projects / gcc.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1fef014) | patch
aarch64: Fix bf16_v(ld|st)n.c failures for big-endian
authorRichard Sandiford <richard.sandiford@arm.com>
Tue, 17 Mar 2020 15:36:37 +0000 (15:36 +0000)
committerRichard Sandiford <richard.sandiford@arm.com>
Tue, 17 Mar 2020 19:23:46 +0000 (19:23 +0000)
commitcf9c3bff39cf973c5c8621ff44199dcb831193a7
treee7d13f9a66092f7a99c5afa018efe6740a0f1b27tree
parent1fef0148be4b40660446d79a2b2dc73e89bfbeffcommit | diff
aarch64: Fix bf16_v(ld|st)n.c failures for big-endian

gcc.target/aarch64/advsimd-intrinsics/bf16_vldn.c and
gcc.target/aarch64/advsimd-intrinsics/bf16_vstn.c were
failing for big-endian targets because the <Vmtype> in
aarch64_be_ld1<mode> and aarch64_be_st1<mode> had no
expansion for the bfloat16 modes.

2020-03-17  Richard Sandiford  <richard.sandiford@arm.com>

gcc/
* config/aarch64/iterators.md (Vmtype): Handle V4BF and V8BF.
gcc/ChangeLog diff | blob | history
gcc/config/aarch64/iterators.md diff | blob | history