[all-commits] [llvm/llvm-project] 136ac8: AMDGPU/GlobalISel: RegBankLegalize rules for pk_f6...
Krzysztof Parzyszek via All-commits
all-commits at lists.llvm.org
Mon Jun 15 09:12:46 PDT 2026
Branch: refs/heads/users/kparzysz/m01-misparsed-call
Home: https://github.com/llvm/llvm-project
Commit: 136ac86f75e1ae3332db1b58ea352f09f16a20c4
https://github.com/llvm/llvm-project/commit/136ac86f75e1ae3332db1b58ea352f09f16a20c4
Author: Petar Avramovic <Petar.Avramovic at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/AMDGPU/AMDGPURegBankLegalizeRules.cpp
M llvm/test/CodeGen/AMDGPU/fcanonicalize.ll
Log Message:
-----------
AMDGPU/GlobalISel: RegBankLegalize rules for pk_f64 fcanonicalize (#203870)
Commit: 9ce06fd29838777aaa89f133fe4b9f764be11d35
https://github.com/llvm/llvm-project/commit/9ce06fd29838777aaa89f133fe4b9f764be11d35
Author: Alexander Weinrauch <alexander.weinrauch at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
M llvm/test/CodeGen/AMDGPU/expert_scheduling_gfx1250.mir
Log Message:
-----------
[AMDGPU] Track LDS DMA VGPR source reads in expert scheduling (#203292)
LDS DMA instructions are marked as VALU but we need to treat them as
FLAT instructions for expert mode waitcnt insertion. This meant we were
missing a `vm_vsrc` wait for WAR and RAW on input registers of the LDS
DMA instruction.
I added lit tests for other memory ops as well to catch future
regressions.
Commit: 7dc09d0d3cf1812d7f845b9ae5c6ed67f1bd1082
https://github.com/llvm/llvm-project/commit/7dc09d0d3cf1812d7f845b9ae5c6ed67f1bd1082
Author: Owen Anderson <resistor at mac.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
A llvm/include/llvm/Support/CHERICapabilityFormat.h
A llvm/lib/Support/CHERICapabilityFormat.cpp
M llvm/lib/Support/CMakeLists.txt
A llvm/unittests/Support/CHERICapabilityFormatTest.cpp
M llvm/unittests/Support/CMakeLists.txt
Log Message:
-----------
[CHERI] Add a Support utility for determining alignment requirements of CHERI capabilities. (#197402)
On CHERI systems with compressed bounds representations (which is all of
them that anyone cares about today), one of the tradeoffs to achieve
that compression is a requirement for larger allocations to be more
highly aligned. This impacts both code generation and linking in places
where globals need to be aligned and/or padded based on this
requirement. The specific alignment requirements vary by capability
format.
Co-authored-by: Alex Richardson <alexrichardson at google.com>
Commit: 5fb08f44c1d2bb430cfa55ec808352885ae216c0
https://github.com/llvm/llvm-project/commit/5fb08f44c1d2bb430cfa55ec808352885ae216c0
Author: Pavel Labath <pavel at labath.sk>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M libc/config/linux/aarch64/entrypoints.txt
M libc/config/linux/riscv/entrypoints.txt
M libc/config/linux/x86_64/entrypoints.txt
M libc/include/netinet/in.yaml
M libc/src/CMakeLists.txt
A libc/src/netinet/CMakeLists.txt
A libc/src/netinet/in6addr_any.cpp
A libc/src/netinet/in6addr_any.h
A libc/src/netinet/in6addr_loopback.cpp
A libc/src/netinet/in6addr_loopback.h
M libc/test/src/netinet/CMakeLists.txt
M libc/test/src/netinet/in_test.cpp
Log Message:
-----------
[libc] Add the in6addr_{any,loopback} variables (#203579)
The existence of the variables is required by POSIX.
Assisted by Gemini.
Commit: 5388feaf865f2d48592588417bdf9d7f93de0755
https://github.com/llvm/llvm-project/commit/5388feaf865f2d48592588417bdf9d7f93de0755
Author: Gil Rapaport <gil.rapaport at mobileye.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
M mlir/lib/Conversion/FuncToEmitC/FuncToEmitC.cpp
M mlir/lib/Dialect/EmitC/IR/EmitC.cpp
M mlir/lib/Target/Cpp/TranslateToCpp.cpp
M mlir/test/Conversion/FuncToEmitC/func-to-emitc.mlir
M mlir/test/Dialect/EmitC/invalid_ops.mlir
M mlir/test/Dialect/EmitC/ops.mlir
M mlir/test/Target/Cpp/func.mlir
M mlir/test/Target/Cpp/member.mlir
Log Message:
-----------
[mlir][emitc] Support member access for values (#203308)
The `emitc.member` op is currently limited to taking lvalues of opaque
types representing structs and returning either lvalues or arrays of its
fields. Accessing members of SSA values of opaque types, therefore,
requires assigning them to an `emitc.variable`, applying `emitc.member`
to it and `emitc.load` to the member's lvalue. For users only wishing to
read members of a struct value, this should be redundant.
This PR extends `emitc.member` to handle struct values directly by
accepting opaque types as argument, as long as the result type is
neither an lvalue nor an array, which imply memory location. This
provides similar semantics to extracting elements out of SSA
tensors/vectors.
