[all-commits] [llvm/llvm-project] 383e5d: [Driver] Don't warn on ObjC constant-literal flags...

Hans Wennborg via All-commits all-commits at lists.llvm.org
Fri Jun 5 09:27:33 PDT 2026


  Branch: refs/heads/revert-201373-users/mizvekov/get-template-inst-args
  Home:   https://github.com/llvm/llvm-project
  Commit: 383e5d2a6907410dc3486604bdaae0c18cbf111c
      https://github.com/llvm/llvm-project/commit/383e5d2a6907410dc3486604bdaae0c18cbf111c
  Author: Akira Hatanaka <ahatanak at gmail.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M clang/include/clang/Options/Options.td
    M clang/test/Driver/objc-constant-literals.m

  Log Message:
  -----------
  [Driver] Don't warn on ObjC constant-literal flags for non-ObjC inputs (#200303)

-fobjc-constant-literals and the per-kind
-f[no-]constant-ns{number,array, dictionary}-literals flags are only
acted on for Objective-C inputs. When a build system passes them
uniformly to every source via a shared response file, they go unclaimed
on assembly and C/C++ inputs and trigger:

  error: argument unused during compilation: '-fobjc-constant-literals'
         [-Werror,-Wunused-command-line-argument]

Mark the options NoArgumentUnused so the unused argument is ignored
regardless of input type or whether the cc1 or cc1as path runs. The
behavior of the flags is unchanged: only the positive forms are
forwarded to cc1, and only for ObjC inputs.

rdar://175647747


  Commit: de1ff3e2db8a207f366f33b8918079a5ed273d77
      https://github.com/llvm/llvm-project/commit/de1ff3e2db8a207f366f33b8918079a5ed273d77
  Author: Ryan Mitchell <Ryan.Mitchell at amd.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/MIMGInstructions.td
    M llvm/test/MC/AMDGPU/gfx1250_asm_vimage.s
    M llvm/test/MC/AMDGPU/gfx1250_asm_vimage_err.s
    M llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vimage.txt

  Log Message:
  -----------
  [AMDGPU] Allow null operands in VImage tensor instructions (#200911)

NULL is equivalent to passing a block of SGPRs that are set to zero, and
is allowed for 3+ opnds.


  Commit: c8faaf9a9ef24873fe97d46554ad5fb909d0fd5d
      https://github.com/llvm/llvm-project/commit/c8faaf9a9ef24873fe97d46554ad5fb909d0fd5d
  Author: Yingwei Zheng <dtcxzyw2333 at gmail.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    A llvm/test/tools/llubi/call_unknown.ll
    M llvm/tools/llubi/lib/Interpreter.cpp

  Log Message:
  -----------
  [llubi] Don't check type after the program exited (#201862)

Closes https://github.com/llvm/llvm-project/issues/201791.


  Commit: 86d0c2b10427110892de794f73c0b0db476a9211
      https://github.com/llvm/llvm-project/commit/86d0c2b10427110892de794f73c0b0db476a9211
  Author: Charles Zablit <c_zablit at apple.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M lldb/test/API/tools/lldb-server/TestGdbRemote_qThreadStopInfo.py

  Log Message:
  -----------
  [lldb][Windows] enable TestGdbRemote_qThreadStopInfo (#201868)


  Commit: fdd3e7cae11fff30a33af79a8814db9c72d5179b
      https://github.com/llvm/llvm-project/commit/fdd3e7cae11fff30a33af79a8814db9c72d5179b
  Author: Miguel A. Arroyo <miguel.arroyo at rockstargames.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/cmake/modules/HandleLLVMOptions.cmake

  Log Message:
  -----------
  [llvm][cmake] Pass /Brepro to linker (#201708)

* The /Brepro flag should also be passed to the linker. The `BUILD.gn`
correctly does
[this](https://github.com/llvm/llvm-project/blob/70f7167ce2622a610a6d6dca4d5934ea8a07a056/llvm/utils/gn/build/BUILD.gn#L332).
* The /Brepro flag in `clang-cl` omits the timestamp for .obj files
while the linker flag omits the timestamp for .exe and .dll files.


