @@ add_mlir_conversion_library(MLIRVectorToLLVMPass
   MLIRVectorToLLVM
 
   MLIRArmNeonDialect
+  MLIRArmNeonTransforms
   MLIRArmSVEDialect
   MLIRArmSVETransforms
   MLIRAMXDialect
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp 
b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
index 0ee6dce9ee94b..293e01a5bf4d4 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
@@ -14,6 +14,7 @@
 #include "mlir/Dialect/AMX/Transforms.h"
 #include "mlir/Dialect/Arith/IR/Arith.h"
 #include "mlir/Dialect/ArmNeon/ArmNeonDialect.h"
+#include "mlir/Dialect/ArmNeon/Transforms.h"
 #include "mlir/Dialect/ArmSVE/IR/ArmSVEDialect.h"
 #include "mlir/Dialect/ArmSVE/Transforms/Transforms.h"
 #include "mlir/Dialect/Func/IR/FuncOps.h"
@@ -82,6 +83,12 @@ void ConvertVectorToLLVMPass::runOnOperation() {
 populateVectorStepLoweringPattern

[llvm-branch-commits] [mlir] [MLIR][ArmSVE] Add an ArmSVE dialect operation which maps to svusmmla (PR #135634)

2025-05-15 Thread Momchil Velikov via llvm-branch-commits


https://github.com/momchil-velikov updated 
https://github.com/llvm/llvm-project/pull/135634

>From e60ca5aadf1043a0cb59d50da5f3dbf68bd50c51 Mon Sep 17 00:00:00 2001
From: Momchil Velikov 
Date: Thu, 10 Apr 2025 14:38:27 +
Subject: [PATCH] [MLIR][ArmSVE] Add an ArmSVE dialect operation which maps to
 `svusmmla`

---
 mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td | 95 +++
 .../Transforms/LegalizeForLLVMExport.cpp  |  4 +
 .../Dialect/ArmSVE/legalize-for-llvm.mlir | 12 +++
 mlir/test/Dialect/ArmSVE/roundtrip.mlir   | 11 +++
 mlir/test/Target/LLVMIR/arm-sve.mlir  | 12 +++
 5 files changed, 96 insertions(+), 38 deletions(-)

diff --git a/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td 
b/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
index 3a990f8464ef8..7385bb73b449a 100644
--- a/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
+++ b/mlir/include/mlir/Dialect/ArmSVE/IR/ArmSVE.td
@@ -147,11 +147,9 @@ class ScalableMaskedIOp,
-   AllTypesMatch<["acc", "dst"]>,
- ]> {
+def SdotOp : ArmSVE_Op<"sdot", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Vector-vector dot product and accumulate op";
   let description = [{
 SDOT: Signed integer addition of dot product.
@@ -178,11 +176,9 @@ def SdotOp : ArmSVE_Op<"sdot",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def SmmlaOp : ArmSVE_Op<"smmla",
-[Pure,
-AllTypesMatch<["src1", "src2"]>,
-AllTypesMatch<["acc", "dst"]>,
-  ]> {
+def SmmlaOp : ArmSVE_Op<"smmla", [Pure,
+  AllTypesMatch<["src1", "src2"]>,
+  AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Matrix-matrix multiply and accumulate op";
   let description = [{
 SMMLA: Signed integer matrix multiply-accumulate.
@@ -210,11 +206,9 @@ def SmmlaOp : ArmSVE_Op<"smmla",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def UdotOp : ArmSVE_Op<"udot",
-   [Pure,
-   AllTypesMatch<["src1", "src2"]>,
-   AllTypesMatch<["acc", "dst"]>,
- ]> {
+def UdotOp : ArmSVE_Op<"udot", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Vector-vector dot product and accumulate op";
   let description = [{
 UDOT: Unsigned integer addition of dot product.
@@ -241,11 +235,9 @@ def UdotOp : ArmSVE_Op<"udot",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
-def UmmlaOp : ArmSVE_Op<"ummla",
-[Pure,
-AllTypesMatch<["src1", "src2"]>,
-AllTypesMatch<["acc", "dst"]>,
-  ]> {
+def UmmlaOp : ArmSVE_Op<"ummla", [Pure,
+  AllTypesMatch<["src1", "src2"]>,
+  AllTypesMatch<["acc", "dst"]>]> {
   let summary = "Matrix-matrix multiply and accumulate op";
   let description = [{
 UMMLA: Unsigned integer matrix multiply-accumulate.
@@ -273,14 +265,42 @@ def UmmlaOp : ArmSVE_Op<"ummla",
 "$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
 }
 
+def UsmmlaOp : ArmSVE_Op<"usmmla", [Pure,
+AllTypesMatch<["src1", "src2"]>,
+AllTypesMatch<["acc", "dst"]>]> {
+  let summary = "Matrix-matrix multiply and accumulate op";
+  let description = [{
+USMMLA: Unsigned by signed integer matrix multiply-accumulate.
+
+The unsigned by signed integer matrix multiply-accumulate operation
+multiplies the 2×8 matrix of unsigned 8-bit integer values held
+the first source vector by the 8×2 matrix of signed 8-bit integer
+values in the second source vector. The resulting 2×2 widened 32-bit
+integer matrix product is then added to the 32-bit integer matrix
+accumulator.
+
+Source:
+https://developer.arm.com/documentation/100987/
+  }];
+  // Supports (vector<16xi8>, vector<16xi8>) -> (vector<4xi32>)
+  let arguments = (ins
+  ScalableVectorOfLengthAndType<[4], [I32]>:$acc,
+  ScalableVectorOfLengthAndType<[16], [I8]>:$src1,
+  ScalableVectorOfLengthAndType<[16], [I8]>:$src2
+  );
+  let results = (outs ScalableVectorOfLengthAndType<[4], [I32]>:$dst);
+  let assemblyFormat =
+"$acc `,` $src1 `,` $src2 attr-dict `:` type($src1) `to` type($dst)";
+}
+
 class SvboolTypeConstraint : TypesMatchWith<
   "expected corresponding svbool type widened to [16]xi1",
   lhsArg, rhsArg,
   
"VectorType(VectorType::Builder(::llvm::cast($_self)).setDim(::llvm::cast($_self).getRank()
 - 1, 16))">;
 
 def ConvertFromSvboolOp : ArmSVE_Op<"convert_from_svbool",
-[Pure, SvboolTypeConstraint<"result", "source">]>
-{
+

[llvm-branch-commits] [mlir] [MLIR][ArmSVE] Add initial lowering of vector.contract to SVE `*MMLA` instructions (PR #135636)

2025-05-15 Thread Momchil Velikov via llvm-branch-commits


https://github.com/momchil-velikov updated 
https://github.com/llvm/llvm-project/pull/135636

>From aa8a667f206874af3b26811ec04d58be12ad43de Mon Sep 17 00:00:00 2001
From: Momchil Velikov 
Date: Tue, 8 Apr 2025 14:43:54 +
Subject: [PATCH 1/3] [MLIR][ArmSVE] Add initial lowering of `vector.contract`
 to SVE `*MMLA` instructions

---
 mlir/include/mlir/Conversion/Passes.td|   4 +
 .../Dialect/ArmSVE/Transforms/Transforms.h|   3 +
 .../Conversion/VectorToLLVM/CMakeLists.txt|   1 +
 .../VectorToLLVM/ConvertVectorToLLVMPass.cpp  |   7 +
 .../LowerContractionToSMMLAPattern.cpp|   5 +-
 .../Dialect/ArmSVE/Transforms/CMakeLists.txt  |   1 +
 .../LowerContractionToSVEI8MMPattern.cpp  | 304 ++
 .../Vector/CPU/ArmSVE/vector-smmla.mlir   |  94 ++
 .../Vector/CPU/ArmSVE/vector-summla.mlir  |  85 +
 .../Vector/CPU/ArmSVE/vector-ummla.mlir   |  94 ++
 .../Vector/CPU/ArmSVE/vector-usmmla.mlir  |  95 ++
 .../CPU/ArmSVE/contraction-smmla-4x8x4.mlir   | 117 +++
 .../ArmSVE/contraction-smmla-8x8x8-vs2.mlir   | 159 +
 .../CPU/ArmSVE/contraction-summla-4x8x4.mlir  | 118 +++
 .../CPU/ArmSVE/contraction-ummla-4x8x4.mlir   | 119 +++
 .../CPU/ArmSVE/contraction-usmmla-4x8x4.mlir  | 117 +++
 16 files changed, 1322 insertions(+), 1 deletion(-)
 create mode 100644 
mlir/lib/Dialect/ArmSVE/Transforms/LowerContractionToSVEI8MMPattern.cpp
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-smmla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-summla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-ummla.mlir
 create mode 100644 mlir/test/Dialect/Vector/CPU/ArmSVE/vector-usmmla.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-smmla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-smmla-8x8x8-vs2.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-summla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-ummla-4x8x4.mlir
 create mode 100644 
mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/contraction-usmmla-4x8x4.mlir

diff --git a/mlir/include/mlir/Conversion/Passes.td 
b/mlir/include/mlir/Conversion/Passes.td
index 10557658d5d7d..b496ee0114910 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -1431,6 +1431,10 @@ def ConvertVectorToLLVMPass : 
Pass<"convert-vector-to-llvm"> {
"bool", /*default=*/"false",
"Enables the use of ArmSVE dialect while lowering the vector "
"dialect.">,
+Option<"armI8MM", "enable-arm-i8mm",
+   "bool", /*default=*/"false",
+   "Enables the use of Arm FEAT_I8MM instructions while lowering "
+   "the vector dialect.">,
 Option<"x86Vector", "enable-x86vector",
"bool", /*default=*/"false",
"Enables the use of X86Vector dialect while lowering the vector "
diff --git a/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h 
b/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
index 8665c8224cc45..232e2be29e574 100644
--- a/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
+++ b/mlir/include/mlir/Dialect/ArmSVE/Transforms/Transforms.h
@@ -20,6 +20,9 @@ class RewritePatternSet;
 void populateArmSVELegalizeForLLVMExportPatterns(
 const LLVMTypeConverter &converter, RewritePatternSet &patterns);
 
+void populateLowerContractionToSVEI8MMPatternPatterns(
+RewritePatternSet &patterns);
+
 /// Configure the target to support lowering ArmSVE ops to ops that map to LLVM
 /// intrinsics.
 void configureArmSVELegalizeForExportTarget(LLVMConversionTarget &target);
diff --git a/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt 
b/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
index 330474a718e30..8e2620029c354 100644
--- a/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
+++ b/mlir/lib/Conversion/VectorToLLVM/CMakeLists.txt
@@ -35,6 +35,7 @@ add_mlir_conversion_library(MLIRVectorToLLVMPass
   MLIRVectorToLLVM
 
   MLIRArmNeonDialect
+  MLIRArmNeonTransforms
   MLIRArmSVEDialect
   MLIRArmSVETransforms
   MLIRAMXDialect
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp 
b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
index 0ee6dce9ee94b..293e01a5bf4d4 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVMPass.cpp
@@ -14,6 +14,7 @@
 #include "mlir/Dialect/AMX/Transforms.h"
 #include "mlir/Dialect/Arith/IR/Arith.h"
 #include "mlir/Dialect/ArmNeon/ArmNeonDialect.h"
+#include "mlir/Dialect/ArmNeon/Transforms.h"
 #include "mlir/Dialect/ArmSVE/IR/ArmSVEDialect.h"
 #include "mlir/Dialect/ArmSVE/Transforms/Transforms.h"
 #include "mlir/Dialect/Func/IR/FuncOps.h"
@@ -82,6 +83,12 @@ void ConvertVectorToLLVMPass::runOnOperation() {
 populateVectorStepLoweringPattern

[llvm-branch-commits] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread via llvm-branch-commits


https://github.com/llvmbot created 
https://github.com/llvm/llvm-project/pull/140051

Backport d5da557782dd47395fb41e03d7663df6319d7ea6

Requested by: @cjacek

>From 7ebd643db843168d1e6b11f4339a0bbeb61e7d89 Mon Sep 17 00:00:00 2001
From: Jacek Caban 
Date: Thu, 15 May 2025 03:28:18 -0700
Subject: [PATCH] [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives
 (#139631)

Make it a no-op for now, which is sufficient for non-hybrid images.

Fixes #131712.

(cherry picked from commit d5da557782dd47395fb41e03d7663df6319d7ea6)
---
 lld/COFF/Driver.cpp   |  6 +++
 lld/COFF/Options.td   |  1 +
 lld/test/COFF/arm64x-sameaddress.test | 56 +++
 3 files changed, 63 insertions(+)
 create mode 100644 lld/test/COFF/arm64x-sameaddress.test

diff --git a/lld/COFF/Driver.cpp b/lld/COFF/Driver.cpp
index f50ca529df4d7..a669b7e9296f6 100644
--- a/lld/COFF/Driver.cpp
+++ b/lld/COFF/Driver.cpp
@@ -492,6 +492,12 @@ void LinkerDriver::parseDirectives(InputFile *file) {
 case OPT_alternatename:
   parseAlternateName(arg->getValue());
   break;
+case OPT_arm64xsameaddress:
+  if (!file->symtab.isEC())
+Warn(ctx) << arg->getSpelling()
+  << " is not allowed in non-ARM64EC files (" << toString(file)
+  << ")";
+  break;
 case OPT_defaultlib:
   if (std::optional path = findLibIfNew(arg->getValue()))
 enqueuePath(*path, false, false);
diff --git a/lld/COFF/Options.td b/lld/COFF/Options.td
index b6fd3d0daaef9..ea2e7ded38043 100644
--- a/lld/COFF/Options.td
+++ b/lld/COFF/Options.td
@@ -355,3 +355,4 @@ def tlbid : P_priv<"tlbid">;
 def tlbout : P_priv<"tlbout">;
 def verbose_all : P_priv<"verbose">;
 def guardsym : P_priv<"guardsym">;
+def arm64xsameaddress : P_priv<"arm64xsameaddress">;
diff --git a/lld/test/COFF/arm64x-sameaddress.test 
b/lld/test/COFF/arm64x-sameaddress.test
new file mode 100644
index 0..c69be9d268c3b
--- /dev/null
+++ b/lld/test/COFF/arm64x-sameaddress.test
@@ -0,0 +1,56 @@
+REQUIRES: aarch64
+RUN: split-file %s %t.dir && cd %t.dir
+
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows func-arm64ec.s -o 
func-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows func-arm64.s -o 
func-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows drectve.s -o drectve.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows drectve.s -o 
drectve-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows 
%S/Inputs/loadconfig-arm64ec.s -o loadconfig-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows 
%S/Inputs/loadconfig-arm64.s -o loadconfig-arm64.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out.dll loadconfig-arm64.obj 
loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-cmd.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj -arm64xsameaddress:func
+
+RUN: lld-link -machine:arm64ec -dll -noentry -out:out-ec.dll 
loadconfig-arm64ec.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-warn.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve-arm64.obj 2>&1 | 
FileCheck --check-prefix=WARN %s
+WARN: lld-link: warning: -arm64xsameaddress: is not allowed in non-ARM64EC 
files (drectve-arm64.obj)
+
+#--- func-arm64.s
+.section .text,"xr",discard,func
+.globl func
+func:
+mov x0, #1
+ret
+
+#--- func-arm64ec.s
+.section .text,"xr",discard,"#func"
+.globl "#func"
+"#func":
+mov x0, #2
+ret
+
+.weak_anti_dep func
+.set func,"#func"
+
+.section .wowthk,"xr",discard,entry_thunk
+.globl entry_thunk
+entry_thunk:
+mov x0, #3
+ret
+
+.section .test,"dr"
+.rva func
+
+   .section .hybmp$x,"yi"
+   .symidx "#func"
+   .symidx entry_thunk
+   .word 1
+
+#--- drectve.s
+.section .drectve, "yn"
+.ascii " -arm64xsameaddress:func"

