public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d
@ 2023-11-15  9:42 theodort at inf dot ethz.ch
  2023-11-15 11:59 ` [Bug tree-optimization/112542] " rguenth at gcc dot gnu.org
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: theodort at inf dot ethz.ch @ 2023-11-15  9:42 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112542

            Bug ID: 112542
           Summary: [14 Regression] Dead Code Elimination Regression since
                    r14-4280-gc3c6f30496d
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: theodort at inf dot ethz.ch
  Target Milestone: ---

https://godbolt.org/z/3vj8x1bPK

Given the following code:

void foo(void);
static short b = -1;
static int c, e;
static int *f = &c;
static char g;
static char(a)(char h, int i) {
    if (!(((i) >= -1) && ((i) <= 0))) {
        __builtin_unreachable();
    }
    return h || i ? h : h < 0;
}
static void(d)(unsigned h) {
    if (!(((h) >= 0) && ((h) <= 4095017279))) {
        foo();
    }
}
int main() {
    *f = b == 0;
    g = a(c, c);
    d(g);
    d(4095017279);
    if (e) b = 0;
    a(0, b);
}

gcc-trunk -O3 does not eliminate the call to foo:

main:
        xorl    %eax, %eax
        cmpw    $0, b(%rip)
        sete    %al
        movl    %eax, c(%rip)
        jne     .L4
        pushq   %rax
        call    foo
        xorl    %eax, %eax
        popq    %rdx
        ret
.L4:
        xorl    %eax, %eax
        ret

gcc-13.2.0 -O3 eliminates the call to foo:

main:
        xorl    %eax, %eax
        cmpw    $0, b(%rip)
        sete    %al
        movl    %eax, c(%rip)
        xorl    %eax, %eax
        ret

Bisects to r14-4280-gc3c6f30496d

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

* [Bug tree-optimization/112542] [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d
  2023-11-15  9:42 [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d theodort at inf dot ethz.ch
@ 2023-11-15 11:59 ` rguenth at gcc dot gnu.org
  2023-11-15 16:48 ` pinskia at gcc dot gnu.org
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: rguenth at gcc dot gnu.org @ 2023-11-15 11:59 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112542

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |missed-optimization
   Target Milestone|---                         |14.0

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

* [Bug tree-optimization/112542] [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d
  2023-11-15  9:42 [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d theodort at inf dot ethz.ch
  2023-11-15 11:59 ` [Bug tree-optimization/112542] " rguenth at gcc dot gnu.org
@ 2023-11-15 16:48 ` pinskia at gcc dot gnu.org
  2024-03-07 21:02 ` law at gcc dot gnu.org
  2024-05-07  7:42 ` [Bug tree-optimization/112542] [14/15 " rguenth at gcc dot gnu.org
  3 siblings, 0 replies; 5+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-11-15 16:48 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112542

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-11-15
     Ever confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
```
  _7 = (char) c.2_5;
...
  _26 = _7 == 0;
  _27 = c.2_5 == 0;
  _28 = _26 & _27;
  if (_28 != 0)
```
we transform this into:
```
  _26 = _7 == 0;
  _27 = c.2_5 == 0;
  if (c.2_5 == 0)
```
During FRE1
Which is correct. and _26 goes away later on.

Then in evpr GCC changes:
```
  # iftmp.8_29 = PHI <_7(4), 0(5)>

```
into:
```
  # iftmp.8_29 = PHI <-1(4), 0(5)>
```

Which is also correct.

Where it goes wrong is vrp1.

We end up with this mess:
```
  b.0_1 = b;
  _2 = b.0_1 == 0;
  _4 = (int) _2;
  c = _4;
  i.7_12 = (unsigned int) _2;
  _16 = b.0_1 != 0;
  _15 = _16 ? 1 : 2;
  if (_15 == 2)
    goto <bb 3>; [0.00%]
  else
    goto <bb 4>; [100.00%]

  <bb 3> [count: 0]:
  __builtin_unreachable ();
```

But that is basically:
```
...
if (b.0_1 == 0) __builtin_unreachable();
```

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

* [Bug tree-optimization/112542] [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d
  2023-11-15  9:42 [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d theodort at inf dot ethz.ch
  2023-11-15 11:59 ` [Bug tree-optimization/112542] " rguenth at gcc dot gnu.org
  2023-11-15 16:48 ` pinskia at gcc dot gnu.org
@ 2024-03-07 21:02 ` law at gcc dot gnu.org
  2024-05-07  7:42 ` [Bug tree-optimization/112542] [14/15 " rguenth at gcc dot gnu.org
  3 siblings, 0 replies; 5+ messages in thread
From: law at gcc dot gnu.org @ 2024-03-07 21:02 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112542

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] 5+ messages in thread

* [Bug tree-optimization/112542] [14/15 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d
  2023-11-15  9:42 [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d theodort at inf dot ethz.ch
                   ` (2 preceding siblings ...)
  2024-03-07 21:02 ` law at gcc dot gnu.org
@ 2024-05-07  7:42 ` rguenth at gcc dot gnu.org
  3 siblings, 0 replies; 5+ messages in thread
From: rguenth at gcc dot gnu.org @ 2024-05-07  7:42 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112542

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|14.0                        |14.2

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 14.1 is being released, retargeting bugs to GCC 14.2.

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

end of thread, other threads:[~2024-05-07  7:42 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-11-15  9:42 [Bug tree-optimization/112542] New: [14 Regression] Dead Code Elimination Regression since r14-4280-gc3c6f30496d theodort at inf dot ethz.ch
2023-11-15 11:59 ` [Bug tree-optimization/112542] " rguenth at gcc dot gnu.org
2023-11-15 16:48 ` pinskia at gcc dot gnu.org
2024-03-07 21:02 ` law at gcc dot gnu.org
2024-05-07  7:42 ` [Bug tree-optimization/112542] [14/15 " rguenth 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).