[all-commits] [llvm/llvm-project] 180281: [AMDGPU] Fix op_sel settings for v_cvt_scale32_* a...

Wed Jul 30 13:43:02 PDT 2025

  Branch: refs/heads/users/pcc/spr/lto-redesign-the-cfi-aliases-metadata
  Home:   https://github.com/llvm/llvm-project
  Commit: 180281b8eccfa8e1ec4ebc58df7685497f9ccb26
      https://github.com/llvm/llvm-project/commit/180281b8eccfa8e1ec4ebc58df7685497f9ccb26
  Author: Changpeng Fang <changpeng.fang at amd.com>
  Date:   2025-07-30 (Wed, 30 Jul 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
    M llvm/lib/Target/AMDGPU/VOP3Instructions.td
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.pk.gfx950.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.sr.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.sr.pk.gfx950.ll

  Log Message:
  -----------
  [AMDGPU] Fix op_sel settings for v_cvt_scale32_* and v_cvt_sr_* (#151286)

For OPF_OPSEL_SRCBYTE: Vector instruction uses OPSEL[1:0] to specify a
byte
select for the first source operand. So op_sel [0, 0], [1, 0], [0, 1]
and [1, 1] should map
to byte 0, 1, 2 and 3, respectively.

For OPF_OPSEL_DSTBYTE: OPSEL is used as a destination byte select.
OPSEL[2:3]
specify which byte of the destination to write to. Note that the order
of the bits is different
from that of OPF_OPSEL_SRCBYT. So the mapping should be: op_sel [0, 0],
[0, 1], [1, 0]
and [1, 1] map to byte 0, 1, 2 and 3, respectively.

Fixes: SWDEV-544901

  Commit: 96b4425669352e240c753c352b2782cbfce0e129
      https://github.com/llvm/llvm-project/commit/96b4425669352e240c753c352b2782cbfce0e129
  Author: Andrzej Warzyński <andrzej.warzynski at arm.com>
  Date:   2025-07-30 (Wed, 30 Jul 2025)

  Changed paths:
    M mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp

  Log Message:
  -----------
  [mlir][linalg][nfc] Clean-up leftover code post #149156 (#151334)

In https://github.com/llvm/llvm-project/pull/149156, I ensured that we
no longer generate spurious `tensor.empty` ops when vectorizing
`linalg.unpack`.

This follow-up removes leftover code that is now redundant but was
missed in the original PR and in #150602 that was also meant to clean-up
left-over code.

Note, this is removing code to compute "write-vector-sizes". Instead,
these are fully inferred from previous Ops.

  Commit: 24f9482abddbc24135d8e0143eb1bac499645648
      https://github.com/llvm/llvm-project/commit/24f9482abddbc24135d8e0143eb1bac499645648
  Author: Snehasish Kumar <snehasishk at google.com>
  Date:   2025-07-30 (Wed, 30 Jul 2025)

  Changed paths:
    M compiler-rt/include/profile/MemProfData.inc
    M compiler-rt/lib/memprof/memprof_rawprofile.cpp
    M compiler-rt/lib/memprof/tests/CMakeLists.txt
    R compiler-rt/lib/memprof/tests/histogram_encoding.cpp
    R compiler-rt/test/memprof/TestCases/memprof_histogram_uint8.cpp
    M llvm/include/llvm/ProfileData/MemProfData.inc
    M llvm/lib/ProfileData/MemProfReader.cpp
    M llvm/test/tools/llvm-profdata/Inputs/basic-histogram.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/basic-histogram.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/basic.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/basic.memprofraw
    R llvm/test/tools/llvm-profdata/Inputs/basic_v4.memprofexe
    R llvm/test/tools/llvm-profdata/Inputs/basic_v4.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/buildid.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/buildid.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/inline.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/inline.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/multi.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/multi.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/padding-histogram.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/padding-histogram.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/pic.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/pic.memprofraw
    M llvm/test/tools/llvm-profdata/memprof-basic-histogram.test
    M llvm/test/tools/llvm-profdata/memprof-basic.test
    R llvm/test/tools/llvm-profdata/memprof-basic_v4.test
    M llvm/test/tools/llvm-profdata/memprof-inline.test
    M llvm/test/tools/llvm-profdata/memprof-multi.test
    M llvm/test/tools/llvm-profdata/memprof-padding-histogram.test
    M llvm/test/tools/llvm-profdata/memprof-pic.test

