public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2
@ 2021-05-20 10:47 zsojka at seznam dot cz
  2021-05-20 11:27 ` [Bug target/100701] " rguenth at gcc dot gnu.org
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: zsojka at seznam dot cz @ 2021-05-20 10:47 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100701

            Bug ID: 100701
           Summary: [12 Regression] wrong code with -O -fschedule-insns2
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: rtl-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zsojka at seznam dot cz
  Target Milestone: ---
              Host: x86_64-pc-linux-gnu

Created attachment 50850
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=50850&action=edit
reduced testcase

Output:
$ x86_64-pc-linux-gnu-gcc -O -fschedule-insns2 testcase.c
$ ./a.out 
Aborted

$ x86_64-pc-linux-gnu-gcc -v
Using built-in specs.
COLLECT_GCC=/repo/gcc-trunk/binary-latest-amd64/bin/x86_64-pc-linux-gnu-gcc
COLLECT_LTO_WRAPPER=/repo/gcc-trunk/binary-trunk-r12-940-20210520111231-g507359e1d4d-checking-yes-rtl-df-extra-amd64/bin/../libexec/gcc/x86_64-pc-linux-gnu/12.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /repo/gcc-trunk//configure --enable-languages=c,c++
--enable-valgrind-annotations --disable-nls --enable-checking=yes,rtl,df,extra
--with-cloog --with-ppl --with-isl --build=x86_64-pc-linux-gnu
--host=x86_64-pc-linux-gnu --target=x86_64-pc-linux-gnu
--with-ld=/usr/bin/x86_64-pc-linux-gnu-ld
--with-as=/usr/bin/x86_64-pc-linux-gnu-as --disable-libstdcxx-pch
--prefix=/repo/gcc-trunk//binary-trunk-r12-940-20210520111231-g507359e1d4d-checking-yes-rtl-df-extra-amd64
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 12.0.0 20210520 (experimental) (GCC)

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug target/100701] [12 Regression] wrong code with -O -fschedule-insns2
  2021-05-20 10:47 [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2 zsojka at seznam dot cz
@ 2021-05-20 11:27 ` rguenth at gcc dot gnu.org
  2021-05-20 13:29 ` ubizjak at gmail dot com
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-05-20 11:27 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100701

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2021-05-20
             Target|                            |x86_64-*-* i?86-*-*
             Status|UNCONFIRMED                 |NEW
          Component|rtl-optimization            |target
                 CC|                            |uros at gcc dot gnu.org
   Target Milestone|---                         |12.0

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
        orq     %rdi, %rsi
        pshuflw $0, %xmm3, %xmm0
        movq    %xmm0, %rbp
        and     %rbx, %rbp
        je      .L4

Confirmed.  Somehow the mmx_andv8qi3 doesn't clobber CC:

(insn 19 88 15 2 (parallel [
            (set (reg:CCZ 17 flags)
                (compare:CCZ (ior:DI (reg:DI 4 si [158])
                        (reg:DI 5 di [orig:101 i ] [101]))
                    (const_int 0 [0])))
            (clobber (reg:DI 4 si [158]))
        ]) "t.c":13:6 562 {*iordi_3}
     (expr_list:REG_DEAD (reg:DI 4 si [158])
        (expr_list:REG_UNUSED (reg:DI 4 si [158])
            (nil))))
(insn:TI 15 19 89 2 (set (reg:V4HI 20 xmm0 [106])
        (vec_duplicate:V4HI (truncate:HI (reg:SI 23 xmm3 [162])))) "t.c":12:5
1459 {*vec_dupv4hi}
     (expr_list:REG_DEAD (reg:SI 23 xmm3 [162])
        (nil)))
(insn:TI 89 15 17 2 (set (reg/v:V8QI 6 bp [orig:95 a ] [95])
        (reg:V8QI 20 xmm0 [106])) "t.c":12:5 1302 {*movv8qi_internal}
     (expr_list:REG_DEAD (reg:V8QI 20 xmm0 [106])
        (nil)))
(insn:TI 17 89 20 2 (set (reg/v:V8QI 6 bp [orig:95 a ] [95])
        (and:V8QI (reg/v:V8QI 6 bp [orig:95 a ] [95])
            (reg/v:V8QI 3 bx [orig:98 a ] [98]))) "t.c":12:5 1422
{*mmx_andv8qi3}
     (nil))
(jump_insn 20 17 25 2 (set (pc)
        (if_then_else (eq (reg:CCZ 17 flags)
                (const_int 0 [0]))
            (label_ref:DI 113)
            (pc))) "t.c":13:6 822 {*jcc}
     (expr_list:REG_DEAD (reg:CCZ 17 flags)
        (int_list:REG_BR_PROB 7 (nil)))