Commit: 6e6c296f20149b0a9e793c2d526566bddb162939
https://github.com/llvm/llvm-project/commit/6e6c296f20149b0a9e793c2d526566bddb162939
Author: Mehdi Amini <joker.eph at gmail.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M flang/lib/Optimizer/Analysis/AliasAnalysis.cpp
M flang/lib/Optimizer/Dialect/FIROps.cpp
M flang/lib/Optimizer/OpenACC/Support/FIROpenACCOpsInterfaces.cpp
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.h
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.td
M mlir/lib/Analysis/AliasAnalysis/LocalAliasAnalysis.cpp
M mlir/lib/Analysis/DataFlow/DeadCodeAnalysis.cpp
M mlir/lib/Analysis/DataFlow/DenseAnalysis.cpp
M mlir/lib/Analysis/DataFlow/LivenessAnalysis.cpp
M mlir/lib/Analysis/DataFlow/SparseAnalysis.cpp
M mlir/lib/Analysis/SliceWalk.cpp
M mlir/lib/Dialect/Affine/IR/AffineOps.cpp
M mlir/lib/Dialect/Async/IR/Async.cpp
M mlir/lib/Dialect/Bufferization/Transforms/OwnershipBasedBufferDeallocation.cpp
M mlir/lib/Dialect/EmitC/IR/EmitC.cpp
M mlir/lib/Dialect/Func/IR/FuncOps.cpp
M mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
M mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp
M mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp
M mlir/lib/Dialect/OpenACC/IR/OpenACCCG.cpp
M mlir/lib/Dialect/SCF/IR/SCF.cpp
M mlir/lib/Dialect/Shape/IR/Shape.cpp
M mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp
M mlir/lib/Dialect/Transform/IR/TransformOps.cpp
M mlir/lib/Dialect/Transform/TuneExtension/TuneExtensionOps.cpp
M mlir/lib/Dialect/XeGPU/Transforms/XeGPULayoutImpl.cpp
M mlir/lib/Interfaces/ControlFlowInterfaces.cpp
M mlir/test/Dialect/SCF/invalid.mlir
M mlir/test/lib/Analysis/DataFlow/TestDenseBackwardDataFlowAnalysis.cpp
M mlir/test/lib/Dialect/Test/TestOpDefs.cpp
M mlir/unittests/Interfaces/ControlFlowInterfacesTest.cpp
Log Message:
-----------
[mlir] Use explicit op targets for region successors (#203552)
Replace the RegionSuccessor parent sentinel with an explicit Operation *
target. Operation successors now model continuing after that operation,
while region successors continue to point at Region.
Update RegionBranchOpInterface helpers, analyses, and dialect
implementations to use operation successors. This removes the ambiguity
of interpreting parent relative to whichever operation produced the
successor when introducing early-exits.
For downstream integration for this API change:
- `successor.isParent()` -> `successor.isOperation()`
- `RegionSuccessor::parent()` -> RegionSuccessor(op)` ; that means that
you need to have access to the target op. In general it is either
`getOperation()` or `getParentOp()` (depending if you're updating a
method on the `RegionBranchOpInterface` operation or on the terminator.
Assisted-by: Codex
Commit: 28f5731cc0b785dddb73c6a12bb4fd35b5cea449
https://github.com/llvm/llvm-project/commit/28f5731cc0b785dddb73c6a12bb4fd35b5cea449
Author: Mel Chen <mel.chen at sifive.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/test/Transforms/LoopVectorize/RISCV/strided-accesses.ll
Log Message:
-----------
[LV] Pre-commit test case for strided accesses derived from lshr exact. nfc (#203488)
Commit: 7bc9882cce313b73b263b4c1377ccd322794f413
https://github.com/llvm/llvm-project/commit/7bc9882cce313b73b263b4c1377ccd322794f413
Author: Nikita Popov <npopov at redhat.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
M llvm/test/Transforms/InstCombine/strcmp-memcmp.ll
Log Message:
-----------
[SimplifyLibCalls] Use context instruction strcmp->memcmp transform (#203893)
Pass the context instruction when checking for dereferenceability.
Commit: 00671e0bdaa14562af80d5fba16ca852242d6c11
https://github.com/llvm/llvm-project/commit/00671e0bdaa14562af80d5fba16ca852242d6c11
Author: David Green <david.green at arm.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp
A llvm/test/CodeGen/AArch64/GlobalISel/knownbits-insert-vector.mir
Log Message:
-----------
[GlobalISel] Add known-bits for G_INSERT_VECTOR_ELT. (#200930)
This adds known-bits for G_INSERT_VECTOR_ELT, mostly a copy of the SDAG equivalent code.
Commit: 46770bfbdb9ab00c58d07f657dcbd168434ee2a8
https://github.com/llvm/llvm-project/commit/46770bfbdb9ab00c58d07f657dcbd168434ee2a8
Author: Pooja Hemashekar <hemashekar at roofline.ai>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M mlir/include/mlir-c/Dialect/Linalg.h
M mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
M mlir/lib/Bindings/Python/DialectLinalg.cpp
M mlir/lib/CAPI/Dialect/Linalg.cpp
M mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
M mlir/test/python/dialects/linalg/utils.py
M mlir/unittests/Dialect/Linalg/InferConvolutionDimsTest.cpp
Log Message:
-----------
[mlir][linalg] Add maps-based inferConvolutionDims overload (#203323)
This PR adds an overload of `inferConvolutionDims` that takes only
indexing maps (input, filter, output), mirroring the maps-based
`inferContractionDims` overload added in #76081, along with the
corresponding C API and python bindings.
