https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80975
--- Comment #3 from Thomas Koenig ---
Created attachment 41469
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41469&action=edit
Attempt at straightforward patch
This one looks rather straightforward. I'll be testing this.
at gcc dot gnu.org |tkoenig at gcc dot
gnu.org
Target Milestone|--- |7.2
--- Comment #4 from Thomas Koenig ---
I'll do this.
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
$ cat foo.f90
program main
integer :: i
do i=1,1
print *,i
end do
end program main
$ gfortran -o os.s -S -Os
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80945
--- Comment #6 from Thomas Koenig ---
If this patch is fixed, please remember to remove the extra
check in frontend-passes.c (traverse_io_block). Just grep
for 80945.
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
See
https://gcc.gnu.org/ml/fortran/2017-06/msg00014.html
and
https://gcc.gnu.org/ml/gcc-testresults/2017-06/msg00530.html
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80988
Thomas Koenig changed:
What|Removed |Added
Status|UNCONFIRMED |WAITING
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80975
--- Comment #5 from Thomas Koenig ---
Author: tkoenig
Date: Tue Jun 6 19:18:37 2017
New Revision: 248932
URL: https://gcc.gnu.org/viewcvs?rev=248932&root=gcc&view=rev
Log:
2017-06-06 Thomas Koenig
PR fortran/80975
* m4/matmu
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80988
--- Comment #3 from Thomas Koenig ---
(In reply to Jerry DeLisle from comment #2)
> Try compiling with -fno-frontend-optimize to confirm.
... and if this fixes the problem (assuming it is in foo.f90),
do the following:
$ gfortran -c -fno-fronte
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80975
--- Comment #6 from Thomas Koenig ---
Author: tkoenig
Date: Tue Jun 6 22:23:07 2017
New Revision: 248935
URL: https://gcc.gnu.org/viewcvs?rev=248935&root=gcc&view=rev
Log:
2017-06-06 Thomas Koenig
Backport from trunk
PR fort
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80975
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80988
--- Comment #5 from Thomas Koenig ---
Simple test case:
$ cat swim.f90
program main
implicit none
integer :: i,j
integer, dimension(3,3) :: a
a = reshape([(((i*10+j),i=1,3),j=1,3)], shape(a))
print *,(a(i,i),i=1,3)
end program main
$
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #6 from Thomas Koenig ---
Created attachment 41508
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41508&action=edit
What an unrolled cshift could look like
This is what an unrolled version of cshift could look like,
for a simpl
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80988
--- Comment #6 from Thomas Koenig ---
Author: tkoenig
Date: Sat Jun 10 10:07:13 2017
New Revision: 249092
URL: https://gcc.gnu.org/viewcvs?rev=249092&root=gcc&view=rev
Log:
2017-06-10 Thomas Koenig
PR fortran/80988
* frontend
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80988
Thomas Koenig changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #8 from Thomas Koenig ---
(In reply to Jerry DeLisle from comment #7)
> (In reply to Thomas Koenig from comment #6)
> > Created attachment 41508 [details]
> > What an unrolled cshift could look like
> >
> > This is what an unrolled v
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #10 from Thomas Koenig ---
Also see the discussion at
https://groups.google.com/forum/#!topic/comp.lang.fortran/AI0F1Vpkc3I
There is one thing that I do not understand. For the following
test code, which compares straightforward DO
Component: fortran
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
program main
real, dimension(:), allocatable :: r
real, dimension(3,3) :: a
call random_number(a)
r = reshape(eoshift(a,1),shape(r))
print *,size(r)
print
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81062
Thomas Koenig changed:
What|Removed |Added
Status|WAITING |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #11 from Thomas Koenig ---
Created attachment 41541
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41541&action=edit
Patch for the library, not yet quite correct
This is a patch which brings a dramatic speedup for
any cshift wi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #12 from Thomas Koenig ---
Created attachment 41542
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41542&action=edit
Test case which still fails
||tkoenig at gcc dot gnu.org
--- Comment #1 from Thomas Koenig ---
Still present on current trunk.
||tkoenig at gcc dot gnu.org
Resolution|--- |FIXED
--- Comment #5 from Thomas Koenig ---
Fixed with the same patch as PR 65542.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=34871
Thomas Koenig changed:
What|Removed |Added
Last reconfirmed|2008-11-18 19:35:34 |2017-6-15
--- Comment #4 from Thomas Koe
||tkoenig at gcc dot gnu.org
--- Comment #5 from Thomas Koenig ---
We should probably do a clobber assignment for two cases:
1. Inside the procedure. A declaration of intent(out) means the
value is undefined inside the procedure.