  Log Message:
  -----------
  Revert "[MemProf] Change histogram storage from uint64_t to uint16_t" (#151382)

Reverts llvm/llvm-project#147854

Test failure when building with gcc.
https://lab.llvm.org/buildbot/#/builders/174/builds/21989

  Commit: a7ac43125eabb5845a2855bbf37fd4485831e860
      https://github.com/llvm/llvm-project/commit/a7ac43125eabb5845a2855bbf37fd4485831e860
  Author: Jonas Devlieghere <jonas at devlieghere.com>
  Date:   2025-07-30 (Wed, 30 Jul 2025)

  Changed paths:
    M lldb/packages/Python/lldbsuite/test/builders/__init__.py
    M lldb/packages/Python/lldbsuite/test/lldbplatformutil.py
    M lldb/packages/Python/lldbsuite/test/lldbtest.py

  Log Message:
  -----------
  Re-land "[lldb] Pick the builder for the target platform (#151262)"

Pick the builder for the target platform, not the host platform. This is
necessary when running the test suite remotely on a different platform.
Unlike for Darwin, both Windows and Linux us the default builder, which
is why this went unnoticed on the remote-linux bots.

  Commit: c063c59c11bb53d02c91b5810c05fd9e8699ebf8
      https://github.com/llvm/llvm-project/commit/c063c59c11bb53d02c91b5810c05fd9e8699ebf8
  Author: Peter Collingbourne <peter at pcc.me.uk>
  Date:   2025-07-30 (Wed, 30 Jul 2025)

  Changed paths:
    M compiler-rt/include/profile/MemProfData.inc
    M compiler-rt/lib/memprof/memprof_rawprofile.cpp
    M compiler-rt/lib/memprof/tests/CMakeLists.txt
    R compiler-rt/lib/memprof/tests/histogram_encoding.cpp
    R compiler-rt/test/memprof/TestCases/memprof_histogram_uint8.cpp
    M lldb/packages/Python/lldbsuite/test/builders/__init__.py
    M lldb/packages/Python/lldbsuite/test/lldbplatformutil.py
    M lldb/packages/Python/lldbsuite/test/lldbtest.py
    M llvm/include/llvm/LTO/LTO.h
    M llvm/include/llvm/ProfileData/MemProfData.inc
    M llvm/lib/LTO/LTO.cpp
    M llvm/lib/ProfileData/MemProfReader.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
    M llvm/lib/Target/AMDGPU/VOP3Instructions.td
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.pk.gfx950.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.sr.ll
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.scalef32.sr.pk.gfx950.ll
    M llvm/test/tools/llvm-profdata/Inputs/basic-histogram.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/basic-histogram.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/basic.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/basic.memprofraw
    R llvm/test/tools/llvm-profdata/Inputs/basic_v4.memprofexe
    R llvm/test/tools/llvm-profdata/Inputs/basic_v4.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/buildid.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/buildid.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/inline.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/inline.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/multi.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/multi.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/padding-histogram.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/padding-histogram.memprofraw
    M llvm/test/tools/llvm-profdata/Inputs/pic.memprofexe
    M llvm/test/tools/llvm-profdata/Inputs/pic.memprofraw
    M llvm/test/tools/llvm-profdata/memprof-basic-histogram.test
    M llvm/test/tools/llvm-profdata/memprof-basic.test
    R llvm/test/tools/llvm-profdata/memprof-basic_v4.test
    M llvm/test/tools/llvm-profdata/memprof-inline.test
    M llvm/test/tools/llvm-profdata/memprof-multi.test
    M llvm/test/tools/llvm-profdata/memprof-padding-histogram.test
    M llvm/test/tools/llvm-profdata/memprof-pic.test
    M mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp

  Log Message:
  -----------
  Rebase

Created using spr 1.3.6-beta.1

Compare: https://github.com/llvm/llvm-project/compare/55b725981cad...c063c59c11bb

To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications