public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: "Roger Sayle" <roger@nextmovesoftware.com>
To: "'GCC Patches'" <gcc-patches@gcc.gnu.org>
Subject: [x86 PATCH] Fix libitm.c/memset-1.c test fails with new peephole2s.
Date: Sat, 12 Mar 2022 18:50:27 -0000	[thread overview]
Message-ID: <003401d83642$0b222c40$216684c0$@nextmovesoftware.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 1402 bytes --]


My sincere apologies for the breakage, but alas handling SImode in the
recently added "xorl;movb -> movzbl" peephole2 turns out to be slightly
more complicated that just using SWI48 as a mode iterator.  I'd failed
to check the machine description carefully, but the *zero_extend<mode>si2
define_insn is conditionally defined, based on x86 target tuning using
TARGET_ZERO_EXTEND_WITH_AND, and therefore unavailable on 486 and pentium
unless optimizing the code for size.  It turns out that the libitm testsuite
specifies -m486 with make check RUNTESTFLAGS="--target_board='unix{-m32}'"
and therefore encounters/catches my oversight.

Fixed by adding the appropriate conditions to the new peephole2 patterns.
It don't think it's worth the effort to provide an equivalent optimization
for
these (very) old architectures.

Tested on x86_64-pc-linux-gnu with make bootstrap and make -k check
with no new failures.  Confirmed using RUNTESTFLAGS that this
fixes the above failure, and the recently added testcase with -march=i486.
Ok for mainline?


2022-03-12  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
	* config/i386/i386.md (peephole2 xorl;movb -> movzbl): Disable
	transformation when *zero_extend<mode>si2 is not available.

gcc/testsuite/ChangeLog
	* gcc.target/i386/pr98335.c: Skip this test if tuning for i486
	or pentium, and not optimizing for size.


Sorry for the noise.
Roger
--


[-- Attachment #2: patchct7.txt --]
[-- Type: text/plain, Size: 1840 bytes --]

diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index c8fbf60..80b5974 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -4316,7 +4316,10 @@
 	      (clobber (reg:CC FLAGS_REG))])
    (set (strict_low_part (match_operand:SWI12 1 "general_reg_operand"))
 	(match_operand:SWI12 2 "nonimmediate_operand"))]
-  "REGNO (operands[0]) == REGNO (operands[1])"
+  "REGNO (operands[0]) == REGNO (operands[1])
+   && (<SWI48:MODE_SIZE> != 4
+       || !TARGET_ZERO_EXTEND_WITH_AND
+       || !optimize_function_for_speed_p (cfun))"
   [(set (match_dup 0) (zero_extend:SWI48 (match_dup 2)))])
 
 ;; Likewise, but preserving FLAGS_REG.
@@ -4324,7 +4327,10 @@
   [(set (match_operand:SWI48 0 "general_reg_operand") (const_int 0))
    (set (strict_low_part (match_operand:SWI12 1 "general_reg_operand"))
 	(match_operand:SWI12 2 "nonimmediate_operand"))]
-  "REGNO (operands[0]) == REGNO (operands[1])"
+  "REGNO (operands[0]) == REGNO (operands[1])
+   && (<SWI48:MODE_SIZE> != 4
+       || !TARGET_ZERO_EXTEND_WITH_AND
+       || !optimize_function_for_speed_p (cfun))"
   [(set (match_dup 0) (zero_extend:SWI48 (match_dup 2)))])
 \f
 ;; Sign extension instructions
diff --git a/gcc/testsuite/gcc.target/i386/pr98335.c b/gcc/testsuite/gcc.target/i386/pr98335.c
index 7fa7ad7..bf731b4 100644
--- a/gcc/testsuite/gcc.target/i386/pr98335.c
+++ b/gcc/testsuite/gcc.target/i386/pr98335.c
@@ -1,5 +1,10 @@
 /* { dg-do compile } */
 /* { dg-options "-O2" } */
+/* { dg-skip-if "" { *-*-* } { "-march=i[45]86" } { "-O[sz]" } } */
+/* { dg-skip-if "" { *-*-* } { "-march=pentium" } { "-O[sz]" } } */
+/* { dg-skip-if "" { *-*-* } { "-mtune=i[45]86" } { "-O[sz]" } } */
+/* { dg-skip-if "" { *-*-* } { "-mtune=pentium" } { "-O[sz]" } } */
+
 union Data { char a; short b; };
 
 char c;

             reply	other threads:[~2022-03-12 18:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-12 18:50 Roger Sayle [this message]
2022-03-13 17:41 ` Uros Bizjak

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='003401d83642$0b222c40$216684c0$@nextmovesoftware.com' \
    --to=roger@nextmovesoftware.com \
    --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).