* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
@ 2021-08-19 18:42 ` pinskia at gcc dot gnu.org
2021-08-19 19:20 ` jakub at gcc dot gnu.org
` (4 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-08-19 18:42 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed|2018-07-29 00:00:00 |2021-8-19
--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
clang:
movabsq $72057589742960640, %rax # imm = 0xFFFFFF00000000
andq %rdi, %rax
bswapq %rax
shrq $56, %rdi
orl %edi, %eax
ICC:
bswap %rdi #11.51
movl %edi, %eax #11.51
ret
ICC looks like it is able to do it correctly.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
2021-08-19 18:42 ` [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards pinskia at gcc dot gnu.org
@ 2021-08-19 19:20 ` jakub at gcc dot gnu.org
2021-08-20 15:38 ` jakub at gcc dot gnu.org
` (3 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2021-08-19 19:20 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org
--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
It boils down to even simpler:
int
bar (unsigned long long value)
{
return ((value & 0x000000ff00000000ull) >> 8)
| ((value & 0x0000ff0000000000ull) >> 24)
| ((value & 0x00ff000000000000ull) >> 40)
| ((value & 0xff00000000000000ull) >> 56);
}
which is what you get from #c2 after optimizations.
The bswap pass tries to ATM recognize just nops - 0x0807060504030201
permutation
markers - and full bswaps - 0x0102030405060708 permutations, where in those
permutation bytes
0 - target byte has the value 0
FF - target byte has an unknown value (eg. due to sign extension)
1..size - marker value is the byte index in the source (0 for lsb).
But we could very well handle also masked bswaps, either just those one can get
from zero extensions, so 0x0000000005060708 or 0x0000000000000708 etc., or
generally with clearing of arbitrary bytes in it, say
0x0100030400060700 by doing __builtin_bswap64 (arg) & 0xff00ffff00ffff00ULL
etc.
Then this optimization would fall out from that, because we'd do
(int) (__builtin_bswap64 (arg) & 0xffffffffULL) and further opts would optimize
away the masking.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
2021-08-19 18:42 ` [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards pinskia at gcc dot gnu.org
2021-08-19 19:20 ` jakub at gcc dot gnu.org
@ 2021-08-20 15:38 ` jakub at gcc dot gnu.org
2021-08-23 9:54 ` cvs-commit at gcc dot gnu.org
` (2 subsequent siblings)
5 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2021-08-20 15:38 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|unassigned at gcc dot gnu.org |jakub at gcc dot gnu.org
Status|NEW |ASSIGNED
--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 51331
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51331&action=edit
gcc12-pr86723.patch
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
` (2 preceding siblings ...)
2021-08-20 15:38 ` jakub at gcc dot gnu.org
@ 2021-08-23 9:54 ` cvs-commit at gcc dot gnu.org
2021-08-23 9:54 ` jakub at gcc dot gnu.org
2021-08-25 2:35 ` pinskia at gcc dot gnu.org
5 siblings, 0 replies; 6+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2021-08-23 9:54 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:
https://gcc.gnu.org/g:b320edc0c29c838b0090c3c9be14187d132f73f2
commit r12-3072-gb320edc0c29c838b0090c3c9be14187d132f73f2
Author: Jakub Jelinek <jakub@redhat.com>
Date: Mon Aug 23 11:52:06 2021 +0200
bswap: Recognize (int) __builtin_bswap64 (arg) idioms or __builtin_bswap??
(arg) & mask [PR86723]
The following patch recognizes in the bswap pass (only there for now,
haven't done it for store merging pass yet) code sequences that can
be handled by (int32) __builtin_bswap64 (arg), i.e. where we have
0x05060708 n->n with 64-bit non-memory argument (if it is memory, we
can just load the 32-bit at 4 bytes into the address and n->n would
be 0x01020304; and only 64 -> 32 bit, because 64 -> 16 bit or 32 -> 16 bit
would mean only two bytes in the result and probably not worth it),
and furthermore the case where we have in the 0x0102030405060708 etc.