2. From the caller. Calling a procedure with an
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44646
Thomas Koenig changed:
What|Removed |Added
Status|NEW |WAITING
--- Comment #7 from Thomas Koeni
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #13 from Thomas Koenig ---
Author: tkoenig
Date: Sun Jun 18 18:04:19 2017
New Revision: 249350
URL: https://gcc.gnu.org/viewcvs?rev=249350&root=gcc&view=rev
Log:
2017-06-18 Thomas Koenig
PR fortran/52473
* m4/cshi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78366
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|NEW
Assignee|tkoenig at gcc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #14 from Thomas Koenig ---
Author: tkoenig
Date: Sat Jun 24 07:07:56 2017
New Revision: 249620
URL: https://gcc.gnu.org/viewcvs?rev=249620&root=gcc&view=rev
Log:
2017-06-24 Thomas Koenig
PR fortran/52473
* Makefil
,
||tkoenig at gcc dot gnu.org
--- Comment #2 from Thomas Koenig ---
I was sort of waiting for the new SPEC suite to expose bugs :-)
The patch looks obvious enough, even in the absence
of a test case.
Could you run a regression test and then submit the patch
to
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81195
--- Comment #8 from Thomas Koenig ---
(In reply to Jim Wilson from comment #7)
> I have reproduced the same problem from the gcc-7 branch.
I fully expect this bug to be in all active branches
of gfortran. We should commit a fix to all of them.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|NEW
Assignee|tkoenig at gcc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=39627
Bug 39627 depends on bug 44646, which changed state.
Bug 44646 Summary: [F08] Implement DO CONCURRENT
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44646
What|Removed |Added
--
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44646
Thomas Koenig changed:
What|Removed |Added
Status|WAITING |RESOLVED
Resolution|---
at gcc dot gnu.org |tkoenig at gcc dot
gnu.org
--- Comment #8 from Thomas Koenig ---
Looks doable.
Severity: normal
Priority: P3
Component: bootstrap
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
Created attachment 41669
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41669&action=edit
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
--- Comment #1 from Thomas Koenig ---
Created attachment 41670
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41670&action=edit
Output of make -j4
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
Thomas Koenig changed:
What|Removed |Added
Target Milestone|--- |7.0
Summary|Bootstrap fails d
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
--- Comment #3 from Thomas Koenig ---
Could be a duplicate of PR 81298.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
Thomas Koenig changed:
What|Removed |Added
Depends on||81295
--- Comment #4 from Thomas Koenig
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81298
Thomas Koenig changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80365
Thomas Koenig changed:
What|Removed |Added
CC||tkoenig at gcc dot gnu.org
--- Comment
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
--- Comment #18 from Thomas Koenig ---
Created attachment 40119
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40119&action=edit
Version that works (AVX only)
Here is a version that should only do AVX stuff on Intel processors.
Optimizatio
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
Thomas Koenig changed:
What|Removed |Added
Attachment #40119|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
--- Comment #22 from Thomas Koenig ---
Author: tkoenig
Date: Sat Dec 3 09:44:35 2016
New Revision: 243219
URL: https://gcc.gnu.org/viewcvs?rev=243219&root=gcc&view=rev
Log:
2016-12-03 Thomas Koenig
PR fortran/78379
* config/
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78226
--- Comment #6 from Thomas Koenig ---
Current regressions with the patch at
https://gcc.gnu.org/ml/fortran/2016-12/msg00039.html :
allocate_with_source_10.f08 allocate_with_source_11.f08
allocate_with_source_13.f03 allocate_with_source_14.f03
a
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78226
--- Comment #8 from Thomas Koenig ---
It appears that the patch at
https://gcc.gnu.org/ml/fortran/2016-12/msg00040.html
actually catches more errors than the original one.
So, more cases to find... Nice thing is, one
fix can fix quite a large
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78226
--- Comment #12 from Thomas Koenig ---
Hi Andre,
the patch at
https://gcc.gnu.org/ml/fortran/2016-12/msg00048.html
is OK (or, at your choice, obvious and simple).
I don't have access to my e-mail at the moment, so I cannot
reply to the list d
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78226
--- Comment #15 from Thomas Koenig ---
Author: tkoenig
Date: Sat Dec 10 22:28:32 2016
New Revision: 243520
URL: https://gcc.gnu.org/viewcvs?rev=243520&root=gcc&view=rev
Log:
2016-12-10 Thomas Koenig
PR fortran/78226
* error.c
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78226
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66189
Thomas Koenig changed:
What|Removed |Added
Status|NEW |WAITING
--- Comment #4 from Thomas Koeni
||tkoenig at gcc dot gnu.org
--- Comment #10 from Thomas Koenig ---
Is there anything left to fix?
