[all-commits] [llvm/llvm-project] 186ae8: [NFC][mlir][linalg] Make conv_3d_ncdhw_fcdhw consi...
Alexander Richardson via All-commits
all-commits at lists.llvm.org
Mon Mar 3 16:16:24 PST 2025
Branch: refs/heads/users/arichardson/spr/lowertypetests-skip-intrinsics-when-determining-thumb-support
Home: https://github.com/llvm/llvm-project
Commit: 186ae8c06ae00d439e5ac4e2a6b80d5aeac7a187
https://github.com/llvm/llvm-project/commit/186ae8c06ae00d439e5ac4e2a6b80d5aeac7a187
Author: Nirvedh Meshram <96096277+nirvedhmeshram at users.noreply.github.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
Log Message:
-----------
[NFC][mlir][linalg] Make conv_3d_ncdhw_fcdhw consistent with 2D variant (#129547)
Other convolutions such as conv_2d_nchw_fchw have a output affine map
with no permutations and the input and the filter map are adjusted
accordingly. This makes conv_3d_ncdhw_fcdhw a stand out so this PR
changes the affine map to be consistent with other variants.
Commit: f98718c9a169d90e676f877c5b6761b22f424b73
https://github.com/llvm/llvm-project/commit/f98718c9a169d90e676f877c5b6761b22f424b73
Author: Mats Jun Larsen <mats at jun.codes>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M clang/lib/CodeGen/CGOpenMPRuntime.cpp
Log Message:
-----------
Remove leftover unused variable from #128711
Commit: f457c56240ee293c7ac72c7e6fc8150adb0a4a36
https://github.com/llvm/llvm-project/commit/f457c56240ee293c7ac72c7e6fc8150adb0a4a36
Author: Alexey Bataev <a.bataev at outlook.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/test/Transforms/SLPVectorizer/RISCV/spillcost.ll
Log Message:
-----------
[SLP][NFC]Update the test to check correctly the spill cost
Commit: 3ce67a81fac301e7a308c7c6c08e0abd883972e9
https://github.com/llvm/llvm-project/commit/3ce67a81fac301e7a308c7c6c08e0abd883972e9
Author: Craig Topper <craig.topper at sifive.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/utils/TableGen/DecoderEmitter.cpp
Log Message:
-----------
[TableGen] Remove unnecessary use of utostr to print a byte. NFC
We can cast to unsigned instead.
Commit: efb880de11f8fa3646c04032c0a2c9bded1935d7
https://github.com/llvm/llvm-project/commit/efb880de11f8fa3646c04032c0a2c9bded1935d7
Author: Craig Topper <craig.topper at sifive.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/utils/TableGen/DecoderEmitter.cpp
Log Message:
-----------
[TableGen] Fix incorrect comment. NFC
Commit: 313b71fc1a9ae17ea5ecba8afcb4e5b80e1f4043
https://github.com/llvm/llvm-project/commit/313b71fc1a9ae17ea5ecba8afcb4e5b80e1f4043
Author: Craig Topper <craig.topper at sifive.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h
M llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
M llvm/lib/Target/RISCV/RISCVMachineFunctionInfo.h
Log Message:
-----------
[RISCV] Simplify tracking of tracking and encoding of push/pop in RISCVFrameLowering. NFC (#129343)
Previously we calculated the max register id. Then converted it to the number
of registers and encoding. Then converted number of registers to stack
size. Then saved number of registers, encoding, and stack size to
MachineFunctionInfo.
This patch removes the calculation of max register id, and instead
calculates the number of registers. The encoding is removed from
MachineFunctionInfo in favor of converting the number of registers to
encoding at the time of use.
Commit: 3c90c900ba5ef9f0cb5808c73c388d3ee21cb683
https://github.com/llvm/llvm-project/commit/3c90c900ba5ef9f0cb5808c73c388d3ee21cb683
Author: TatWai Chong <tatwai.chong at arm.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp
M mlir/test/Dialect/Tosa/level_check.mlir
Log Message:
-----------
[mlir][tosa] Add several level checks (#129580)
Add the following types of level check to consolidate the level validity
- Complete rank level checks for operations.
- Add MAX_LOG2_SIZE level check: The maximum value is 63 when the level
is set to "none" and 31 when the level is set to "8K".
- Add MAX_TENSOR_LIST_SIZE level check : The maximum value is 256 when
the level is set to "none" and 64 when the level is set to "8K".
- TOSA 1.0 spec does not allow operations with dynamic shapes, so an
error should be raised instead.
