* Invalid "dg-do run" in the testsuite (with 2 blanks)
@ 2024-03-24 20:19 Harald Anlauf
2024-03-24 23:26 ` Jerry D
` (2 more replies)
0 siblings, 3 replies; 13+ messages in thread
From: Harald Anlauf @ 2024-03-24 20:19 UTC (permalink / raw)
To: fortran
Dear all,
I noticed by chance that we have quite a lot of improperly specified do-do
directives in the testsuite.
% grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
83
Note that with two blanks instead of just one a testcase does not get executed.
Does anybody want to earn the honors to change the directives and
check for "fallout" in the testsuite?
Cheers,
Harald
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-24 20:19 Invalid "dg-do run" in the testsuite (with 2 blanks) Harald Anlauf
@ 2024-03-24 23:26 ` Jerry D
2024-03-25 4:53 ` Jerry D
2024-04-02 6:46 ` Andrew Pinski
2 siblings, 0 replies; 13+ messages in thread
From: Jerry D @ 2024-03-24 23:26 UTC (permalink / raw)
To: Harald Anlauf, fortran
On 3/24/24 1:19 PM, Harald Anlauf wrote:
> Dear all,
>
> I noticed by chance that we have quite a lot of improperly specified do-do
> directives in the testsuite.
>
> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> 83
>
> Note that with two blanks instead of just one a testcase does not get executed.
>
> Does anybody want to earn the honors to change the directives and
> check for "fallout" in the testsuite?
>
> Cheers,
> Harald
>
I will do this for the betterment of the order of all good people. I
need the therapy after beating my head against a wall last week eating
separators. :)
Cheers,
Jerry -
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-24 20:19 Invalid "dg-do run" in the testsuite (with 2 blanks) Harald Anlauf
2024-03-24 23:26 ` Jerry D
@ 2024-03-25 4:53 ` Jerry D
2024-03-25 8:27 ` Manfred Schwarb
2024-04-02 6:46 ` Andrew Pinski
2 siblings, 1 reply; 13+ messages in thread
From: Jerry D @ 2024-03-25 4:53 UTC (permalink / raw)
To: Harald Anlauf, fortran
On 3/24/24 1:19 PM, Harald Anlauf wrote:
> Dear all,
>
> I noticed by chance that we have quite a lot of improperly specified do-do
> directives in the testsuite.
>
> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> 83
>
> Note that with two blanks instead of just one a testcase does not get executed.
>
> Does anybody want to earn the honors to change the directives and
> check for "fallout" in the testsuite?
>
> Cheers,
> Harald
>
One failure after fixing all the spaces ( sed is our friend ).
FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times
optimized "_gfortran_matmul" 0
Jerry -
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 4:53 ` Jerry D
@ 2024-03-25 8:27 ` Manfred Schwarb
2024-03-25 17:38 ` Jerry D
0 siblings, 1 reply; 13+ messages in thread
From: Manfred Schwarb @ 2024-03-25 8:27 UTC (permalink / raw)
To: Jerry D, Harald Anlauf, fortran
Am 25.03.24 um 05:53 schrieb Jerry D:
> On 3/24/24 1:19 PM, Harald Anlauf wrote:
>> Dear all,
>>
>> I noticed by chance that we have quite a lot of improperly specified do-do
>> directives in the testsuite.
>>
>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
>> 83
>>
I think this is on purpose.
The idea to use a "feature" in dejagnu to only iterate once and not
over all possible options. So execution time can be lowered a bit.
But I don't know if this hack still works, it definitely did work some years ago.
Cheers,
Manfred
>> Note that with two blanks instead of just one a testcase does not get executed.
>>
>> Does anybody want to earn the honors to change the directives and
>> check for "fallout" in the testsuite?
>>
>> Cheers,
>> Harald
>>
>
> One failure after fixing all the spaces ( sed is our friend ).
>
> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
>
> Jerry -
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 8:27 ` Manfred Schwarb
@ 2024-03-25 17:38 ` Jerry D
2024-03-25 19:53 ` Harald Anlauf
0 siblings, 1 reply; 13+ messages in thread
From: Jerry D @ 2024-03-25 17:38 UTC (permalink / raw)
To: Manfred Schwarb, Harald Anlauf, fortran
On 3/25/24 1:27 AM, Manfred Schwarb wrote:
> Am 25.03.24 um 05:53 schrieb Jerry D:
>> On 3/24/24 1:19 PM, Harald Anlauf wrote:
>>> Dear all,
>>>
>>> I noticed by chance that we have quite a lot of improperly specified do-do
>>> directives in the testsuite.