A dump with recent trunk shows
integer(kind=4) D.3468;
D.3468 = (integer(kind=4)) status2;
_gfortran_signal_sub_int (&i1, &i2, &D.3468);
||tkoenig at gcc dot gnu.org
Summary|formal argument cannot be |formal argument cannot be
|same as procedure name |same as procedure name in
|(r178939) |ENTRY
--- Comment #2 from Thomas Koenig ---
Still
||tkoenig at gcc dot gnu.org
--- Comment #2 from Thomas Koenig ---
Still present.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53957
Thomas Koenig changed:
What|Removed |Added
Last reconfirmed|2012-07-18 00:00:00 |2016-12-11
--- Comment #16 from Thomas K
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55814
Thomas Koenig changed:
What|Removed |Added
Last reconfirmed|2012-12-26 00:00:00 |2016-12-11
--- Comment #4 from Thomas Ko
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
Thomas Koenig changed:
What|Removed |Added
CC||tkoenig at gcc dot gnu.org
--- Comment
at gcc dot gnu.org |tkoenig at gcc dot
gnu.org
--- Comment #5 from Thomas Koenig ---
Trivial patch:
Index: decl.c
===
--- decl.c (Revision 243516)
+++ decl.c (Arbeitskopie)
@@ -922,7 +922,8 @@ char_len_param_value
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53957
--- Comment #18 from Thomas Koenig ---
Under Preferences/Email Preferences, you can select "Disable All Mail",
which should work and keep you from getting unwanted mail.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
--- Comment #6 from Thomas Koenig ---
Author: tkoenig
Date: Sat Dec 17 11:13:16 2016
New Revision: 243773
URL: https://gcc.gnu.org/viewcvs?rev=243773&root=gcc&view=rev
Log:
2016-12-17 Thomas Koenig
PR fortran/78239
* decl.c(c
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
--- Comment #9 from Thomas Koenig ---
Author: tkoenig
Date: Sat Dec 17 17:03:49 2016
New Revision: 243776
URL: https://gcc.gnu.org/viewcvs?rev=243776&root=gcc&view=rev
Log:
2016-12-17 Thomas Koenig
PR fortran/78239
* decl.c (
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
--- Comment #10 from Thomas Koenig ---
It helps to actually commit the fix, not only the test
case and the ChangeLog entry :-)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
--- Comment #11 from Thomas Koenig ---
Author: tkoenig
Date: Thu Dec 22 17:05:13 2016
New Revision: 243891
URL: https://gcc.gnu.org/viewcvs?rev=243891&root=gcc&view=rev
Log:
2016-12-22 Thomas Koenig
Backport from trunk
PR for
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
--- Comment #12 from Thomas Koenig ---
Author: tkoenig
Date: Thu Dec 22 20:27:52 2016
New Revision: 243895
URL: https://gcc.gnu.org/viewcvs?rev=243895&root=gcc&view=rev
Log:
2016-12-22 Thomas Koenig
Backport from trunk
PR for
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78239
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66189
Thomas Koenig changed:
What|Removed |Added
Status|WAITING |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=37131
Bug 37131 depends on bug 66189, which changed state.
Bug 66189 Summary: Block loops for inline matmul
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66189
What|Removed |Added
--
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51119
Bug 51119 depends on bug 66189, which changed state.
Bug 66189 Summary: Block loops for inline matmul
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66189
What|Removed |Added
--
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52473
--- Comment #4 from Thomas Koenig ---
Looking at the code, inlining cshift for a constant shift
could already be a good idea.
So, change
b = cshift(b,1) + cshift(b,-1)
to
a = b
b(1) = a(2) + a(n)
b(2:n-1) = a(1:n-2) + a(3:n)
b(n) = a(1) + a(
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78972
Thomas Koenig changed:
What|Removed |Added
Target Milestone|--- |7.0
--- Comment #6 from Thomas Koenig -
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57042
Thomas Koenig changed:
What|Removed |Added
Summary|ICE/Segfault with |Strange typespec with
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79127
Thomas Koenig changed:
What|Removed |Added
CC||tkoenig at gcc dot gnu.org
Target
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
Consider the following code. The function "scalar" contains two formulas in a
function which are identical, except for the coefficients which
dif
: enhancement
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
There is a special trigonometric identity that comes up
when solving cubic equations with three real roots
(see
||tkoenig at gcc dot gnu.org
Severity|normal |enhancement
--- Comment #3 from Thomas Koenig ---
This is not mandated by the standard, but I can see that
it could be useful.
Confirming as enhancement request.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66089
Thomas Koenig changed:
What|Removed |Added
Status|ASSIGNED|WAITING
--- Comment #21 from Thomas Koen
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65542
Thomas Koenig changed:
What|Removed |Added
CC||tkoenig at gcc dot gnu.org
--- Comment
at gcc dot gnu.org |tkoenig at gcc dot
gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65542
--- Comment #6 from Thomas Koenig ---
Author: tkoenig
Date: Sun Feb 12 16:10:22 2017
New Revision: 245376
URL: https://gcc.gnu.org/viewcvs?rev=245376&root=gcc&view=rev
Log:
2017-02-12 Thomas Koenig
PR fortran/65542
* intrinsi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79151
--- Comment #2 from Thomas Koenig ---
Another test case.
