Severity: normal
Priority: P3
Component: inline-asm
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 45265
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=45265&action=edit
Failing pre
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88559
--- Comment #3 from James Almer ---
This code has compiled and the resulting assembly worked without issues for
like a dozen major GCC releases, and now it's suddenly invalid?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88559
--- Comment #4 from James Almer ---
To expand, as i mentioned and showed in the OP this code has compiled without
issues up to some point between December 10 and December 17 with gcc trunk. I'm
not arguing about clobbering the stack pointer being
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 37636
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=37636&action=edit
Test case
[jamrial@ArchVM ~]$ gcc -v
Using built-in specs.
COLLECT_
: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 38332
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=38332&action=edit
Preprocessed output
The attached preprocessed output (Created from the source using GCC 5, 6 an
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70768
--- Comment #2 from James Almer ---
(In reply to Markus Trippelsdorf from comment #1)
> Well, you simply need more RAM to compile this testcase, because gcc-6 uses
> slightly over 2GB peak, so your machine starts swapping.
> On my machine (with 8
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 41942
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41942&action=edit
Preprocessed output created with -freport-bug
[jamrial@ArchVM
: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 42105
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42105&action=edit
Preprocessed output created with -freport-bug
[jamrial@ArchVM ~]$ c++ -std=c++17 -c ccBqWFGo.i
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82080
James Almer changed:
What|Removed |Added
Attachment #42105|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82080
--- Comment #4 from James Almer ---
The regression seems to have started with revision r251433.
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 41379
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41379&action=edit
Preprocessed output creat
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80812
--- Comment #1 from James Almer ---
This started with r248153
: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
$ cat suggest.c
enum {
TYPE_A,
}
int fn(void)
{
int b = TYPE_B;
int c = TYPE_C
++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 39820
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=39820&action=edit
Preprocessed output as created by -freport-bug
/home/jamrial/range-v3/
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60902
James Almer changed:
What|Removed |Added
CC||jamrial at gmail dot com
--- Comment #32
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
https://raw.githubusercontent.com/foo86/dcadec/e884d9a0d9127528bdc47c1f8db392363fb50876/libdcadec/idct_float.c
gcc -O3 -mavx512f -mavx512vl -c -o libdcadec
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Created attachment 35328
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35328&action=edit
Assembly output
IRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
unsigned foo(void);
int main(void)
{
if (__builtin_ia32_bzhi_si(foo(), foo()))
return 1;
return 0;
}
Compiled with -mbmi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65871
--- Comment #1 from James Almer ---
The same apparently happens with bextr, blsi, blsr, and most (if not all) of
AMD's tbm instructions. They set the ZF flag but gcc still generates a test
instruction.
http://www.felixcloutier.com/x86/BEXTR.html
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65871
--- Comment #4 from James Almer ---
(In reply to Uroš Bizjak from comment #3)
> Please see the patch, attached in Comment #2.
>
> While I can see the use (and benefit) to model the patterns that also set CC
> register internally for BEXTR and BZ
: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 35428
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35428&action=edit
Preprocessed source as generated by -freport-bug
In file included from
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65871
--- Comment #7 from James Almer ---
Thanks for the above fix.
I forgot to test BMI1's andn. That one should have an insn modeled this way as
well.
int foo (unsigned int x, unsigned int y)
{
if (~x & y)
return 1;
return 0;
}
gc
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 35494
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35494&acti
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66216
--- Comment #2 from James Almer ---
(In reply to Jonathan Wakely from comment #1)
> And now it gives an ICE on trunk, so it's regressed from rejects-valid to
> ice-on-valid-code:
>
> a.cc:1:7: internal compiler error: canonical types differ for
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
James Almer changed:
What|Removed |Added
Known to work|5.1.0 |
Summary|[6 Regression] tree ch
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
--- Comment #2 from James Almer ---
Created attachment 35594
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35594&action=edit
Preprocessed source as generated by -freport-bug, from the second test case
++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 35608
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35608&action=edit
Preprocessed source as generated by -freport-bug
GCC 6.0.0 svn 223625
++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 35674
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35674&action=edit
Preprocessed source as generated by -freport-bug
GCC 6.0.0 svn 223906
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 36432
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=36432&action=edit
Preprocessed source as generated by -freport-bug
In file includ
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Some change to trunk made in the last 18 days is making gcc miscompile the
files libavcodec/jpeg2000dec.c libavcodec/j2kenc.c and libavcodec/avuidec.c
from ffmpeg git head when -O3 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67920
--- Comment #1 from James Almer ---
For reference you can also check
http://fate.ffmpeg.org/report.cgi?time=20151010052205&slot=x86_64-archlinux-gcc-experimental
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: sanitizer
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
CC: dodji at gcc dot gnu.org, dvyukov at gcc dot gnu.org,
jakub at gcc d
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67920
--- Comment #3 from James Almer ---
(In reply to Andrew Pinski from comment #2)
> Can you also compile with -fsanitize=undefined and try that? Can you also
> try -fsanitize=address ?