Co-authored-by: Tai Ly <tai.ly at arm.com>
Commit: dfc5f37e3a62a93ce16935cea1d9822bec649a2b
https://github.com/llvm/llvm-project/commit/dfc5f37e3a62a93ce16935cea1d9822bec649a2b
Author: Florian Hahn <flo at fhahn.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/lib/Transforms/Vectorize/VPlan.h
Log Message:
-----------
[VPlan] Move onlyFirstLaneUsed to VPWidenInductionRecipe (NFC).
Move onlyFirstLaneUsed from VPWidenIntOrFpInductionRecipe and
VPWidenPointerInduction to VPWidenInductionRecipe. Also mark step value
as having only its first lane used.
Commit: b2ba43a9c1193f1d90ad9d30dada85caebd2c56d
https://github.com/llvm/llvm-project/commit/b2ba43a9c1193f1d90ad9d30dada85caebd2c56d
Author: Peter Klausler <pklausler at nvidia.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M flang/lib/Semantics/check-declarations.cpp
A flang/test/Semantics/generic13.f90
Log Message:
-----------
[flang] Refine checking of type-bound generics (#129292)
I merged a patch yesterday
(https://github.com/llvm/llvm-project/pull/128980) that strengthened
error detection of indistinguishable specific procedures in a type-bound
generic procedure, and broke a couple of tests. Refine the check so that
it doesn't flag valid cases of overridden bindings, and add a thorough
test with all of the boundary cases that I can think of.
Commit: 79a25e11fe119520e7cb70118df18e199217c891
https://github.com/llvm/llvm-project/commit/79a25e11fe119520e7cb70118df18e199217c891
Author: Peter Klausler <pklausler at nvidia.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M flang/include/flang/Evaluate/characteristics.h
M flang/include/flang/Evaluate/tools.h
M flang/lib/Evaluate/check-expression.cpp
M flang/lib/Evaluate/fold-logical.cpp
M flang/lib/Evaluate/fold.cpp
M flang/lib/Evaluate/intrinsics.cpp
M flang/lib/Evaluate/shape.cpp
M flang/lib/Evaluate/tools.cpp
M flang/lib/Lower/ConvertConstant.cpp
M flang/lib/Semantics/check-call.cpp
M flang/lib/Semantics/data-to-inits.cpp
M flang/lib/Semantics/definable.cpp
M flang/lib/Semantics/expression.cpp
M flang/lib/Semantics/pointer-assignment.cpp
M flang/lib/Semantics/resolve-names.cpp
M flang/test/Evaluate/folding06.f90
M flang/test/Lower/HLFIR/null.f90
M flang/test/Semantics/associated.f90
M flang/test/Semantics/call27.f90
M flang/test/Semantics/null01.f90
Log Message:
-----------
[flang] Further work on NULL(MOLD=allocatable) (#129345)
Refine handling of NULL(...) in semantics to properly distinguish
NULL(), NULL(objectPointer), NULL(procPointer), and NULL(allocatable)
from each other in relevant contexts.
Add IsNullAllocatable() and IsNullPointerOrAllocatable() utility
functions. IsNullAllocatable() is true only for NULL(allocatable); it is
false for a bare NULL(), which can be detected independently with
IsBareNullPointer().
IsNullPointer() now returns false for NULL(allocatable).
ALLOCATED(NULL(allocatable)) now works, and folds to .FALSE.
These utilities were modified to accept const pointer arguments rather
than const references; I usually prefer this style when the result
should clearly be false for a null argument (in the C sense), and it
helped me find all of their use sites in the code.
Commit: f6e83664e01e8dcb9bf241dd0154ff3adcc2e876
https://github.com/llvm/llvm-project/commit/f6e83664e01e8dcb9bf241dd0154ff3adcc2e876
Author: Peter Klausler <pklausler at nvidia.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M flang/lib/Semantics/check-declarations.cpp
M flang/test/Semantics/coarrays02.f90
M flang/test/Semantics/init01.f90
Log Message:
-----------
[flang] Improve two coarray error messages (#129597)
Two messages that complain about local variables mention that they don't
have the SAVE attribute; in both cases, it would be okay if they were
ALLOCATABLE instead. Clarify the messages.