>>>
>>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
>>> 83
>>>
>
> I think this is on purpose.
> The idea to use a "feature" in dejagnu to only iterate once and not
> over all possible options. So execution time can be lowered a bit.
>
> But I don't know if this hack still works, it definitely did work some years ago.
>
> Cheers,
> Manfred
>
>
>>> Note that with two blanks instead of just one a testcase does not get executed.
>>>
>>> Does anybody want to earn the honors to change the directives and
>>> check for "fallout" in the testsuite?
>>>
>>> Cheers,
>>> Harald
>>>
>>
>> One failure after fixing all the spaces ( sed is our friend ).
>>
>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
>>
>> Jerry -
>
Yes it still works, I noticed the tests will do two passes each rather
than 12 per file.
I will hold off on doing anything with these.
Regards,
Jerry -
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 17:38 ` Jerry D
@ 2024-03-25 19:53 ` Harald Anlauf
2024-03-25 20:52 ` Jerry D
0 siblings, 1 reply; 13+ messages in thread
From: Harald Anlauf @ 2024-03-25 19:53 UTC (permalink / raw)
To: Jerry D; +Cc: Manfred Schwarb, fortran
> >>> I noticed by chance that we have quite a lot of improperly specified do-do
> >>> directives in the testsuite.
> >>>
> >>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> >>> 83
> >>>
> >
> > I think this is on purpose.
> > The idea to use a "feature" in dejagnu to only iterate once and not
> > over all possible options. So execution time can be lowered a bit.
> >
> > But I don't know if this hack still works, it definitely did work some years ago.
> >
> > Cheers,
> > Manfred
Is this "feature" documented somewhere? I don't see it on
https://gcc.gnu.org/onlinedocs/gcc-13.2.0/gccint/Directives.html
Given that the dg-directives are important and possibly fragile,
and since we had issues in the past, can we check that a test
that was added works the way intended?
> >>> Note that with two blanks instead of just one a testcase does not get executed.
> >>>
> >>> Does anybody want to earn the honors to change the directives and
> >>> check for "fallout" in the testsuite?
> >>>
> >>> Cheers,
> >>> Harald
> >>>
> >>
> >> One failure after fixing all the spaces ( sed is our friend ).
> >>
> >> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
This does actually point to an issue with the testcase:
it only works properly with optimization enabled.
Manual inspection of this test and the expected dump suggests
that e.g. -O1 could have been added to the dg-options directive.
Shouldn't we fix at least the dg-options of that testcase?
Cheers,
Harald
> >> Jerry -
> >
>
> Yes it still works, I noticed the tests will do two passes each rather
> than 12 per file.
>
> I will hold off on doing anything with these.
>
> Regards,
>
> Jerry -
>
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 19:53 ` Harald Anlauf
@ 2024-03-25 20:52 ` Jerry D
2024-03-25 22:30 ` Aw: " Harald Anlauf
0 siblings, 1 reply; 13+ messages in thread
From: Jerry D @ 2024-03-25 20:52 UTC (permalink / raw)
To: Harald Anlauf; +Cc: Manfred Schwarb, fortran
On 3/25/24 12:53 PM, Harald Anlauf wrote:
>>>>> I noticed by chance that we have quite a lot of improperly specified do-do
>>>>> directives in the testsuite.
>>>>>
>>>>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
>>>>> 83
>>>>>
>>>
>>> I think this is on purpose.
>>> The idea to use a "feature" in dejagnu to only iterate once and not
>>> over all possible options. So execution time can be lowered a bit.
>>>
>>> But I don't know if this hack still works, it definitely did work some years ago.
>>>
>>> Cheers,
>>> Manfred
>
> Is this "feature" documented somewhere? I don't see it on
>
> https://gcc.gnu.org/onlinedocs/gcc-13.2.0/gccint/Directives.html
>
> Given that the dg-directives are important and possibly fragile,
> and since we had issues in the past, can we check that a test
> that was added works the way intended?
