[all-commits] [llvm/llvm-project] 8b4902: [Compiler-rt][test] Fix circular link dependency b...
Jan Patrick Lehr via All-commits
all-commits at lists.llvm.org
Tue Jun 9 03:04:49 PDT 2026
Branch: refs/heads/revert-200312-emit-cbuffer-globals-as-internal
Home: https://github.com/llvm/llvm-project
Commit: 8b4902300521d4a0980d9d35210c02b405f0df86
https://github.com/llvm/llvm-project/commit/8b4902300521d4a0980d9d35210c02b405f0df86
Author: Garvit Gupta <garvgupt at qti.qualcomm.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M compiler-rt/test/builtins/Unit/lit.cfg.py
Log Message:
-----------
[Compiler-rt][test] Fix circular link dependency between builtins and libc (#199482)
Currently, the link order is `libclang_rt.builtins.a -lc -lm`. Builtins
are scanned first after which symbols like `abort` are unresolved
references that are resolved through libc.a. However, resolving the
references to these symbols further lead to undefined references to
`_aeabi_uldivmod` etc. that can only resolved through builtins.
Reversing the order also wont fix the issue because `libc.a` introduces
`__aeabi_uldivmod` which is resolved by builtins but it introduces
`abort` which can only be resolved libc.a.
This patch fixes this by wrapping the archives in a linker group
(--start-group/--end-group), which instructs the linker to rescan all
archives in the group until no new symbols can be resolved.
This error is exposed only when bfd like linkers are used.
Commit: ab31c28892a9ad5e016e94500861c93018736e7b
https://github.com/llvm/llvm-project/commit/ab31c28892a9ad5e016e94500861c93018736e7b
Author: Srinivasa Ravi <srinivasar at nvidia.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
M mlir/lib/Dialect/LLVMIR/IR/NVVMDialect.cpp
M mlir/test/Target/LLVMIR/nvvm/convert_fp4x2.mlir
M mlir/test/Target/LLVMIR/nvvm/convert_fp6x2.mlir
M mlir/test/Target/LLVMIR/nvvm/convert_fp8x2.mlir
M mlir/test/Target/LLVMIR/nvvmir-invalid.mlir
Log Message:
-----------
[MLIR][NVVM] Add support for narrow-fp to bf16x2 conversions (#200157)
This change adds the following NVVM Ops to support narrow-fp to bf16x2
conversions:
- `nvvm.convert.f6x2.to.bf16x2`
- `nvvm.convert.f4x2.to.bf16x2`
- `nvvm.convert.f8x2.to.bf16x2` (updated to allow `E4M3FN` and `E5M2`
types)
Also removes unnecessary verifiers for narrow-fp to `f16x2` conversions
to instead use `TypeAttrOf` to validate the source type in the ODS
definition.
Commit: c3d13ceb8a894312211c88bf2ef947708442a855
https://github.com/llvm/llvm-project/commit/c3d13ceb8a894312211c88bf2ef947708442a855
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M llvm/lib/Target/AMDGPU/AMDGPULowerBufferFatPointers.cpp
M llvm/test/CodeGen/AMDGPU/lower-buffer-fat-pointers-contents-legalization.ll
Log Message:
-----------
[AMDGPU] Use alloc size for array stride in LowerBufferFatPointers (#202530)
Array elements are laid out at multiples of getTypeAllocSize, not
getTypeStoreSize
LLVM memory model lays out array element `i` at `i * allocSize`
(reflected in `DataLayout::getTypeAllocSize`), apply it for fat pointers
to prevent miscompile
Commit: 3fec9c702b597e1a0204ff30137ae8d1557e395c
https://github.com/llvm/llvm-project/commit/3fec9c702b597e1a0204ff30137ae8d1557e395c
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M mlir/lib/Conversion/SPIRVToLLVM/SPIRVToLLVM.cpp
M mlir/test/Conversion/SPIRVToLLVM/cast-ops-to-llvm.mlir
A mlir/test/Conversion/SPIRVToLLVM/cl-ops-to-llvm.mlir
M mlir/test/Conversion/SPIRVToLLVM/gl-ops-to-llvm.mlir
M mlir/test/Conversion/SPIRVToLLVM/logical-ops-to-llvm.mlir
Log Message:
-----------
[mlir][SPIR-V] Add SPIRVToLLVM direct conversions for cast, CL, GL and logical ops (#202506)
Lower the OpenCL extended instruction set math ops, GL math ops (Trunc,
Asin, Acos, Atan), logical Ordered/Unordered, and the pointer cast ops
to their LLVM dialect equivalents
Commit: 6bf8d4b059bd5cd4b660be6063769a9f162e5ec0
https://github.com/llvm/llvm-project/commit/6bf8d4b059bd5cd4b660be6063769a9f162e5ec0
Author: Ramkumar Ramachandra <artagnon at tenstorrent.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
M llvm/test/Transforms/LoopVectorize/X86/epilog-vectorization-inductions.ll
M llvm/test/Transforms/LoopVectorize/X86/induction-step.ll
M llvm/test/Transforms/LoopVectorize/expand-scev-after-invoke.ll
M llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
M llvm/test/Transforms/LoopVectorize/induction.ll
M llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
M llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll
M llvm/test/Transforms/LoopVectorize/iv-select-cmp-trunc.ll
M llvm/test/Transforms/LoopVectorize/nested-loops-scev-expansion.ll
Log Message:
-----------
[VPlan] Look through BCast when folding live-ins (#202527)
This gives us some minor improvements.