___
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] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread via llvm-branch-commits


https://github.com/llvmbot milestoned 
https://github.com/llvm/llvm-project/pull/140051
___
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] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:

@mstorsjo What do you think about merging this PR to the release branch?

https://github.com/llvm/llvm-project/pull/140051
___
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] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-lld

Author: None (llvmbot)


Changes

Backport d5da557782dd47395fb41e03d7663df6319d7ea6

Requested by: @cjacek

---
Full diff: https://github.com/llvm/llvm-project/pull/140051.diff


3 Files Affected:

- (modified) lld/COFF/Driver.cpp (+6) 
- (modified) lld/COFF/Options.td (+1) 
- (added) lld/test/COFF/arm64x-sameaddress.test (+56) 


``diff
diff --git a/lld/COFF/Driver.cpp b/lld/COFF/Driver.cpp
index f50ca529df4d7..a669b7e9296f6 100644
--- a/lld/COFF/Driver.cpp
+++ b/lld/COFF/Driver.cpp
@@ -492,6 +492,12 @@ void LinkerDriver::parseDirectives(InputFile *file) {
 case OPT_alternatename:
   parseAlternateName(arg->getValue());
   break;
+case OPT_arm64xsameaddress:
+  if (!file->symtab.isEC())
+Warn(ctx) << arg->getSpelling()
+  << " is not allowed in non-ARM64EC files (" << toString(file)
+  << ")";
+  break;
 case OPT_defaultlib:
   if (std::optional path = findLibIfNew(arg->getValue()))
 enqueuePath(*path, false, false);
diff --git a/lld/COFF/Options.td b/lld/COFF/Options.td
index b6fd3d0daaef9..ea2e7ded38043 100644
--- a/lld/COFF/Options.td
+++ b/lld/COFF/Options.td
@@ -355,3 +355,4 @@ def tlbid : P_priv<"tlbid">;
 def tlbout : P_priv<"tlbout">;
 def verbose_all : P_priv<"verbose">;
 def guardsym : P_priv<"guardsym">;
+def arm64xsameaddress : P_priv<"arm64xsameaddress">;
diff --git a/lld/test/COFF/arm64x-sameaddress.test 
b/lld/test/COFF/arm64x-sameaddress.test
new file mode 100644
index 0..c69be9d268c3b
--- /dev/null
+++ b/lld/test/COFF/arm64x-sameaddress.test
@@ -0,0 +1,56 @@
+REQUIRES: aarch64
+RUN: split-file %s %t.dir && cd %t.dir
+
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows func-arm64ec.s -o 
func-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows func-arm64.s -o 
func-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows drectve.s -o drectve.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows drectve.s -o 
drectve-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows 
%S/Inputs/loadconfig-arm64ec.s -o loadconfig-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows 
%S/Inputs/loadconfig-arm64.s -o loadconfig-arm64.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out.dll loadconfig-arm64.obj 
loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-cmd.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj -arm64xsameaddress:func
+
+RUN: lld-link -machine:arm64ec -dll -noentry -out:out-ec.dll 
loadconfig-arm64ec.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-warn.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve-arm64.obj 2>&1 | 
FileCheck --check-prefix=WARN %s
+WARN: lld-link: warning: -arm64xsameaddress: is not allowed in non-ARM64EC 
files (drectve-arm64.obj)
+
+#--- func-arm64.s
+.section .text,"xr",discard,func
+.globl func
+func:
+mov x0, #1
+ret
+
+#--- func-arm64ec.s
+.section .text,"xr",discard,"#func"
+.globl "#func"
+"#func":
+mov x0, #2
+ret
+
+.weak_anti_dep func
+.set func,"#func"
+
+.section .wowthk,"xr",discard,entry_thunk
+.globl entry_thunk
+entry_thunk:
+mov x0, #3
+ret
+
+.section .test,"dr"
+.rva func
+
+   .section .hybmp$x,"yi"
+   .symidx "#func"
+   .symidx entry_thunk
+   .word 1
+
+#--- drectve.s
+.section .drectve, "yn"
+.ascii " -arm64xsameaddress:func"

``




https://github.com/llvm/llvm-project/pull/140051
___
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] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:



@llvm/pr-subscribers-platform-windows

@llvm/pr-subscribers-lld-coff

Author: None (llvmbot)


Changes

Backport d5da557782dd47395fb41e03d7663df6319d7ea6

Requested by: @cjacek

---
Full diff: https://github.com/llvm/llvm-project/pull/140051.diff


3 Files Affected:

- (modified) lld/COFF/Driver.cpp (+6) 
- (modified) lld/COFF/Options.td (+1) 
- (added) lld/test/COFF/arm64x-sameaddress.test (+56) 


``diff
diff --git a/lld/COFF/Driver.cpp b/lld/COFF/Driver.cpp
index f50ca529df4d7..a669b7e9296f6 100644
--- a/lld/COFF/Driver.cpp
+++ b/lld/COFF/Driver.cpp
@@ -492,6 +492,12 @@ void LinkerDriver::parseDirectives(InputFile *file) {
 case OPT_alternatename:
   parseAlternateName(arg->getValue());
   break;
+case OPT_arm64xsameaddress:
+  if (!file->symtab.isEC())
+Warn(ctx) << arg->getSpelling()
+  << " is not allowed in non-ARM64EC files (" << toString(file)
+  << ")";
+  break;
 case OPT_defaultlib:
   if (std::optional path = findLibIfNew(arg->getValue()))
 enqueuePath(*path, false, false);
diff --git a/lld/COFF/Options.td b/lld/COFF/Options.td
index b6fd3d0daaef9..ea2e7ded38043 100644
--- a/lld/COFF/Options.td
+++ b/lld/COFF/Options.td
@@ -355,3 +355,4 @@ def tlbid : P_priv<"tlbid">;
 def tlbout : P_priv<"tlbout">;
 def verbose_all : P_priv<"verbose">;
 def guardsym : P_priv<"guardsym">;
+def arm64xsameaddress : P_priv<"arm64xsameaddress">;
diff --git a/lld/test/COFF/arm64x-sameaddress.test 
b/lld/test/COFF/arm64x-sameaddress.test
new file mode 100644
index 0..c69be9d268c3b
--- /dev/null
+++ b/lld/test/COFF/arm64x-sameaddress.test
@@ -0,0 +1,56 @@
+REQUIRES: aarch64
+RUN: split-file %s %t.dir && cd %t.dir
+
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows func-arm64ec.s -o 
func-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows func-arm64.s -o 
func-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows drectve.s -o drectve.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows drectve.s -o 
drectve-arm64.obj
+RUN: llvm-mc -filetype=obj -triple=arm64ec-windows 
%S/Inputs/loadconfig-arm64ec.s -o loadconfig-arm64ec.obj
+RUN: llvm-mc -filetype=obj -triple=aarch64-windows 
%S/Inputs/loadconfig-arm64.s -o loadconfig-arm64.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out.dll loadconfig-arm64.obj 
loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-cmd.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj -arm64xsameaddress:func
+
+RUN: lld-link -machine:arm64ec -dll -noentry -out:out-ec.dll 
loadconfig-arm64ec.obj func-arm64ec.obj drectve.obj
+
+RUN: lld-link -machine:arm64x -dll -noentry -out:out-warn.dll 
loadconfig-arm64.obj loadconfig-arm64ec.obj \
+RUN:  func-arm64.obj func-arm64ec.obj drectve-arm64.obj 2>&1 | 
FileCheck --check-prefix=WARN %s
+WARN: lld-link: warning: -arm64xsameaddress: is not allowed in non-ARM64EC 
files (drectve-arm64.obj)
+
+#--- func-arm64.s
+.section .text,"xr",discard,func
+.globl func
+func:
+mov x0, #1
+ret
+
+#--- func-arm64ec.s
+.section .text,"xr",discard,"#func"
+.globl "#func"
+"#func":
+mov x0, #2
+ret
+
+.weak_anti_dep func
+.set func,"#func"
+
+.section .wowthk,"xr",discard,entry_thunk
+.globl entry_thunk
+entry_thunk:
+mov x0, #3
+ret
+
+.section .test,"dr"
+.rva func
+
+   .section .hybmp$x,"yi"
+   .symidx "#func"
+   .symidx entry_thunk
+   .word 1
+
+#--- drectve.s
+.section .drectve, "yn"
+.ascii " -arm64xsameaddress:func"

``




https://github.com/llvm/llvm-project/pull/140051
___
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] [lld] release/20.x: [LLD][COFF] Allow -arm64xsameaddress in ARM64EC directives (#139631) (PR #140051)

2025-05-15 Thread Martin Storsjö via llvm-branch-commits


https://github.com/mstorsjo approved this pull request.

LGTM

https://github.com/llvm/llvm-project/pull/140051
___
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] [release/20.x] Update release notes for SystemZ changes (PR #140060)

2025-05-15 Thread Ulrich Weigand via llvm-branch-commits


https://github.com/uweigand created 
https://github.com/llvm/llvm-project/pull/140060

None

>From c97e80756a221267af0faebafaa956674f49d732 Mon Sep 17 00:00:00 2001
From: Ulrich Weigand 
Date: Thu, 15 May 2025 16:00:30 +0200
Subject: [PATCH] Update release notes for SystemZ changes

---
 llvm/docs/ReleaseNotes.md | 9 +
 1 file changed, 9 insertions(+)

diff --git a/llvm/docs/ReleaseNotes.md b/llvm/docs/ReleaseNotes.md
index f34003eaf0fe2..ef4ec9b56f364 100644
--- a/llvm/docs/ReleaseNotes.md
+++ b/llvm/docs/ReleaseNotes.md
@@ -349,6 +349,15 @@ Changes to the RISC-V Backend
   extension.
 * Added ``Sdext`` and ``Sdtrig`` extensions.
 
+Changes to the SystemZ Backend
+--
+
+* Added support for the IBM z17 processor and the arch15 cpu architecture.
+* Added support for `__builtin_setjump` and `__builtin_longjmp`.
+* Improve inlining heuristics to fix compile time explosion in certain cases.
+* Improve various cost functions.
+* Improve compatibility of the assembler parser with the GNU assembler.
+
 Changes to the WebAssembly Backend
 --
 

___
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] [AArch64] Correct position of CFI Instruction for Pointer Authentication (PR #137800)

2025-05-15 Thread Daniel Kiss via llvm-branch-commits


DanielKristofKiss wrote:

ping for review.
Counterpart landed in main already.

https://github.com/llvm/llvm-project/pull/137800
___
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] [AArch64] Correct position of CFI Instruction for Pointer Authentication (PR #137800)

2025-05-15 Thread Jack Styles via llvm-branch-commits


https://github.com/Stylie777 approved this pull request.

LGTM

https://github.com/llvm/llvm-project/pull/137800
___
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] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Paul Kirth via llvm-branch-commits


ilovepi wrote:

I thought we switched away from cloning to avoid redundancy. I'll take a look 
through previous patches though.

https://github.com/llvm/llvm-project/pull/13
___
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] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Paul Kirth via llvm-branch-commits


ilovepi wrote:

nuts, I forgot about https://github.com/llvm/llvm-project/pull/72180, I guess 
I'll need to think about this some more.

https://github.com/llvm/llvm-project/pull/13
___
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] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Paul Kirth via llvm-branch-commits


ilovepi wrote:

oh https://github.com/llvm/llvm-project/pull/139223 may avoid the issue w/ 
BlockAddress

https://github.com/llvm/llvm-project/pull/13
___
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] [llvm] [HLSL][RootSignature] Add parsing of flags to RootParam (PR #140152)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-clang

Author: Finn Plummer (inbelic)


Changes

- defines RootDescriptorFlags in-memory representation
- defines parseRootDescriptorFlags to be DXC compatible. This is why we support 
multiple `|` flags even validation will assert that only one flag is set...
- add unit tests to demonstrate functionality

Final part of and resolves https://github.com/llvm/llvm-project/issues/126577

---
Full diff: https://github.com/llvm/llvm-project/pull/140152.diff


4 Files Affected:

- (modified) clang/include/clang/Parse/ParseHLSLRootSignature.h (+3) 
- (modified) clang/lib/Parse/ParseHLSLRootSignature.cpp (+60) 
- (modified) clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp (+17-3) 
- (modified) llvm/include/llvm/Frontend/HLSL/HLSLRootSignature.h (+25) 


``diff
diff --git a/clang/include/clang/Parse/ParseHLSLRootSignature.h 
b/clang/include/clang/Parse/ParseHLSLRootSignature.h
index 436d217cec5b1..7b9168290d62a 100644
--- a/clang/include/clang/Parse/ParseHLSLRootSignature.h
+++ b/clang/include/clang/Parse/ParseHLSLRootSignature.h
@@ -93,6 +93,7 @@ class RootSignatureParser {
 std::optional Reg;
 std::optional Space;
 std::optional Visibility;
+std::optional Flags;
   };
   std::optional
   parseRootParamParams(RootSignatureToken::Kind RegType);
@@ -113,6 +114,8 @@ class RootSignatureParser {
 