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug target/100701] [12 Regression] wrong code with -O -fschedule-insns2
  2021-05-20 10:47 [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2 zsojka at seznam dot cz
  2021-05-20 11:27 ` [Bug target/100701] " rguenth at gcc dot gnu.org
@ 2021-05-20 13:29 ` ubizjak at gmail dot com
  2021-05-20 16:50 ` cvs-commit at gcc dot gnu.org
  2021-05-20 16:53 ` ubizjak at gmail dot com
  3 siblings, 0 replies; 5+ messages in thread
From: ubizjak at gmail dot com @ 2021-05-20 13:29 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100701

Uroš Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |ubizjak at gmail dot com
                 CC|uros at gcc dot gnu.org            |

--- Comment #2 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Richard Biener from comment #1)
>         orq     %rdi, %rsi
>         pshuflw $0, %xmm3, %xmm0
>         movq    %xmm0, %rbp
>         and     %rbx, %rbp
>         je      .L4
> 
> Confirmed.  Somehow the mmx_andv8qi3 doesn't clobber CC:

Eh, stupid me. Let's remove integer regs from the insn patterns for now.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug target/100701] [12 Regression] wrong code with -O -fschedule-insns2
  2021-05-20 10:47 [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2 zsojka at seznam dot cz
  2021-05-20 11:27 ` [Bug target/100701] " rguenth at gcc dot gnu.org
  2021-05-20 13:29 ` ubizjak at gmail dot com
@ 2021-05-20 16:50 ` cvs-commit at gcc dot gnu.org
  2021-05-20 16:53 ` ubizjak at gmail dot com
  3 siblings, 0 replies; 5+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-05-20 16:50 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100701

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Uros Bizjak <uros@gcc.gnu.org>:

https://gcc.gnu.org/g:a71f55c482ada2c6c31d450ac22494b547512127

commit r12-945-ga71f55c482ada2c6c31d450ac22494b547512127
Author: Uros Bizjak <ubizjak@gmail.com>
Date:   Thu May 20 18:48:16 2021 +0200

    i386: Avoid integer logic insns for 32bit and 64bit vector modes [PR100701]

    Integer logic instructions clobber flags, do not use them for
    32bit and 64bit vector modes.

    2021-05-20  Uroš Bizjak  <ubizjak@gmail.com>

    gcc/
            PR target/100701
            * config/i386/i386.md (isa): Remove x64_bmi.
            (enabled): Remove x64_bmi.
            * config/i386/mmx.md (mmx_andnot<MMXMODEI:mode>3):
            Remove general register alternative.
            (*andnot<VI_32:mode>3): Ditto.
            (*mmx_<any_logic:code><MMXMODEI:mode>3): Ditto.
            (*<any_logic:code><VI_32:mode>3): Ditto.

    gcc/testsuite/

            PR target/100701
            * gcc.target/i386/pr100701.c: New test.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Bug target/100701] [12 Regression] wrong code with -O -fschedule-insns2
  2021-05-20 10:47 [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2 zsojka at seznam dot cz
                   ` (2 preceding siblings ...)
  2021-05-20 16:50 ` cvs-commit at gcc dot gnu.org
@ 2021-05-20 16:53 ` ubizjak at gmail dot com
  3 siblings, 0 replies; 5+ messages in thread
From: ubizjak at gmail dot com @ 2021-05-20 16:53 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100701

Uroš Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #4 from Uroš Bizjak <ubizjak at gmail dot com> ---
Fixed.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2021-05-20 16:53 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-05-20 10:47 [Bug rtl-optimization/100701] New: [12 Regression] wrong code with -O -fschedule-insns2 zsojka at seznam dot cz
2021-05-20 11:27 ` [Bug target/100701] " rguenth at gcc dot gnu.org
2021-05-20 13:29 ` ubizjak at gmail dot com
2021-05-20 16:50 ` cvs-commit at gcc dot gnu.org
2021-05-20 16:53 ` ubizjak at gmail dot com

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).