public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2)
@ 2021-04-09  9:54 zhendong.su at inf dot ethz.ch
  2021-04-09 10:52 ` [Bug tree-optimization/99987] " rguenth at gcc dot gnu.org
                   ` (6 more replies)
  0 siblings, 7 replies; 8+ messages in thread
From: zhendong.su at inf dot ethz.ch @ 2021-04-09  9:54 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

            Bug ID: 99987
           Summary: missed optimization for dead code elimination at -O3
                    (vs. -O2)
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: ipa
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zhendong.su at inf dot ethz.ch
                CC: marxin at gcc dot gnu.org
  Target Milestone: ---

[748] % 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 20210409 (experimental) [master revision
96292c3e343:4e14cad25b9:019a922063f26784d5a070d9198a1f937b8a8343] (GCC) 
[749] % 
[749] % 
[749] % gcctk -O2 -S -o O2.s small.c
[750] % gcctk -O3 -S -o O3.s small.c
[751] % 
[751] % wc O2.s O3.s
  68  150  961 O2.s
  89  195 1278 O3.s
 157  345 2239 total
[752] % 
[752] % grep foo O2.s
[753] % grep foo O3.s
        call    foo
[754] % 
[754] % cat small.c
extern void foo(void);
static int a[8] = {0,0,0,0,0,0,0,0};
int b, c, j;
static int *e(int k) {
  if (k && b)
    j = 0;
  while (j)
    if (!k)
      foo();
  return 0;
}
static void d() {
  long g;
  unsigned h = 8;
  if (b)
    for (g = 2; 1; g--) {
      int i = 0;
      h--;
      c = (a[g] == (&i != e(h)));
    }
}
void f() { d(); }

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
@ 2021-04-09 10:52 ` rguenth at gcc dot gnu.org
  2023-05-05  7:50 ` [Bug tree-optimization/99987] [12/13/14 Regression] " pinskia at gcc dot gnu.org
                   ` (5 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-04-09 10:52 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
            Version|unknown                     |11.0
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |missed-optimization
          Component|ipa                         |tree-optimization
   Last reconfirmed|                            |2021-04-09

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed.  With -O2 DOM3 removes the call, with -O3 it does not.

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
  2021-04-09 10:52 ` [Bug tree-optimization/99987] " rguenth at gcc dot gnu.org
@ 2023-05-05  7:50 ` pinskia at gcc dot gnu.org
  2023-05-08 12:21 ` rguenth at gcc dot gnu.org
                   ` (4 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-05-05  7:50 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |12.3
           Keywords|                            |needs-bisection
            Summary|missed optimization for     |[12/13/14 Regression]
                   |dead code elimination at    |missed optimization for
                   |-O3 (vs. -O2)               |dead code elimination at
                   |                            |-O3 (vs. -O2)

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Note GCC 12+ can't remove the call to foo even at -O2. I have not looked into
why though.

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
  2021-04-09 10:52 ` [Bug tree-optimization/99987] " rguenth at gcc dot gnu.org
  2023-05-05  7:50 ` [Bug tree-optimization/99987] [12/13/14 Regression] " pinskia at gcc dot gnu.org
@ 2023-05-08 12:21 ` rguenth at gcc dot gnu.org
  2023-05-09 11:39 ` marxin at gcc dot gnu.org
                   ` (3 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: rguenth at gcc dot gnu.org @ 2023-05-08 12:21 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|12.3                        |12.4

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 12.3 is being released, retargeting bugs to GCC 12.4.

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
                   ` (2 preceding siblings ...)
  2023-05-08 12:21 ` rguenth at gcc dot gnu.org
@ 2023-05-09 11:39 ` marxin at gcc dot gnu.org
  2023-08-18  6:13 ` pinskia at gcc dot gnu.org
                   ` (2 subsequent siblings)
  6 siblings, 0 replies; 8+ messages in thread
From: marxin at gcc dot gnu.org @ 2023-05-09 11:39 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|needs-bisection             |

--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #2)
> Note GCC 12+ can't remove the call to foo even at -O2. I have not looked
> into why though.

It's due to r12-4240-g2b8453c401b699.

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
                   ` (3 preceding siblings ...)
  2023-05-09 11:39 ` marxin at gcc dot gnu.org
@ 2023-08-18  6:13 ` pinskia at gcc dot gnu.org
  2023-08-18  6:36 ` pinskia at gcc dot gnu.org
  2024-03-10  2:45 ` law at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-08-18  6:13 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Note turning off ivopts, DOM3 can't optimize away the call either ...

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
                   ` (4 preceding siblings ...)
  2023-08-18  6:13 ` pinskia at gcc dot gnu.org
@ 2023-08-18  6:36 ` pinskia at gcc dot gnu.org
  2024-03-10  2:45 ` law at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-08-18  6:36 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|                            |7.5.0

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
So this is bascially the missed optimization:
```
void foo(void);
int main()
{
  unsigned h = 123+5;
  for(unsigned g = 123; g != 0; g--)
  {
    h--;
    if(h == 0)
       foo();
  }
}
```
The relationship between h and g is `h = g + 5` so you can never get `h == 0`
in the loop as the range of g is [123,0].

Note In the original testcase, only GCC 8-11 dom3 (on x86_64; but not on
aarch64 due to IVOPTS differences) could optimize away the call to foo.

Also note LLVM can optimize it away both the above testcase and the original
one.

^ permalink raw reply	[flat|nested] 8+ messages in thread

* [Bug tree-optimization/99987] [12/13/14 Regression] missed optimization for dead code elimination at -O3 (vs. -O2)
  2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
                   ` (5 preceding siblings ...)
  2023-08-18  6:36 ` pinskia at gcc dot gnu.org
@ 2024-03-10  2:45 ` law at gcc dot gnu.org
  6 siblings, 0 replies; 8+ messages in thread
From: law at gcc dot gnu.org @ 2024-03-10  2:45 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99987

Jeffrey A. Law <law at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
                 CC|                            |law at gcc dot gnu.org

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2024-03-10  2:45 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-09  9:54 [Bug ipa/99987] New: missed optimization for dead code elimination at -O3 (vs. -O2) zhendong.su at inf dot ethz.ch
2021-04-09 10:52 ` [Bug tree-optimization/99987] " rguenth at gcc dot gnu.org
2023-05-05  7:50 ` [Bug tree-optimization/99987] [12/13/14 Regression] " pinskia at gcc dot gnu.org
2023-05-08 12:21 ` rguenth at gcc dot gnu.org
2023-05-09 11:39 ` marxin at gcc dot gnu.org
2023-08-18  6:13 ` pinskia at gcc dot gnu.org
2023-08-18  6:36 ` pinskia at gcc dot gnu.org
2024-03-10  2:45 ` law 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).