   /// Parsing methods of various enums
   std::optional parseShaderVisibility();
+  std::optional
+  parseRootDescriptorFlags();
   std::optional
   parseDescriptorRangeFlags();
 
diff --git a/clang/lib/Parse/ParseHLSLRootSignature.cpp 
b/clang/lib/Parse/ParseHLSLRootSignature.cpp
index edb61f29f10d7..faf261cc9b7fe 100644
--- a/clang/lib/Parse/ParseHLSLRootSignature.cpp
+++ b/clang/lib/Parse/ParseHLSLRootSignature.cpp
@@ -193,6 +193,7 @@ std::optional 
RootSignatureParser::parseRootParam() {
 ExpectedReg = TokenKind::uReg;
 break;
   }
+  Param.setDefaultFlags();
 
   auto Params = parseRootParamParams(ExpectedReg);
   if (!Params.has_value())
@@ -214,6 +215,9 @@ std::optional 
RootSignatureParser::parseRootParam() {
   if (Params->Visibility.has_value())
 Param.Visibility = Params->Visibility.value();
 
+  if (Params->Flags.has_value())
+Param.Flags = Params->Flags.value();
+
   if (consumeExpectedToken(TokenKind::pu_r_paren,
diag::err_hlsl_unexpected_end_of_params,
/*param of=*/TokenKind::kw_RootConstants))
@@ -475,6 +479,23 @@ RootSignatureParser::parseRootParamParams(TokenKind 
RegType) {
 return std::nullopt;
   Params.Visibility = Visibility;
 }
+
+// `flags` `=` ROOT_DESCRIPTOR_FLAGS
+if (tryConsumeExpectedToken(TokenKind::kw_flags)) {
+  if (Params.Flags.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+
+  if (consumeExpectedToken(TokenKind::pu_equal))
+return std::nullopt;
+
+  auto Flags = parseRootDescriptorFlags();
+  if (!Flags.has_value())
+return std::nullopt;
+  Params.Flags = Flags;
+}
   } while (tryConsumeExpectedToken(TokenKind::pu_comma));
 
   return Params;
@@ -654,6 +675,45 @@ RootSignatureParser::parseShaderVisibility() {
   return std::nullopt;
 }
 
+std::optional
+RootSignatureParser::parseRootDescriptorFlags() {
+  assert(CurToken.TokKind == TokenKind::pu_equal &&
+ "Expects to only be invoked starting at given keyword");
+
+  // Handle the edge-case of '0' to specify no flags set
+  if (tryConsumeExpectedToken(TokenKind::int_literal)) {
+if (!verifyZeroFlag()) {
+  getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_non_zero_flag);
+  return std::nullopt;
+}
+return RootDescriptorFlags::None;
+  }
+
+  TokenKind Expected[] = {
+#define ROOT_DESCRIPTOR_FLAG_ENUM(NAME, LIT) TokenKind::en_##NAME,
+#include "clang/Lex/HLSLRootSignatureTokenKinds.def"
+  };
+
+  std::optional Flags;
+
+  do {
+if (tryConsumeExpectedToken(Expected)) {
+  switch (CurToken.TokKind) {
+#define ROOT_DESCRIPTOR_FLAG_ENUM(NAME, LIT)   
\
+  case TokenKind::en_##NAME:   
\
+Flags =
\
+maybeOrFlag(Flags, RootDescriptorFlags::NAME);
\
+break;
+#include "clang/Lex/HLSLRootSignatureTokenKinds.def"
+  default:
+llvm_unreachable("Switch for consumed enum token was not provided");
+  }
+}
+  } while (tryConsumeExpectedToken(TokenKind::pu_or));
+
+  return Flags;
+}
+
 std::optional
 RootSignatureParser::parseDescriptorRangeFlags() {
   assert(CurToken.TokKind == TokenKind::pu_equal &&
diff --git a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp 
b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
index 02bf38dcb110f..7ed286589f8fa 100644
--- a/clang/unittests/Parse/ParseHLSLRootSi

[llvm-branch-commits] [llvm] [AMDGPU] Automate creation of byte_sel dags. NFCI. (PR #140155)

2025-05-15 Thread Stanislav Mekhanoshin via llvm-branch-commits


https://github.com/rampitec created 
https://github.com/llvm/llvm-project/pull/140155

None

>From 071898b2e2b1f23e67ad5471df2088a0db167555 Mon Sep 17 00:00:00 2001
From: Stanislav Mekhanoshin 
Date: Thu, 15 May 2025 15:41:55 -0700
Subject: [PATCH] [AMDGPU] Automate creation of byte_sel dags. NFCI.

---
 llvm/lib/Target/AMDGPU/SIInstrInfo.td  | 21 +++--
 llvm/lib/Target/AMDGPU/VOP1Instructions.td | 11 ---
 llvm/lib/Target/AMDGPU/VOP3Instructions.td | 11 +--
 3 files changed, 16 insertions(+), 27 deletions(-)

diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.td 
b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
index 47ee0a7f60ee3..a5814dd7cf9f4 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
@@ -1293,6 +1293,7 @@ def WaitVMVSrc : NamedIntOperand<"wait_vm_vsrc"> {
 def ByteSel : NamedIntOperand<"byte_sel"> {
   let Validator = "isUInt<2>";
 }
+def ByteSel0 : DefaultOperand;
 
 let PrintMethod = "printBitOp3" in
 def BitOp3 : NamedIntOperand<"bitop3">;
@@ -1971,7 +1972,8 @@ class getIns32 
 class getIns64  {
+Operand Src0Mod, Operand Src1Mod, Operand Src2Mod,
+bit HasFP8ByteSel = 0, bit HasFP8DstByteSel = 0> {
   dag src0 = !if(!ge(NumSrcArgs, 1),
  !if (HasModifiers,
   (ins Src0Mod:$src0_modifiers, Src0RC:$src0),
@@ -1989,18 +1991,23 @@ class getIns64  {
+Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit 
HasOpSel,
+bit HasFP8ByteSel = 0, bit HasFP8DstByteSel = 0> {
   // getInst64 handles clamp and omod. implicit mutex between vop3p and omod
   dag base = getIns64 .ret;
+Src0Mod, Src1Mod, Src2Mod, HasFP8ByteSel, 
HasFP8DstByteSel>.ret;
   dag opsel = (ins op_sel0:$op_sel);
   dag ret = !con(base, !if(HasOpSel, opsel, (ins)));
 }
@@ -2612,7 +2619,8 @@ class VOPProfile  _ArgVT, bit 
_EnableClamp = 0> {
   field dag Ins32 = getIns32.ret;
   field dag Ins64 = getIns64.ret;
+ HasOMod, Src0Mod, Src1Mod, Src2Mod,
+ HasFP8ByteSel, HasFP8DstByteSel>.ret;
   field dag InsVOP3P = getInsVOP3P.ret;
@@ -2630,7 +2638,8 @@ class VOPProfile  _ArgVT, bit 
_EnableClamp = 0> {
  Src0ModDPP, Src1ModDPP, Src2ModDPP>.ret;
   defvar InsVOP3DPPBase = getInsVOP3Base.ret;
+  Src0ModVOP3DPP, Src1ModVOP3DPP, Src2ModVOP3DPP, HasOpSel,
+  HasFP8ByteSel, HasFP8DstByteSel>.ret;
   defvar InsVOP3PDPPBase = getInsVOP3P.ret;
diff --git a/llvm/lib/Target/AMDGPU/VOP1Instructions.td 
b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
index f885de3c13b12..7fdd951ecbd3c 100644
--- a/llvm/lib/Target/AMDGPU/VOP1Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
@@ -674,17 +674,6 @@ class VOPProfile_Base_CVT_F_F8_ByteSel : 
VOPProfile<[DstVT, i32
   let HasClamp = 0;
   let HasOMod = 0;
   let HasModifiers = 0;
-
-  defvar bytesel = (ins ByteSel:$byte_sel);
-  let Ins64 = !con(getIns64.ret,
-   bytesel);
-  let InsVOP3Base = !con(getInsVOP3Base.ret,
- bytesel);
 }
 
 let SubtargetPredicate = isGFX12Plus, OtherPredicates = 
[HasFP8ConversionInsts],
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td 
b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
index a7b90b9e319da..0252c4f1b0929 100644
--- a/llvm/lib/Target/AMDGPU/VOP3Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -593,6 +593,7 @@ def VOP3_CVT_SR_F8_F32_Profile : 
VOP3_Profile,
   let HasExtVOP3DPP = 1;
   let HasOpSel = 1;
   let HasFP8DstByteSel = 1;
+  let HasFP8ByteSel = 0; // It works as a dst-bytesel, but does not have 
byte_sel operand.
   let AsmVOP3OpSel = !subst(", $src2_modifiers", "",
 getAsmVOP3OpSel<3, HasClamp, HasOMod,
 HasSrc0FloatMods, HasSrc1FloatMods,
@@ -607,16 +608,6 @@ class VOP3_CVT_SR_F8_ByteSel_Profile :
   VOP3_Profile> {
   let HasFP8DstByteSel = 1;
   let HasClamp = 0;
-  defvar bytesel = (ins VGPR_32:$vdst_in, ByteSel:$byte_sel);
-  let Ins64 = !con(getIns64.ret,
-   bytesel);
-  let InsVOP3Base = !con(
-getInsVOP3Base.ret,
-bytesel);
 }
 
 def IsPow2Plus1: PatLeaf<(i32 imm), [{

___
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] Support pre-aggregated basic sample profile (PR #140196)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-bolt

Author: Amir Ayupov (aaupov)


Changes

Define a pre-aggregated basic sample format:
```
S  
```

Test Plan: update pre-aggregated-perf.test


---
Full diff: https://github.com/llvm/llvm-project/pull/140196.diff


5 Files Affected:

- (modified) bolt/include/bolt/Profile/DataAggregator.h (+20-14) 
- (modified) bolt/lib/Profile/DataAggregator.cpp (+49-39) 
- (added) bolt/test/X86/Inputs/pre-aggregated-basic.txt (+18) 
- (modified) bolt/test/X86/pre-aggregated-perf.test (+8) 
- (modified) bolt/test/link_fdata.py (+3-3) 


``diff
diff --git a/bolt/include/bolt/Profile/DataAggregator.h 
b/bolt/include/bolt/Profile/DataAggregator.h
index 6d918134137d5..7d60b4689fb04 100644
--- a/bolt/include/bolt/Profile/DataAggregator.h
+++ b/bolt/include/bolt/Profile/DataAggregator.h
@@ -370,26 +370,25 @@ class DataAggregator : public DataReader {
   /// memory.
   ///
   /// File format syntax:
-  /// {B|F|f|T} [:] [:] []
-  ///[]
+  /// {S|B|F|f|T}  [] []  []
   ///
-  /// B - indicates an aggregated branch
-  /// F - an aggregated fall-through
+  /// where , ,  have the format [:]
+  ///
+  /// S - indicates an aggregated basic sample at 
+  /// B - indicates an aggregated branch from  to 
+  /// F - an aggregated fall-through from  to 
   /// f - an aggregated fall-through with external origin - used to 
disambiguate
   ///   between a return hitting a basic block head and a regular internal
   ///   jump to the block
-  /// T - an aggregated trace: branch with a fall-through (from, to, ft_end)
-  ///
-  ///  - build id of the object containing the start address. We can
-  /// skip it for the main binary and use "X" for an unknown object. This will
-  /// save some space and facilitate human parsing.
-  ///
-  ///  - hex offset from the object base load address (0 for the
-  /// main executable unless it's PIE) to the start address.
+  /// T - an aggregated trace: branch from  to  with a fall-through
+  ///   to 
   ///
-  /// ,  - same for the end address.
+  ///  - build id of the object containing the address. We can skip it for
+  /// the main binary and use "X" for an unknown object. This will save some
+  /// space and facilitate human parsing.
   ///
-  ///  - same for the fallthrough_end address.
+  ///  - hex offset from the object base load address (0 for the
+  /// main executable unless it's PIE) to the address.
   ///
   ///  - total aggregated count of the branch or a fall-through.
   ///
@@ -397,10 +396,17 @@ class DataAggregator : public DataReader {
   /// Omitted for fall-throughs.
   ///
   /// Example:
+  /// Basic samples profile:
+  /// S 41be50 3
+  ///
+  /// Soft-deprecated branch profile with separate branches and fall-throughs:
   /// F 41be50 41be50 3
   /// F 41be90 41be90 4
   /// B 4b1942 39b57f0 3 0
   /// B 4b196f 4b19e0 2 0
+  ///
+  /// Recommended branch profile with pre-aggregated traces:
+  /// T 4b196f 4b19e0 4b19ef 2
   void parsePreAggregated();
 
   /// Parse the full output of pre-aggregated LBR samples generated by
diff --git a/bolt/lib/Profile/DataAggregator.cpp 
b/bolt/lib/Profile/DataAggregator.cpp
index aa681e633c0d8..6e889908003da 100644
--- a/bolt/lib/Profile/DataAggregator.cpp
+++ b/bolt/lib/Profile/DataAggregator.cpp
@@ -1216,54 +1216,54 @@ std::error_code 
DataAggregator::parseAggregatedLBREntry() {
   ErrorOr TypeOrErr = parseString(FieldSeparator);
   if (std::error_code EC = TypeOrErr.getError())
 return EC;
-  enum AggregatedLBREntry { TRACE, BRANCH, FT, FT_EXTERNAL_ORIGIN, INVALID };
+  enum AggregatedLBREntry {
+TRACE,
+SAMPLE,
+BRANCH,
+FT,
+FT_EXTERNAL_ORIGIN,
+INVALID
+  };
   auto Type = StringSwitch(TypeOrErr.get())
   .Case("T", TRACE)
+  .Case("S", SAMPLE)
   .Case("B", BRANCH)
   .Case("F", FT)
   .Case("f", FT_EXTERNAL_ORIGIN)
   .Default(INVALID);
   if (Type == INVALID) {
-reportError("expected T, B, F or f");
+reportError("expected T, S, B, F or f");
 return make_error_code(llvm::errc::io_error);
   }
 
-  while (checkAndConsumeFS()) {
-  }
-  ErrorOr From = parseLocationOrOffset();
-  if (std::error_code EC = From.getError())
-return EC;
+  std::optional Addrs[3];
+  int AddrNum = 2;
+  if (Type == TRACE)
+AddrNum = 3;
+  else if (Type == SAMPLE)
+AddrNum = 1;
 
-  while (checkAndConsumeFS()) {
-  }
-  ErrorOr To = parseLocationOrOffset();
-  if (std::error_code EC = To.getError())
-return EC;
+  int64_t Counters[2];
+  int CounterNum = 1;
+  if (Type == BRANCH)
+CounterNum = 2;
 
-  ErrorOr TraceFtEnd = std::error_code();
-  if (Type == AggregatedLBREntry::TRACE) {
+  for (int I = 0; I < AddrNum; ++I) {
 while (checkAndConsumeFS()) {
 }
-TraceFtEnd = parseLocationOrOffset();
-if (std::error_code EC = TraceFtEnd.getError())
-  return EC;
-  }
-
-  while (checkAndConsumeFS()) {
+if (Er

[llvm-branch-commits] [BOLT] Support pre-aggregated basic sample profile (PR #140196)

2025-05-15 Thread Amir Ayupov via llvm-branch-commits


https://github.com/aaupov created 
https://github.com/llvm/llvm-project/pull/140196

Define a pre-aggregated basic sample format:
```
S  
```

Test Plan: update pre-aggregated-perf.test




  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] Support pre-aggregated basic sample profile (PR #140196)

2025-05-15 Thread via llvm-branch-commits


github-actions[bot] wrote:




:warning: C/C++ code formatter, clang-format found issues in your code. 
:warning:



You can test this locally with the following command:


``bash
git-clang-format --diff HEAD~1 HEAD --extensions cpp,h -- 
bolt/include/bolt/Profile/DataAggregator.h bolt/lib/Profile/DataAggregator.cpp
``





View the diff from clang-format here.


