[PATCH] D60618: [ARM] Add bitcast/extract_subvec. of fp16 vectors
Diogo N. Sampaio via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Apr 25 09:43:05 PDT 2019
dnsampaio updated this revision to Diff 196660.
dnsampaio added a comment.
Fixes requested
- Removed duplicated line
- Changed tests as to not use vector shuffling. However, as vector are always loaded / passed as a f64 or v2f64, it is required to always perform some operation of the original type before bitcasting and performing an operation in the new type, else it always performs vrev64.16
Repository:
rL LLVM
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D60618/new/
https://reviews.llvm.org/D60618
Files:
lib/Target/ARM/ARMInstrNEON.td
test/CodeGen/ARM/big-endian-neon-fp16-bitconv.ll
test/CodeGen/ARM/fp16-vector-argument.ll
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D60618.196660.patch
Type: text/x-patch
Size: 45110 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190425/8312e1f9/attachment.bin>
More information about the llvm-commits
mailing list