Hi all,
Here is v2 of patch 11/14 from the unpacked FP series [1]. Apologies for
the delay in posting.
Changes since V1:
* Applied Richard's suggestions RE comments, and the code change
for aarch64_sve_emit_masked_fp_pred.
* The tests now use -moverride=sve_width=2048, and check for the
extending-load patterns (in keeping with the rest of the series).
Thanks,
Spencer
[1]
https://inbox.sourceware.org/gcc-patches/20250602100707.2659-1-spencer.ab...@arm.com/
Spencer Abson (1):
aarch64: Add support for unpacked SVE FP conditional binary arithmetic
gcc/config/aarch64/aarch64-protos.h | 1 +
gcc/config/aarch64/aarch64-sve.md | 152 +++++++++---------
gcc/config/aarch64/aarch64.cc | 27 ++++
.../aarch64/sve/unpacked_cond_binary_bf16_2.C | 18 +++
.../sve/unpacked_cond_builtin_fmax_2.c | 24 +++
.../sve/unpacked_cond_builtin_fmin_2.c | 24 +++
.../aarch64/sve/unpacked_cond_fadd_2.c | 28 ++++
.../aarch64/sve/unpacked_cond_fdiv_2.c | 22 +++
.../aarch64/sve/unpacked_cond_fmaxnm_2.c | 24 +++
.../aarch64/sve/unpacked_cond_fminnm_2.c | 24 +++
.../aarch64/sve/unpacked_cond_fmul_2.c | 22 +++
.../aarch64/sve/unpacked_cond_fsubr_2.c | 26 +++
12 files changed, 319 insertions(+), 73 deletions(-)
create mode 100644
gcc/testsuite/g++.target/aarch64/sve/unpacked_cond_binary_bf16_2.C
create mode 100644
gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_builtin_fmax_2.c
create mode 100644
gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_builtin_fmin_2.c
create mode 100644 gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fadd_2.c
create mode 100644 gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fdiv_2.c
create mode 100644
gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fmaxnm_2.c
create mode 100644
gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fminnm_2.c
create mode 100644 gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fmul_2.c
create mode 100644 gcc/testsuite/gcc.target/aarch64/sve/unpacked_cond_fsubr_2.c
--
2.34.1