Assisted by: Claude-code
Commit: d9f14cafbbe7e8a4e34a48128539c15099f28563
https://github.com/llvm/llvm-project/commit/d9f14cafbbe7e8a4e34a48128539c15099f28563
Author: Pavel Labath <pavel at labath.sk>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M libc/src/__support/endian_internal.h
M libc/src/arpa/inet/ntohl.cpp
M libc/src/arpa/inet/ntohs.cpp
M libc/test/src/__support/endian_internal_test.cpp
Log Message:
-----------
[libc] Add Endian::from_{big,little}_endian (#203895)
This is actually the same function as its to_ counterpart, but it lets
us correctly express the intent.
The functions are already useful for implementing ntoh?, but I'll add
other uses of them soon.
Commit: 1b51a09b73e9b1868830522c38d109a70135dbd7
https://github.com/llvm/llvm-project/commit/1b51a09b73e9b1868830522c38d109a70135dbd7
Author: Timm Baeder <tbaeder at redhat.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M clang/lib/AST/ByteCode/Compiler.cpp
M clang/lib/AST/ByteCode/Interp.cpp
M clang/lib/AST/ByteCode/Interp.h
M clang/lib/AST/ByteCode/InterpState.h
M clang/lib/AST/ByteCode/Opcodes.td
M clang/lib/AST/ByteCode/Pointer.h
A clang/test/AST/ByteCode/dynamic-cast.cpp
Log Message:
-----------
[clang][bytecode] Actually implement `dynamic_cast` (#203489)
Do a full type search if necessary.
Commit: 688bc5341814524478f7dcb14d5cd3dd9d231b0c
https://github.com/llvm/llvm-project/commit/688bc5341814524478f7dcb14d5cd3dd9d231b0c
Author: fineg74 <61437305+fineg74 at users.noreply.github.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
M llvm/test/CodeGen/SPIRV/transcoding/atomic-load-store-unsupported.ll
A llvm/test/CodeGen/SPIRV/transcoding/load-atomic-ptr.ll
Log Message:
-----------
[SPIRV] Let atomic load load pointers (#202823)
SPIRV atomic load permits only integer or floats as per SPIRV spec. When
compiling libc there several places in the code where pointers are
atomically loaded causing compilation to break. It can be fixed by using
casting in the libc code but in order to keep the libc code clean it is
preferrable to do it in SPIRV backend. This change will cast pointer
parameter to a pointer to integer of appropriate size and generate
atomic load instruction that uses integers per SPIRV spec. The value
returned by atomic load is casted to an original pointer type
---------
Co-authored-by: Dmitry Sidorov <dsidorov at amd.com>
Commit: e26e9ba06f46f9619bb6d2df40850a54559640ce
https://github.com/llvm/llvm-project/commit/e26e9ba06f46f9619bb6d2df40850a54559640ce
Author: Nikolas Klauser <nikolasklauser at berlin.de>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/include/llvm/Analysis/Loads.h
M llvm/lib/Analysis/Loads.cpp
M llvm/lib/CodeGen/MachineOperand.cpp
M llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
M llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
M llvm/lib/Transforms/Scalar/MergeICmps.cpp
M llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
Log Message:
-----------
[Loads][NFC] Add isDereferenceablePointer overload taking a size (#203905)
Quite a few places just end up passing `Align(1)` as the alignment.
Instead, provide an overload of `isDereferenceablePointer` taking a size
instead of a `Type*` and use that.
Commit: f1f79b53a57792af2120f02d97fab98902fc6747
https://github.com/llvm/llvm-project/commit/f1f79b53a57792af2120f02d97fab98902fc6747
Author: Anshil Gandhi <95053726+gandhi56 at users.noreply.github.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/test/CodeGen/AMDGPU/usubsat.ll
Log Message:
-----------
[NFC][AMDGPU] Precommit tests for usubsat pattern (#203453)
Add tests to illustrate codegen changes in PR #203155.
Commit: 2f0a17cc05242d8b98d43a4c60a60f6bcbf5ed10
https://github.com/llvm/llvm-project/commit/2f0a17cc05242d8b98d43a4c60a60f6bcbf5ed10
Author: Yao Qi <yao_qi at apple.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M clang/lib/Frontend/TextDiagnosticPrinter.cpp
M clang/unittests/Frontend/TextDiagnosticTest.cpp
M lldb/test/API/commands/expression/diagnostics/TestExprDiagnostics.py
M lldb/test/Shell/Expr/TestObjCxxEnumConflict.test
Log Message:
-----------
[clang] Honor ShowLevel for diagnostics without a source location (#203520)
When lldb evaluates expression and gets error, it prints double
`error: ` as below,
```
(lldb) expression v1::withImplicitTag(Simple{.mem = 6})
note: Ran expression as 'C++11'.
error: error: Multiple internal symbols found for 'v1'
```
The first `error:` is from lldb and the second `error:` is from
clang's diagnostic. LLDB's `ClangDiagnosticManagerAdapter` sets
`ShowLevel=false`, so that its own rendering
layer (`RenderDiagnosticDetails`) can add the severity prefix with
color. However, clang still adds "error:" in the output buffer means
the guard is missing.