  Commit: ae3ef5367dd37e7b6b80339fcaa267dead9ad223
      https://github.com/llvm/llvm-project/commit/ae3ef5367dd37e7b6b80339fcaa267dead9ad223
  Author: Joshua Rodriguez <josh.rodriguez at arm.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64InstrInfo.td
    M llvm/lib/Target/AArch64/GISel/AArch64RegisterBankInfo.cpp
    M llvm/test/CodeGen/AArch64/arm64-arith-saturating.ll

  Log Message:
  -----------
  [AArch64][GlobalISel] Add patterns for signed scalar extend intrinsics (#201617)

Follow on from https://github.com/llvm/llvm-project/pull/201546
Add patterns for signed versions of scalar extend intrinsics as well.


  Commit: c941964c58cf8ba2b65c1358f42576ab1eff67b0
      https://github.com/llvm/llvm-project/commit/c941964c58cf8ba2b65c1358f42576ab1eff67b0
  Author: Manuel Carrasco <Manuel.Carrasco at amd.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/lib/Target/SPIRV/SPIRVNonSemanticDebugHandler.cpp
    M llvm/lib/Target/SPIRV/SPIRVNonSemanticDebugHandler.h
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-int-string-dedup.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-multi-scalar-params.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-omit.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-pointer-debug-none-base.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-pointer-param.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-scalar-returns.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-subroutine-type-flags.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-void-prototypes.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-pointer-composite-pointee.ll

  Log Message:
  -----------
  [SPIRV] Extend NSDI debug handling for DebugTypeFunction. (#197003)

Extend NSDI handling for
[DebugTypeFunction](https://github.khronos.org/SPIRV-Registry/nonsemantic/NonSemantic.Shader.DebugInfo.html#DebugTypeFunction).

Changes:

- Collect required types with DebugInfoFinder instead of walking
DbgVariableRecords only. This allows processing types that might not be
available in these records, and avoids code duplication for the
traversals.
- Emit DebugTypeFunction for DISubroutineType when every signature slot
maps to an emitted debug type.
- Reset per-module state in beginModule().
- Replace parallel FileStringRegs/BasicTypeNameRegs with a StringMap
OpString cache, deduplicating identical strings.
- Add LLVM DINode flag lowering for NSDI.
- Cache DebugInfoNone, OpTypeVoid and OpTypeInt32 registers to avoid
duplicate instructions.


  Commit: c4f2f2535438cbdf23d9f84689ea0f518ffa4a95
      https://github.com/llvm/llvm-project/commit/c4f2f2535438cbdf23d9f84689ea0f518ffa4a95
  Author: Charles Zablit <c_zablit at apple.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M lldb/source/Plugins/Process/Windows/Common/CMakeLists.txt
    A lldb/source/Plugins/Process/Windows/Common/IOHandlerProcessSTDIOWindows.cpp
    A lldb/source/Plugins/Process/Windows/Common/IOHandlerProcessSTDIOWindows.h
    M lldb/source/Plugins/Process/Windows/Common/ProcessWindows.cpp
    M lldb/source/Plugins/Process/Windows/Common/ProcessWindows.h

  Log Message:
  -----------
  [NFC][lldb][Windows] extract IOHandlerProcessSTDIOWindows (#201651)

Co-authored-by: Nerixyz <nero.9 at hotmail.de>


  Commit: 89f4b84d8b2cd0a5c54dc9a819090f131bf87d0d
      https://github.com/llvm/llvm-project/commit/89f4b84d8b2cd0a5c54dc9a819090f131bf87d0d
  Author: Justin Lebar <justin.lebar at gmail.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp
    M llvm/test/Transforms/InstCombine/ptr-replace-alloca.ll

  Log Message:
  -----------
  [InstCombine] Use copyMetadata in PointerReplacer::replace (#201827)

PointerReplacer::replace creates a new load that differs from the
original only in its pointer operand; the loaded type is unchanged.  It
was using copyMetadataForLoad(), which is meant for the case where the
load's *type* changes.  Since the type is the same here, plain
copyMetadata() is correct and preserves all metadata directly.