``diff
diff --git a/bolt/lib/Profile/DataAggregator.cpp 
b/bolt/lib/Profile/DataAggregator.cpp
index 6e8899080..02ae9e0b8 100644
--- a/bolt/lib/Profile/DataAggregator.cpp
+++ b/bolt/lib/Profile/DataAggregator.cpp
@@ -1260,7 +1260,8 @@ std::error_code DataAggregator::parseAggregatedLBREntry() 
{
   for (int I = 0; I < CounterNum; ++I) {
 while (checkAndConsumeFS()) {
 }
-if (ErrorOr Count = parseNumberField(FieldSeparator, I + 1 == 
CounterNum))
+if (ErrorOr Count =
+parseNumberField(FieldSeparator, I + 1 == CounterNum))
   Counters[I] = Count.get();
 else
   return Count.getError();

``




https://github.com/llvm/llvm-project/pull/140196
___
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] [flang] cdc4b7c - Revert "[Flang] Turn on alias analysis for locally allocated objects (#139682)"

2025-05-15 Thread via llvm-branch-commits


Author: Dominik Adamski
Date: 2025-05-16T08:51:36+02:00
New Revision: cdc4b7c6467188d3a69a55cadd2fd318bf6d4f66

URL: 
https://github.com/llvm/llvm-project/commit/cdc4b7c6467188d3a69a55cadd2fd318bf6d4f66
DIFF: 
https://github.com/llvm/llvm-project/commit/cdc4b7c6467188d3a69a55cadd2fd318bf6d4f66.diff

LOG: Revert "[Flang] Turn on alias analysis for locally allocated objects 
(#139682)"

This reverts commit cf16c97bfa1416672d8990862369e86f360aa11e.

Added: 


Modified: 
flang/lib/Optimizer/Transforms/AddAliasTags.cpp
flang/test/Fir/tbaa-codegen2.fir
flang/test/Transforms/tbaa-with-dummy-scope2.fir
flang/test/Transforms/tbaa2.fir
flang/test/Transforms/tbaa3.fir

Removed: 




diff  --git a/flang/lib/Optimizer/Transforms/AddAliasTags.cpp 
b/flang/lib/Optimizer/Transforms/AddAliasTags.cpp
index 5cfbdc33285f9..66b4b84998801 100644
--- a/flang/lib/Optimizer/Transforms/AddAliasTags.cpp
+++ b/flang/lib/Optimizer/Transforms/AddAliasTags.cpp
@@ -43,10 +43,13 @@ static llvm::cl::opt
 static llvm::cl::opt
 enableDirect("direct-tbaa", llvm::cl::init(true), llvm::cl::Hidden,
  llvm::cl::desc("Add TBAA tags to direct variables"));
-static llvm::cl::opt
-enableLocalAllocs("local-alloc-tbaa", llvm::cl::init(true),
-  llvm::cl::Hidden,
-  llvm::cl::desc("Add TBAA tags to local allocations."));
+// This is **known unsafe** (misscompare in spec2017/wrf_r). It should
+// not be enabled by default.
+// The code is kept so that these may be tried with new benchmarks to see if
+// this is worth fixing in the future.
+static llvm::cl::opt enableLocalAllocs(
+"local-alloc-tbaa", llvm::cl::init(false), llvm::cl::Hidden,
+llvm::cl::desc("Add TBAA tags to local allocations. UNSAFE."));
 
 namespace {
 

diff  --git a/flang/test/Fir/tbaa-codegen2.fir 
b/flang/test/Fir/tbaa-codegen2.fir
index e4bfa9087ec75..8f8b6a29129e7 100644
--- a/flang/test/Fir/tbaa-codegen2.fir
+++ b/flang/test/Fir/tbaa-codegen2.fir
@@ -100,7 +100,7 @@ module attributes {fir.defaultkind = "a1c4d8i4l4r4", 
fir.kindmap = "", llvm.targ
 // [...]
 // CHECK:  %[[VAL50:.*]] = getelementptr i32, ptr %{{.*}}, i64 %{{.*}}
 // store to the temporary:
-// CHECK:  store i32 %{{.*}}, ptr %[[VAL50]], align 4, !tbaa 
![[TMP_DATA_ACCESS_TAG:.*]]
+// CHECK:  store i32 %{{.*}}, ptr %[[VAL50]], align 4, !tbaa 
![[DATA_ACCESS_TAG:.*]]
 // [...]
 
 // CHECK: [[BOX_ACCESS_TAG]] = !{![[BOX_ACCESS_TYPE:.*]], 
![[BOX_ACCESS_TYPE]], i64 0}
@@ -111,7 +111,4 @@ module attributes {fir.defaultkind = "a1c4d8i4l4r4", 
fir.kindmap = "", llvm.targ
 // CHECK: ![[A_ACCESS_TYPE]] = !{!"dummy arg data/_QFfuncEa", 
![[ARG_ACCESS_TYPE:.*]], i64 0}
 // CHECK: ![[ARG_ACCESS_TYPE]] = !{!"dummy arg data", 
![[DATA_ACCESS_TYPE:.*]], i64 0}
 // CHECK: ![[DATA_ACCESS_TYPE]] = !{!"any data access", ![[ANY_ACCESS_TYPE]], 
i64 0}
-// CHECK: ![[TMP_DATA_ACCESS_TAG]] = !{![[TMP_DATA_ACCESS_TYPE:.*]], 
![[TMP_DATA_ACCESS_TYPE]], i64 0}
-// CHECK: ![[TMP_DATA_ACCESS_TYPE]] = !{!"allocated data/", 
![[TMP_ACCESS_TYPE:.*]], i64 0}
-// CHECK: ![[TMP_ACCESS_TYPE]] = !{!"allocated data", 
![[TARGET_ACCESS_TAG:.*]], i64 0}
-// CHECK: ![[TARGET_ACCESS_TAG]] = !{!"target data", ![[DATA_ACCESS_TYPE]], 
i64 0}
+// CHECK: ![[DATA_ACCESS_TAG]] = !{![[DATA_ACCESS_TYPE]], 
![[DATA_ACCESS_TYPE]], i64 0}

diff  --git a/flang/test/Transforms/tbaa-with-dummy-scope2.fir 
b/flang/test/Transforms/tbaa-with-dummy-scope2.fir
index 249471de458d3..c8f419fbee652 100644
--- a/flang/test/Transforms/tbaa-with-dummy-scope2.fir
+++ b/flang/test/Transforms/tbaa-with-dummy-scope2.fir
@@ -43,15 +43,12 @@ func.func @_QPtest1() attributes {noinline} {
 // CHECK: #[[$ATTR_0:.+]] = #llvm.tbaa_root
 // CHECK: #[[$ATTR_1:.+]] = #llvm.tbaa_type_desc}>
 // CHECK: #[[$ATTR_2:.+]] = #llvm.tbaa_type_desc}>
-// CHECK: #[[$TARGETDATA:.+]] = #llvm.tbaa_type_desc}>
 // CHECK: #[[$ATTR_3:.+]] = #llvm.tbaa_type_desc}>
-// CHECK: #[[$LOCAL_ATTR_0:.+]] = #llvm.tbaa_type_desc}>
+// CHECK: #[[TARGETDATA:.+]] = #llvm.tbaa_type_desc}>
 // CHECK: #[[$ATTR_5:.+]] = #llvm.tbaa_type_desc}>
-// CHECK: #[[$ATTR_4:.+]] = #llvm.tbaa_type_desc}>
+// CHECK: #[[$ATTR_4:.+]] = #llvm.tbaa_type_desc}>
 // CHECK: #[[$ATTR_7:.+]] = #llvm.tbaa_tag
-// CHECK: #[[$LOCAL_ATTR_1:.+]] = #llvm.tbaa_type_desc}>
 // CHECK: #[[$ATTR_6:.+]] = #llvm.tbaa_type_desc}>
-// CHECK: #[[$LOCAL_ATTR_2:.+]] = #llvm.tbaa_tag
 // CHECK: #[[$ATTR_8:.+]] = #llvm.tbaa_tag
 // CHECK-LABEL:   func.func @_QPtest1() attributes {noinline} {
 // CHECK:   %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "y", 
uniq_name = "_QFtest1FinnerEy"}
@@ -60,8 +57,8 @@ func.func @_QPtest1() attributes {noinline} {
 // CHECK:   %[[VAL_5:.*]] = fir.dummy_scope : !fir.dscope
 // CHECK:   %[[VAL_6:.*]] = fir.declare %[[VAL_4]] dummy_scope 
%[[VAL_5]] {uniq_name = "_QFtest1FinnerEx"} : (!fir.ref, !fir.dscope) -> 
!fir.ref
 // CHECK:

[llvm-branch-commits] [llvm] [DirectX] adding support to read/write descriptor table data using obj2yaml/yaml2obj (PR #138315)

2025-05-15 Thread via llvm-branch-commits


https://github.com/joaosaffran converted_to_draft 
https://github.com/llvm/llvm-project/pull/138315
___
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] Fix the unit test errors. (PR #140184)

2025-05-15 Thread Qinkun Bao via llvm-branch-commits


https://github.com/qinkunbao created 
https://github.com/llvm/llvm-project/pull/140184

None


___
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] Fix the unit test errors. (PR #140184)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-llvm-support

Author: Qinkun Bao (qinkunbao)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/140184.diff


2 Files Affected:

- (modified) llvm/include/llvm/Support/SpecialCaseList.h (+2-3) 
- (modified) llvm/lib/Support/SpecialCaseList.cpp (+2-1) 


``diff
diff --git a/llvm/include/llvm/Support/SpecialCaseList.h 
b/llvm/include/llvm/Support/SpecialCaseList.h
index 4f4c097c7162a..fc6dc93651f38 100644
--- a/llvm/include/llvm/Support/SpecialCaseList.h
+++ b/llvm/include/llvm/Support/SpecialCaseList.h
@@ -132,9 +132,8 @@ class SpecialCaseList {
   using SectionEntries = StringMap>;
 
   struct Section {
-Section(std::unique_ptr M, StringRef str)
-: SectionMatcher(std::move(M)), SectionStr(SectionStr) {};
-Section(StringRef str) : Section(std::make_unique(), str) {};
+Section(std::unique_ptr M) : SectionMatcher(std::move(M)) {};
+Section() : Section(std::make_unique()) {};
 
 std::unique_ptr SectionMatcher;
 SectionEntries Entries;
diff --git a/llvm/lib/Support/SpecialCaseList.cpp 
b/llvm/lib/Support/SpecialCaseList.cpp
index 76c705c097aaa..9f044ced3a17c 100644
--- a/llvm/lib/Support/SpecialCaseList.cpp
+++ b/llvm/lib/Support/SpecialCaseList.cpp
@@ -132,8 +132,9 @@ bool SpecialCaseList::createInternal(const MemoryBuffer *MB,
 Expected
 SpecialCaseList::addSection(StringRef SectionStr, unsigned LineNo,
 bool UseGlobs) {
-  Sections.emplace_back(SectionStr);
+  Sections.emplace_back();
   auto &Section = Sections.back();
+  Section.SectionStr = SectionStr;
 
   if (auto Err = Section.SectionMatcher->insert(SectionStr, LineNo, UseGlobs)) 
{
 return createStringError(errc::invalid_argument,

``




https://github.com/llvm/llvm-project/pull/140184
___
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] Fix the unit test errors. (PR #140184)

2025-05-15 Thread Qinkun Bao via llvm-branch-commits


https://github.com/qinkunbao closed 
https://github.com/llvm/llvm-project/pull/140184
___
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] [SPARC][IAS] Add definitions for OSA 2011 instructions (PR #138403)

2025-05-15 Thread Brad Smith via llvm-branch-commits


brad0 wrote:

cc @s-barannikov 

https://github.com/llvm/llvm-project/pull/138403
___
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] [flang] [flang][OpenMP] use attribute for delayed privatization barrier (PR #140092)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:



@llvm/pr-subscribers-flang-fir-hlfir

@llvm/pr-subscribers-flang-openmp

Author: Tom Eccles (tblah)


Changes

Fixes #136357

The barrier needs to go between the copying into firstprivate variables and the 
initialization call for the OpenMP construct (e.g. wsloop). There is no way of 
expressing this in MLIR because for delayed privatization that is all implicit 
(added in MLIR->LLVMIR conversion).

The previous approach put the barrier immediately before the wsloop (or 
similar). For delayed privatization, the firstprivate copy code would then be 
inserted after that, opening the possibility for the race observed in the bug 
report.

This patch solves the issue by instead setting an attribute on the mlir 
operation, which will instruct openmp dialect to llvm ir conversion to insert a 
barrier in the correct place.

