procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

RISC-V: Fix Dynamic LMUL compile option

Browse source 
gcc/ChangeLog:

	* config/riscv/riscv-opts.h (enum riscv_autovec_lmul_enum): Fix Dynamic status.
	* config/riscv/riscv-v.cc (preferred_simd_mode): Ditto.
	(autovectorize_vector_modes): Ditto.
	(vectorize_related_mode): Ditto.
This commit is contained in:
Juzhe-Zhong 2023-09-04 17:08:34 +08:00 • committed by Lehua Ding
parent 99ec76eb8b
commit 6f94ef6c86
2 changed files with 9 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
gcc/config/riscv/riscv-opts.h
View file

@ -81,7 +81,7 @@ enum riscv_autovec_lmul_enum {
RVV_M4 = 4,
RVV_M8 = 8,
/* For dynamic LMUL, we compare COST start with LMUL8. */
RVV_DYNAMIC = RVV_M8
RVV_DYNAMIC = 9
};
enum riscv_multilib_select_kind {

15
gcc/config/riscv/riscv-v.cc
View file

@ -1978,16 +1978,16 @@ preferred_simd_mode (scalar_mode mode)
vectorizer when we enable them in this target hook. Currently, we can
support auto-vectorization in -march=rv32_zve32x_zvl128b. Wheras,
-march=rv32_zve32x_zvl32b or -march=rv32_zve32x_zvl64b are disabled. */
int lmul = riscv_autovec_lmul == RVV_DYNAMIC ? RVV_M8 : riscv_autovec_lmul;
if (autovec_use_vlmax_p ())
{
if (TARGET_MIN_VLEN < 128 && riscv_autovec_lmul < RVV_M2)
if (TARGET_MIN_VLEN < 128 && lmul < RVV_M2)
return word_mode;
/* We use LMUL = 1 as base bytesize which is BYTES_PER_RISCV_VECTOR and
riscv_autovec_lmul as multiply factor to calculate the the NUNITS to
get the auto-vectorization mode. */
poly_uint64 nunits;
poly_uint64 vector_size
= BYTES_PER_RISCV_VECTOR * ((int) riscv_autovec_lmul);
poly_uint64 vector_size = BYTES_PER_RISCV_VECTOR * lmul;
poly_uint64 scalar_size = GET_MODE_SIZE (mode);
gcc_assert (multiple_p (vector_size, scalar_size, &nunits));
machine_mode rvv_mode;
@ -2161,10 +2161,10 @@ get_cmp_insn_code (rtx_code code, machine_mode mode)
unsigned int
autovectorize_vector_modes (vector_modes *modes, bool)
{
int lmul = riscv_autovec_lmul == RVV_DYNAMIC ? RVV_M8 : riscv_autovec_lmul;
if (autovec_use_vlmax_p ())
{
poly_uint64 full_size
= BYTES_PER_RISCV_VECTOR * ((int) riscv_autovec_lmul);
poly_uint64 full_size = BYTES_PER_RISCV_VECTOR * lmul;
/* Start with a RVV<LMUL>QImode where LMUL is the number of units that
fit a whole vector.
@ -2194,7 +2194,7 @@ autovectorize_vector_modes (vector_modes *modes, bool)
{
/* Push all VLSmodes according to TARGET_MIN_VLEN. */
unsigned int i = 0;
unsigned int base_size = TARGET_MIN_VLEN * riscv_autovec_lmul / 8;
unsigned int base_size = TARGET_MIN_VLEN * lmul / 8;
unsigned int size = base_size;
machine_mode mode;
while (size > 0 && get_vector_mode (QImode, size).exists (&mode))
@ -2219,8 +2219,9 @@ vectorize_related_mode (machine_mode vector_mode, scalar_mode element_mode,
{
/* TODO: We will support RVV VLS auto-vectorization mode in the future. */
poly_uint64 min_units;
int lmul = riscv_autovec_lmul == RVV_DYNAMIC ? RVV_M8 : riscv_autovec_lmul;
if (autovec_use_vlmax_p () && riscv_v_ext_vector_mode_p (vector_mode)
&& multiple_p (BYTES_PER_RISCV_VECTOR * ((int) riscv_autovec_lmul),
&& multiple_p (BYTES_PER_RISCV_VECTOR * lmul,
GET_MODE_SIZE (element_mode), &min_units))
{
machine_mode rvv_mode;