public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled
@ 2020-05-26 16:44 rsandifo at gcc dot gnu.org
2020-05-26 16:55 ` [Bug target/95341] " ktkachov at gcc dot gnu.org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: rsandifo at gcc dot gnu.org @ 2020-05-26 16:44 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
Bug ID: 95341
Summary: Poor vector_size decomposition when SVE is enabled
Product: gcc
Version: 11.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: rsandifo at gcc dot gnu.org
Target Milestone: ---
Target: aarch64*-*-*
Compiling the following with -O2 -march=armv8-a:
typedef unsigned int uint_vec __attribute__((vector_size(32)));
uint_vec f1(uint_vec x, uint_vec y) { return x + y; }
generates nice Advanced SIMD code:
ld1 {v4.16b - v5.16b}, [x0]
ld1 {v2.16b - v3.16b}, [x1]
add v0.4s, v4.4s, v2.4s
add v1.4s, v5.4s, v3.4s
st1 {v0.16b - v1.16b}, [x8]
ret
But compiling with -march=armv8.2-a+sve generates extremely bad
scalar code, so bad that I'll spare people's eyes by not quoting
it here.
I haven't yet analysed this or checked whether it's a regression.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/95341] Poor vector_size decomposition when SVE is enabled
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
@ 2020-05-26 16:55 ` ktkachov at gcc dot gnu.org
2020-05-26 17:13 ` ktkachov at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2020-05-26 16:55 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
ktkachov at gcc dot gnu.org changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
CC| |ktkachov at gcc dot gnu.org
Status|UNCONFIRMED |NEW
Last reconfirmed| |2020-05-26
--- Comment #1 from ktkachov at gcc dot gnu.org ---
Confirmed.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/95341] Poor vector_size decomposition when SVE is enabled
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
2020-05-26 16:55 ` [Bug target/95341] " ktkachov at gcc dot gnu.org
@ 2020-05-26 17:13 ` ktkachov at gcc dot gnu.org
2020-05-26 17:21 ` rsandifo at gcc dot gnu.org
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2020-05-26 17:13 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
--- Comment #2 from ktkachov at gcc dot gnu.org ---
Interestingly, -msve-vector-bits gives good codegen for 128, 256, 512 but bad
for 1024, 2048 and VLA code
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/95341] Poor vector_size decomposition when SVE is enabled
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
2020-05-26 16:55 ` [Bug target/95341] " ktkachov at gcc dot gnu.org
2020-05-26 17:13 ` ktkachov at gcc dot gnu.org
@ 2020-05-26 17:21 ` rsandifo at gcc dot gnu.org
2020-05-27 7:38 ` rguenth at gcc dot gnu.org
2024-02-27 8:08 ` pinskia at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: rsandifo at gcc dot gnu.org @ 2020-05-26 17:21 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
--- Comment #3 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
(In reply to ktkachov from comment #2)
> Interestingly, -msve-vector-bits gives good codegen for 128, 256, 512 but
> bad for 1024, 2048 and VLA code
That's because addition is the one operation that we currently support
on unpacked vectors. For subtraction it's a different story (just 128-bit
and 256-bit is good, as expected).
But the surprising thing for me was that we didn't fall back
to Advanced SIMD in VLA mode...
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/95341] Poor vector_size decomposition when SVE is enabled
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
` (2 preceding siblings ...)
2020-05-26 17:21 ` rsandifo at gcc dot gnu.org
@ 2020-05-27 7:38 ` rguenth at gcc dot gnu.org
2024-02-27 8:08 ` pinskia at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: rguenth at gcc dot gnu.org @ 2020-05-27 7:38 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |missed-optimization
--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
Does vector lowering already make a mess out of it?
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/95341] Poor vector_size decomposition when SVE is enabled
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
` (3 preceding siblings ...)
2020-05-27 7:38 ` rguenth at gcc dot gnu.org
@ 2024-02-27 8:08 ` pinskia at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu.org @ 2024-02-27 8:08 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95341
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution|--- |FIXED
Target Milestone|--- |14.0
See Also| |https://gcc.gnu.org/bugzill
| |a/show_bug.cgi?id=112787
--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Fixed by r14-6752-ga3ff76278efe00 for GCC 14.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-02-27 8:08 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-26 16:44 [Bug target/95341] New: Poor vector_size decomposition when SVE is enabled rsandifo at gcc dot gnu.org
2020-05-26 16:55 ` [Bug target/95341] " ktkachov at gcc dot gnu.org
2020-05-26 17:13 ` ktkachov at gcc dot gnu.org
2020-05-26 17:21 ` rsandifo at gcc dot gnu.org
2020-05-27 7:38 ` rguenth at gcc dot gnu.org
2024-02-27 8:08 ` 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).