public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/40927] Code fails to vectorize in g++ but does vectorize in gcc.
[not found] <bug-40927-4@http.gcc.gnu.org/bugzilla/>
@ 2011-11-05 23:45 ` paolo.carlini at oracle dot com
0 siblings, 0 replies; 2+ messages in thread
From: paolo.carlini at oracle dot com @ 2011-11-05 23:45 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40927
Paolo Carlini <paolo.carlini at oracle dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
CC|gcc-bugs at gcc dot gnu.org |
Known to work| |4.6.2, 4.7.0
Resolution| |FIXED
--- Comment #2 from Paolo Carlini <paolo.carlini at oracle dot com> 2011-11-05 23:44:51 UTC ---
Apparently fixed a lot of time ago, in 4.5.x. Otherwise, please re-open,
thanks.
^ permalink raw reply [flat|nested] 2+ messages in thread
* [Bug c++/40927] Code fails to vectorize in g++ but does vectorize in gcc.
2009-08-01 0:26 [Bug c++/40927] New: " public at dgmo dot org
@ 2009-08-01 14:58 ` rguenth at gcc dot gnu dot org
0 siblings, 0 replies; 2+ messages in thread
From: rguenth at gcc dot gnu dot org @ 2009-08-01 14:58 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from rguenth at gcc dot gnu dot org 2009-08-01 14:58 -------
This is because of the differences in the IL C vs. C++.
C has
r0 = (short unsigned int) *pcr - (short unsigned int) *ppr;
r1 = (short unsigned int) *pcg - (short unsigned int) *ppg;
r2 = (short unsigned int) *pcb - (short unsigned int) *ppb;
r0 = (r0 >> 8 | 65280) & r0;
r1 = (r1 >> 8 | 65280) & r1;
r2 = (r2 >> 8 | 65280) & r2;
short unsigned int r = (r0 + r1) + r2;
*n = (*n + r) - (*n >> 5);
*o = (unsigned char) r;
while C++ does (EH stuff stripped):
r0 = (short unsigned int) *(unsigned char *) pcr - (short unsigned int)
*(unsigned char *) ppr;
r1 = (short unsigned int) *(unsigned char *) pcg - (short unsigned int)
*(unsigned char *) ppg;
r2 = (short unsigned int) *(unsigned char *) pcb - (short unsigned int)
*(unsigned char *) ppb;
r0 = (short unsigned int) (((short int) ((int) r0 >> 8) | -256) & (short int)
r0);
r1 = (short unsigned int) (((short int) ((int) r1 >> 8) | -256) & (short int)
r1);
r2 = (short unsigned int) (((short int) ((int) r2 >> 8) | -256) & (short int)
r2);
r = (r0 + r1) + r2;
*(short unsigned int *) n = (*(short unsigned int *) n + r) - (short unsigned
int) ((int) *(short unsigned int *) n >> 5);
*(unsigned char *) o = (unsigned char) r;
which looks unfolded compared to what C ends up with (the C FE shortens
the BIT_IOR_EXPR to unsigned short, something fold should do instead).
--
rguenth at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Ever Confirmed|0 |1
Last reconfirmed|0000-00-00 00:00:00 |2009-08-01 14:58:34
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40927
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2011-11-05 23:45 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <bug-40927-4@http.gcc.gnu.org/bugzilla/>
2011-11-05 23:45 ` [Bug c++/40927] Code fails to vectorize in g++ but does vectorize in gcc paolo.carlini at oracle dot com
2009-08-01 0:26 [Bug c++/40927] New: " public at dgmo dot org
2009-08-01 14:58 ` [Bug c++/40927] " rguenth at gcc dot gnu dot 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).