public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/100095] New: missed optimization for dead code elimination at -O3 (vs. -O2)
@ 2021-04-15 11:07 zhendong.su at inf dot ethz.ch
  2021-04-15 12:06 ` [Bug tree-optimization/100095] " rguenth at gcc dot gnu.org
  2023-08-09 21:57 ` pinskia at gcc dot gnu.org
  0 siblings, 2 replies; 3+ messages in thread
From: zhendong.su at inf dot ethz.ch @ 2021-04-15 11:07 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100095

            Bug ID: 100095
           Summary: missed optimization for dead code elimination at -O3
                    (vs. -O2)
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zhendong.su at inf dot ethz.ch
  Target Milestone: ---

[654] % gcctk -v
Using built-in specs.
COLLECT_GCC=gcctk
COLLECT_LTO_WRAPPER=/local/suz-local/software/local/gcc-trunk/libexec/gcc/x86_64-pc-linux-gnu/11.0.1/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-trunk/configure --disable-bootstrap
--prefix=/local/suz-local/software/local/gcc-trunk --enable-languages=c,c++
--disable-werror --enable-multilib --with-system-zlib
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 11.0.1 20210415 (experimental) [master revision
4dd9e1c541e:7315804b0a0:b5f644a98b3f3543d3a8d2dfea7785c22879013f] (GCC) 
[655] % 
[655] % gcctk -O2 -S -c -o O2.s small.c
[656] % gcctk -O3 -S -c -o O3.s small.c
[657] % 
[657] % wc O2.s O3.s
  75  164  982 O2.s
 121  269 1617 O3.s
 196  433 2599 total
[658] % 
[658] % grep foo O2.s
[659] % grep foo O3.s
        jmp     foo
[660] % 
[660] % cat small.c
extern void foo(void);
int a, b, c, d, g, h;
static int *e = &d;
volatile int f;
void i() {
  for (d = 5; d >= 0; d--)
    for (c = 0; c < 4; c++)
      while (1) {
        h = b ? a % b : 0;
        if (g)
          f;
        if (*e)
          break;
      }
  foo();
}

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug tree-optimization/100095] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-15 11:07 [Bug tree-optimization/100095] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
@ 2021-04-15 12:06 ` rguenth at gcc dot gnu.org
  2023-08-09 21:57 ` pinskia at gcc dot gnu.org
  1 sibling, 0 replies; 3+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-04-15 12:06 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100095

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |85316
            Version|unknown                     |11.0
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-04-15
           Keywords|                            |missed-optimization

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
VRP optimizes this at -O2 but fails at -O3 where we fail to do some
final value replacement as we unswitch the loop nest on the b != 0 condition
which makes us end up with multiple exits which final_value_replacement_loop
doesn't like (it's somewhat artificial what we end up with - leaving
code commoning on the plate).

In the end the trigger of the missing optimization is unswitching.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85316
[Bug 85316] [meta-bug] VRP range propagation missed cases

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug tree-optimization/100095] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-15 11:07 [Bug tree-optimization/100095] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
  2021-04-15 12:06 ` [Bug tree-optimization/100095] " rguenth at gcc dot gnu.org
@ 2023-08-09 21:57 ` pinskia at gcc dot gnu.org
  1 sibling, 0 replies; 3+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-08-09 21:57 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100095

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|                            |12.1.0
           Keywords|                            |needs-bisection

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Looks to be fixed for GCC 12.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2023-08-09 21:57 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-15 11:07 [Bug tree-optimization/100095] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
2021-04-15 12:06 ` [Bug tree-optimization/100095] " rguenth at gcc dot gnu.org
2023-08-09 21:57 ` pinskia at gcc dot gnu.org

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).