`TextDiagnosticPrinter::HandleDiagnostic` has two paths: one for
diagnostics with a valid source location (delegating to
`TextDiagnostic::emitDiagnostic`) and a simpler path for diagnostics
without one. The with-location path already guarded
`printDiagnosticLevel`
behind `DiagOpts.ShowLevel`; the no-location path called it
unconditionally.
Add a parameterized unit test covering `ShowLevel=true` and
`ShowLevel=false`
for no-location diagnostics to prevent regression.
---------
Co-authored-by: Michael Buch <michaelbuch12 at gmail.com>
Commit: 86add58afb1c2de7339e779ecec5d812810350f3
https://github.com/llvm/llvm-project/commit/86add58afb1c2de7339e779ecec5d812810350f3
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
M llvm/test/CodeGen/SPIRV/hlsl-resources/SampleErrors.ll
Log Message:
-----------
[SPIR-V] Return false after non-constant offset error in generateSampleImage (#203512)
generateSampleImage emitted the diagnostic but kept building the
instruction, producing a malformed OpImageSample with the offset
silently dropped
Commit: e919148ec7b3d288dc91ab770b49f43b45cb7877
https://github.com/llvm/llvm-project/commit/e919148ec7b3d288dc91ab770b49f43b45cb7877
Author: Asher Mancinelli <ashermancinelli at gmail.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
Log Message:
-----------
[mlir][nfc] Fix typo in tcgen05.ld.red docs (#203630)
Signed-off-by: Asher Mancinelli <amancinelli at nvidia.com>
Commit: d038850fde42c69028b84135437cb2be994d0ada
https://github.com/llvm/llvm-project/commit/d038850fde42c69028b84135437cb2be994d0ada
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
A llvm/test/CodeGen/SPIRV/pointers/getelementptr-subbyte-array.ll
Log Message:
-----------
[SPIR-V] Fix division by zero on byte addressing GEP into a subbyte array (#202593)
walkLogicalAccessChainConstant divided the byte offset by the element
size which is zero for sub-byte element types ([N x i1])
Commit: 2a789821b0d723bc92f61563e74e67d69d660927
https://github.com/llvm/llvm-project/commit/2a789821b0d723bc92f61563e74e67d69d660927
Author: Stephen Tozer <stephen.tozer at sony.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M cross-project-tests/debuginfo-tests/dexter/dex/debugger/DebuggerControllers/ScriptDebuggerController.py
M cross-project-tests/debuginfo-tests/dexter/dex/evaluation/StateMatch.py
M cross-project-tests/debuginfo-tests/dexter/dex/test_script/Nodes.py
M cross-project-tests/debuginfo-tests/dexter/dex/test_script/Script.py
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/Inputs/header.h
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/invalid_label.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/offset.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/simple_labels.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/source_root_dir.cpp
Log Message:
-----------
Revert "[Dexter] Add label nodes for line references" (#203920)
Reverts llvm/llvm-project#202544
Commit appears to be causing failures in the pre-merge checks for
AArch64 specifically, though this is not entirely consistent. Reverting
to fix pre-merge checks immediately, will aim to reland with either a
fix or disabling tests for AArch64.
Commit: f32b9eb416beaa13cceac849905a9cc0a993620d
https://github.com/llvm/llvm-project/commit/f32b9eb416beaa13cceac849905a9cc0a993620d
Author: Björn Pettersson <bjorn.a.pettersson at ericsson.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Transforms/Scalar/ConstantHoisting.cpp
A llvm/test/Transforms/ConstantHoisting/AArch64/consthoist-unreachable-phi-edge.ll
M llvm/test/Transforms/ConstantHoisting/X86/pr52689-not-all-uses-rebased.ll
Log Message:
-----------
[ConstantHoisting] Skip PHI edges from unreachable blocks (#203892)
When collecting constant candidates, skip incoming PHI edges from blocks
that are unreachable from entry. This avoids assertion failures when the
pass later tries to find insertion points or update users for constants
that only appear on unreachable edges.
While touching this part of the code, also remove an older XFAIL from
test/Transforms/ConstantHoisting/X86/pr52689-not-all-uses-rebased.ll
That test case also triggered the same assert once upon a time, but it
has been set to XFAIL for some time since the reproducer no longer
triggered the bug. This patch turns it into a normal test case instead
of an XFAIL test. Afaict the original problem may have been the same. We
have PHI nodes with edges from unreachable blocks. One difference
compared to the new aarch64 test is that here the involved constants are
GEPs and not simple scalars.
Closes https://github.com/llvm/llvm-project/issues/52689
Commit: 5fbc81a2f2b00c21e84bb8f450233713acb6edda
https://github.com/llvm/llvm-project/commit/5fbc81a2f2b00c21e84bb8f450233713acb6edda
Author: Nikita Popov <npopov at redhat.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Transforms/Scalar/JumpThreading.cpp
Log Message:
-----------
[JumpThreading] Use isGuaranteedToTransferExecutionToSuccessor() with range (#203918)
Use the overload that accepts a range of instructions. This is not NFC
because the scan is now subject to ScanLimit.
