https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122103
--- Comment #6 from Andrew Pinski ---
Created attachment 62484
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62484&action=edit
Here is a modified testcase which can show it in GCC 14.3.0 also
The only difference between this one and the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122117
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Severity|normal
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122103
--- Comment #5 from Andrew Pinski ---
(In reply to Tamar Christina from comment #2)
> Note that __builtin_sqrt is correctly marked as trapping. So this doesn't
> vectorize.
>
> void f (float *c, int n)
> {
> for (int i = 0; i < n; i++)
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122118
Andrew Pinski changed:
What|Removed |Added
Keywords||missed-optimization
Target|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113682
Andrew Pinski changed:
What|Removed |Added
CC||rockeet at gmail dot com
--- Comment #1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122116
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122116
Andrew Pinski changed:
What|Removed |Added
Severity|normal |enhancement
Keywords|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122116
--- Comment #2 from Andrew Pinski ---
That is because gcc does not produce 2 cmov as you have
If (a)
B =
Else
C =
That is unrelated to __builtin_expect_with_probability really.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122116
Andrew Pinski changed:
What|Removed |Added
Component|c++ |rtl-optimization
--- Comment #1 from An
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122068
--- Comment #4 from Andrew Pinski ---
(In reply to Richard Biener from comment #3)
> Is that the same as hardbool?
No. Hardbool traps when the value is neither true or false value.
>From the patch
(https://patchwork.sourceware.org/project/gcc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122110
Andrew Pinski changed:
What|Removed |Added
Component|middle-end |tree-optimization
Severity|no
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122110
Andrew Pinski changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53947
Bug 53947 depends on bug 122110, which changed state.
Bug 122110 Summary: bool prevents vectorization of simple loops
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122110
What|Removed |Added
-
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101639
Andrew Pinski changed:
What|Removed |Added
CC||manu at gcc dot gnu.org
--- Comment #5
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122103
Andrew Pinski changed:
What|Removed |Added
CC||pinskia at gcc dot gnu.org
--- Comment
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #15 from Andrew Pinski ---
Oh yes and https://github.com/sg16-unicode/sg16/issues/67 .
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122096
--- Comment #2 from Andrew Pinski ---
>I am not sure if it is a dup for PR122058
I doubt it. (without more debug there is a change GCC is doing the right thing
there and the code in ICU is undefined).
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #19 from Andrew Pinski ---
(In reply to Andrew Pinski from comment #18)
> (In reply to Sam James from comment #17)
> > OK, tried -fno-strict-aliasing, didn't help. -fno-ipa-modref does not help
> > either. Guess I need to look to see
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122103
--- Comment #1 from Andrew Pinski ---
vect__98.143_326 = .SQRT (vect_prephitmp_93.130_310);
mask__15.131_311 = vect_prephitmp_93.130_310 < {
1.2081668171172168513294309377670288085938e-2, ... };
Those are the only 2 which
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122102
Andrew Pinski changed:
What|Removed |Added
Summary|[16 Regression] [RISCV] |[16 Regression] [RISCV]
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122101
Andrew Pinski changed:
What|Removed |Added
Severity|normal |enhancement
--- Comment #1 from Andrew
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #18 from Andrew Pinski ---
(In reply to Sam James from comment #17)
> OK, tried -fno-strict-aliasing, didn't help. -fno-ipa-modref does not help
> either. Guess I need to look to see which pass ends up doing more as a
> result of the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122100
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #13 from Andrew Pinski ---
(In reply to Andrew Pinski from comment #11)
>
> Without -fno-semantic-interposition, the functions that cause the undefined
> behavior is not inlinined so it does not show up there.
I should also mention
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122096
--- Comment #4 from Andrew Pinski ---
(In reply to Haochen Jiang from comment #3)
>
> > I am not sure what more should be done here since -O3/-Ofast is
> > defined/documtned as it might increase the code size without any gain over
> > -O2.
>
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122102
--- Comment #3 from Andrew Pinski ---
(In reply to Andrew Pinski from comment #2)
> >First bad commit: 850c1ad7e3cd240e97e41fa3f1e6b38c2700106b
>
>
> This can't be right, as that only touched the Ada front-end.
Even worse that commit only tou
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80151
--- Comment #7 from Andrew Pinski ---
So what about:
if (vec.size())
Yes it could/should be if(!vec.empty())
But still.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114450
Andrew Pinski changed:
What|Removed |Added
CC||olaf.krzikalla at dlr dot de
--- Commen
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122102
Andrew Pinski changed:
What|Removed |Added
Summary|[RISCV] [Miscompile] GCC - |[16 Regression] [RISCV]
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122096
Andrew Pinski changed:
What|Removed |Added
Target Milestone|--- |16.0
Keywords|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #14 from Andrew Pinski ---
The whole template IcuCApiHelper looks very much questionable.
