public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/26656] Optimization flaw on conditional set of a bit.
[not found] <bug-26656-4@http.gcc.gnu.org/bugzilla/>
@ 2011-12-24 3:12 ` pinskia at gcc dot gnu.org
2021-07-26 19:18 ` [Bug target/26656] " pinskia at gcc dot gnu.org
1 sibling, 0 replies; 2+ messages in thread
From: pinskia at gcc dot gnu.org @ 2011-12-24 3:12 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26656
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |missed-optimization
Status|UNCONFIRMED |NEW
Last reconfirmed| |2011-12-24
Component|target |tree-optimization
Ever Confirmed|0 |1
--- Comment #8 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-12-24 03:08:17 UTC ---
limit time ?: time ?: wa time if time if wa
18 40000 40000 350000 40000
28 30000 40000 440000 40000
38 30000 40000 460000 40000
48 40000 40000 400000 30000
58 30000 40000 460000 30000
68 40000 30000 540000 40000
78 30000 30000 530000 40000
88 30000 40000 530000 40000
98 40000 30000 530000 40000
108 30000 40000 580000 40000
118 40000 30000 550000 40000
128 30000 30000 520000 40000
138 30000 40000 520000 30000
148 30000 40000 460000 40000
158 30000 30000 470000 30000
168 30000 30000 450000 40000
178 30000 30000 430000 40000
188 30000 30000 380000 30000
198 30000 30000 400000 30000
208 40000 30000 370000 30000
218 30000 40000 260000 30000
228 30000 30000 290000 40000
238 30000 40000 210000 40000
248 30000 40000 210000 40000
This is at -O3. So only flagIf is slower.
Without the vectorizer:
limit time ?: time ?: wa time if time if wa
18 180000 120000 420000 110000
28 170000 110000 450000 110000
38 130000 120000 460000 120000
48 160000 110000 480000 110000
58 210000 110000 530000 120000
68 150000 120000 500000 120000
78 180000 110000 540000 120000
88 170000 120000 550000 110000
98 180000 120000 560000 120000
108 180000 120000 580000 110000
118 190000 120000 580000 110000
128 160000 120000 550000 120000
138 160000 120000 540000 110000
148 160000 120000 540000 110000
158 160000 110000 520000 110000
168 190000 110000 490000 120000
178 170000 120000 450000 120000
188 150000 110000 430000 110000
198 150000 120000 390000 110000
208 130000 120000 370000 110000
218 150000 120000 330000 120000
228 140000 120000 300000 120000
238 180000 110000 280000 110000
248 150000 110000 260000 110000
^ permalink raw reply [flat|nested] 2+ messages in thread
* [Bug target/26656] Optimization flaw on conditional set of a bit.
[not found] <bug-26656-4@http.gcc.gnu.org/bugzilla/>
2011-12-24 3:12 ` [Bug tree-optimization/26656] Optimization flaw on conditional set of a bit pinskia at gcc dot gnu.org
@ 2021-07-26 19:18 ` pinskia at gcc dot gnu.org
1 sibling, 0 replies; 2+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-07-26 19:18 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26656
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Component|tree-optimization |target
Target| |x86_64
--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
flagQuestion:
.L2:
cmpb %bpl, (%r12,%rax)
setb %cl
addl %ecx, %ecx
orb %cl, (%rbx,%rax)
addq $1, %rax
cmpq $100000000, %rax
jne .L2
flagQuestionWorkaround:
.L7:
cmpb %bpl, (%r12,%rax)
setb %cl
addl %ecx, %ecx
orb %cl, (%rbx,%rax)
addq $1, %rax
cmpq $100000000, %rax
jne .L7
flagIf:
.L12:
cmpb %bl, (%r12,%rax)
jnb .L11
orb $2, 0(%rbp,%rax)
.L11:
addq $1, %rax
cmpq $100000000, %rax
jne .L12
flagIfWorkaround:
.L16:
cmpb %bpl, (%r12,%rax)
setb %cl
addl %ecx, %ecx
orb %cl, (%rbx,%rax)
addq $1, %rax
cmpq $100000000, %rax
jne .L16
There is not much be done with flagIf really since it is a conditional
load/store.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-07-26 19:18 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <bug-26656-4@http.gcc.gnu.org/bugzilla/>
2011-12-24 3:12 ` [Bug tree-optimization/26656] Optimization flaw on conditional set of a bit pinskia at gcc dot gnu.org
2021-07-26 19:18 ` [Bug target/26656] " 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).