Commit: 2d8a394129b3b538071f7774e34c31d7cbf3c882
https://github.com/llvm/llvm-project/commit/2d8a394129b3b538071f7774e34c31d7cbf3c882
Author: Robert Imschweiler <robert.imschweiler at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M openmp/runtime/src/CMakeLists.txt
A openmp/runtime/src/kmp_adt.cpp
A openmp/runtime/src/kmp_adt.h
M openmp/runtime/src/kmp_str.cpp
M openmp/runtime/src/kmp_str.h
A openmp/runtime/unittests/ADT/CMakeLists.txt
A openmp/runtime/unittests/ADT/TestStringRef.cpp
M openmp/runtime/unittests/CMakeLists.txt
M openmp/runtime/unittests/String/TestKmpStr.cpp
Log Message:
-----------
[libomp] Add kmp_str_ref (ADT 1/2) (#176162)
libomp currently has two limitations:
1) although it's C++, it doesn't link against the C++ stdlib 2) it
cannot link against the implementation of LLVM ADTs
These limitations shall not be altered at the moment.
As a result, this commit introduces kmp_str_ref, which is similar to
LLVM's StringRef. It currently only includes methods I need at the
moment, but it's easily extensible.
Commit: e57ebfdaefa7c2ea660878984cfe5c739f626d6b
https://github.com/llvm/llvm-project/commit/e57ebfdaefa7c2ea660878984cfe5c739f626d6b
Author: Jason Van Beusekom <jason.van-beusekom at hpe.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M flang/include/flang/Support/Flags.h
M flang/lib/Lower/OpenMP/OpenMP.cpp
M flang/lib/Semantics/resolve-directives.cpp
M flang/lib/Support/Flags.cpp
M flang/test/Integration/OpenMP/map-types-and-sizes.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/distribute-standalone-private.f90
A flang/test/Lower/OpenMP/DelayedPrivatization/target-firstprivate.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-allocatable.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-implicit-scalar-map-2.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-implicit-scalar-map.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-simple.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-distribute-parallel-do-simd-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-distribute-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-nested-distribute-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-private-implicit-scalar-map.f90
R flang/test/Lower/OpenMP/Todo/firstprivate-target.f90
M flang/test/Lower/OpenMP/defaultmap.f90
M flang/test/Lower/OpenMP/different_vars_lastprivate_barrier.f90
M flang/test/Lower/OpenMP/optional-argument-map-2.f90
M flang/test/Lower/OpenMP/private-derived-type.f90
M flang/test/Lower/OpenMP/same_var_first_lastprivate.f90
M flang/test/Lower/OpenMP/target-map-complex.f90
M flang/test/Lower/OpenMP/target-scope.f90
M flang/test/Lower/OpenMP/target.f90
M flang/test/Lower/do_concurrent_delayed_locality.f90
M flang/test/Lower/do_concurrent_local_assoc_entity.f90
M flang/test/Lower/do_concurrent_local_default_init.f90
M flang/test/Lower/do_concurrent_loop_in_nested_block.f90
M flang/test/Lower/do_concurrent_reduce.f90
M flang/test/Semantics/OpenMP/symbol08.f90
A offload/test/offloading/fortran/target-firstprivate.f90
Log Message:
-----------
[Flang][OpenMP] remove enable-delayed-privatization-staging to suppor… (#203626)
…t target first private default
This commit follows the decision in #182356 to remove the not yet
implemented for delayed privatization for firstprivate and private in
`omp target` regions in flang
Fixes #182356
Assisted with Opus
Commit: 5385418df01200d7388d3c42e60c14e938645838
https://github.com/llvm/llvm-project/commit/5385418df01200d7388d3c42e60c14e938645838
Author: jay0x <90309873+blazie2004 at users.noreply.github.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M flang-rt/lib/runtime/io-api.cpp
M flang-rt/unittests/Runtime/ExternalIOTest.cpp
Log Message:
-----------
[Flang] [Runtime ]Fix write endfile abort (#191633)
A WRITE after ENDFILE with ERR= or IOSTAT= was crashing instead of
handling the error properly. Earlier, the program was crashing because
the error was triggered too early (before error handling was ready).
---------
Co-authored-by: Jay Satish Kumar Patel <kumarpat at hpe.com>
Commit: edb7ad812dab579ac6d869f34a5a9e3f4d5992de
https://github.com/llvm/llvm-project/commit/edb7ad812dab579ac6d869f34a5a9e3f4d5992de
Author: Julian Klappenbach <julian at twilight.digital>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
A llvm/test/CodeGen/SPIRV/pointers/type-deduce-global-array-poison.ll
Log Message:
-----------
[SPIR-V] Deduce pointee type for all global variables, not only initialized ones (#202047)
SPIRVEmitIntrinsics::processGlobalValue only recorded a global variable's
element type when hasInitializer() was true. Now it's recording it unconditionally.
