[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339
--- Comment #16 from GCC Commits ---
The master branch has been updated by Jakub Jelinek :
https://gcc.gnu.org/g:ab2da8fb67b1aa0557a16b62689a888730dba610
commit r14-9494-gab2da8fb67b1aa0557a16b62689a888730dba610
Author: Jakub Jelinek
Date: Fri Mar 15 10:46:47 2024 +0100
i386: Fix a pasto in ix86_expand_int_sse_cmp [PR114339]
In r13-3803-gfa271afb58 I've added an optimization for LE/LEU/GE/GEU
comparison against CONST_VECTOR. As the comments say:
/* x <= cst can be handled as x < cst + 1 unless there is
wrap around in cst + 1. */
...
/* For LE punt if some element is signed maximum. */
...
/* For LEU punt if some element is unsigned maximum. */
and
/* x >= cst can be handled as x > cst - 1 unless there is
wrap around in cst - 1. */
...
/* For GE punt if some element is signed minimum. */
...
/* For GEU punt if some element is zero. */
Apparently I wrote the GE/GEU (second case) first and then
copied/adjusted it for LE/LEU, most of the adjustments look correct, but
I've left if (code == GE) comparison when testing if it should punt for
signed maximum. That condition is never true, because this is in
switch (code) { ... case LE: case LEU: block and we really meant to
be what the comment says, for LE punt if some element is signed maximum,
as then cst + 1 wraps around.
The following patch fixes the pasto.
2024-03-15 Jakub Jelinek
PR target/114339
* config/i386/i386-expand.cc (ix86_expand_int_sse_cmp) :
Fix
a pasto, compare code against LE rather than GE.
* gcc.target/i386/pr114339.c: New test.
[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339 --- Comment #15 from Jakub Jelinek --- Created attachment 57707 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=57707=edit gcc14-pr114339.patch Untested fix.
[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339 Jakub Jelinek changed: What|Removed |Added Assignee|unassigned at gcc dot gnu.org |jakub at gcc dot gnu.org --- Comment #14 from Jakub Jelinek --- Indeed, r13-3803-gfa271afb584230, so mine.
[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339
--- Comment #13 from Jakub Jelinek ---
Nice, further cleaned up:
/* PR target/114339 */
/* { dg-do run } */
/* { dg-options "-O2 -Wno-psabi" } */
/* { dg-additional-options "-mavx" { target avx_runtime } } */
typedef long V __attribute__((vector_size (16)));
__attribute__((noipa)) V
foo (V a)
{
return a <= (V) {0, __LONG_MAX__ };
}
int
main ()
{
V t = foo ((V) { 0, 0 });
if (t[0] != -1L || t[1] != -1L)
__builtin_abort ();
}
[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339 --- Comment #12 from Andrew Pinski --- I suspecting r13-3803-gfa271afb584230 which missed the border case of INT_MAX/INT_MIN .
[Bug target/114339] [13/14 regression] Tor miscompiled with -O2 -mavx -fno-vect-cost-model since r14-6822
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114339 Andrew Pinski changed: What|Removed |Added Known to work||10.1.0, 12.1.0, 12.3.0, ||7.1.0 Summary|[14 regression] Tor |[13/14 regression] Tor |miscompiled with -O2 -mavx |miscompiled with -O2 -mavx |-fno-vect-cost-model since |-fno-vect-cost-model since |r14-6822|r14-6822 Known to fail||13.1.0 Target Milestone|14.0|13.3 --- Comment #11 from Andrew Pinski --- My reduced testcase started to fail in GCC 13.
