[PATCH] D144911: adding bf16 support to NVPTX

jpienaar added inline comments. Comment at: llvm/lib/Target/NVPTX/NVPTXInstrInfo.td:3352 - // Coalesce two bf16 registers into bf16x2 - def BuildBF16x2 : NVPTXInst<(outs BFloat16x2Regs:$dst), - (ins BFloat16Regs:$a, BFloat16Regs:$b), -

[PATCH] D144911: adding bf16 support to NVPTX

jpienaar added a comment. In D144911#4347029 , @kushanam wrote: > In D144911#4340097 , @jpienaar > wrote: > >> Thanks Artem for explaining, > > I think somehow, Phabricator pushes my changes to >

[PATCH] D140896: [WIP] Move from llvm::makeArrayRef to ArrayRef deduction guides

jpienaar added inline comments. Comment at: llvm/include/llvm/ADT/ArrayRef.h:502 + /// Deduction guide to construct an ArrayRef from a C array. + template ArrayRef(const T ()[N]) -> ArrayRef; mehdi_amini wrote: > Can we keep the makeArrayRef functions for

[PATCH] D138668: Correct typos

jpienaar closed this revision. jpienaar added a comment. Landed in https://github.com/llvm/llvm-project/commit/a9f9f3dff474b7bdb19129eaf625d3ef0084a975 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138668/new/ https://reviews.llvm.org/D138668

[PATCH] D138668: Correct typos

jpienaar added a comment. Do you need help in landing change? Or waiting for all reviewers? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138668/new/ https://reviews.llvm.org/D138668 ___ cfe-commits

[PATCH] D128049: [mlir] move SCF headers to SCF/{IR,Transforms} respectively

jpienaar accepted this revision. jpienaar added a comment. This revision is now accepted and ready to land. More consistency is good and update for downstream users mechanical, so LGTM Comment at: mlir/include/mlir/Dialect/SCF/Transforms/Transforms.h:13 -#ifndef

[PATCH] D124750: [MLIR] Add a utility to sort the operands of commutative ops

jpienaar added a comment. > (1) the operands defined by non-constant-like ops come first, followed by (2) > block arguments, and these are followed by (3) the operands defined by > constant-like ops. I would have thought block-arguments would come first as we don't know their values, while

[PATCH] D123836: [Driver] Move Lanai IAS enabling to Generic_GCC::IsIntegratedAssemblerDefault, NFC

jpienaar accepted this revision. jpienaar added a comment. Thanks Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D123836/new/ https://reviews.llvm.org/D123836 ___ cfe-commits mailing list

[PATCH] D116488: Add a misc-unused-parameters.CommentOutUnusedParameters to clang-tidy

jpienaar accepted this revision. jpienaar added a comment. This revision is now accepted and ready to land. Looks reasonable as an option/this is another way to make it explicit that param is unused. Comment at:

[PATCH] D116488: Add a misc-unused-parameters.CommentOutUnusedParameters to clang-tidy

jpienaar added inline comments. Comment at: clang-tools-extra/docs/clang-tidy/checks/misc-unused-parameters.rst:25 + + void a(int ) { /*some code that doesn't use `i`*/ } + OOC why the extra space after the type? Repository: rG LLVM Github Monorepo

[PATCH] D91410: [llvm][clang][mlir] Add checks for the return values from Target::createXXX to prevent protential null deref

jpienaar accepted this revision. jpienaar added a comment. LG MLIR changes CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91410/new/ https://reviews.llvm.org/D91410 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D82860: Port ObjCMTAction to new option parsing system

jpienaar added a comment. This seems to breaking clang-5 builds: utils/TableGen/CMakeFiles/llvm-tblgen.dir/OptParserEmitter.cpp.o: In function `llvm::EmitOptParser(llvm::RecordKeeper&, llvm::raw_ostream&)':

[PATCH] D81422: Change filecheck default to dump input on failure

jpienaar accepted this revision. jpienaar added a comment. I think this is a good default given it provides very useful info in failure case. For cases where folks expect it to fail + don't want it logged, then setting never (if that is a concern) seems better behavior IMHO. Repository: rG

[PATCH] D81045: [MLIR] Modify HasParent trait to allow one of several op's as a parent

jpienaar added inline comments. Comment at: llvm/include/llvm/Support/Casting.h:147 +template +LLVM_NODISCARD inline typename std::enable_if::type +isa(const Y ) { I'm worried that folks may not notice this change based on the change description (e.g., the

[PATCH] D76262: [NFC] Add UsedDeclVisitor

jpienaar added a comment. This does not appear to be NFC: git checkout 704cd4d5d0754904361823588f203369c309deca ; ninja check-mlir passes git checkout 08ab8c9af4dd27cb306b449edc9a9c50ed11194a

[PATCH] D70172: [CUDA][HIP][OpenMP] Emit deferred diagnostics by a post-parsing AST travese

jpienaar added a comment. This seems to result in triggering clang/lib/CodeGen/CGExpr.cpp:2626 when compiling mlir/lib/Transforms/AffineDataCopyGeneration.cpp with clang build with assertions on (clean build at e8e078c just