aboutsummaryrefslogtreecommitdiff
path: root/clang/tools/include-mapping/cppreference_parser.py
diff options
context:
space:
mode:
authorLuke Lau <luke@igalia.com>2024-11-04 20:53:51 +0800
committerGitHub <noreply@github.com>2024-11-04 20:53:51 +0800
commit7bf0d6d032c5ef04a0f4966df8760664aaefc871 (patch)
treeb94ba431c21ceb4b482c298bcaaaf177a3742e9e /clang/tools/include-mapping/cppreference_parser.py
parent7d3536840528bb91b6ff90c86084f7879459c450 (diff)
downloadllvm-7bf0d6d032c5ef04a0f4966df8760664aaefc871.zip
llvm-7bf0d6d032c5ef04a0f4966df8760664aaefc871.tar.gz
llvm-7bf0d6d032c5ef04a0f4966df8760664aaefc871.tar.bz2
[RISCV] Lower fixed-length strided VP loads and stores for zvfhmin/zvfbfmin (#114750)
Similarly to #114731, these don't actually require any instructions from the extensions. The motivation for this and #114731 is to eventually enable isLegalElementTypeForRVV for f16 with zvfhmin and bf16 with zvfbfmin in order to enable scalable vectorization. Although the scalable codegen support for f16 and bf16 is now complete enough for anything the loop vectorizer may emit, enabling isLegalElementTypeForRVV would make certian hooks like isLegalInterleavedAccessType and isLegalStridedLoadStore return true for f16 and bf16. This means SLP would start emitting these intrinsics, so we need to add fixed-length codegen support.
Diffstat (limited to 'clang/tools/include-mapping/cppreference_parser.py')
0 files changed, 0 insertions, 0 deletions