---
Full diff: https://github.com/llvm/llvm-project/pull/140092.diff


5 Files Affected:

- (modified) flang/lib/Lower/OpenMP/DataSharingProcessor.cpp (+10-3) 
- (modified) flang/lib/Lower/OpenMP/DataSharingProcessor.h (+1-1) 
- (modified) flang/test/Lower/OpenMP/lastprivate-allocatable.f90 (+1-1) 
- (modified) flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90 
(+1-2) 
- (modified) flang/test/Lower/OpenMP/same_var_first_lastprivate.f90 (+1-2) 


``diff
diff --git a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp 
b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
index 7eec598645eac..0949fe84f209f 100644
--- a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
@@ -62,7 +62,7 @@ void DataSharingProcessor::processStep1(
 
   privatize(clauseOps);
 
-  insertBarrier();
+  insertBarrier(clauseOps);
 }
 
 void DataSharingProcessor::processStep2(mlir::Operation *op, bool isLoop) {
@@ -230,8 +230,15 @@ bool DataSharingProcessor::needBarrier() {
   return false;
 }
 
-void DataSharingProcessor::insertBarrier() {
-  if (needBarrier())
+void DataSharingProcessor::insertBarrier(
+mlir::omp::PrivateClauseOps *clauseOps) {
+  if (!needBarrier())
+return;
+
+  if (useDelayedPrivatization)
+clauseOps->privateNeedsBarrier =
+mlir::UnitAttr::get(&converter.getMLIRContext());
+  else
 firOpBuilder.create(converter.getCurrentLocation());
 }
 
diff --git a/flang/lib/Lower/OpenMP/DataSharingProcessor.h 
b/flang/lib/Lower/OpenMP/DataSharingProcessor.h
index 54a42fd199831..7787e4ffb03c2 100644
--- a/flang/lib/Lower/OpenMP/DataSharingProcessor.h
+++ b/flang/lib/Lower/OpenMP/DataSharingProcessor.h
@@ -100,7 +100,7 @@ class DataSharingProcessor {
   const omp::ObjectList &objects,
   llvm::SetVector &symbolSet);
   void collectSymbolsForPrivatization();
-  void insertBarrier();
+  void insertBarrier(mlir::omp::PrivateClauseOps *clauseOps);
   void collectDefaultSymbols();
   void collectImplicitSymbols();
   void collectPreDeterminedSymbols();
diff --git a/flang/test/Lower/OpenMP/lastprivate-allocatable.f90 
b/flang/test/Lower/OpenMP/lastprivate-allocatable.f90
index 1d31edd16efea..c2626e14b51c7 100644
--- a/flang/test/Lower/OpenMP/lastprivate-allocatable.f90
+++ b/flang/test/Lower/OpenMP/lastprivate-allocatable.f90
@@ -8,7 +8,7 @@
 ! CHECK:   fir.store %[[VAL_2]] to %[[VAL_0]] : 
!fir.ref>>
 ! CHECK:   %[[VAL_3:.*]]:2 = hlfir.declare %[[VAL_0]] {fortran_attrs = 
{{.*}}, uniq_name = "_QFEa"} : 
(!fir.ref>>) -> (!fir.ref>>, 
!fir.ref>>)
 ! CHECK:   omp.parallel {
-! CHECK: omp.wsloop private(@{{.*}} %{{.*}} -> %{{.*}}, @{{.*}} 
%{{.*}} -> %[[VAL_17:.*]] : !fir.ref>>, !fir.ref) {
+! CHECK: omp.wsloop private(@{{.*}} %{{.*}} -> %{{.*}}, @{{.*}} 
%{{.*}} -> %[[VAL_17:.*]] : !fir.ref>>, !fir.ref) 
private_barrier {
 ! CHECK:   omp.loop_nest
 ! CHECK:   %[[VAL_16:.*]]:2 = hlfir.declare %{{.*}} 
{fortran_attrs = {{.*}}, uniq_name = "_QFEa"} : 
(!fir.ref>>) -> (!fir.ref>>, 
!fir.ref>>)
 ! CHECK:   %[[VAL_18:.*]]:2 = hlfir.declare %[[VAL_17]] 
{uniq_name = "_QFEi"} : (!fir.ref) -> (!fir.ref, !fir.ref)
diff --git a/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90 
b/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
index 60de8fa6f46a2..5d37010f4095b 100644
--- a/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
+++ b/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
@@ -226,8 +226,7 @@ subroutine firstpriv_lastpriv_int(arg1, arg2)
 ! Firstprivate update
 
 
-!CHECK-NEXT: omp.barrier
-!CHECK: omp.wsloop private(@{{.*}} %{{.*}}#0 -> %[[CLONE1:.*]], @{{.*}} 
%{{.*}}#0 -> %[[IV:.*]] : !fir.ref, !fir.ref) {
+!CHECK: omp.wsloop private(@{{.*}} %{{.*}}#0 -> %[[CLONE1:.*]], @{{.*}} 
%{{.*}}#0 -> %[[IV:.*]] : !fir.ref, !fir.ref) private_barrier {
 !CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 !CHECK: %[[CLONE1_DECL:.*]]:2 = hlfir.declare %[[CLONE1]] {uniq_name = 
"_QFfirstpriv_lastpriv_int2Earg1"} : (!fir.ref) -> (!fir.ref, 
!fir.ref)
 
diff --git a/flang/test/Lowe

[llvm-branch-commits] [flang] [flang][OpenMP] use attribute for delayed privatization barrier (PR #140092)

2025-05-15 Thread Tom Eccles via llvm-branch-commits


tblah wrote:

PR stack:

1. https://github.com/llvm/llvm-project/pull/140089
2. https://github.com/llvm/llvm-project/pull/140090
3. https://github.com/llvm/llvm-project/pull/140092

https://github.com/llvm/llvm-project/pull/140092
___
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] [mlir][OpenMP] Add translation of private_barrier attr to LLVMIR (PR #140090)

2025-05-15 Thread Tom Eccles via llvm-branch-commits


tblah wrote:

PR stack:

1. https://github.com/llvm/llvm-project/pull/140089
2. https://github.com/llvm/llvm-project/pull/140090
3. https://github.com/llvm/llvm-project/pull/140092

https://github.com/llvm/llvm-project/pull/140090
___
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] [StaticDataLayout][PGO]Implement reader and writer change for data access profiles (PR #139997)

2025-05-15 Thread Mingming Liu via llvm-branch-commits


https://github.com/mingmingl-llvm edited 
https://github.com/llvm/llvm-project/pull/139997
___
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] [StaticDataLayout][PGO]Implement reader and writer change for data access profiles (PR #139997)

2025-05-15 Thread Mingming Liu via llvm-branch-commits


https://github.com/mingmingl-llvm ready_for_review 
https://github.com/llvm/llvm-project/pull/139997
___
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] [StaticDataLayout][PGO]Implement reader and writer change for data access profiles (PR #139997)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-pgo

Author: Mingming Liu (mingmingl-llvm)


Changes

https://github.com/llvm/llvm-project/pull/138170 introduces classes to operate 
on data access profiles. This change supports the read and write of 
`DataAccessProfData` in indexed format of MemProf (v4) as well as its the text 
(yaml) format. 

For indexed format:
* InstrProfWriter owns (by `std::unique_ptr`) the 
data access profiles, and gives a non-owned copy when it calls `writeMemProf`.
  * MemProf v4 header has a new `uint64_t` to record the byte offset of data 
access profiles. This `uint64_t` field is zero if data access profile is not 
set (nullptr).
* MemProfReader reads the offset from v4 header and de-serializes in-memory 
bytes into class `DataAccessProfData`.
* MemProfYAML.h adds the mapping for DAP class, and make DAP optional for both 
read and write.


---

Patch is 26.46 KiB, truncated to 20.00 KiB below, full version: 
https://github.com/llvm/llvm-project/pull/139997.diff


14 Files Affected:

- (modified) llvm/include/llvm/ProfileData/DataAccessProf.h (+9-3) 
- (modified) llvm/include/llvm/ProfileData/IndexedMemProfData.h (+9-3) 
- (modified) llvm/include/llvm/ProfileData/InstrProfReader.h (+5-1) 
- (modified) llvm/include/llvm/ProfileData/InstrProfWriter.h (+6) 
- (modified) llvm/include/llvm/ProfileData/MemProfReader.h (+15) 
- (modified) llvm/include/llvm/ProfileData/MemProfYAML.h (+58) 
- (modified) llvm/lib/ProfileData/DataAccessProf.cpp (+4-2) 
- (modified) llvm/lib/ProfileData/IndexedMemProfData.cpp (+48-13) 
- (modified) llvm/lib/ProfileData/InstrProfReader.cpp (+14) 
- (modified) llvm/lib/ProfileData/InstrProfWriter.cpp (+16-4) 
- (modified) llvm/lib/ProfileData/MemProfReader.cpp (+30) 
- (modified) llvm/test/tools/llvm-profdata/memprof-yaml.test (+11) 
- (modified) llvm/tools/llvm-profdata/llvm-profdata.cpp (+4) 
- (modified) llvm/unittests/ProfileData/DataAccessProfTest.cpp (+6-5) 


``diff
diff --git a/llvm/include/llvm/ProfileData/DataAccessProf.h 
b/llvm/include/llvm/ProfileData/DataAccessProf.h
index e8504102238d1..f5f6abf0a2817 100644
--- a/llvm/include/llvm/ProfileData/DataAccessProf.h
+++ b/llvm/include/llvm/ProfileData/DataAccessProf.h
@@ -41,6 +41,8 @@ namespace data_access_prof {
 struct SourceLocation {
   SourceLocation(StringRef FileNameRef, uint32_t Line)
   : FileName(FileNameRef.str()), Line(Line) {}
+
+  SourceLocation() {}
   /// The filename where the data is located.
   std::string FileName;
   /// The line number in the source code.
@@ -53,6 +55,8 @@ namespace internal {
 // which strings are owned by `DataAccessProfData`. Used by 
`DataAccessProfData`
 // to represent data locations internally.
 struct SourceLocationRef {
+  SourceLocationRef(StringRef FileNameRef, uint32_t Line)
+  : FileName(FileNameRef), Line(Line) {}
   // The filename where the data is located.
   StringRef FileName;
   // The line number in the source code.
@@ -100,8 +104,9 @@ using SymbolHandle = std::variant;
 /// The data access profiles for a symbol.
 struct DataAccessProfRecord {
 public:
-  DataAccessProfRecord(SymbolHandleRef SymHandleRef,
-   ArrayRef LocRefs) {
+  DataAccessProfRecord(SymbolHandleRef SymHandleRef, uint64_t AccessCount,
+   ArrayRef LocRefs)
+  : AccessCount(AccessCount) {
 if (std::holds_alternative(SymHandleRef)) {
   SymHandle = std::get(SymHandleRef).str();
 } else
@@ -110,8 +115,9 @@ struct DataAccessProfRecord {
 for (auto Loc : LocRefs)
   Locations.push_back(SourceLocation(Loc.FileName, Loc.Line));
   }
+  DataAccessProfRecord() {}
   SymbolHandle SymHandle;
-
+  uint64_t AccessCount;
   // The locations of data in the source code. Optional.
   SmallVector Locations;
 };
diff --git a/llvm/include/llvm/ProfileData/IndexedMemProfData.h 
b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
index 3c6c329d1c49d..66fa38472059b 100644
--- a/llvm/include/llvm/ProfileData/IndexedMemProfData.h
+++ b/llvm/include/llvm/ProfileData/IndexedMemProfData.h
@@ -10,14 +10,20 @@
 //
 
//===--===//
 
+#include "llvm/ProfileData/DataAccessProf.h"
 #include "llvm/ProfileData/InstrProf.h"
 #include "llvm/ProfileData/MemProf.h"
 
+#include 
+#include 
+
 namespace llvm {
 
 // Write the MemProf data to OS.
-Error writeMemProf(ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
-   memprof::IndexedVersion MemProfVersionRequested,
-   bool MemProfFullSchema);
+Error writeMemProf(
+ProfOStream &OS, memprof::IndexedMemProfData &MemProfData,
+memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema,
+std::optional>
+DataAccessProfileData);
 
 } // namespace llvm
diff --git a/llvm/include/llvm/ProfileData/InstrProfReader.h 
b/llvm/include/llvm/ProfileData/InstrProfReader.h
index c250a9ede39bc..a3436e1dfe711 100644
--- a/llvm/include/llvm/ProfileDat

[llvm-branch-commits] [llvm] [AMDGPU] Automate creation of byte_sel dags. NFCI. (PR #140155)

2025-05-15 Thread Stanislav Mekhanoshin via llvm-branch-commits


https://github.com/rampitec ready_for_review 
https://github.com/llvm/llvm-project/pull/140155
___
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] Automate creation of byte_sel dags. NFCI. (PR #140155)

2025-05-15 Thread Stanislav Mekhanoshin via llvm-branch-commits


rampitec wrote:

> [!WARNING]
> This pull request is not mergeable via GitHub because a downstack PR is 
> open. Once all requirements are satisfied, merge this PR as a stack  href="https://app.graphite.dev/github/pr/llvm/llvm-project/140155?utm_source=stack-comment-downstack-mergeability-warning";
>  >on Graphite.
> https://graphite.dev/docs/merge-pull-requests";>Learn more

* **#140155** https://app.graphite.dev/github/pr/llvm/llvm-project/140155?utm_source=stack-comment-icon";
 target="_blank">https://static.graphite.dev/graphite-32x32-black.png"; alt="Graphite" 
width="10px" height="10px"/> 👈 https://app.graphite.dev/github/pr/llvm/llvm-project/140155?utm_source=stack-comment-view-in-graphite";
 target="_blank">(View in Graphite)
* **#140131** https://app.graphite.dev/github/pr/llvm/llvm-project/140131?utm_source=stack-comment-icon";
 target="_blank">https://static.graphite.dev/graphite-32x32-black.png"; alt="Graphite" 
width="10px" height="10px"/>
* `main`




This stack of pull requests is managed by https://graphite.dev?utm-source=stack-comment";>Graphite. Learn 
more about https://stacking.dev/?utm_source=stack-comment";>stacking.


https://github.com/llvm/llvm-project/pull/140155
___
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] Automate creation of byte_sel dags. NFCI. (PR #140155)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-backend-amdgpu

Author: Stanislav Mekhanoshin (rampitec)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/140155.diff


3 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/SIInstrInfo.td (+15-6) 
- (modified) llvm/lib/Target/AMDGPU/VOP1Instructions.td (-11) 
- (modified) llvm/lib/Target/AMDGPU/VOP3Instructions.td (+1-10) 