It converts between a "C" struct to a "C++" type and I suspect there might be
some aliasing issues around that but I am not 100% sure that is t
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #11 from Andrew Pinski ---
Does adding -fno-strict-aliasing fix the issue? I am wondering if we
devirtualizing some extra calls now that we didn't before and start to inline
some other of those and there is some violation of strict a
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #12 from Andrew Pinski ---
>10 works and >=11 fails. I haven't bisected between 10 and 11 yet.
So modref was introduced in GCC 11, at r11-3308-gd119f34c952f87 and this has
the feeling of an aliasing violation.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122095
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122095
--- Comment #1 from Andrew Pinski ---
Created attachment 62468
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62468&action=edit
Full testcase
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122087
Andrew Pinski changed:
What|Removed |Added
Resolution|--- |DUPLICATE
Status|UNCONFIRME
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122088
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Target Milestone|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=43374
Andrew Pinski changed:
What|Removed |Added
CC||qingren2hxb at gmail dot com
--- Comment
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122071
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122070
--- Comment #2 from Andrew Pinski ---
Created attachment 62463
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62463&action=edit
testcase from godbolt
Please next time attach the testcase, bugzilla has a way to paste it into a
text field i
|ASSIGNED
Target Milestone|--- |16.0
Last reconfirmed||2025-09-28
Assignee|unassigned at gcc dot gnu.org |pinskia at gcc dot
gnu.org
: missed-optimization, testsuite-fail
Severity: normal
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: pinskia at gcc dot gnu.org
Target Milestone: ---
Target: riscv
As mentioned in
https
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117576
Andrew Pinski changed:
What|Removed |Added
CC||wilco at gcc dot gnu.org
--- Comment #2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=40770
Andrew Pinski changed:
What|Removed |Added
CC||wilco at gcc dot gnu.org
--- Comment #27
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=40770
Andrew Pinski changed:
What|Removed |Added
CC||blaffablaffa at gmail dot com
--- Commen
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70901
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101295
Andrew Pinski changed:
What|Removed |Added
CC||blubban at gmail dot com
--- Comment #2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122084
Andrew Pinski changed:
What|Removed |Added
Resolution|--- |DUPLICATE
Status|UNCONFIRME
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122083
Andrew Pinski changed:
What|Removed |Added
Keywords||missed-optimization
--- Comment #4 from
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122083
Andrew Pinski changed:
What|Removed |Added
Target Milestone|--- |15.3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122074
Andrew Pinski changed:
What|Removed |Added
Severity|normal |enhancement
Status|WAITING
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122081
Andrew Pinski changed:
What|Removed |Added
Target Milestone|--- |16.0
Keywords|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121999
--- Comment #7 from Andrew Pinski ---
(In reply to Filip Kastl from comment #6)
> (In reply to Richard Biener from comment #5)
> > hoping for a non-FDO testcase from fuzzers/rebuilders ;)
>
> Alright. I'll be on the lookout for that :)
PR 12
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122079
Andrew Pinski changed:
What|Removed |Added
Summary|ICE on valid code at|[16 regression] ICE on
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122073
Andrew Pinski changed:
What|Removed |Added
See Also||https://gcc.gnu.org/bugzill
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122073
Andrew Pinski changed:
What|Removed |Added
Keywords||diagnostic
--- Comment #1 from Andrew P
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122074
--- Comment #2 from Andrew Pinski ---
Icx at -O0 has a similar issue
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122074
Andrew Pinski changed:
What|Removed |Added
Last reconfirmed||2025-09-26
Status|UNCONFIRM
|1
Assignee|unassigned at gcc dot gnu.org |pinskia at gcc dot
gnu.org
Status|UNCONFIRMED |ASSIGNED
--- Comment #2 from Andrew Pinski ---
So maybe arguments needs to be done at expand but the rest can be done via an
early pass.
Let
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122068
--- Comment #1 from Andrew Pinski ---
The only problem with doing at expand phase is we might be loading via byte for
bools, e.g vectorization.
So doing at the front-end in gimplification or an early pass will get away from
that issue.
Priority: P3
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: pinskia at gcc dot gnu.org
Target Milestone: ---
https://discourse.llvm.org/t/defining-what-happens-when-a-bool-isn-t-0-or-1/86778/34?u=pinskia
This is nice security option
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122064
--- Comment #3 from Andrew Pinski ---
This was broken on the 16th but fixed on the 22nd.
