public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug middle-end/95208] New: missed switch optimization as bit test
@ 2020-05-19 13:42 nathan at gcc dot gnu.org
2020-05-20 11:02 ` [Bug middle-end/95208] " marxin at gcc dot gnu.org
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: nathan at gcc dot gnu.org @ 2020-05-19 13:42 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95208
Bug ID: 95208
Summary: missed switch optimization as bit test
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: nathan at gcc dot gnu.org
Target Milestone: ---
Created attachment 48565
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48565&action=edit
The if case is optimized better
This comes from libcpp/lex.c's raw string lexer. We're testing whether a
character falls into a particular set of values. The switch is emitted as the
usual dispatch table. The if case, after range checking, turns into
'if ((1ul << (c - BASE)) & MAGIC_VALUE)' Which is somewhat better. Why
doesn't the switch form do that?
_Z3bazc:
.LFB1:
.cfi_startproc
leal -97(%rdi), %eax
cmpb $29, %al
jbe .L7
subl $33, %edi
cmpb $62, %dil
ja .L6
movabsq $8646911282403868279, %rax
btq %rdi, %rax
jc .L7
.L6:
ret
.p2align 4,,10
.p2align 3
.L7:
jmp _Z3barv
.cfi_endproc
^ permalink raw reply [flat|nested] 4+ messages in thread* [Bug middle-end/95208] missed switch optimization as bit test
2020-05-19 13:42 [Bug middle-end/95208] New: missed switch optimization as bit test nathan at gcc dot gnu.org
@ 2020-05-20 11:02 ` marxin at gcc dot gnu.org
2021-07-21 0:14 ` pinskia at gcc dot gnu.org
2021-12-07 8:15 ` marxin at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-05-20 11:02 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95208
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed| |2020-05-20
Status|UNCONFIRMED |NEW
Ever confirmed|0 |1
--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
Looking at the dump files, we select a jump table:
switch (c_2(D)) <default: <L88> [67.00%], case 33 ... 35: <L0> [33.00%], case
37 ... 39: <L0> [33.00%], case 42 ... 63: <L0> [33.00%], case 65 ... 91: <L0>
[33.00%], case 93 ... 95: <L0> [33.00%], case 97 ... 126: <L0> [33.00%]>
;; GIMPLE switch case clusters: JT(values:88 comparisons:12 range:94 density:
12.77%):33-126
We prefer a smaller number of clusters (in this case) one because we don't have
to build a decision tree on top of them. In this case one can't handle all in
of bit test as the range of values is >= 64.
^ permalink raw reply [flat|nested] 4+ messages in thread* [Bug middle-end/95208] missed switch optimization as bit test
2020-05-19 13:42 [Bug middle-end/95208] New: missed switch optimization as bit test nathan at gcc dot gnu.org
2020-05-20 11:02 ` [Bug middle-end/95208] " marxin at gcc dot gnu.org
@ 2021-07-21 0:14 ` pinskia at gcc dot gnu.org
2021-12-07 8:15 ` marxin at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-07-21 0:14 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95208
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Severity|normal |enhancement
--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
This looks fixed in GCC 11.
^ permalink raw reply [flat|nested] 4+ messages in thread* [Bug middle-end/95208] missed switch optimization as bit test
2020-05-19 13:42 [Bug middle-end/95208] New: missed switch optimization as bit test nathan at gcc dot gnu.org
2020-05-20 11:02 ` [Bug middle-end/95208] " marxin at gcc dot gnu.org
2021-07-21 0:14 ` pinskia at gcc dot gnu.org
@ 2021-12-07 8:15 ` marxin at gcc dot gnu.org
2 siblings, 0 replies; 4+ messages in thread
From: marxin at gcc dot gnu.org @ 2021-12-07 8:15 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95208
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |FIXED
Status|NEW |RESOLVED
--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
Fixed since r11-5049-g5e303cdee1ff01e4.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-12-07 8:15 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-19 13:42 [Bug middle-end/95208] New: missed switch optimization as bit test nathan at gcc dot gnu.org
2020-05-20 11:02 ` [Bug middle-end/95208] " marxin at gcc dot gnu.org
2021-07-21 0:14 ` pinskia at gcc dot gnu.org
2021-12-07 8:15 ` marxin 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).