``diff
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.td 
b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
index 47ee0a7f60ee3..a5814dd7cf9f4 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
@@ -1293,6 +1293,7 @@ def WaitVMVSrc : NamedIntOperand<"wait_vm_vsrc"> {
 def ByteSel : NamedIntOperand<"byte_sel"> {
   let Validator = "isUInt<2>";
 }
+def ByteSel0 : DefaultOperand;
 
 let PrintMethod = "printBitOp3" in
 def BitOp3 : NamedIntOperand<"bitop3">;
@@ -1971,7 +1972,8 @@ class getIns32 
 class getIns64  {
+Operand Src0Mod, Operand Src1Mod, Operand Src2Mod,
+bit HasFP8ByteSel = 0, bit HasFP8DstByteSel = 0> {
   dag src0 = !if(!ge(NumSrcArgs, 1),
  !if (HasModifiers,
   (ins Src0Mod:$src0_modifiers, Src0RC:$src0),
@@ -1989,18 +1991,23 @@ class getIns64  {
+Operand Src0Mod, Operand Src1Mod, Operand Src2Mod, bit 
HasOpSel,
+bit HasFP8ByteSel = 0, bit HasFP8DstByteSel = 0> {
   // getInst64 handles clamp and omod. implicit mutex between vop3p and omod
   dag base = getIns64 .ret;
+Src0Mod, Src1Mod, Src2Mod, HasFP8ByteSel, 
HasFP8DstByteSel>.ret;
   dag opsel = (ins op_sel0:$op_sel);
   dag ret = !con(base, !if(HasOpSel, opsel, (ins)));
 }
@@ -2612,7 +2619,8 @@ class VOPProfile  _ArgVT, bit 
_EnableClamp = 0> {
   field dag Ins32 = getIns32.ret;
   field dag Ins64 = getIns64.ret;
+ HasOMod, Src0Mod, Src1Mod, Src2Mod,
+ HasFP8ByteSel, HasFP8DstByteSel>.ret;
   field dag InsVOP3P = getInsVOP3P.ret;
@@ -2630,7 +2638,8 @@ class VOPProfile  _ArgVT, bit 
_EnableClamp = 0> {
  Src0ModDPP, Src1ModDPP, Src2ModDPP>.ret;
   defvar InsVOP3DPPBase = getInsVOP3Base.ret;
+  Src0ModVOP3DPP, Src1ModVOP3DPP, Src2ModVOP3DPP, HasOpSel,
+  HasFP8ByteSel, HasFP8DstByteSel>.ret;
   defvar InsVOP3PDPPBase = getInsVOP3P.ret;
diff --git a/llvm/lib/Target/AMDGPU/VOP1Instructions.td 
b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
index f885de3c13b12..7fdd951ecbd3c 100644
--- a/llvm/lib/Target/AMDGPU/VOP1Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP1Instructions.td
@@ -674,17 +674,6 @@ class VOPProfile_Base_CVT_F_F8_ByteSel : 
VOPProfile<[DstVT, i32
   let HasClamp = 0;
   let HasOMod = 0;
   let HasModifiers = 0;
-
-  defvar bytesel = (ins ByteSel:$byte_sel);
-  let Ins64 = !con(getIns64.ret,
-   bytesel);
-  let InsVOP3Base = !con(getInsVOP3Base.ret,
- bytesel);
 }
 
 let SubtargetPredicate = isGFX12Plus, OtherPredicates = 
[HasFP8ConversionInsts],
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td 
b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
index a7b90b9e319da..0252c4f1b0929 100644
--- a/llvm/lib/Target/AMDGPU/VOP3Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -593,6 +593,7 @@ def VOP3_CVT_SR_F8_F32_Profile : 
VOP3_Profile,
   let HasExtVOP3DPP = 1;
   let HasOpSel = 1;
   let HasFP8DstByteSel = 1;
+  let HasFP8ByteSel = 0; // It works as a dst-bytesel, but does not have 
byte_sel operand.
   let AsmVOP3OpSel = !subst(", $src2_modifiers", "",
 getAsmVOP3OpSel<3, HasClamp, HasOMod,
 HasSrc0FloatMods, HasSrc1FloatMods,
@@ -607,16 +608,6 @@ class VOP3_CVT_SR_F8_ByteSel_Profile :
   VOP3_Profile> {
   let HasFP8DstByteSel = 1;
   let HasClamp = 0;
-  defvar bytesel = (ins VGPR_32:$vdst_in, ByteSel:$byte_sel);
-  let Ins64 = !con(getIns64.ret,
-   bytesel);
-  let InsVOP3Base = !con(
-getInsVOP3Base.ret,
-bytesel);
 }
 
 def IsPow2Plus1: PatLeaf<(i32 imm), [{

``




https://github.com/llvm/llvm-project/pull/140155
___
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] [llvm][EmbedBitcodePass] Prevent modifying the module with ThinLTO (PR #139999)

2025-05-15 Thread Nikita Popov via llvm-branch-commits


nikic wrote:

@ilovepi Does it also work on the release branch? I'd mainly see the clone 
module approach as something easily backportable for the release branch, but I 
assume for main we'll want a different solution?

https://github.com/llvm/llvm-project/pull/13
___
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] [NFCI] [ValueTracking] Add pass to print ConstantRanges (PR #140144)

2025-05-15 Thread Florian Mayer via llvm-branch-commits


https://github.com/fmayer created 
https://github.com/llvm/llvm-project/pull/140144

None


___
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] [NFCI] [ValueTracking] Add pass to print ConstantRanges (PR #140144)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-testing-tools

Author: Florian Mayer (fmayer)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/140144.diff


6 Files Affected:

- (modified) llvm/include/llvm/Analysis/ValueTracking.h (+15-1) 
- (modified) llvm/lib/Analysis/ValueTracking.cpp (+17) 
- (modified) llvm/lib/Passes/PassBuilder.cpp (+1) 
- (modified) llvm/lib/Passes/PassRegistry.def (+1) 
- (added) llvm/test/Analysis/ValueTracking/print-constant-range.ll (+234) 
- (modified) llvm/utils/UpdateTestChecks/common.py (+1) 


``diff
diff --git a/llvm/include/llvm/Analysis/ValueTracking.h 
b/llvm/include/llvm/Analysis/ValueTracking.h
index 61dbb07e7128e..341c20931761a 100644
--- a/llvm/include/llvm/Analysis/ValueTracking.h
+++ b/llvm/include/llvm/Analysis/ValueTracking.h
@@ -19,9 +19,10 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/FMF.h"
-#include "llvm/IR/Instructions.h"
 #include "llvm/IR/InstrTypes.h"
+#include "llvm/IR/Instructions.h"
 #include "llvm/IR/Intrinsics.h"
+#include "llvm/IR/PassManager.h"
 #include 
 #include 
 
@@ -1001,6 +1002,19 @@ std::optional isImpliedByDomCondition(CmpPredicate 
Pred, const Value *LHS,
 void findValuesAffectedByCondition(Value *Cond, bool IsAssume,
function_ref InsertAffected);
 
+/// Printer pass for \p computeConstantRange results.
+class ConstantRangePrinterPass
+: public PassInfoMixin {
+  raw_ostream &OS;
+
+public:
+  explicit ConstantRangePrinterPass(raw_ostream &OS) : OS(OS) {}
+
+  PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
+
+  static bool isRequired() { return true; }
+};
+
 } // end namespace llvm
 
 #endif // LLVM_ANALYSIS_VALUETRACKING_H
diff --git a/llvm/lib/Analysis/ValueTracking.cpp 
b/llvm/lib/Analysis/ValueTracking.cpp
index 3d403531cea2f..3ebbf887b3e6c 100644
--- a/llvm/lib/Analysis/ValueTracking.cpp
+++ b/llvm/lib/Analysis/ValueTracking.cpp
@@ -49,6 +49,7 @@
 #include "llvm/IR/GlobalAlias.h"
 #include "llvm/IR/GlobalValue.h"
 #include "llvm/IR/GlobalVariable.h"
+#include "llvm/IR/InstIterator.h"
 #include "llvm/IR/InstrTypes.h"
 #include "llvm/IR/Instruction.h"
 #include "llvm/IR/Instructions.h"
@@ -10451,3 +10452,19 @@ void llvm::findValuesAffectedByCondition(
 }
   }
 }
+
+PreservedAnalyses ConstantRangePrinterPass::run(Function &F,
+FunctionAnalysisManager &AM) {
+  // For compatibility with opt's -analyze feature under legacy pass manager
+  // which was not ported to NPM. This keeps tests using
+  // update_analyze_test_checks.py working.
+  OS << "Printing analysis 'ConstantRange' for function '" << F.getName()
+ << "':\n";
+  for (const Instruction &I : instructions(F)) {
+if (!I.getType()->isIntOrIntVectorTy())
+  continue;
+auto CR = computeConstantRange(&I, false);
+OS << I << ": " << CR << "\n";
+  }
+  return PreservedAnalyses::all();
+}
diff --git a/llvm/lib/Passes/PassBuilder.cpp b/llvm/lib/Passes/PassBuilder.cpp
index 7740f622ede7c..060c1b535befd 100644
--- a/llvm/lib/Passes/PassBuilder.cpp
+++ b/llvm/lib/Passes/PassBuilder.cpp
@@ -78,6 +78,7 @@
 #include "llvm/Analysis/TargetTransformInfo.h"
 #include "llvm/Analysis/TypeBasedAliasAnalysis.h"
 #include "llvm/Analysis/UniformityAnalysis.h"
+#include "llvm/Analysis/ValueTracking.h"
 #include "llvm/CodeGen/AssignmentTrackingAnalysis.h"
 #include "llvm/CodeGen/AtomicExpand.h"
 #include "llvm/CodeGen/BasicBlockSectionsProfileReader.h"
diff --git a/llvm/lib/Passes/PassRegistry.def b/llvm/lib/Passes/PassRegistry.def
index ea792280ed975..8cbc6a1156c51 100644
--- a/llvm/lib/Passes/PassRegistry.def
+++ b/llvm/lib/Passes/PassRegistry.def
@@ -462,6 +462,7 @@ FUNCTION_PASS("print", 
PhiValuesPrinterPass(errs()))
 FUNCTION_PASS("print", PostDominatorTreePrinterPass(errs()))
 FUNCTION_PASS("print", RegionInfoPrinterPass(errs()))
 FUNCTION_PASS("print", ScalarEvolutionPrinterPass(errs()))
+FUNCTION_PASS("print", ConstantRangePrinterPass(errs()))
 FUNCTION_PASS("print", StackSafetyPrinterPass(errs()))
 FUNCTION_PASS("print", UniformityInfoPrinterPass(errs()))
 FUNCTION_PASS("reassociate", ReassociatePass())
diff --git a/llvm/test/Analysis/ValueTracking/print-constant-range.ll 
b/llvm/test/Analysis/ValueTracking/print-constant-range.ll
new file mode 100644
index 0..ad444551e9adc
--- /dev/null
+++ b/llvm/test/Analysis/ValueTracking/print-constant-range.ll
@@ -0,0 +1,234 @@
+; NOTE: Assertions have been autogenerated by 
utils/update_analyze_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -disable-output '-passes=print' < %s -S 2>&1 | 
FileCheck %s
+
+define i1 @shl_C_X_ugt(i8 %x) {
+; CHECK-LABEL: 'shl_C_X_ugt'
+; CHECK-NEXT:%shl = shl i8 7, %x: [1,-31)
+; CHECK-NEXT:%r = icmp ugt i8 %shl, -32: full-set
+;
+  %shl = shl i8 7, %x
+  %r = icmp ugt i8 %shl, 224
+  ret i1 %r
+}
+
+define i1 @shl_C_X_ugt2(i8 %x) {
+; CHECK-LABEL: 'shl_C_X_ugt2'
+; CHECK-NEXT:%shl = shl i8 5, %x: [1,-63

[llvm-branch-commits] [NFCI] [ValueTracking] Add pass to print ConstantRanges (PR #140144)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-llvm-analysis

Author: Florian Mayer (fmayer)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/140144.diff


6 Files Affected:

- (modified) llvm/include/llvm/Analysis/ValueTracking.h (+15-1) 
- (modified) llvm/lib/Analysis/ValueTracking.cpp (+17) 
- (modified) llvm/lib/Passes/PassBuilder.cpp (+1) 
- (modified) llvm/lib/Passes/PassRegistry.def (+1) 
- (added) llvm/test/Analysis/ValueTracking/print-constant-range.ll (+234) 
- (modified) llvm/utils/UpdateTestChecks/common.py (+1) 


``diff
diff --git a/llvm/include/llvm/Analysis/ValueTracking.h 
b/llvm/include/llvm/Analysis/ValueTracking.h
index 61dbb07e7128e..341c20931761a 100644
--- a/llvm/include/llvm/Analysis/ValueTracking.h
+++ b/llvm/include/llvm/Analysis/ValueTracking.h
@@ -19,9 +19,10 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/FMF.h"
-#include "llvm/IR/Instructions.h"
 #include "llvm/IR/InstrTypes.h"
+#include "llvm/IR/Instructions.h"
 #include "llvm/IR/Intrinsics.h"
+#include "llvm/IR/PassManager.h"
 #include 
 #include 
 
@@ -1001,6 +1002,19 @@ std::optional isImpliedByDomCondition(CmpPredicate 
Pred, const Value *LHS,
 void findValuesAffectedByCondition(Value *Cond, bool IsAssume,
function_ref InsertAffected);
 
+/// Printer pass for \p computeConstantRange results.
+class ConstantRangePrinterPass
+: public PassInfoMixin {
+  raw_ostream &OS;
+
+public:
+  explicit ConstantRangePrinterPass(raw_ostream &OS) : OS(OS) {}
+
+  PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
+
+  static bool isRequired() { return true; }
+};
+
 } // end namespace llvm
 
