public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/101752] New: vectorizer failed due to comilattion tme alias of same pointer
@ 2021-08-03 12:55 crazylht at gmail dot com
  2021-08-03 13:08 ` [Bug tree-optimization/101752] " rguenth at gcc dot gnu.org
  0 siblings, 1 reply; 2+ messages in thread
From: crazylht at gmail dot com @ 2021-08-03 12:55 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101752

            Bug ID: 101752
           Summary: vectorizer failed due to comilattion tme alias of same
                    pointer
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: crazylht at gmail dot com
                CC: rguenth at gcc dot gnu.org
  Target Milestone: ---
              Host: x86_64-pc-linux-gnu

double a[1024], b[1024],c[1024],d[1024],e[1024];
void
foo ()
{
    for (int i = 0; i != 1024; i++)
      if (b[i] < c[i])
        a[i] = __builtin_fma (d[i], e[i], a[i]);
}

gcc -O3 -march=skylake-avx512 -S

/export/users2/liuhongt/tools-build/build_intel-innersource_condop/../../gcc/intel-innersource/condop/gcc/testsuite/gcc.target/i386/cond_op_fma_double-1.c:46:1:
note:   can tell at compile time that a[i_17] and *_33 alias
/export/users2/liuhongt/tools-build/build_intel-innersource_condop/../../gcc/intel-innersource/condop/gcc/testsuite/gcc.target/i386/cond_op_fma_double-1.c:46:1:
missed:   not vectorized: compilation time alias: _3 = a[i_17];
.MASK_STORE (_33, 64B, _32, _7);

a[i_17] is exactly the same as *_33.

# i_17 = PHI <i_13(7), 0(15)>
# ivtmp_16 = PHI <ivtmp_15(7), 800(15)>
_1 = b[i_17];
_2 = c[i_17];
_3 = a[i_17];
_4 = -_3;
_5 = e[i_17];
_6 = d[i_17];
_7 = __builtin_fma (_6, _5, _4);
_32 = _1 < _2;
_33 = &a[i_17];
.MASK_STORE (_33, 64B, _32, _7);
i_13 = i_17 + 1;
ivtmp_15 = ivtmp_16 - 1;
if (ivtmp_15 != 0)
  goto <bb 7>; [98.99%]
else
  goto <bb 6>; [1.01%]

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug tree-optimization/101752] vectorizer failed due to comilattion tme alias of same pointer
  2021-08-03 12:55 [Bug tree-optimization/101752] New: vectorizer failed due to comilattion tme alias of same pointer crazylht at gmail dot com
@ 2021-08-03 13:08 ` rguenth at gcc dot gnu.org
  0 siblings, 0 replies; 2+ messages in thread
From: rguenth at gcc dot gnu.org @ 2021-08-03 13:08 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101752

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |DUPLICATE
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Dup.

*** This bug has been marked as a duplicate of bug 101548 ***

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-08-03 13:08 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-03 12:55 [Bug tree-optimization/101752] New: vectorizer failed due to comilattion tme alias of same pointer crazylht at gmail dot com
2021-08-03 13:08 ` [Bug tree-optimization/101752] " 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).