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.
Tested on aarch64-linux-gnu and aarch64_be-elf, pushed.
Richard
2020-03-17 Richard Sandiford <richard.sandif...@arm.com>
gcc/
* config/aarch64/iterators.md (Vmtype): Handle V4BF and V8BF.
---
gcc/config/aarch64/iterators.md | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md
index 95fa3e4baa1..8e434389e59 100644
--- a/gcc/config/aarch64/iterators.md
+++ b/gcc/config/aarch64/iterators.md
@@ -1000,7 +1000,8 @@ (define_mode_attr Vmtype [(V8QI ".8b") (V16QI ".16b")
(V4HI ".4h") (V8HI ".8h")
(V2SI ".2s") (V4SI ".4s")
(V2DI ".2d") (V4HF ".4h")
- (V8HF ".8h") (V2SF ".2s")
+ (V8HF ".8h") (V4BF ".4h")
+ (V8BF ".8h") (V2SF ".2s")
(V4SF ".4s") (V2DF ".2d")
(DI "") (SI "")
(HI "") (QI "")