 #endif // LLVM_ANALYSIS_VALUETRACKING_H
diff --git a/llvm/lib/Analysis/ValueTracking.cpp 
b/llvm/lib/Analysis/ValueTracking.cpp
index 3d403531cea2f..3ebbf887b3e6c 100644
--- a/llvm/lib/Analysis/ValueTracking.cpp
+++ b/llvm/lib/Analysis/ValueTracking.cpp
@@ -49,6 +49,7 @@
 #include "llvm/IR/GlobalAlias.h"
 #include "llvm/IR/GlobalValue.h"
 #include "llvm/IR/GlobalVariable.h"
+#include "llvm/IR/InstIterator.h"
 #include "llvm/IR/InstrTypes.h"
 #include "llvm/IR/Instruction.h"
 #include "llvm/IR/Instructions.h"
@@ -10451,3 +10452,19 @@ void llvm::findValuesAffectedByCondition(
 }
   }
 }
+
+PreservedAnalyses ConstantRangePrinterPass::run(Function &F,
+FunctionAnalysisManager &AM) {
+  // For compatibility with opt's -analyze feature under legacy pass manager
+  // which was not ported to NPM. This keeps tests using
+  // update_analyze_test_checks.py working.
+  OS << "Printing analysis 'ConstantRange' for function '" << F.getName()
+ << "':\n";
+  for (const Instruction &I : instructions(F)) {
+if (!I.getType()->isIntOrIntVectorTy())
+  continue;
+auto CR = computeConstantRange(&I, false);
+OS << I << ": " << CR << "\n";
+  }
+  return PreservedAnalyses::all();
+}
diff --git a/llvm/lib/Passes/PassBuilder.cpp b/llvm/lib/Passes/PassBuilder.cpp
index 7740f622ede7c..060c1b535befd 100644
--- a/llvm/lib/Passes/PassBuilder.cpp
+++ b/llvm/lib/Passes/PassBuilder.cpp
@@ -78,6 +78,7 @@
 #include "llvm/Analysis/TargetTransformInfo.h"
 #include "llvm/Analysis/TypeBasedAliasAnalysis.h"
 #include "llvm/Analysis/UniformityAnalysis.h"
+#include "llvm/Analysis/ValueTracking.h"
 #include "llvm/CodeGen/AssignmentTrackingAnalysis.h"
 #include "llvm/CodeGen/AtomicExpand.h"
 #include "llvm/CodeGen/BasicBlockSectionsProfileReader.h"
diff --git a/llvm/lib/Passes/PassRegistry.def b/llvm/lib/Passes/PassRegistry.def
index ea792280ed975..8cbc6a1156c51 100644
--- a/llvm/lib/Passes/PassRegistry.def
+++ b/llvm/lib/Passes/PassRegistry.def
@@ -462,6 +462,7 @@ FUNCTION_PASS("print", 
PhiValuesPrinterPass(errs()))
 FUNCTION_PASS("print", PostDominatorTreePrinterPass(errs()))
 FUNCTION_PASS("print", RegionInfoPrinterPass(errs()))
 FUNCTION_PASS("print", ScalarEvolutionPrinterPass(errs()))
+FUNCTION_PASS("print", ConstantRangePrinterPass(errs()))
 FUNCTION_PASS("print", StackSafetyPrinterPass(errs()))
 FUNCTION_PASS("print", UniformityInfoPrinterPass(errs()))
 FUNCTION_PASS("reassociate", ReassociatePass())
diff --git a/llvm/test/Analysis/ValueTracking/print-constant-range.ll 
b/llvm/test/Analysis/ValueTracking/print-constant-range.ll
new file mode 100644
index 0..ad444551e9adc
--- /dev/null
+++ b/llvm/test/Analysis/ValueTracking/print-constant-range.ll
@@ -0,0 +1,234 @@
+; NOTE: Assertions have been autogenerated by 
utils/update_analyze_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -disable-output '-passes=print' < %s -S 2>&1 | 
FileCheck %s
+
+define i1 @shl_C_X_ugt(i8 %x) {
+; CHECK-LABEL: 'shl_C_X_ugt'
+; CHECK-NEXT:%shl = shl i8 7, %x: [1,-31)
+; CHECK-NEXT:%r = icmp ugt i8 %shl, -32: full-set
+;
+  %shl = shl i8 7, %x
+  %r = icmp ugt i8 %shl, 224
+  ret i1 %r
+}
+
+define i1 @shl_C_X_ugt2(i8 %x) {
+; CHECK-LABEL: 'shl_C_X_ugt2'
+; CHECK-NEXT:%shl = shl i8 5, %x: [1,-63

[llvm-branch-commits] [NFCI] [ValueTracking] Add pass to print ConstantRanges (PR #140144)

2025-05-15 Thread Florian Mayer via llvm-branch-commits


https://github.com/fmayer updated 
https://github.com/llvm/llvm-project/pull/140144



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] [NFCI] [ValueTracking] Add pass to print ConstantRanges (PR #140144)

2025-05-15 Thread Florian Mayer via llvm-branch-commits


https://github.com/fmayer updated 
https://github.com/llvm/llvm-project/pull/140144


___
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] [llvm] [HLSL][RootSignature] Add parsing of Register in params for RootParams (PR #140148)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-hlsl

Author: Finn Plummer (inbelic)


Changes

- defines the `parseRootParamParams` infastructure for parsing the params of a 
RootParam
- defines the register type to illustrate use
- add tests to demonstrate functionality

Part 2 of https://github.com/llvm/llvm-project/issues/126577

---
Full diff: https://github.com/llvm/llvm-project/pull/140148.diff


4 Files Affected:

- (modified) clang/include/clang/Parse/ParseHLSLRootSignature.h (+6) 
- (modified) clang/lib/Parse/ParseHLSLRootSignature.cpp (+42) 
- (modified) clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp (+9-4) 
- (modified) llvm/include/llvm/Frontend/HLSL/HLSLRootSignature.h (+1) 


``diff
diff --git a/clang/include/clang/Parse/ParseHLSLRootSignature.h 
b/clang/include/clang/Parse/ParseHLSLRootSignature.h
index b96416ab6c6c4..92b8e7389426a 100644
--- a/clang/include/clang/Parse/ParseHLSLRootSignature.h
+++ b/clang/include/clang/Parse/ParseHLSLRootSignature.h
@@ -89,6 +89,12 @@ class RootSignatureParser {
   };
   std::optional parseRootConstantParams();
 
+  struct ParsedRootParamParams {
+std::optional Reg;
+  };
+  std::optional
+  parseRootParamParams(RootSignatureToken::Kind RegType);
+
   struct ParsedClauseParams {
 std::optional Reg;
 std::optional NumDescriptors;
diff --git a/clang/lib/Parse/ParseHLSLRootSignature.cpp 
b/clang/lib/Parse/ParseHLSLRootSignature.cpp
index 3ed60442eaa14..de7c2452b3fd4 100644
--- a/clang/lib/Parse/ParseHLSLRootSignature.cpp
+++ b/clang/lib/Parse/ParseHLSLRootSignature.cpp
@@ -176,20 +176,37 @@ std::optional 
RootSignatureParser::parseRootParam() {
 return std::nullopt;
 
   RootParam Param;
+  TokenKind ExpectedReg;
   switch (ParamKind) {
   default:
 llvm_unreachable("Switch for consumed token was not provided");
   case TokenKind::kw_CBV:
 Param.Type = ParamType::CBuffer;
+ExpectedReg = TokenKind::bReg;
 break;
   case TokenKind::kw_SRV:
 Param.Type = ParamType::SRV;
+ExpectedReg = TokenKind::tReg;
 break;
   case TokenKind::kw_UAV:
 Param.Type = ParamType::UAV;
+ExpectedReg = TokenKind::uReg;
 break;
   }
 
+  auto Params = parseRootParamParams(ExpectedReg);
+  if (!Params.has_value())
+return std::nullopt;
+
+  // Check mandatory parameters were provided
+  if (!Params->Reg.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_missing_param)
+<< ExpectedReg;
+return std::nullopt;
+  }
+
+  Param.Reg = Params->Reg.value();
+
   if (consumeExpectedToken(TokenKind::pu_r_paren,
diag::err_hlsl_unexpected_end_of_params,
/*param of=*/TokenKind::kw_RootConstants))
@@ -398,6 +415,31 @@ RootSignatureParser::parseRootConstantParams() {
   return Params;
 }
 
+std::optional
+RootSignatureParser::parseRootParamParams(TokenKind RegType) {
+  assert(CurToken.TokKind == TokenKind::pu_l_paren &&
+ "Expects to only be invoked starting at given token");
+
+  ParsedRootParamParams Params;
+  do {
+// ( `b` | `t` | `u`) POS_INT
+if (tryConsumeExpectedToken(RegType)) {
+  if (Params.Reg.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+  auto Reg = parseRegister();
+  if (!Reg.has_value())
+return std::nullopt;
+  Params.Reg = Reg;
+}
+
+  } while (tryConsumeExpectedToken(TokenKind::pu_comma));
+
+  return Params;
+}
+
 std::optional
 RootSignatureParser::parseDescriptorTableClauseParams(TokenKind RegType) {
   assert(CurToken.TokKind == TokenKind::pu_l_paren &&
diff --git a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp 
b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
index 876dcd17f0389..0f32523493a53 100644
--- a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
+++ b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
@@ -346,9 +346,9 @@ TEST_F(ParseHLSLRootSignatureTest, ValidParseRootFlagsTest) 
{
 
 TEST_F(ParseHLSLRootSignatureTest, ValidParseRootParamsTest) {
   const llvm::StringLiteral Source = R"cc(
-CBV(),
-SRV(),
-UAV()
+CBV(b0),
+SRV(t42),
+UAV(u34893247)
   )cc";
 
   TrivialModuleLoader ModLoader;
@@ -368,15 +368,20 @@ TEST_F(ParseHLSLRootSignatureTest, 
ValidParseRootParamsTest) {
 
   RootElement Elem = Elements[0];
   ASSERT_TRUE(std::holds_alternative(Elem));
-  ASSERT_EQ(std::get(Elem).Type, ParamType::CBuffer);
+  ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::BReg);
+  ASSERT_EQ(std::get(Elem).Reg.Number, 0u);
 
   Elem = Elements[1];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::SRV);
+  ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::TReg);
+  ASSERT_EQ(std::get(Elem).Reg.Number, 42u);
 
   Elem = Elements[2];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::UAV);
+  ASSERT_EQ(std::get(Elem).Reg.View

[llvm-branch-commits] [clang] [llvm] [HLSL][RootSignature] Add parsing of Register in params for RootParams (PR #140148)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-clang

Author: Finn Plummer (inbelic)


Changes

- defines the `parseRootParamParams` infastructure for parsing the params of a 
RootParam
- defines the register type to illustrate use
- add tests to demonstrate functionality

Part 2 of https://github.com/llvm/llvm-project/issues/126577

---
Full diff: https://github.com/llvm/llvm-project/pull/140148.diff


4 Files Affected:

- (modified) clang/include/clang/Parse/ParseHLSLRootSignature.h (+6) 
- (modified) clang/lib/Parse/ParseHLSLRootSignature.cpp (+42) 
- (modified) clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp (+9-4) 
- (modified) llvm/include/llvm/Frontend/HLSL/HLSLRootSignature.h (+1) 


``diff
diff --git a/clang/include/clang/Parse/ParseHLSLRootSignature.h 
b/clang/include/clang/Parse/ParseHLSLRootSignature.h
index b96416ab6c6c4..92b8e7389426a 100644
--- a/clang/include/clang/Parse/ParseHLSLRootSignature.h
+++ b/clang/include/clang/Parse/ParseHLSLRootSignature.h
@@ -89,6 +89,12 @@ class RootSignatureParser {
   };
   std::optional parseRootConstantParams();
 
+  struct ParsedRootParamParams {
+std::optional Reg;
+  };
+  std::optional
+  parseRootParamParams(RootSignatureToken::Kind RegType);
+
   struct ParsedClauseParams {
 std::optional Reg;
 std::optional NumDescriptors;
diff --git a/clang/lib/Parse/ParseHLSLRootSignature.cpp 
b/clang/lib/Parse/ParseHLSLRootSignature.cpp
index 3ed60442eaa14..de7c2452b3fd4 100644
--- a/clang/lib/Parse/ParseHLSLRootSignature.cpp
+++ b/clang/lib/Parse/ParseHLSLRootSignature.cpp
@@ -176,20 +176,37 @@ std::optional 
RootSignatureParser::parseRootParam() {
 return std::nullopt;
 
   RootParam Param;
+  TokenKind ExpectedReg;
   switch (ParamKind) {
   default:
 llvm_unreachable("Switch for consumed token was not provided");
   case TokenKind::kw_CBV:
 Param.Type = ParamType::CBuffer;
+ExpectedReg = TokenKind::bReg;
 break;
   case TokenKind::kw_SRV:
 Param.Type = ParamType::SRV;
+ExpectedReg = TokenKind::tReg;
 break;
   case TokenKind::kw_UAV:
 Param.Type = ParamType::UAV;
+ExpectedReg = TokenKind::uReg;
 break;
   }
 
+  auto Params = parseRootParamParams(ExpectedReg);
+  if (!Params.has_value())
+return std::nullopt;
+
+  // Check mandatory parameters were provided
+  if (!Params->Reg.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_missing_param)
+<< ExpectedReg;
+return std::nullopt;
+  }
+
+  Param.Reg = Params->Reg.value();
+
   if (consumeExpectedToken(TokenKind::pu_r_paren,
diag::err_hlsl_unexpected_end_of_params,
/*param of=*/TokenKind::kw_RootConstants))
@@ -398,6 +415,31 @@ RootSignatureParser::parseRootConstantParams() {
   return Params;
 }
 
+std::optional
+RootSignatureParser::parseRootParamParams(TokenKind RegType) {
+  assert(CurToken.TokKind == TokenKind::pu_l_paren &&
+ "Expects to only be invoked starting at given token");
+
+  ParsedRootParamParams Params;
+  do {
+// ( `b` | `t` | `u`) POS_INT
+if (tryConsumeExpectedToken(RegType)) {
+  if (Params.Reg.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+  auto Reg = parseRegister();
+  if (!Reg.has_value())
+return std::nullopt;
+  Params.Reg = Reg;
+}
+
+  } while (tryConsumeExpectedToken(TokenKind::pu_comma));
+
+  return Params;
+}
+
 std::optional
 RootSignatureParser::parseDescriptorTableClauseParams(TokenKind RegType) {
   assert(CurToken.TokKind == TokenKind::pu_l_paren &&
diff --git a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp 
b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
index 876dcd17f0389..0f32523493a53 100644
--- a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
+++ b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
@@ -346,9 +346,9 @@ TEST_F(ParseHLSLRootSignatureTest, ValidParseRootFlagsTest) 
{
 
 TEST_F(ParseHLSLRootSignatureTest, ValidParseRootParamsTest) {
   const llvm::StringLiteral Source = R"cc(
-CBV(),
-SRV(),
-UAV()
+CBV(b0),
+SRV(t42),
+UAV(u34893247)
   )cc";
 
   TrivialModuleLoader ModLoader;
@@ -368,15 +368,20 @@ TEST_F(ParseHLSLRootSignatureTest, 
ValidParseRootParamsTest) {
 
   RootElement Elem = Elements[0];
   ASSERT_TRUE(std::holds_alternative(Elem));
-  ASSERT_EQ(std::get(Elem).Type, ParamType::CBuffer);
+  ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::BReg);
+  ASSERT_EQ(std::get(Elem).Reg.Number, 0u);
 