  Commit: 3691cf9d025a1d37f69da8abf8577926e8fe9b84
      https://github.com/llvm/llvm-project/commit/3691cf9d025a1d37f69da8abf8577926e8fe9b84
  Author: Timur Golubovich <timur.golubovich at intel.com>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M llvm/test/CodeGen/AArch64/fp-conversion-to-tbl.ll
    M llvm/test/CodeGen/AArch64/pr164181.ll
    M llvm/test/CodeGen/AArch64/sitofp-to-tbl.ll
    M llvm/test/CodeGen/AArch64/trunc-to-tbl.ll
    M llvm/test/CodeGen/X86/mbp-false-cfg-break.ll
    M llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll

  Log Message:
  -----------
  [Test] Fix loop exit conditions to prevent trivial optimizations (#201867)

Several tests had 'br i1 %ec, label %loop, label %exit' which exits on
the first iteration instead of looping so I swapped them. Also changed
predicates to keep the loops, otherwise they are going to be eliminated
by https://github.com/llvm/llvm-project/pull/201839.


  Commit: b235617112039202675221d5f450b67994428cca
      https://github.com/llvm/llvm-project/commit/b235617112039202675221d5f450b67994428cca
  Author: Hans Wennborg <hans at hanshq.net>
  Date:   2026-06-05 (Fri, 05 Jun 2026)

  Changed paths:
    M clang/include/clang/Options/Options.td
    M clang/test/Driver/objc-constant-literals.m
    M lldb/source/Plugins/Process/Windows/Common/CMakeLists.txt
    A lldb/source/Plugins/Process/Windows/Common/IOHandlerProcessSTDIOWindows.cpp
    A lldb/source/Plugins/Process/Windows/Common/IOHandlerProcessSTDIOWindows.h
    M lldb/source/Plugins/Process/Windows/Common/ProcessWindows.cpp
    M lldb/source/Plugins/Process/Windows/Common/ProcessWindows.h
    M lldb/test/API/tools/lldb-server/TestGdbRemote_qThreadStopInfo.py
    M llvm/cmake/modules/HandleLLVMOptions.cmake
    M llvm/lib/Target/AArch64/AArch64InstrInfo.td
    M llvm/lib/Target/AArch64/GISel/AArch64RegisterBankInfo.cpp
    M llvm/lib/Target/AMDGPU/MIMGInstructions.td
    M llvm/lib/Target/SPIRV/SPIRVNonSemanticDebugHandler.cpp
    M llvm/lib/Target/SPIRV/SPIRVNonSemanticDebugHandler.h
    M llvm/lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp
    M llvm/test/CodeGen/AArch64/arm64-arith-saturating.ll
    M llvm/test/CodeGen/AArch64/fp-conversion-to-tbl.ll
    M llvm/test/CodeGen/AArch64/pr164181.ll
    M llvm/test/CodeGen/AArch64/sitofp-to-tbl.ll
    M llvm/test/CodeGen/AArch64/trunc-to-tbl.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-int-string-dedup.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-multi-scalar-params.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-omit.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-pointer-debug-none-base.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-pointer-param.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-scalar-returns.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-subroutine-type-flags.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-function-void-prototypes.ll
    A llvm/test/CodeGen/SPIRV/debug-info/debug-type-pointer-composite-pointee.ll
    M llvm/test/CodeGen/X86/mbp-false-cfg-break.ll
    M llvm/test/MC/AMDGPU/gfx1250_asm_vimage.s
    M llvm/test/MC/AMDGPU/gfx1250_asm_vimage_err.s
    M llvm/test/MC/Disassembler/AMDGPU/gfx1250_dasm_vimage.txt
    M llvm/test/Transforms/InstCombine/ptr-replace-alloca.ll
    M llvm/test/Transforms/LoopStrengthReduce/X86/pr62660-normalization-failure.ll
    A llvm/test/tools/llubi/call_unknown.ll
    M llvm/tools/llubi/lib/Interpreter.cpp

  Log Message:
  -----------
  Merge branch 'main' into revert-201373-users/mizvekov/get-template-inst-args


Compare: https://github.com/llvm/llvm-project/compare/19b0cecd52e0...b23561711203

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