public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/100858] New: Simple common code hoisting is not performed
@ 2021-06-01 15:16 nok.raven at gmail dot com
  2021-06-02  8:01 ` [Bug tree-optimization/100858] " rguenth at gcc dot gnu.org
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: nok.raven at gmail dot com @ 2021-06-01 15:16 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100858

            Bug ID: 100858
           Summary: Simple common code hoisting is not performed
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: nok.raven at gmail dot com
  Target Milestone: ---

void bar();

int foo(bool f)
{
    if (f) {
        bar();
        return 1;
    }
    else {
        bar();
        return 2;
    }
}

; GCC
foo(bool):
        sub     rsp, 8
        test    dil, dil
        je      .L2
        call    bar()
        mov     eax, 1
        add     rsp, 8
        ret
.L2:
        call    bar()
        mov     eax, 2
        add     rsp, 8
        ret

; Clang
foo(bool):
        push    rbx
        mov     ebx, edi
        call    bar()
        mov     eax, 2
        sub     eax, ebx
        pop     rbx
        ret

https://godbolt.org/z/j9eKG1dds

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug tree-optimization/100858] Simple common code hoisting is not performed
  2021-06-01 15:16 [Bug tree-optimization/100858] New: Simple common code hoisting is not performed nok.raven at gmail dot com
@ 2021-06-02  8:01 ` rguenth at gcc dot gnu.org
  2021-06-02 23:18 ` nok.raven at gmail dot com
  2021-07-28 20:17 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-06-02  8:01 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100858

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-06-02
     Ever confirmed|0                           |1

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
That's really a duplicate of 100858 - this case can be handled by sinking as
well
since we "sink" the return.  Make it

void bar();

int foo(bool f)
{
    if (f) {
        bar();
        __builtin_abort ();
    }
    else {
        bar();
        return 2;
    }
}

to force hoisting.  Hoisting is done by PRE but that requires a LHS and
doesn't handle calls with side-effects.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug tree-optimization/100858] Simple common code hoisting is not performed
  2021-06-01 15:16 [Bug tree-optimization/100858] New: Simple common code hoisting is not performed nok.raven at gmail dot com
  2021-06-02  8:01 ` [Bug tree-optimization/100858] " rguenth at gcc dot gnu.org
@ 2021-06-02 23:18 ` nok.raven at gmail dot com
  2021-07-28 20:17 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: nok.raven at gmail dot com @ 2021-06-02 23:18 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100858

--- Comment #2 from Nikita Kniazev <nok.raven at gmail dot com> ---
(In reply to Richard Biener from comment #1)
> That's really a duplicate of 100858 - this case can be handled by sinking as
> well
> since we "sink" the return.  Make it
> 
> void bar();
> 
> int foo(bool f)
> {
>     if (f) {
>         bar();
>         __builtin_abort ();
>     }
>     else {
>         bar();
>         return 2;
>     }
> }
> to force hoisting.

I have a hard time in finding this as equivalent to the original code.

> Hoisting is done by PRE but that requires a LHS and doesn't handle calls with side-effects.

That seems like an artificial limitation. Why we cannot hoist the bar call?

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug tree-optimization/100858] Simple common code hoisting is not performed
  2021-06-01 15:16 [Bug tree-optimization/100858] New: Simple common code hoisting is not performed nok.raven at gmail dot com
  2021-06-02  8:01 ` [Bug tree-optimization/100858] " rguenth at gcc dot gnu.org
  2021-06-02 23:18 ` nok.raven at gmail dot com
@ 2021-07-28 20:17 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-07-28 20:17 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100858

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement
                 CC|                            |pinskia at gcc dot gnu.org

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-07-28 20:17 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-01 15:16 [Bug tree-optimization/100858] New: Simple common code hoisting is not performed nok.raven at gmail dot com
2021-06-02  8:01 ` [Bug tree-optimization/100858] " rguenth at gcc dot gnu.org
2021-06-02 23:18 ` nok.raven at gmail dot com
2021-07-28 20:17 ` 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).