Always try a latest git sources when reporting a bug, it might already been
fixed like this one.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122021
Andrew Pinski changed:
What|Removed |Added
CC||xxs_chy at outlook dot com
--- Comment
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122064
Andrew Pinski changed:
What|Removed |Added
Resolution|--- |DUPLICATE
Status|WAITING
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122061
--- Comment #2 from Andrew Pinski ---
Expensive in compile time that is.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122061
Andrew Pinski changed:
What|Removed |Added
Keywords||diagnostic, false-negative
Se
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122060
Andrew Pinski changed:
What|Removed |Added
Status|WAITING |UNCONFIRMED
Ever confirmed|1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122060
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #10 from Andrew Pinski ---
> Other advantage of the patch is that the multi-target devirtualizatoin gets
tested without profile feedback so we more likely notice problems with it.
So Jan must have known that multi-target devirtuali
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #5 from Andrew Pinski ---
Can you modify the inplace operator new and add a asm("":"+m"(r)); for the
return value?
I am trying to figure out if there is a lifetime change that got missed.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122058
--- Comment #3 from Andrew Pinski ---
Does fno-devirtualize-speculatively fix it?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122057
Andrew Pinski changed:
What|Removed |Added
See Also||https://gcc.gnu.org/bugzill
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122054
--- Comment #2 from Andrew Pinski ---
Created attachment 62445
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62445&action=edit
Preprocessed source
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122054
Andrew Pinski changed:
What|Removed |Added
Known to fail||15.1.0, 15.2.0
Known to work|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122052
Andrew Pinski changed:
What|Removed |Added
Component|middle-end |fortran
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122050
--- Comment #3 from Andrew Pinski ---
(In reply to Richard Biener from comment #2)
> What does ({1;({});}) do?
void type for both C and C++ front-ends.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122050
--- Comment #1 from Andrew Pinski ---
Note the C front-end behavior didn't change between GCC 4.0 and 4.1 so the
behavior was there with the older bison compiler.
: UNCONFIRMED
Keywords: stmt-expr, wrong-code
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: pinskia at gcc dot gnu.org
Target Milestone: ---
Take:
```
int foo() {
return ({1; 2; 3;;});
}
```
The C
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122049
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102439
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Keywords|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108456
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122048
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122035
--- Comment #16 from Andrew Pinski ---
As I mentioned at -O2, the undefined behavior is removed as we can figure out
strlen(&a) is always 0.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122035
--- Comment #14 from Andrew Pinski ---
(In reply to Krzysztof Pecka from comment #13)
> Can you give me the command that prints you this? I am not getting anything,
> after adding `-fsanitize=undefined -fsanitize=address`
This was at runtime.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122043
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122045
Andrew Pinski changed:
What|Removed |Added
Last reconfirmed||2025-09-23
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122042
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120341
--- Comment #9 from Andrew Pinski ---
*** Bug 122042 has been marked as a duplicate of this bug. ***
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122040
--- Comment #2 from Andrew Pinski ---
(In reply to Xi Ruoyao from comment #1)
> I'm unsure if there are more undocumented code patterns causing
> -fisolate-erroneous-paths-dereference to produce ud2.
No that is it currently.
stmt_uses_name_in_
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122035
--- Comment #12 from Andrew Pinski ---
(In reply to Krzysztof Pecka from comment #11)
> Interesting. I honestly forgot about null character.
> Still for me, interesting to see those different behaviors with different
> optimization flags.
Well
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122024
Andrew Pinski changed:
What|Removed |Added
Resolution|--- |DUPLICATE
Status|UNCONFIRME
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122021
Andrew Pinski changed:
What|Removed |Added
CC||zsojka at seznam dot cz
--- Comment #9
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122039
Andrew Pinski changed:
What|Removed |Added
Ever confirmed|0 |1
Status|UNCONFIRMED
: UNCONFIRMED
Keywords: missed-optimization
Severity: normal
Priority: P3
Component: tree-optimization
Assignee: pinskia at gcc dot gnu.org
Reporter: pinskia at gcc dot gnu.org
Target Milestone: ---
The code does:
```
for (i = gsi_start_bb
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122036
Andrew Pinski changed:
What|Removed |Added
See Also||https://gcc.gnu.org/bugzill
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111875
--- Comment #7 from Andrew Pinski ---
(In reply to Richard Biener from comment #3)
> git blame might tell -
It was done for PR 56977. __builtin_constant_p folding in fab.
> there was the alternative to
> perform this propagation in fold_built
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122013
Andrew Pinski changed:
What|Removed |Added
Blocks|101758 |
Keywords|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119416
Andrew Pinski changed:
What|Removed |Added
Status|ASSIGNED|NEW
--- Comment #4 from Andrew Pinski
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101758
Andrew Pinski changed:
What|Removed |Added
Status|ASSIGNED|NEW
Assignee|pinskia at gcc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111875
--- Comment #6 from Andrew Pinski ---
Created attachment 62439
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62439&action=edit
patch which moves the folding of assume_aligned to gimple fold from fab
I should have looked for this bug earl
1 - 100 of 12120 matches
Mail list logo