Commit: edb7292a511171fb1fe75f85fc85464b91130a8f
https://github.com/llvm/llvm-project/commit/edb7292a511171fb1fe75f85fc85464b91130a8f
Author: Jacques Pienaar <jpienaar at google.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M mlir/include/mlir/IR/OperationSupport.h
M mlir/lib/IR/AsmPrinter.cpp
M mlir/unittests/IR/OperationSupportTest.cpp
Log Message:
-----------
[mlir] Add use nameloc to OpPrintingFlags (#129584)
Commit: cb4f24b0e5c4e7c463e59120af4f13ab81519047
https://github.com/llvm/llvm-project/commit/cb4f24b0e5c4e7c463e59120af4f13ab81519047
Author: TatWai Chong <tatwai.chong at arm.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp
M mlir/test/Dialect/Tosa/level_check.mlir
Log Message:
-----------
Revert "[mlir][tosa] Add several level checks" (#129602)
Reverts llvm/llvm-project#129580
There is a lit test failure reported in
https://lab.llvm.org/buildbot/#/builders/116/builds/11023
Commit: e1fd681837b85563e186f3739623cfa6653a722c
https://github.com/llvm/llvm-project/commit/e1fd681837b85563e186f3739623cfa6653a722c
Author: Douglas Gliner <Douglas.Gliner at sony.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M clang/include/clang/Basic/DiagnosticIDs.h
Log Message:
-----------
Bump DIAG_SIZE_LEX since we've hit the limit downstream as of 1c4e0f6.
Commit: 829401fe3ca1486285f5ef32865d189747a91305
https://github.com/llvm/llvm-project/commit/829401fe3ca1486285f5ef32865d189747a91305
Author: Aiden Grossman <aidengrossman at google.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
A llvm/docs/CIBestPractices.rst
M llvm/docs/Reference.rst
Log Message:
-----------
[Github][Docs] Add CI Best Practices Docs (#129462)
This is something that Tom and I have discussed briefly for a while now,
a doc that lists out all of the best practices we want to adhere to
surrounding CI things along with their associated motivations/any other
relevant info. This patch adds that doc along with three best practices
surrounding Github Workflows that we try and adhere to (although more
work needs to be done to get 100% adherance).
Commit: ee09df8e73d8575542283d8cfe351764af290f96
https://github.com/llvm/llvm-project/commit/ee09df8e73d8575542283d8cfe351764af290f96
Author: Uday Bondhugula <uday at polymagelabs.com>
Date: 2025-03-04 (Tue, 04 Mar 2025)
Changed paths:
M mlir/include/mlir/Dialect/Affine/Passes.td
M mlir/lib/Dialect/Affine/Transforms/AffineDataCopyGeneration.cpp
M mlir/test/Dialect/Affine/affine-data-copy.mlir
M mlir/test/Dialect/Affine/dma-generate.mlir
Log Message:
-----------
[MLIR][Affine] Switch default option for affine data copy generate pass (#129445)
The more common use of this pass for testing and other purposes is the
pointwise copy generation as opposed to DMA generation. Switch the
default pass option to pointwise copy generation for easier testing. NFC
otherwise.
Users who were relying on `generate-dma` to be on by default will have
to pass it explicitly: `-affine-data-copy-generate='generate-dma=1'`.
Commit: b44fbdee00bbead186baf4109ea7ca440bd8a0b3
https://github.com/llvm/llvm-project/commit/b44fbdee00bbead186baf4109ea7ca440bd8a0b3
Author: Petr Penzin <penzin.dev at gmail.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M llvm/docs/ReleaseNotes.md
M llvm/lib/Target/RISCV/RISCVFeatures.td
M llvm/lib/Target/RISCV/RISCVISelLowering.cpp
M llvm/lib/Target/RISCV/RISCVProcessors.td
M llvm/lib/Target/RISCV/RISCVSubtarget.h
M llvm/test/Analysis/CostModel/RISCV/shuffle-permute.ll
M llvm/test/CodeGen/RISCV/features-info.ll
Log Message:
-----------
[RISCV] Tune flag for fast vrgather.vv (#124664)
Add tune knob for N*Log2(N) vrgather.vv cost.
Commit: a59b17c8adde34e26b0f101fc2942637b945e1e5
https://github.com/llvm/llvm-project/commit/a59b17c8adde34e26b0f101fc2942637b945e1e5
Author: Derek Schuff <dschuff at chromium.org>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M lld/test/wasm/emit-relocs.s
M lld/wasm/InputChunks.cpp
M lld/wasm/InputChunks.h
M lld/wasm/OutputSections.cpp
M lld/wasm/OutputSections.h
M lld/wasm/Symbols.cpp
Log Message:
-----------
[lld][WebAssembly] Do not emit relocs against dead symbols (#129346)
When emitting relocs with linked output (i.e. --emit-relocs)
skip relocs against dead symbols (which do not appear in the output)
and do not emit them.
