public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc(refs/vendors/redhat/heads/gcc-8-branch)] AArch64: Add Neoverse N2 tuning model
@ 2021-04-23 10:16 Jakub Jelinek
0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2021-04-23 10:16 UTC (permalink / raw)
To: gcc-cvs
https://gcc.gnu.org/g:c25d669bc1486d66c63c3213fb37475f16c25150
commit c25d669bc1486d66c63c3213fb37475f16c25150
Author: Kyrylo Tkachov <kyrylo.tkachov@arm.com>
Date: Thu Oct 8 15:20:23 2020 +0100
AArch64: Add Neoverse N2 tuning model
This is the GCC 8 patch for the Neoverse N2 tuning struct.
It sets the AARCH64_EXTRA_TUNE_PREFER_ADVSIMD_AUTOVEC tune flag as well.
Bootstrapped and tested on the branch.
gcc/
* config/aarch64/aarch64.c (neoversen2_tunings): Define.
* config/aarch64/aarch64-cores.def (neoverse-n2): Use it.
Diff:
---
gcc/config/aarch64/aarch64-cores.def | 2 +-
gcc/config/aarch64/aarch64.c | 26 ++++++++++++++++++++++++++
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/gcc/config/aarch64/aarch64-cores.def b/gcc/config/aarch64/aarch64-cores.def
index 29727fabc2a..dfb839c01cc 100644
--- a/gcc/config/aarch64/aarch64-cores.def
+++ b/gcc/config/aarch64/aarch64-cores.def
@@ -101,7 +101,7 @@ AARCH64_CORE("zeus", zeus, cortexa57, 8_4A, AARCH64_FL_FOR_ARCH8_4 | AARCH64_FL
AARCH64_CORE("neoverse-v1", neoversev1, cortexa57, 8_4A, AARCH64_FL_FOR_ARCH8_4 | AARCH64_FL_F16 | AARCH64_FL_RCPC | AARCH64_FL_SVE | AARCH64_FL_RNG, neoversev1, 0x41, 0xd40, -1)
/* Armv8.5-A Architecture Processors. */
-AARCH64_CORE("neoverse-n2", neoversen2, cortexa57, 8_4A, AARCH64_FL_FOR_ARCH8_4 | AARCH64_FL_F16 | AARCH64_FL_SVE | AARCH64_FL_RNG, cortexa72, 0x41, 0xd49, -1)
+AARCH64_CORE("neoverse-n2", neoversen2, cortexa57, 8_4A, AARCH64_FL_FOR_ARCH8_4 | AARCH64_FL_F16 | AARCH64_FL_SVE | AARCH64_FL_RNG, neoversen2, 0x41, 0xd49, -1)
/* ARMv8-A big.LITTLE implementations. */
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 3d218b6b951..72c11e35659 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -786,6 +786,32 @@ static const struct tune_params neoversev1_tunings =
&generic_prefetch_tune
};
+static const struct tune_params neoversen2_tunings =
+{
+ &cortexa57_extra_costs,
+ &generic_addrcost_table,
+ &cortexa57_regmove_cost,
+ &cortexa57_vector_cost,
+ &generic_branch_cost,
+ &generic_approx_modes,
+ 4, /* memmov_cost */
+ 3, /* issue_rate */
+ (AARCH64_FUSE_AES_AESMC | AARCH64_FUSE_MOV_MOVK | AARCH64_FUSE_ADRP_ADD
+ | AARCH64_FUSE_MOVK_MOVK), /* fusible_ops */
+ 16, /* function_align. */
+ 4, /* jump_align. */
+ 8, /* loop_align. */
+ 2, /* int_reassoc_width. */
+ 4, /* fp_reassoc_width. */
+ 1, /* vec_reassoc_width. */
+ 2, /* min_div_recip_mul_sf. */
+ 2, /* min_div_recip_mul_df. */
+ 0, /* max_case_values. */
+ tune_params::AUTOPREFETCHER_WEAK, /* autoprefetcher_model. */
+ (AARCH64_EXTRA_TUNE_PREFER_ADVSIMD_AUTOVEC), /* tune_flags. */
+ &generic_prefetch_tune
+};
+
static const struct tune_params cortexa73_tunings =
{
&cortexa57_extra_costs,
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-04-23 10:16 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-23 10:16 [gcc(refs/vendors/redhat/heads/gcc-8-branch)] AArch64: Add Neoverse N2 tuning model Jakub Jelinek
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).