   Elem = Elements[1];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::SRV);
+  ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::TReg);
+  ASSERT_EQ(std::get(Elem).Reg.Number, 42u);
 
   Elem = Elements[2];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::UAV);
+  ASSERT_EQ(std::get(Elem).Reg.Vie

[llvm-branch-commits] [clang] [llvm] [HLSL][RootSignature] Add parsing of Register in params for RootParams (PR #140148)

2025-05-15 Thread Finn Plummer via llvm-branch-commits


https://github.com/inbelic created 
https://github.com/llvm/llvm-project/pull/140148

- defines the `parseRootParamParams` infastructure for parsing the params of a 
RootParam
- defines the register type to illustrate use
- add tests to demonstrate functionality

Part 2 of https://github.com/llvm/llvm-project/issues/126577



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] [llvm] [HLSL][RootSignature] Add parsing of optional parameters for RootParam (PR #140151)

2025-05-15 Thread Finn Plummer via llvm-branch-commits


https://github.com/inbelic created 
https://github.com/llvm/llvm-project/pull/140151

- define in-memory representation of optional non-flag parameters to RootParam
- fill in data to parse these params in `parseRootParamParams`
- add unit tests

Part 3 of https://github.com/llvm/llvm-project/issues/126577



  



Rate limit · GitHub


  body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe 
UI,Helvetica,Arial,sans-serif,Apple Color Emoji,Segoe UI Emoji,Segoe UI Symbol;
font-size: 14px;
line-height: 1.5;
margin: 0;
  }

  .container { margin: 50px auto; max-width: 600px; text-align: center; 
padding: 0 24px; }

  a { color: #0366d6; text-decoration: none; }
  a:hover { text-decoration: underline; }

  h1 { line-height: 60px; font-size: 48px; font-weight: 300; margin: 0px; 
text-shadow: 0 1px 0 #fff; }
  p { color: rgba(0, 0, 0, 0.5); margin: 20px 0 40px; }

  ul { list-style: none; margin: 25px 0; padding: 0; }
  li { display: table-cell; font-weight: bold; width: 1%; }

  .logo { display: inline-block; margin-top: 35px; }
  .logo-img-2x { display: none; }
  @media
  only screen and (-webkit-min-device-pixel-ratio: 2),
  only screen and (   min--moz-device-pixel-ratio: 2),
  only screen and ( -o-min-device-pixel-ratio: 2/1),
  only screen and (min-device-pixel-ratio: 2),
  only screen and (min-resolution: 192dpi),
  only screen and (min-resolution: 2dppx) {
.logo-img-1x { display: none; }
.logo-img-2x { display: inline-block; }
  }

  #suggestions {
margin-top: 35px;
color: #ccc;
  }
  #suggestions a {
color: #66;
font-weight: 200;
font-size: 14px;
margin: 0 10px;
  }


  
  



  Whoa there!
  You have exceeded a secondary rate limit.
Please wait a few minutes before you try again;
in some cases this may take up to an hour.
  
  
https://support.github.com/contact";>Contact Support —
https://githubstatus.com";>GitHub Status —
https://twitter.com/githubstatus";>@githubstatus
  

  

  

  

  

  


___
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] [llvm] [HLSL][RootSignature] Add parsing of optional parameters for RootParam (PR #140151)

2025-05-15 Thread via llvm-branch-commits


llvmbot wrote:




@llvm/pr-subscribers-clang

Author: Finn Plummer (inbelic)


Changes

- define in-memory representation of optional non-flag parameters to RootParam
- fill in data to parse these params in `parseRootParamParams`
- add unit tests

Part 3 of https://github.com/llvm/llvm-project/issues/126577

---
Full diff: https://github.com/llvm/llvm-project/pull/140151.diff


4 Files Affected:

- (modified) clang/include/clang/Parse/ParseHLSLRootSignature.h (+2) 
- (modified) clang/lib/Parse/ParseHLSLRootSignature.cpp (+40) 
- (modified) clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp (+8-2) 
- (modified) llvm/include/llvm/Frontend/HLSL/HLSLRootSignature.h (+2) 


``diff
diff --git a/clang/include/clang/Parse/ParseHLSLRootSignature.h 
b/clang/include/clang/Parse/ParseHLSLRootSignature.h
index 92b8e7389426a..436d217cec5b1 100644
--- a/clang/include/clang/Parse/ParseHLSLRootSignature.h
+++ b/clang/include/clang/Parse/ParseHLSLRootSignature.h
@@ -91,6 +91,8 @@ class RootSignatureParser {
 
   struct ParsedRootParamParams {
 std::optional Reg;
+std::optional Space;
+std::optional Visibility;
   };
   std::optional
   parseRootParamParams(RootSignatureToken::Kind RegType);
diff --git a/clang/lib/Parse/ParseHLSLRootSignature.cpp 
b/clang/lib/Parse/ParseHLSLRootSignature.cpp
index de7c2452b3fd4..edb61f29f10d7 100644
--- a/clang/lib/Parse/ParseHLSLRootSignature.cpp
+++ b/clang/lib/Parse/ParseHLSLRootSignature.cpp
@@ -207,6 +207,13 @@ std::optional 
RootSignatureParser::parseRootParam() {
 
   Param.Reg = Params->Reg.value();
 
+  // Fill in optional values
+  if (Params->Space.has_value())
+Param.Space = Params->Space.value();
+
+  if (Params->Visibility.has_value())
+Param.Visibility = Params->Visibility.value();
+
   if (consumeExpectedToken(TokenKind::pu_r_paren,
diag::err_hlsl_unexpected_end_of_params,
/*param of=*/TokenKind::kw_RootConstants))
@@ -435,6 +442,39 @@ RootSignatureParser::parseRootParamParams(TokenKind 
RegType) {
   Params.Reg = Reg;
 }
 
+// `space` `=` POS_INT
+if (tryConsumeExpectedToken(TokenKind::kw_space)) {
+  if (Params.Space.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+
+  if (consumeExpectedToken(TokenKind::pu_equal))
+return std::nullopt;
+
+  auto Space = parseUIntParam();
+  if (!Space.has_value())
+return std::nullopt;
+  Params.Space = Space;
+}
+
+// `visibility` `=` SHADER_VISIBILITY
+if (tryConsumeExpectedToken(TokenKind::kw_visibility)) {
+  if (Params.Visibility.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+
+  if (consumeExpectedToken(TokenKind::pu_equal))
+return std::nullopt;
+
+  auto Visibility = parseShaderVisibility();
+  if (!Visibility.has_value())
+return std::nullopt;
+  Params.Visibility = Visibility;
+}
   } while (tryConsumeExpectedToken(TokenKind::pu_comma));
 
   return Params;
diff --git a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp 
b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
index 0f32523493a53..02bf38dcb110f 100644
--- a/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
+++ b/clang/unittests/Parse/ParseHLSLRootSignatureTest.cpp
@@ -347,8 +347,8 @@ TEST_F(ParseHLSLRootSignatureTest, ValidParseRootFlagsTest) 
{
 TEST_F(ParseHLSLRootSignatureTest, ValidParseRootParamsTest) {
   const llvm::StringLiteral Source = R"cc(
 CBV(b0),
-SRV(t42),
-UAV(u34893247)
+SRV(space = 4, t42, visibility = SHADER_VISIBILITY_GEOMETRY),
+UAV(visibility = SHADER_VISIBILITY_HULL, u34893247)
   )cc";
 
   TrivialModuleLoader ModLoader;
@@ -370,18 +370,24 @@ TEST_F(ParseHLSLRootSignatureTest, 
ValidParseRootParamsTest) {
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::BReg);
   ASSERT_EQ(std::get(Elem).Reg.Number, 0u);
+  ASSERT_EQ(std::get(Elem).Space, 0u);
+  ASSERT_EQ(std::get(Elem).Visibility, ShaderVisibility::All);
 
   Elem = Elements[1];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::SRV);
   ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::TReg);
   ASSERT_EQ(std::get(Elem).Reg.Number, 42u);
+  ASSERT_EQ(std::get(Elem).Space, 4u);
+  ASSERT_EQ(std::get(Elem).Visibility, ShaderVisibility::Geometry);
 
   Elem = Elements[2];
   ASSERT_TRUE(std::holds_alternative(Elem));
   ASSERT_EQ(std::get(Elem).Type, ParamType::UAV);
   ASSERT_EQ(std::get(Elem).Reg.ViewType, RegisterType::UReg);
   ASSERT_EQ(std::get(Elem).Reg.Number, 34893247u);
+  ASSERT_EQ(std::get(Elem).Space, 0u);
+  ASSERT_EQ(std::get(Elem).Visibility, ShaderVisibility::Hull);
 
   ASSERT_TRUE(Consumer->isSatisfied());
 }
diff --git a/llv

[llvm-branch-commits] [clang] [llvm] [HLSL][RootSignature] Add parsing of flags to RootParam (PR #140152)

2025-05-15 Thread Finn Plummer via llvm-branch-commits


https://github.com/inbelic created 
https://github.com/llvm/llvm-project/pull/140152

- defines RootDescriptorFlags in-memory representation
- defines parseRootDescriptorFlags to be DXC compatible. This is why we support 
multiple `|` flags even validation will assert that only one flag is set...
- add unit tests to demonstrate functionality

Final part of and resolves https://github.com/llvm/llvm-project/issues/126577

>From ab3904222ef1d21a6040a27e753598d8525c068a Mon Sep 17 00:00:00 2001
From: Finn Plummer 
Date: Thu, 15 May 2025 20:31:30 +
Subject: [PATCH] [HLSL][RootSignature] Add parsing of flags to RootParam

- defines RootDescriptorFlags in-memory representation
- defines parseRootDescriptorFlags to be DXC compatible. This is why we
support multiple `|` flags even validation will assert that only one
flag is set...
- add unit tests to demonstrate functionality
---
 .../clang/Parse/ParseHLSLRootSignature.h  |  3 +
 clang/lib/Parse/ParseHLSLRootSignature.cpp| 60 +++
 .../Parse/ParseHLSLRootSignatureTest.cpp  | 20 ++-
 .../llvm/Frontend/HLSL/HLSLRootSignature.h| 25 
 4 files changed, 105 insertions(+), 3 deletions(-)

diff --git a/clang/include/clang/Parse/ParseHLSLRootSignature.h 
b/clang/include/clang/Parse/ParseHLSLRootSignature.h
index 436d217cec5b1..7b9168290d62a 100644
--- a/clang/include/clang/Parse/ParseHLSLRootSignature.h
+++ b/clang/include/clang/Parse/ParseHLSLRootSignature.h
@@ -93,6 +93,7 @@ class RootSignatureParser {
 std::optional Reg;
 std::optional Space;
 std::optional Visibility;
+std::optional Flags;
   };
   std::optional
   parseRootParamParams(RootSignatureToken::Kind RegType);
@@ -113,6 +114,8 @@ class RootSignatureParser {
 
   /// Parsing methods of various enums
   std::optional parseShaderVisibility();
+  std::optional
+  parseRootDescriptorFlags();
   std::optional
   parseDescriptorRangeFlags();
 
diff --git a/clang/lib/Parse/ParseHLSLRootSignature.cpp 
b/clang/lib/Parse/ParseHLSLRootSignature.cpp
index edb61f29f10d7..faf261cc9b7fe 100644
--- a/clang/lib/Parse/ParseHLSLRootSignature.cpp
+++ b/clang/lib/Parse/ParseHLSLRootSignature.cpp
@@ -193,6 +193,7 @@ std::optional 
RootSignatureParser::parseRootParam() {
 ExpectedReg = TokenKind::uReg;
 break;
   }
+  Param.setDefaultFlags();
 
   auto Params = parseRootParamParams(ExpectedReg);
   if (!Params.has_value())
@@ -214,6 +215,9 @@ std::optional 
RootSignatureParser::parseRootParam() {
   if (Params->Visibility.has_value())
 Param.Visibility = Params->Visibility.value();
 
+  if (Params->Flags.has_value())
+Param.Flags = Params->Flags.value();
+
   if (consumeExpectedToken(TokenKind::pu_r_paren,
diag::err_hlsl_unexpected_end_of_params,
/*param of=*/TokenKind::kw_RootConstants))
@@ -475,6 +479,23 @@ RootSignatureParser::parseRootParamParams(TokenKind 
RegType) {
 return std::nullopt;
   Params.Visibility = Visibility;
 }
+
+// `flags` `=` ROOT_DESCRIPTOR_FLAGS
+if (tryConsumeExpectedToken(TokenKind::kw_flags)) {
+  if (Params.Flags.has_value()) {
+getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_repeat_param)
+<< CurToken.TokKind;
+return std::nullopt;
+  }
+
+  if (consumeExpectedToken(TokenKind::pu_equal))
+return std::nullopt;
+
+  auto Flags = parseRootDescriptorFlags();
+  if (!Flags.has_value())
+return std::nullopt;
+  Params.Flags = Flags;
+}
   } while (tryConsumeExpectedToken(TokenKind::pu_comma));
 
   return Params;
@@ -654,6 +675,45 @@ RootSignatureParser::parseShaderVisibility() {
   return std::nullopt;
 }
 
+std::optional
+RootSignatureParser::parseRootDescriptorFlags() {
+  assert(CurToken.TokKind == TokenKind::pu_equal &&
+ "Expects to only be invoked starting at given keyword");
+
+  // Handle the edge-case of '0' to specify no flags set
+  if (tryConsumeExpectedToken(TokenKind::int_literal)) {
+if (!verifyZeroFlag()) {
+  getDiags().Report(CurToken.TokLoc, diag::err_hlsl_rootsig_non_zero_flag);
+  return std::nullopt;
+}
+return RootDescriptorFlags::None;
+  }
+
+  TokenKind Expected[] = {
+#define ROOT_DESCRIPTOR_FLAG_ENUM(NAME, LIT) TokenKind::en_##NAME,
+#include "clang/Lex/HLSLRootSignatureTokenKinds.def"
+  };
+
+  std::optional Flags;
+
+  do {
+if (tryConsumeExpectedToken(Expected)) {
+  switch (CurToken.TokKind) {
+#define ROOT_DESCRIPTOR_FLAG_ENUM(NAME, LIT)   
\
+  case TokenKind::en_##NAME:   
\
+Flags =
\
+maybeOrFlag(Flags, RootDescriptorFlags::NAME);
\
+break;
+#include "clang/Lex/HLSLRootSignatureTokenKinds.def"
+  default:
+llvm_unreachable("Switch for consumed enum token was not provided");
+  }
+}
+  } while (tryConsumeExpectedTo

74 matches

Mail list logo