>
>>>>> Note that with two blanks instead of just one a testcase does not get executed.
>>>>>
>>>>> Does anybody want to earn the honors to change the directives and
>>>>> check for "fallout" in the testsuite?
>>>>>
>>>>> Cheers,
>>>>> Harald
>>>>>
>>>>
>>>> One failure after fixing all the spaces ( sed is our friend ).
>>>>
>>>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
>
> This does actually point to an issue with the testcase:
> it only works properly with optimization enabled.
>
> Manual inspection of this test and the expected dump suggests
> that e.g. -O1 could have been added to the dg-options directive.
>
> Shouldn't we fix at least the dg-options of that testcase?
>
> Cheers,
> Harald
>
I restored the one test that appeared to fail so that it had the two
spaces 'trick'. When run in the test suite, it is compiled with -O
which does invoke the optimization. I manually checked the tree dump
with this option and it indeed has all the _gfortran_matmul calls removed.
I am inclined to leave these all untouched with the two spaces in place.
From the test log:
PASS: gfortran.dg/inline_matmul_1.f90 -O execution test
PASS: gfortran.dg/inline_matmul_1.f90 -O scan-tree-dump-times
optimized "_gfortran_matmul" 0
Regards,
Jerry
^ permalink raw reply [flat|nested] 13+ messages in thread
* Aw: Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 20:52 ` Jerry D
@ 2024-03-25 22:30 ` Harald Anlauf
2024-03-26 1:24 ` Jerry D
2024-03-26 8:53 ` Mikael Morin
0 siblings, 2 replies; 13+ messages in thread
From: Harald Anlauf @ 2024-03-25 22:30 UTC (permalink / raw)
To: Jerry D; +Cc: Manfred Schwarb, fortran
> On 3/25/24 12:53 PM, Harald Anlauf wrote:
> >>>>> I noticed by chance that we have quite a lot of improperly specified do-do
> >>>>> directives in the testsuite.
> >>>>>
> >>>>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> >>>>> 83
> >>>>>
> >>>
> >>> I think this is on purpose.
> >>> The idea to use a "feature" in dejagnu to only iterate once and not
> >>> over all possible options. So execution time can be lowered a bit.
> >>>
> >>> But I don't know if this hack still works, it definitely did work some years ago.
> >>>
> >>> Cheers,
> >>> Manfred
> >
> > Is this "feature" documented somewhere? I don't see it on
> >
> > https://gcc.gnu.org/onlinedocs/gcc-13.2.0/gccint/Directives.html
> >
> > Given that the dg-directives are important and possibly fragile,
> > and since we had issues in the past, can we check that a test
> > that was added works the way intended?
> >
> >>>>> Note that with two blanks instead of just one a testcase does not get executed.
> >>>>>
> >>>>> Does anybody want to earn the honors to change the directives and
> >>>>> check for "fallout" in the testsuite?
> >>>>>
> >>>>> Cheers,
> >>>>> Harald
> >>>>>
> >>>>
> >>>> One failure after fixing all the spaces ( sed is our friend ).
> >>>>
> >>>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
> >
> > This does actually point to an issue with the testcase:
> > it only works properly with optimization enabled.
> >
> > Manual inspection of this test and the expected dump suggests
> > that e.g. -O1 could have been added to the dg-options directive.
> >
> > Shouldn't we fix at least the dg-options of that testcase?
> >
> > Cheers,
> > Harald
> >
>
> I restored the one test that appeared to fail so that it had the two
> spaces 'trick'. When run in the test suite, it is compiled with -O
> which does invoke the optimization. I manually checked the tree dump
> with this option and it indeed has all the _gfortran_matmul calls removed.
>
> I am inclined to leave these all untouched with the two spaces in place.
>
> From the test log:
>
> PASS: gfortran.dg/inline_matmul_1.f90 -O execution test
> PASS: gfortran.dg/inline_matmul_1.f90 -O scan-tree-dump-times
> optimized "_gfortran_matmul" 0
Alright, then leave it that way.
I find it somewhat unsatisfactory though, to have a behavior of the
testsuite harness that is so intranparent.
If it is a simple oversight that the behavior of double space was
never documented, it could simply be fixed, for the sake of everybody.
