public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug middle-end/55653] New: Unnecessary initialization of vector register
@ 2012-12-11 19:34 josh.m.conner at gmail dot com
2012-12-11 20:06 ` [Bug middle-end/55653] " pinskia at gcc dot gnu.org
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: josh.m.conner at gmail dot com @ 2012-12-11 19:34 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55653
Bug #: 55653
Summary: Unnecessary initialization of vector register
Classification: Unclassified
Product: gcc
Version: 4.8.0
Status: UNCONFIRMED
Severity: enhancement
Priority: P3
Component: middle-end
AssignedTo: unassigned@gcc.gnu.org
ReportedBy: josh.m.conner@gmail.com
When initializing all lanes of a vector register, I notice that the register is
first initialized to zero and then all lanes of the vector are independently
initialized, resulting in extra code.
Specifically, I'm looking at the aarch64 target, with the following source:
void
fmla_loop (double * restrict result, double * restrict mul1,
double mul2, int size)
{
int i;
for (i = 0; i < size; i++)
result[i] = result[i] + mul1[i] * mul2;
}
Compiled with:
aarch64-linux-gnu-gcc -std=c99 -O3 -ftree-vectorize -S -o test.s test.c
The resultant code to initialize a vector register with two instances of mul2
is:
adr x3, .LC0
ld1 {v3.2d}, [x3]
ins v3.d[0], v0.d[0]
ins v3.d[1], v0.d[0]
...
.LC0:
.word 0
.word 0
.word 0
.word 0
Where the first two instructions (that initialize the vector register) are
unnecessary, as is the space for .LC0.
Note that this initialization is being performed here in store_constructor:
/* Inform later passes that the old value is dead. */
if (!cleared && !vector && REG_P (target))
emit_move_insn (target, CONST0_RTX (GET_MODE (target)));
right after another check to see if the vector needs to be cleared out (and
determine that it doesn't).
Instead of the emit_move_insn, that code used to be:
emit_insn (gen_rtx_CLOBBER (VOIDmode, target));
But was changed in r101169, with the comment:
"The expr.c change elides an extra move that's creeped in since we
changed clobbered values to get new registers in reload."
(see full checkin text here:
http://gcc.gnu.org/ml/gcc-patches/2005-06/msg01584.html)
It's not clear to me whether this can be changed back, or if later passes
should be recognizing this initialization as redundant, or whether we need a
new expand pattern to match vector fill (vector duplicate). At any rate, the
code is certainly not ideal as it stands.
Thanks!
^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/55653] Unnecessary initialization of vector register
2012-12-11 19:34 [Bug middle-end/55653] New: Unnecessary initialization of vector register josh.m.conner at gmail dot com
@ 2012-12-11 20:06 ` pinskia at gcc dot gnu.org
2013-10-07 9:18 ` rearnsha at gcc dot gnu.org
2013-10-07 9:31 ` rearnsha at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: pinskia at gcc dot gnu.org @ 2012-12-11 20:06 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55653
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |missed-optimization
Target| |aarch64-*-*
CC| |pinskia at gcc dot gnu.org
--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2012-12-11 20:06:00 UTC ---
I think there are two different issues here. The first issue is a target issue
as {0.0, 0.0} should be done as just an xor which really resolves this issue as
it is able to optimize that out (though it is a good thing to do even without
the other part).
The other issue shouldn't !vector be false here as we have a vector?
^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/55653] Unnecessary initialization of vector register
2012-12-11 19:34 [Bug middle-end/55653] New: Unnecessary initialization of vector register josh.m.conner at gmail dot com
2012-12-11 20:06 ` [Bug middle-end/55653] " pinskia at gcc dot gnu.org
@ 2013-10-07 9:18 ` rearnsha at gcc dot gnu.org
2013-10-07 9:31 ` rearnsha at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: rearnsha at gcc dot gnu.org @ 2013-10-07 9:18 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55653
Richard Earnshaw <rearnsha at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution|--- |FIXED
Target Milestone|--- |4.8.0
--- Comment #2 from Richard Earnshaw <rearnsha at gcc dot gnu.org> ---
Fixed with:
2013-01-08 Tejas Belagod <tejas.belagod@arm.com>
* config/aarch64/aarch64-simd.md (vec_init<mode>): New.
* config/aarch64/aarch64-protos.h (aarch64_expand_vector_init):
Declare.
* config/aarch64/aarch64.c (aarch64_simd_dup_constant,
aarch64_simd_make_constant, aarch64_expand_vector_init): New.
^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug middle-end/55653] Unnecessary initialization of vector register
2012-12-11 19:34 [Bug middle-end/55653] New: Unnecessary initialization of vector register josh.m.conner at gmail dot com
2012-12-11 20:06 ` [Bug middle-end/55653] " pinskia at gcc dot gnu.org
2013-10-07 9:18 ` rearnsha at gcc dot gnu.org
@ 2013-10-07 9:31 ` rearnsha at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: rearnsha at gcc dot gnu.org @ 2013-10-07 9:31 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55653
--- Comment #3 from Richard Earnshaw <rearnsha at gcc dot gnu.org> ---
On the secondary issue of initializing FP vectors to zero, we now generate for
typedef double f __attribute__((vector_size(16)));
f g()
{
f a = {0.0, 0.0};
return a;
}
g:
movi v0.2d, 0
ret
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2013-10-07 9:31 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-12-11 19:34 [Bug middle-end/55653] New: Unnecessary initialization of vector register josh.m.conner at gmail dot com
2012-12-11 20:06 ` [Bug middle-end/55653] " pinskia at gcc dot gnu.org
2013-10-07 9:18 ` rearnsha at gcc dot gnu.org
2013-10-07 9:31 ` rearnsha 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).