Assisted-by: Claude (Anthropic)
Commit: 53e36fb7833acd81a5d07eeffa7ff8ae6968125e
https://github.com/llvm/llvm-project/commit/53e36fb7833acd81a5d07eeffa7ff8ae6968125e
Author: Maryam Moghadas <maryammo at ca.ibm.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M clang/lib/Driver/Driver.cpp
M clang/test/Preprocessor/zos-target.c
Log Message:
-----------
[z/OS][Clang] Raise minimum supported z/OS target level to 3.1 (#203580)
Update z/OS target validation and tests to reject targets older than
z/OS 3.1
Commit: fff4896df38cd1decf6361a8a98e750befb759d9
https://github.com/llvm/llvm-project/commit/fff4896df38cd1decf6361a8a98e750befb759d9
Author: forking-google-bazel-bot[bot] <265904573+forking-google-bazel-bot[bot]@users.noreply.github.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M utils/bazel/llvm-project-overlay/clang/unittests/BUILD.bazel
Log Message:
-----------
[Bazel] Fixes 63c9264 (#203875)
This fixes 63c9264c770b4ffb7b5b3816547db45969026c3f.
Co-authored-by: Google Bazel Bot <google-bazel-bot at google.com>
Commit: 0919b1e54c6eff1841948717035eeffe21b27c6c
https://github.com/llvm/llvm-project/commit/0919b1e54c6eff1841948717035eeffe21b27c6c
Author: Jonas Devlieghere <jonas at devlieghere.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M lldb/include/lldb/Core/Statusline.h
M lldb/source/Core/Statusline.cpp
M lldb/source/Host/common/Editline.cpp
M lldb/test/API/functionalities/statusline/TestStatusline.py
Log Message:
-----------
[lldb] Recompute the statusline on resize without clearing the screen (#202691)
On a terminal resize the statusline cleared the whole screen (ESC[2J)
and redrew, because recomputing in place was buggy: the statusline
wrapped and duplicated. The clear also wiped the visible scrollback on
every resize. I got lots of feedback that this wasn't a great user
experience so I spent some time taking another stab at this.
This PR reverts back to recomputing the statusline. After a resize the
terminal still shows the old statusline:
- Making the terminal smaller (horizontally) reflows the full-width line
into ceil(prev_width / width) rows at the bottom
- Making the terminal larger (vertically) leaves it stranded at its old
row
Clear only the rows it can still occupy and redraw, preserving the
scrollback above. Disable autowrap while drawing so a line briefly wider
than the terminal is clipped at the margin rather than wrapping onto the
row above.
Removing the clear exposed that Editline::Refresh() redrew the prompt
via libedit's EL_REFRESH, which uses a cursor model that is stale once
the statusline has moved the cursor, tiling the prompt once per resize.
Repaint the prompt and input from our own tracked position instead.
Fixes #146919
Commit: 8af74a033b667a9af256744a0c701a535694f6d3
https://github.com/llvm/llvm-project/commit/8af74a033b667a9af256744a0c701a535694f6d3
Author: Alex Duran <alejandro.duran at intel.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M offload/plugins-nextgen/level_zero/include/L0Options.h
Log Message:
-----------
[OFFLOAD][L0] Switch to use inorder queues by default (#203897)
Now that other pieces are in place we can switch to using Level Zero
inorder queues by default.
Commit: 132adf753ac1a78ced1839678bb03070c1ca25df
https://github.com/llvm/llvm-project/commit/132adf753ac1a78ced1839678bb03070c1ca25df
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
M llvm/test/CodeGen/SPIRV/llvm-intrinsics/fp-intrinsics.ll
M llvm/test/CodeGen/SPIRV/llvm-intrinsics/frexp.ll
Log Message:
-----------
[SPIR-V] Skip dead second result load for frexp and modf (#201891)
Skip the load for the result in case it is in fact not used
Commit: d4ca116af2e4f792b7967817cd7856dab6ed87ac
https://github.com/llvm/llvm-project/commit/d4ca116af2e4f792b7967817cd7856dab6ed87ac
Author: Dave Lee <davelee.com at gmail.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M lldb/include/lldb/lldb-enumerations.h
Log Message:
-----------
[lldb] Reformat doxygen comments in lldb-enumerations.h (NFC) (#203079)
Convert doxygen comments to precede the enumerator to which they apply
(using `///`). This placement of documentation is more consistent with
how functions and classes are documented. Additionally, with the column
limit, the documentation was quite crammed as it was. Lastly, comments
have been reflowed, so that make full use of horizontal space.