Commit: bf860889a8ff3fea28dbace895cd633ccbe79608
https://github.com/llvm/llvm-project/commit/bf860889a8ff3fea28dbace895cd633ccbe79608
Author: Dmitry Vasilyev <dvassiliev at accesssoftek.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M llvm/unittests/Support/DynamicLibrary/DynamicLibraryTest.cpp
Log Message:
-----------
Revert "[test][Support] Disable CFI-icall for DynamicLibrary Overload test" (#202550)
Reverts llvm/llvm-project#202446
Commit: 383db8a1ddc9f8db49c8305e463b3a80b7cd8638
https://github.com/llvm/llvm-project/commit/383db8a1ddc9f8db49c8305e463b3a80b7cd8638
Author: Jan Patrick Lehr <JanPatrick.Lehr at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M .ci/compute_projects.py
Log Message:
-----------
[CI][Offload] Fix offload depends on openmp (#202541)
It appears that Offload depends on OpenMP. Thus, enable OpenMP as a
runtime to test when offload has changes.
Commit: 08636d456502fb7404e632598e16f3698367a6f7
https://github.com/llvm/llvm-project/commit/08636d456502fb7404e632598e16f3698367a6f7
Author: Davide Grohmann <davide.grohmann at arm.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
A mlir/include/mlir/Dialect/SPIRV/IR/SPIRVExperimentalMLOps.td
M mlir/include/mlir/Dialect/SPIRV/IR/SPIRVOps.td
A mlir/test/Dialect/SPIRV/IR/experimental-ml-ops.mlir
A mlir/test/Target/SPIRV/experimental-ml-ops.mlir
Log Message:
-----------
[mlir][spirv] Add Arm.ExperimentalMLOperations.1 extended inst set (#202283)
This instruction set provides a mechanism to encode experimental ML
operations in SPIR-V modules. Such instructions are encoded via the
single CALL operator in the instruction set by specifying an op_code and
customized inputs values.
Reference:
https://github.com/KhronosGroup/SPIRV-Registry/blob/main/extended/Arm.ExperimentalMLOperations.asciidoc
Signed-off-by: Niklas Lithammer <niklas.lithammer at arm.com>
Signed-off-by: Davide Grohmann <davide.grohmann at arm.com>
Commit: 514f5b766667c4c696b260f783403e8b41bc8798
https://github.com/llvm/llvm-project/commit/514f5b766667c4c696b260f783403e8b41bc8798
Author: lijinpei-amd <jinpli at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M llvm/lib/Analysis/ConstantFolding.cpp
M llvm/test/Transforms/InstSimplify/bitcast-vector-fold.ll
Log Message:
-----------
[ConstantFolding] Fix dropped bits in non-integer-ratio bitcast with undef lane (#202282)
When constant-folding a vector bitcast(e.g. <4 x i24> -> <3 x i32>), an
undef source element inserted a DstBitSize-wide zero placeholder into
the bit buffer. This could clobber defined source element, producing a
wrong result on big-endian targets.
Fix by inserting SrcBitSize-wide zero instead.