Commit: cf05b6e25ac8dcbf6e0ea1524d86bc6d190bf8c0
https://github.com/llvm/llvm-project/commit/cf05b6e25ac8dcbf6e0ea1524d86bc6d190bf8c0
Author: JaydeepChauhan14 <chauhan.jaydeep.ashwinbhai at intel.com>
Date: 2025-03-04 (Tue, 04 Mar 2025)
Changed paths:
M llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
M llvm/lib/Target/X86/X86InstrArithmetic.td
M llvm/lib/Target/X86/X86InstrFragments.td
M llvm/lib/Target/X86/X86InstrOperands.td
A llvm/test/CodeGen/X86/lea-16bit.ll
A llvm/test/CodeGen/X86/lea-8bit.ll
M llvm/utils/TableGen/X86RecognizableInstr.cpp
Log Message:
-----------
[X86] Added support for 8 and 16bit LEA instructions (#122102)
Commit: cc08f2a9d86a36d9d766e07eda423a34cabc1a82
https://github.com/llvm/llvm-project/commit/cc08f2a9d86a36d9d766e07eda423a34cabc1a82
Author: Alex Richardson <alexrichardson at google.com>
Date: 2025-03-03 (Mon, 03 Mar 2025)
Changed paths:
M clang/include/clang/Basic/DiagnosticIDs.h
M clang/lib/CodeGen/CGOpenMPRuntime.cpp
M flang/include/flang/Evaluate/characteristics.h
M flang/include/flang/Evaluate/tools.h
M flang/lib/Evaluate/check-expression.cpp
M flang/lib/Evaluate/fold-logical.cpp
M flang/lib/Evaluate/fold.cpp
M flang/lib/Evaluate/intrinsics.cpp
M flang/lib/Evaluate/shape.cpp
M flang/lib/Evaluate/tools.cpp
M flang/lib/Lower/ConvertConstant.cpp
M flang/lib/Semantics/check-call.cpp
M flang/lib/Semantics/check-declarations.cpp
M flang/lib/Semantics/data-to-inits.cpp
M flang/lib/Semantics/definable.cpp
M flang/lib/Semantics/expression.cpp
M flang/lib/Semantics/pointer-assignment.cpp
M flang/lib/Semantics/resolve-names.cpp
M flang/test/Evaluate/folding06.f90
M flang/test/Lower/HLFIR/null.f90
M flang/test/Semantics/associated.f90
M flang/test/Semantics/call27.f90
M flang/test/Semantics/coarrays02.f90
A flang/test/Semantics/generic13.f90
M flang/test/Semantics/init01.f90
M flang/test/Semantics/null01.f90
M lld/test/wasm/emit-relocs.s
M lld/wasm/InputChunks.cpp
M lld/wasm/InputChunks.h
M lld/wasm/OutputSections.cpp
M lld/wasm/OutputSections.h
M lld/wasm/Symbols.cpp
A llvm/docs/CIBestPractices.rst
M llvm/docs/Reference.rst
M llvm/docs/ReleaseNotes.md
M llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h
M llvm/lib/Target/RISCV/RISCVFeatures.td
M llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
M llvm/lib/Target/RISCV/RISCVISelLowering.cpp
M llvm/lib/Target/RISCV/RISCVMachineFunctionInfo.h
M llvm/lib/Target/RISCV/RISCVProcessors.td
M llvm/lib/Target/RISCV/RISCVSubtarget.h
M llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
M llvm/lib/Target/X86/X86InstrArithmetic.td
M llvm/lib/Target/X86/X86InstrFragments.td
M llvm/lib/Target/X86/X86InstrOperands.td
M llvm/lib/Transforms/IPO/LowerTypeTests.cpp
M llvm/lib/Transforms/Vectorize/VPlan.h
M llvm/test/Analysis/CostModel/RISCV/shuffle-permute.ll
M llvm/test/CodeGen/RISCV/features-info.ll
A llvm/test/CodeGen/X86/lea-16bit.ll
A llvm/test/CodeGen/X86/lea-8bit.ll
M llvm/test/Transforms/SLPVectorizer/RISCV/spillcost.ll
M llvm/utils/TableGen/DecoderEmitter.cpp
M llvm/utils/TableGen/X86RecognizableInstr.cpp
M mlir/include/mlir/Dialect/Affine/Passes.td
M mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
M mlir/include/mlir/IR/OperationSupport.h
M mlir/lib/Dialect/Affine/Transforms/AffineDataCopyGeneration.cpp
M mlir/lib/IR/AsmPrinter.cpp
M mlir/test/Dialect/Affine/affine-data-copy.mlir
M mlir/test/Dialect/Affine/dma-generate.mlir
M mlir/unittests/IR/OperationSupportTest.cpp
Log Message:
-----------
skip all declarations not just intrinsics
Created using spr 1.3.6-beta.1
Compare: https://github.com/llvm/llvm-project/compare/ce838e2aba54...cc08f2a9d86a
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