Assisted-by: claude
Commit: 62f85dee13df277609bcf7d02d91e6823b6d16cd
https://github.com/llvm/llvm-project/commit/62f85dee13df277609bcf7d02d91e6823b6d16cd
Author: Krzysztof Parzyszek <Krzysztof.Parzyszek at amd.com>
Date: 2026-06-15 (Mon, 15 Jun 2026)
Changed paths:
M clang/lib/AST/ByteCode/Compiler.cpp
M clang/lib/AST/ByteCode/Interp.cpp
M clang/lib/AST/ByteCode/Interp.h
M clang/lib/AST/ByteCode/InterpState.h
M clang/lib/AST/ByteCode/Opcodes.td
M clang/lib/AST/ByteCode/Pointer.h
M clang/lib/Driver/Driver.cpp
M clang/lib/Frontend/TextDiagnosticPrinter.cpp
A clang/test/AST/ByteCode/dynamic-cast.cpp
M clang/test/Preprocessor/zos-target.c
M clang/unittests/Frontend/TextDiagnosticTest.cpp
M cross-project-tests/debuginfo-tests/dexter/dex/debugger/DebuggerControllers/ScriptDebuggerController.py
M cross-project-tests/debuginfo-tests/dexter/dex/evaluation/StateMatch.py
M cross-project-tests/debuginfo-tests/dexter/dex/test_script/Nodes.py
M cross-project-tests/debuginfo-tests/dexter/dex/test_script/Script.py
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/Inputs/header.h
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/invalid_label.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/offset.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/simple_labels.cpp
R cross-project-tests/debuginfo-tests/dexter/feature_tests/scripts/labels/source_root_dir.cpp
M flang-rt/lib/runtime/io-api.cpp
M flang-rt/unittests/Runtime/ExternalIOTest.cpp
M flang/include/flang/Support/Flags.h
M flang/lib/Lower/OpenMP/OpenMP.cpp
M flang/lib/Optimizer/Analysis/AliasAnalysis.cpp
M flang/lib/Optimizer/Dialect/FIROps.cpp
M flang/lib/Optimizer/OpenACC/Support/FIROpenACCOpsInterfaces.cpp
M flang/lib/Semantics/resolve-directives.cpp
M flang/lib/Support/Flags.cpp
M flang/test/Integration/OpenMP/map-types-and-sizes.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/distribute-standalone-private.f90
A flang/test/Lower/OpenMP/DelayedPrivatization/target-firstprivate.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-allocatable.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-implicit-scalar-map-2.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-implicit-scalar-map.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-multiple-variables.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-private-simple.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-distribute-parallel-do-simd-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-distribute-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-nested-distribute-private-adjustable-array.f90
M flang/test/Lower/OpenMP/DelayedPrivatization/target-teams-private-implicit-scalar-map.f90
R flang/test/Lower/OpenMP/Todo/firstprivate-target.f90
M flang/test/Lower/OpenMP/defaultmap.f90
M flang/test/Lower/OpenMP/different_vars_lastprivate_barrier.f90
M flang/test/Lower/OpenMP/optional-argument-map-2.f90
M flang/test/Lower/OpenMP/private-derived-type.f90
M flang/test/Lower/OpenMP/same_var_first_lastprivate.f90
M flang/test/Lower/OpenMP/target-map-complex.f90
M flang/test/Lower/OpenMP/target-scope.f90
M flang/test/Lower/OpenMP/target.f90
M flang/test/Lower/do_concurrent_delayed_locality.f90
M flang/test/Lower/do_concurrent_local_assoc_entity.f90
M flang/test/Lower/do_concurrent_local_default_init.f90
M flang/test/Lower/do_concurrent_loop_in_nested_block.f90
M flang/test/Lower/do_concurrent_reduce.f90
M flang/test/Semantics/OpenMP/symbol08.f90
M libc/config/linux/aarch64/entrypoints.txt
M libc/config/linux/riscv/entrypoints.txt
M libc/config/linux/x86_64/entrypoints.txt
M libc/include/netinet/in.yaml
M libc/src/CMakeLists.txt
M libc/src/__support/endian_internal.h
M libc/src/arpa/inet/ntohl.cpp
M libc/src/arpa/inet/ntohs.cpp
A libc/src/netinet/CMakeLists.txt
A libc/src/netinet/in6addr_any.cpp
A libc/src/netinet/in6addr_any.h
A libc/src/netinet/in6addr_loopback.cpp
A libc/src/netinet/in6addr_loopback.h
M libc/test/src/__support/endian_internal_test.cpp
M libc/test/src/netinet/CMakeLists.txt
M libc/test/src/netinet/in_test.cpp
M lldb/include/lldb/Core/Statusline.h
M lldb/include/lldb/lldb-enumerations.h
M lldb/source/Core/Statusline.cpp
M lldb/source/Host/common/Editline.cpp
M lldb/test/API/commands/expression/diagnostics/TestExprDiagnostics.py
M lldb/test/API/functionalities/statusline/TestStatusline.py
M lldb/test/Shell/Expr/TestObjCxxEnumConflict.test
M llvm/include/llvm/Analysis/Loads.h
A llvm/include/llvm/Support/CHERICapabilityFormat.h
M llvm/lib/Analysis/Loads.cpp
M llvm/lib/CodeGen/GlobalISel/GISelValueTracking.cpp