Alive2 proof:
before (unsound): https://alive2.llvm.org/ce/z/R_ZQ75
after (verified): https://alive2.llvm.org/ce/z/VuV3mz
Commit: 6746898d2bfc086947d86715e065f8dbf74e9690
https://github.com/llvm/llvm-project/commit/6746898d2bfc086947d86715e065f8dbf74e9690
Author: Juan Manuel Martinez Caamaño <jmartinezcaamao at gmail.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M clang/include/clang/Basic/LangOptions.def
M clang/include/clang/Options/Options.td
M clang/lib/CodeGen/CGHLSLRuntime.cpp
M clang/lib/Driver/ToolChains/Clang.cpp
R clang/test/CodeGenHLSL/preserve-interface-dce.hlsl
R clang/test/CodeGenHLSL/preserve-interface.hlsl
R clang/test/Driver/dxc_fspv_preserve_interface.hlsl
M llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
R llvm/test/CodeGen/SPIRV/preserve-interface-dce.ll
R llvm/test/CodeGen/SPIRV/preserve-interface.ll
Log Message:
-----------
Revert "[clang][SPIR-V] Implement -fspv-preserve-interface (#196404)" (#202558)
This reverts commit 95c24830265cba8c4844dda3384025bac0bf96b4.
Fail log:
https://lab.llvm.org/buildbot/#/builders/226/builds/9015/steps/7/logs/stdio
```
FAILED: lib/libLLVMFrontendHLSL.so.23.0git
: && /opt/rh/gcc-toolset-13/root/usr/bin/c++ -fPIC -fPIC -fno-semantic-interposition -fvisibility-inlines-hidden -Werror=date-time -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wimplicit-fallthrough -Wno-uninitialized -Wno-nonnull -Wno-class-memaccess -Wno-dangling-reference -Wno-redundant-move -Wno-pessimizing-move -Wno-array-bounds -Wno-stringop-overread -Wno-dangling-pointer -Wno-noexcept-type -Wdelete-non-virtual-dtor -Wsuggest-override -Wno-comment -Wno-misleading-indentation -Wctad-maybe-unsupported -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -Wl,-z,defs -Wl,-z,nodelete -Wl,-rpath-link,/home/botworker/bbot/amdgpu-offload-build-only/build/./lib -Wl,--gc-sections -shared -Wl,-soname,libLLVMFrontendHLSL.so.23.0git -o lib/libLLVMFrontendHLSL.so.23.0git lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/CBuffer.cpp.o lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/HLSLBinding.cpp.o lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/HLSLResource.cpp.o lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/HLSLRootSignature.cpp.o lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/RootSignatureMetadata.cpp.o lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/RootSignatureValidations.cpp.o -Wl,-rpath,"\$ORIGIN/../lib:\$ORIGIN/../lib/x86_64-unknown-linux-gnu:/home/botworker/bbot/amdgpu-offload-build-only/build/lib:" lib/libLLVMCore.so.23.0git lib/libLLVMBinaryFormat.so.23.0git lib/libLLVMSupport.so.23.0git -Wl,-rpath-link,/home/botworker/bbot/amdgpu-offload-build-only/build/lib && :
/opt/rh/gcc-toolset-13/root/usr/libexec/gcc/x86_64-redhat-linux/13/ld: lib/Frontend/HLSL/CMakeFiles/LLVMFrontendHLSL.dir/CBuffer.cpp.o: in function `llvm::hlsl::CBufferMetadata::removeCBufferGlobalsFromUseList(llvm::Module&)':
CBuffer.cpp:(.text._ZN4llvm4hlsl15CBufferMetadata31removeCBufferGlobalsFromUseListERNS_6ModuleE+0xce): undefined reference to `llvm::removeFromUsedLists(llvm::Module&, llvm::function_ref<bool (llvm::Constant*)>)'
collect2: error: ld returned 1 exit status
```
Commit: a172eb9fcdebc93ad123b1f9d8b3ba953b904e7d
https://github.com/llvm/llvm-project/commit/a172eb9fcdebc93ad123b1f9d8b3ba953b904e7d
Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M llvm/lib/CodeGen/GlobalISel/InlineAsmLowering.cpp
M llvm/test/CodeGen/AMDGPU/GlobalISel/inline-asm-mismatched-size.ll
Log Message:
-----------
[GlobalISel][AMDGPU] Emit proper diagnostic when inline asm register allocation fails (#201380)
Replace the silent fallback return with a DiagnosticInfoInlineAsm error
and undef result values, so the failure is reported to the user instead
of relying on -global-isel-abort
discussed in https://github.com/llvm/llvm-project/pull/200771
Commit: 944284fe6329eb133bdbf9af77af6fa0cd280070
https://github.com/llvm/llvm-project/commit/944284fe6329eb133bdbf9af77af6fa0cd280070
Author: Matthias Springer <me at m-sp.org>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.td
Log Message:
-----------
[mlir][Interfaces] Document completeness requirement of `RegionBranchOpInterface` (#202018)
Document that interface implementations must report all possible control
flow edges. Failure to report a possible edge may break
analyses/transformations/APIs such as
`RegionBranchOpInterface::isRepetitiveRegion`.