numbers some bytes 0 (i.e. known to contain zeros rather than source
bytes),
as long as we have at least two original bytes in the right
positions (and no unknown bytes). This can be handled by
__builtin_bswap64 (arg) & 0xff0000ffffff00ffULL etc.
The latter change is the reason why counting the bswap messages doesn't
work
too well in optimize-bswap* tests anymore, while the pass iterates from end
of basic block towards start, it will often match both the bswap at the end
and some of the earlier bswaps with some masks (not a problem generally,
we'll just DCE it away whenever possible). The pass right now doesn't
handle __builtin_bswap* calls in the pattern matching (which is the reason
why it operates backwards), but it uses FOR_EACH_BB_FN (bb, fun) order
of handling blocks and matched sequences can span multiple blocks, so I was
worried about cases like:
void bar (unsigned long long);
unsigned long long
foo (unsigned long long value, int x)
{
unsigned long long tmp = (((value & 0x00000000000000ffull) << 56)
| ((value & 0x000000000000ff00ull) << 40)
| ((value & 0x00000000ff000000ull) << 8));
if (x)
bar (tmp);
return (tmp
| ((value & 0x000000ff00000000ull) >> 8)
| ((value & 0x0000ff0000000000ull) >> 24)
| ((value & 0x0000000000ff0000ull) << 24)
| ((value & 0x00ff000000000000ull) >> 40)
| ((value & 0xff00000000000000ull) >> 56));
}
but it seems we handle even that fine, while bb2 ending in GIMPLE_COND
is processed first, we recognize there a __builtin_bswap64 (value) & mask1,
in the last bb we recognize tmp | (__builtin_bswap64 (value) & mask2) and
PRE optimizes that into t = __builtin_bswap64 (value); tmp = t & mask1;
in the first bb and return t; in the last one.
2021-08-23 Jakub Jelinek <jakub@redhat.com>
PR tree-optimization/86723
* gimple-ssa-store-merging.c (find_bswap_or_nop_finalize): Add
cast64_to_32 argument, set *cast64_to_32 to false, unless n is
non-memory permutation of 64-bit src which only has bytes of
0 or [5..8] and n->range is 4.
(find_bswap_or_nop): Add cast64_to_32 and mask arguments, adjust
find_bswap_or_nop_finalize caller, support bswap with some bytes
zeroed, as long as at least two bytes are not zeroed.
(bswap_replace): Add mask argument and handle masking of bswap
result.
(maybe_optimize_vector_constructor): Adjust find_bswap_or_nop
caller, punt if cast64_to_32 or mask is not all ones.
(pass_optimize_bswap::execute): Adjust find_bswap_or_nop_finalize
caller, for now punt if cast64_to_32.
* gcc.dg/pr86723.c: New test.
* gcc.target/i386/pr86723.c: New test.
* gcc.dg/optimize-bswapdi-1.c: Use -fdump-tree-optimized instead of
-fdump-tree-bswap and scan for number of __builtin_bswap64 calls.
* gcc.dg/optimize-bswapdi-2.c: Likewise.
* gcc.dg/optimize-bswapsi-1.c: Use -fdump-tree-optimized instead of
-fdump-tree-bswap and scan for number of __builtin_bswap32 calls.
* gcc.dg/optimize-bswapsi-5.c: Likewise.
* gcc.dg/optimize-bswapsi-3.c: Likewise. Expect one
__builtin_bswap32
call instead of zero.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
` (3 preceding siblings ...)
2021-08-23 9:54 ` cvs-commit at gcc dot gnu.org
@ 2021-08-23 9:54 ` jakub at gcc dot gnu.org
2021-08-25 2:35 ` pinskia at gcc dot gnu.org
5 siblings, 0 replies; 6+ messages in thread
From: jakub at gcc dot gnu.org @ 2021-08-23 9:54 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution|--- |FIXED
--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed for GCC 12.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug tree-optimization/86723] not optimizing with bswap, that casts to int aftwards
[not found] <bug-86723-4@http.gcc.gnu.org/bugzilla/>
` (4 preceding siblings ...)
2021-08-23 9:54 ` jakub at gcc dot gnu.org
@ 2021-08-25 2:35 ` pinskia at gcc dot gnu.org
5 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-08-25 2:35 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86723
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |12.0
^ permalink raw reply [flat|nested] 6+ messages in thread