M llvm/lib/CodeGen/MachineOperand.cpp
A llvm/lib/Support/CHERICapabilityFormat.cpp
M llvm/lib/Support/CMakeLists.txt
M llvm/lib/Target/AMDGPU/AMDGPURegBankLegalizeRules.cpp
M llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
M llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
M llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
M llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
M llvm/lib/Transforms/Scalar/ConstantHoisting.cpp
M llvm/lib/Transforms/Scalar/JumpThreading.cpp
M llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
M llvm/lib/Transforms/Scalar/MergeICmps.cpp
M llvm/lib/Transforms/Utils/SimplifyLibCalls.cpp
A llvm/test/CodeGen/AArch64/GlobalISel/knownbits-insert-vector.mir
M llvm/test/CodeGen/AMDGPU/expert_scheduling_gfx1250.mir
M llvm/test/CodeGen/AMDGPU/fcanonicalize.ll
M llvm/test/CodeGen/AMDGPU/usubsat.ll
M llvm/test/CodeGen/SPIRV/hlsl-resources/SampleErrors.ll
M llvm/test/CodeGen/SPIRV/llvm-intrinsics/fp-intrinsics.ll
M llvm/test/CodeGen/SPIRV/llvm-intrinsics/frexp.ll
A llvm/test/CodeGen/SPIRV/pointers/getelementptr-subbyte-array.ll
A llvm/test/CodeGen/SPIRV/pointers/type-deduce-global-array-poison.ll
M llvm/test/CodeGen/SPIRV/transcoding/atomic-load-store-unsupported.ll
A llvm/test/CodeGen/SPIRV/transcoding/load-atomic-ptr.ll
A llvm/test/Transforms/ConstantHoisting/AArch64/consthoist-unreachable-phi-edge.ll
M llvm/test/Transforms/ConstantHoisting/X86/pr52689-not-all-uses-rebased.ll
M llvm/test/Transforms/InstCombine/strcmp-memcmp.ll
M llvm/test/Transforms/LoopVectorize/RISCV/strided-accesses.ll
A llvm/unittests/Support/CHERICapabilityFormatTest.cpp
M llvm/unittests/Support/CMakeLists.txt
M mlir/include/mlir-c/Dialect/Linalg.h
M mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
M mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
M mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.h
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.td
M mlir/lib/Analysis/AliasAnalysis/LocalAliasAnalysis.cpp
M mlir/lib/Analysis/DataFlow/DeadCodeAnalysis.cpp
M mlir/lib/Analysis/DataFlow/DenseAnalysis.cpp
M mlir/lib/Analysis/DataFlow/LivenessAnalysis.cpp
M mlir/lib/Analysis/DataFlow/SparseAnalysis.cpp
M mlir/lib/Analysis/SliceWalk.cpp
M mlir/lib/Bindings/Python/DialectLinalg.cpp
M mlir/lib/CAPI/Dialect/Linalg.cpp
M mlir/lib/Conversion/FuncToEmitC/FuncToEmitC.cpp
M mlir/lib/Dialect/Affine/IR/AffineOps.cpp
M mlir/lib/Dialect/Async/IR/Async.cpp
M mlir/lib/Dialect/Bufferization/Transforms/OwnershipBasedBufferDeallocation.cpp
M mlir/lib/Dialect/EmitC/IR/EmitC.cpp
M mlir/lib/Dialect/Func/IR/FuncOps.cpp
M mlir/lib/Dialect/GPU/IR/GPUDialect.cpp
M mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp
M mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp
M mlir/lib/Dialect/OpenACC/IR/OpenACC.cpp
M mlir/lib/Dialect/OpenACC/IR/OpenACCCG.cpp
M mlir/lib/Dialect/SCF/IR/SCF.cpp
M mlir/lib/Dialect/Shape/IR/Shape.cpp
M mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp
M mlir/lib/Dialect/Transform/IR/TransformOps.cpp
M mlir/lib/Dialect/Transform/TuneExtension/TuneExtensionOps.cpp
M mlir/lib/Dialect/XeGPU/Transforms/XeGPULayoutImpl.cpp
M mlir/lib/Interfaces/ControlFlowInterfaces.cpp
M mlir/lib/Target/Cpp/TranslateToCpp.cpp
M mlir/test/Conversion/FuncToEmitC/func-to-emitc.mlir
M mlir/test/Dialect/EmitC/invalid_ops.mlir
M mlir/test/Dialect/EmitC/ops.mlir
M mlir/test/Dialect/SCF/invalid.mlir
M mlir/test/Target/Cpp/func.mlir
M mlir/test/Target/Cpp/member.mlir
M mlir/test/lib/Analysis/DataFlow/TestDenseBackwardDataFlowAnalysis.cpp
M mlir/test/lib/Dialect/Test/TestOpDefs.cpp
M mlir/test/python/dialects/linalg/utils.py
M mlir/unittests/Dialect/Linalg/InferConvolutionDimsTest.cpp
M mlir/unittests/Interfaces/ControlFlowInterfacesTest.cpp
M offload/plugins-nextgen/level_zero/include/L0Options.h
A offload/test/offloading/fortran/target-firstprivate.f90
M openmp/runtime/src/CMakeLists.txt
A openmp/runtime/src/kmp_adt.cpp
A openmp/runtime/src/kmp_adt.h
M openmp/runtime/src/kmp_str.cpp
M openmp/runtime/src/kmp_str.h
A openmp/runtime/unittests/ADT/CMakeLists.txt
A openmp/runtime/unittests/ADT/TestStringRef.cpp
M openmp/runtime/unittests/CMakeLists.txt
M openmp/runtime/unittests/String/TestKmpStr.cpp
M utils/bazel/llvm-project-overlay/clang/unittests/BUILD.bazel
Log Message:
-----------
Merge branch 'main' into users/kparzysz/m01-misparsed-call
Compare: https://github.com/llvm/llvm-project/compare/989c2a463391...62f85dee13df
To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications
More information about the All-commits
mailing list