Commit: bfb1cd6e2d6d58eb8c30c20dc404b32cb74dbec1
https://github.com/llvm/llvm-project/commit/bfb1cd6e2d6d58eb8c30c20dc404b32cb74dbec1
Author: Jan Patrick Lehr <JanPatrick.Lehr at amd.com>
Date: 2026-06-09 (Tue, 09 Jun 2026)
Changed paths:
M .ci/compute_projects.py
M clang/include/clang/Basic/LangOptions.def
M clang/include/clang/Options/Options.td
M clang/lib/CodeGen/CGHLSLRuntime.cpp
M clang/lib/Driver/ToolChains/Clang.cpp
R clang/test/CodeGenHLSL/preserve-interface-dce.hlsl
R clang/test/CodeGenHLSL/preserve-interface.hlsl
R clang/test/Driver/dxc_fspv_preserve_interface.hlsl
M compiler-rt/test/builtins/Unit/lit.cfg.py
M llvm/lib/Analysis/ConstantFolding.cpp
M llvm/lib/CodeGen/GlobalISel/InlineAsmLowering.cpp
M llvm/lib/Target/AMDGPU/AMDGPULowerBufferFatPointers.cpp
M llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
M llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
M llvm/test/CodeGen/AMDGPU/GlobalISel/inline-asm-mismatched-size.ll
M llvm/test/CodeGen/AMDGPU/lower-buffer-fat-pointers-contents-legalization.ll
R llvm/test/CodeGen/SPIRV/preserve-interface-dce.ll
R llvm/test/CodeGen/SPIRV/preserve-interface.ll
M llvm/test/Transforms/InstSimplify/bitcast-vector-fold.ll
M llvm/test/Transforms/LoopVectorize/X86/epilog-vectorization-inductions.ll
M llvm/test/Transforms/LoopVectorize/X86/induction-step.ll
M llvm/test/Transforms/LoopVectorize/expand-scev-after-invoke.ll
M llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
M llvm/test/Transforms/LoopVectorize/induction.ll
M llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
M llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll
M llvm/test/Transforms/LoopVectorize/iv-select-cmp-trunc.ll
M llvm/test/Transforms/LoopVectorize/nested-loops-scev-expansion.ll
M llvm/unittests/Support/DynamicLibrary/DynamicLibraryTest.cpp
M mlir/include/mlir/Dialect/LLVMIR/NVVMOps.td
A mlir/include/mlir/Dialect/SPIRV/IR/SPIRVExperimentalMLOps.td
M mlir/include/mlir/Dialect/SPIRV/IR/SPIRVOps.td
M mlir/include/mlir/Interfaces/ControlFlowInterfaces.td
M mlir/lib/Conversion/SPIRVToLLVM/SPIRVToLLVM.cpp
M mlir/lib/Dialect/LLVMIR/IR/NVVMDialect.cpp
M mlir/test/Conversion/SPIRVToLLVM/cast-ops-to-llvm.mlir
A mlir/test/Conversion/SPIRVToLLVM/cl-ops-to-llvm.mlir
M mlir/test/Conversion/SPIRVToLLVM/gl-ops-to-llvm.mlir
M mlir/test/Conversion/SPIRVToLLVM/logical-ops-to-llvm.mlir
A mlir/test/Dialect/SPIRV/IR/experimental-ml-ops.mlir
M mlir/test/Target/LLVMIR/nvvm/convert_fp4x2.mlir
M mlir/test/Target/LLVMIR/nvvm/convert_fp6x2.mlir
M mlir/test/Target/LLVMIR/nvvm/convert_fp8x2.mlir
M mlir/test/Target/LLVMIR/nvvmir-invalid.mlir
A mlir/test/Target/SPIRV/experimental-ml-ops.mlir
Log Message:
-----------
Merge branch 'main' into revert-200312-emit-cbuffer-globals-as-internal
Compare: https://github.com/llvm/llvm-project/compare/f2d2491707e6...bfb1cd6e2d6d
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