>
> This might detect if it is a bug in the code vs a bug in
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67920
James Almer changed:
What|Removed |Added
Component|rtl-optimization|tree-optimization
--- Comment #5 from Jame
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67920
--- Comment #6 from James Almer ---
Created attachment 36491
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=36491&action=edit
-save-temps output for all three files
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67920
--- Comment #9 from James Almer ---
(In reply to Yuri Rumyantsev from comment #8)
> Please check that revision 228760 will cure your issue.
Looks like it did. Thanks.
: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
The code generated is a simple AND instruction that zeroes the high bits based
on the index value starting from the highest
IRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
#include
int fn (uint64_t a) {
return __builtin_popcountll(a);
}
gcc -O2 -mpopcnt
xor
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68780
--- Comment #1 from James Almer ---
What i assume you want is _mm256_mullo_epi32(a, b), which maps to the vpmulld
instruction (Multiply the packed 32-bit integers in a and b, producing
intermediate 64-bit integers, and store the low 32 bits of th
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 36967
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=36967&action=edit
Preprocessed source as generated by -freport-bug
/home/jamri
: inline-asm
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Regression since GCC 4.9.0. -O0, -O1, -O2 and -O3 are all affected. -Os works
as intended.
float clipf(float a, float amin, float amax)
{
float ret;
__asm__
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 37282
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=37282&action=edit
Preprocessed source as generated by -
-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 37377
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=37377&action=edit
Assembly and preprocessed output of miscompiled file
http://fate.ffm
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69203
James Almer changed:
What|Removed |Added
Attachment #37282|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
--- Comment #5 from James Almer ---
Created attachment 35683
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35683&action=edit
Preprocessed source as generated by -freport-bug, third test case, gcc 5.1.1
svn 223417
How about this one? Crash
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
--- Comment #7 from James Almer ---
(In reply to Markus Trippelsdorf from comment #6)
> (In reply to James Almer from comment #5)
> > Created attachment 35683 [details]
> > Preprocessed source as generated by -freport-bug, third test case, gcc 5.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
--- Comment #8 from James Almer ---
(In reply to James Almer from comment #7)
> (In reply to Markus Trippelsdorf from comment #6)
> > (In reply to James Almer from comment #5)
> > > Created attachment 35683 [details]
> > > Preprocessed source as
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66067
--- Comment #10 from James Almer ---
(In reply to Markus Trippelsdorf from comment #9)
> I've tested latest gcc-5 branch and still cannot reproduce the issue.
> The only ICE I get building range-v3 is PR66405.
Odd. I'll see about reporting the g
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
[jamrial@archVM ~]$ cat zip.c
template struct A;
template struct enable_if;
template struct B;
template using and_c = A
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66426
--- Comment #1 from James Almer ---
FWIW, testcase from Comment 2 of pr66405 unedited:
[jamrial@archVM ~]$ cat pr66405_c2.cpp
template struct A;
template struct enable_if;
template struct B;
template using and_c = A>;
template using Constru
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65871
--- Comment #9 from James Almer ---
Created attachment 35804
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35804&action=edit
Preprocessed code where a test instruction is still generated
Please look at the attachment.
[jamrial@archVM ~]$
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59767
James Almer changed:
What|Removed |Added
CC||jamrial at gmail dot com
--- Comment #1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66872
James Almer changed:
What|Removed |Added
CC||jamrial at gmail dot com
--- Comment #1
++
Assignee: unassigned at gcc dot gnu.org
Reporter: jamrial at gmail dot com
Target Milestone: ---
Created attachment 39010
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=39010&action=edit
Preprocessed output as created by -freport-bug
/home/jamrial/range-v3/includ
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66426
James Almer changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
55 matches
Mail list logo
