From: Richard Biener <rguenther@suse.de>
To: gcc-patches@gcc.gnu.org
Subject: [PATCH] tree-optimization/39821 - fix cost classification for widening arith
Date: Tue, 27 Jul 2021 10:41:18 +0200 (CEST) [thread overview]
Message-ID: <122q6r7-351o-74os-87ss-q3rp695qnq3r@fhfr.qr> (raw)
This adjusts the vectorizer to cost vector_stmt for widening
arithmetic instead of vec_promote_demote in the line of telling
the target that stmt_info->stmt is the meaningful piece we cost.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
2021-07-27 Richard Biener <rguenther@suse.de>
PR tree-optimization/39821
* tree-vect-stmts.c (vect_model_promotion_demotion_cost): Use
vector_stmt for widening arithmetic.
(vectorizable_conversion): Adjust.
---
gcc/tree-vect-stmts.c | 25 ++++++++++++++++---------
1 file changed, 16 insertions(+), 9 deletions(-)
diff --git a/gcc/tree-vect-stmts.c b/gcc/tree-vect-stmts.c
index d71552296bb..074dfdcf385 100644
--- a/gcc/tree-vect-stmts.c
+++ b/gcc/tree-vect-stmts.c
@@ -836,20 +836,24 @@ vect_model_simple_cost (vec_info *,
one if two-step promotion/demotion is required, and so on. NCOPIES
is the number of vector results (and thus number of instructions)
for the narrowest end of the operation chain. Each additional
- step doubles the number of instructions required. */
+ step doubles the number of instructions required. If WIDEN_ARITH
+ is true the stmt is doing widening arithmetic. */
static void
vect_model_promotion_demotion_cost (stmt_vec_info stmt_info,
enum vect_def_type *dt,
unsigned int ncopies, int pwr,
- stmt_vector_for_cost *cost_vec)
+ stmt_vector_for_cost *cost_vec,
+ bool widen_arith)
{
int i;
int inside_cost = 0, prologue_cost = 0;
for (i = 0; i < pwr + 1; i++)
{
- inside_cost += record_stmt_cost (cost_vec, ncopies, vec_promote_demote,
+ inside_cost += record_stmt_cost (cost_vec, ncopies,
+ widen_arith
+ ? vector_stmt : vec_promote_demote,
stmt_info, 0, vect_body);
ncopies *= 2;
}
@@ -4690,6 +4694,10 @@ vectorizable_conversion (vec_info *vinfo,
&& code != WIDEN_LSHIFT_EXPR)
return false;
+ bool widen_arith = (code == WIDEN_PLUS_EXPR
+ || code == WIDEN_MINUS_EXPR
+ || code == WIDEN_MULT_EXPR
+ || code == WIDEN_LSHIFT_EXPR);
op_type = TREE_CODE_LENGTH (code);
/* Check types of lhs and rhs. */
@@ -4779,10 +4787,7 @@ vectorizable_conversion (vec_info *vinfo,
nunits_in = TYPE_VECTOR_SUBPARTS (vectype_in);
nunits_out = TYPE_VECTOR_SUBPARTS (vectype_out);
if (known_eq (nunits_out, nunits_in))
- if (code == WIDEN_MINUS_EXPR
- || code == WIDEN_PLUS_EXPR
- || code == WIDEN_LSHIFT_EXPR
- || code == WIDEN_MULT_EXPR)
+ if (widen_arith)
modifier = WIDEN;
else
modifier = NONE;
@@ -4959,7 +4964,8 @@ vectorizable_conversion (vec_info *vinfo,
unsigned int nvectors
= (slp_node ? SLP_TREE_NUMBER_OF_VEC_STMTS (slp_node) : ncopies);
vect_model_promotion_demotion_cost (stmt_info, dt, nvectors,
- multi_step_cvt, cost_vec);
+ multi_step_cvt, cost_vec,
+ widen_arith);
}
else
{
@@ -4972,7 +4978,8 @@ vectorizable_conversion (vec_info *vinfo,
? SLP_TREE_NUMBER_OF_VEC_STMTS (slp_node) >> multi_step_cvt
: ncopies * 2);
vect_model_promotion_demotion_cost (stmt_info, dt, nvectors,
- multi_step_cvt, cost_vec);
+ multi_step_cvt, cost_vec,
+ widen_arith);
}
interm_types.release ();
return true;
--
2.26.2
reply other threads:[~2021-07-27 8:41 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=122q6r7-351o-74os-87ss-q3rp695qnq3r@fhfr.qr \
--to=rguenther@suse.de \
--cc=gcc-patches@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).