public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
* [Bug c++/98429] New: Some FMA expressions are evaluated less efficiently when -ffast-math is active
@ 2020-12-23 17:22 chtz at informatik dot uni-bremen.de
2021-01-05 9:45 ` [Bug middle-end/98429] " rguenth at gcc dot gnu.org
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: chtz at informatik dot uni-bremen.de @ 2020-12-23 17:22 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98429
Bug ID: 98429
Summary: Some FMA expressions are evaluated less efficiently
when -ffast-math is active
Product: gcc
Version: 10.2.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: chtz at informatik dot uni-bremen.de
Target Milestone: ---
If you compile an expression like
x - x*y
using fast-math optimizations it appears to get re-arranged to
x*(1-y)
which is (probably) never more efficient than the original expression,
especially if the target supports FMA operations.
Simple godbolt demonstration: https://godbolt.org/z/4q1oj9
(This is very likely not C++ specific, but I'm not sure what the correct
component is -- I also just tried this for AVX-FMA)
^ permalink raw reply [flat|nested] 4+ messages in thread* [Bug middle-end/98429] Some FMA expressions are evaluated less efficiently when -ffast-math is active 2020-12-23 17:22 [Bug c++/98429] New: Some FMA expressions are evaluated less efficiently when -ffast-math is active chtz at informatik dot uni-bremen.de @ 2021-01-05 9:45 ` rguenth at gcc dot gnu.org 2021-09-06 18:35 ` pinskia at gcc dot gnu.org 2021-09-06 18:37 ` pinskia at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: rguenth at gcc dot gnu.org @ 2021-01-05 9:45 UTC (permalink / raw) To: gcc-bugs https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98429 Richard Biener <rguenth at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Last reconfirmed| |2021-01-05 Ever confirmed|0 |1 Status|UNCONFIRMED |NEW Component|tree-optimization |middle-end --- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> --- This probably asks for a special case during RTL expansion or FMA forming phase. In itself this is first canonicalization (the user could have written x*(1-y) which should also expand to FMA I suppose). ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/98429] Some FMA expressions are evaluated less efficiently when -ffast-math is active 2020-12-23 17:22 [Bug c++/98429] New: Some FMA expressions are evaluated less efficiently when -ffast-math is active chtz at informatik dot uni-bremen.de 2021-01-05 9:45 ` [Bug middle-end/98429] " rguenth at gcc dot gnu.org @ 2021-09-06 18:35 ` pinskia at gcc dot gnu.org 2021-09-06 18:37 ` pinskia at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: pinskia at gcc dot gnu.org @ 2021-09-06 18:35 UTC (permalink / raw) To: gcc-bugs https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98429 Andrew Pinski <pinskia at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mkretz at gcc dot gnu.org --- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> --- *** Bug 102219 has been marked as a duplicate of this bug. *** ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/98429] Some FMA expressions are evaluated less efficiently when -ffast-math is active 2020-12-23 17:22 [Bug c++/98429] New: Some FMA expressions are evaluated less efficiently when -ffast-math is active chtz at informatik dot uni-bremen.de 2021-01-05 9:45 ` [Bug middle-end/98429] " rguenth at gcc dot gnu.org 2021-09-06 18:35 ` pinskia at gcc dot gnu.org @ 2021-09-06 18:37 ` pinskia at gcc dot gnu.org 2 siblings, 0 replies; 4+ messages in thread From: pinskia at gcc dot gnu.org @ 2021-09-06 18:37 UTC (permalink / raw) To: gcc-bugs https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98429 Andrew Pinski <pinskia at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |DUPLICATE Status|NEW |RESOLVED --- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> --- Dup of bug 56547. *** This bug has been marked as a duplicate of bug 56547 *** ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-09-06 18:37 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-12-23 17:22 [Bug c++/98429] New: Some FMA expressions are evaluated less efficiently when -ffast-math is active chtz at informatik dot uni-bremen.de 2021-01-05 9:45 ` [Bug middle-end/98429] " rguenth at gcc dot gnu.org 2021-09-06 18:35 ` pinskia at gcc dot gnu.org 2021-09-06 18:37 ` 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).