It might even be profitable just to look for divisions, because these
are so expensive that packing/unpacking should always be
profitable.
double foo(double a, double b)
{
return 1/a +
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79151
--- Comment #5 from Thomas Koenig ---
(In reply to Richard Biener from comment #3)
> The question is of course whether vector division has comparable latency /
> throughput as the scalar one.
Here's a test case on a rather old CPU, a Core 2 Q820
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79151
--- Comment #6 from Thomas Koenig ---
A few more test cases with a relatively recent trunk.
POWER7:
[tkoenig@gcc1-power7 ~]$ gcc -mcpu=power7 -O3 foo.c && time ./a.out
41.987257
real0m3.688s
user0m3.685s
sys 0m0.002s
[tkoenig@gcc1-
: target
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
For the following code
typedef double v4do __attribute__((vector_size (32)));
typedef long int v4i __attribute__((vector_size (32)));
#define VSET(vect,val) do { vect[0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79709
Thomas Koenig changed:
What|Removed |Added
Target||x86_64-pc-linux-gnu
Severity|n
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79612
--- Comment #3 from Thomas Koenig ---
I cannot think of this happening with normal code. An
internal error might be better, but internal_error does not
take printf-style arguments.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=51119
--- Comment #49 from Thomas Koenig ---
Author: tkoenig
Date: Sun Feb 26 13:22:43 2017
New Revision: 245745
URL: https://gcc.gnu.org/viewcvs?rev=245745&root=gcc&view=rev
Log:
2017-02-26 Thomas Koenig
PR fortran/51119
* options
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
The following test case
$ cat recip.f90
module try
implicit none
contains
complex function r1(a)
complex, value :: a
r1 = 1./a;
end function r1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
Thomas Koenig changed:
What|Removed |Added
Summary|floating point result |floating point result
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
--- Comment #4 from Thomas Koenig ---
(In reply to Dominique d'Humieres from comment #3)
> Where is computed 1./a? AFAICT the roundoff errors difference with
> optimization is restricted to this computation.
Yep, you're right... seems that const
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
Thomas Koenig changed:
What|Removed |Added
Keywords||wrong-code
Status|WAITING
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
Thomas Koenig changed:
What|Removed |Added
Summary|[6/7 Regression] complex|[5/6/7 Regression] complex
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
--- Comment #8 from Thomas Koenig ---
Test case for double complex:
$ cat re-d.c
#include
#include
#include
char input[] = "1.2e20 -3.2";
int main()
{
double complex c1, c2, r1, r2;
double re, im;
c1 = 1.2e20 - 3.2*I;
sscanf(input,"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79720
--- Comment #12 from Thomas Koenig ---
(In reply to Jakub Jelinek from comment #11)
> Yeah, exactly. But then this is really NOTABUG. The compile time
> evaluation needs to be .5ulp precise, not emulate whatever precision issues
> the library
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: tkoenig at gcc dot gnu.org
Target Milestone: ---
Created attachment 40835
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40835&action=edit
Output of gcc -Ofast -mavx2 -S -o bar-gcc.s bar.c
The test
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79722
--- Comment #1 from Thomas Koenig ---
Created attachment 40836
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=40836&action=edit
Output of icc -O3 -march=core-avx2 -S -o bar-intel.s bar.c
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79722
Thomas Koenig changed:
What|Removed |Added
Keywords||missed-optimization
Target|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79722
Thomas Koenig changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
--- Comment #24 from Thomas Koenig ---
Could be a good idea to add a version with -mfma to the flags for AVX2.
I'll see what I can do. It might be too late for gcc 7, and I also
don't have an AVX2 machine to test on.
Might also be a good idea t
||2017-02-27
CC||tkoenig at gcc dot gnu.org
Ever confirmed|0 |1
--- Comment #4 from Thomas Koenig ---
Same thing happened to me.
Is it possible to get the patch committed?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
--- Comment #25 from Thomas Koenig ---
Author: tkoenig
Date: Thu Mar 2 11:04:01 2017
New Revision: 245836
URL: https://gcc.gnu.org/viewcvs?rev=245836&root=gcc&view=rev
Log:
2017-03-02 Thomas Koenig
PR fortran/78379
* m4/matm
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78379
--- Comment #27 from Thomas Koenig ---
(In reply to # David Edelsohn from comment #26)
> What is AVX-specific, as opposed to SIMD vector size-specific, about this
> feature? It seems that this should be enabled for all SIMD architectures of
> the
2401 - 2500 of 3755 matches
Mail list logo