The take-home message for me is - whenever I write a testcase that
relies on this behavior - to add a comment in the header that this
is intended behavior, and set all compiler flags appropriately...
Cheers,
Harald
> Regards,
>
> Jerry
>
>
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Aw: Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 22:30 ` Aw: " Harald Anlauf
@ 2024-03-26 1:24 ` Jerry D
2024-03-26 7:49 ` Richard Biener
2024-03-26 8:53 ` Mikael Morin
1 sibling, 1 reply; 13+ messages in thread
From: Jerry D @ 2024-03-26 1:24 UTC (permalink / raw)
To: Harald Anlauf, gfortran
On 3/25/24 3:30 PM, Harald Anlauf wrote:
>> On 3/25/24 12:53 PM, Harald Anlauf wrote:
>>>>>>> I noticed by chance that we have quite a lot of improperly specified do-do
>>>>>>> directives in the testsuite.
>>>>>>>
>>>>>>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
>>>>>>> 83
>>>>>>>
>>>>>
>>>>> I think this is on purpose.
>>>>> The idea to use a "feature" in dejagnu to only iterate once and not
>>>>> over all possible options. So execution time can be lowered a bit.
>>>>>
>>>>> But I don't know if this hack still works, it definitely did work some years ago.
>>>>>
>>>>> Cheers,
>>>>> Manfred
>>>
>>> Is this "feature" documented somewhere? I don't see it on
>>>
>>> https://gcc.gnu.org/onlinedocs/gcc-13.2.0/gccint/Directives.html
>>>
>>> Given that the dg-directives are important and possibly fragile,
>>> and since we had issues in the past, can we check that a test
>>> that was added works the way intended?
>>>
>>>>>>> Note that with two blanks instead of just one a testcase does not get executed.
>>>>>>>
>>>>>>> Does anybody want to earn the honors to change the directives and
>>>>>>> check for "fallout" in the testsuite?
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Harald
>>>>>>>
>>>>>>
>>>>>> One failure after fixing all the spaces ( sed is our friend ).
>>>>>>
>>>>>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
>>>
>>> This does actually point to an issue with the testcase:
>>> it only works properly with optimization enabled.
>>>
>>> Manual inspection of this test and the expected dump suggests
>>> that e.g. -O1 could have been added to the dg-options directive.
>>>
>>> Shouldn't we fix at least the dg-options of that testcase?
>>>
>>> Cheers,
>>> Harald
>>>
>>
>> I restored the one test that appeared to fail so that it had the two
>> spaces 'trick'. When run in the test suite, it is compiled with -O
>> which does invoke the optimization. I manually checked the tree dump
>> with this option and it indeed has all the _gfortran_matmul calls removed.
>>
>> I am inclined to leave these all untouched with the two spaces in place.
>>
>> From the test log:
>>
>> PASS: gfortran.dg/inline_matmul_1.f90 -O execution test
>> PASS: gfortran.dg/inline_matmul_1.f90 -O scan-tree-dump-times
>> optimized "_gfortran_matmul" 0
>
> Alright, then leave it that way.
>
> I find it somewhat unsatisfactory though, to have a behavior of the
> testsuite harness that is so intranparent.
>
> If it is a simple oversight that the behavior of double space was
> never documented, it could simply be fixed, for the sake of everybody.
>
> The take-home message for me is - whenever I write a testcase that
> relies on this behavior - to add a comment in the header that this
> is intended behavior, and set all compiler flags appropriately...
>
> Cheers,
> Harald
>
>
With two spaces there are 9 tests run at -O, with one space there are 54
variations executed. Specifying the optimization option in the
dg-options also runs 54 variations. It appears that the "feature" was
done on purpose, but it is frustrating that this is not documented anywhere.
I do remember discussions way back about some tests taking a long time
to run and needing a way to speed up the overall testing. I will look
over the documentation and see if we can get this addressed.
For my own test cases, I have always specified the options I want and
for front-end work, optimizations don't usually matter except in this
case. I don't have time to go through all 83 tests.
From my system here:
Testing of trunk complete..... one space
real 4m56.294s
user 35m51.082s
sys 12m51.283s
Testing of trunk complete..... two spaces
real 4m44.421s
user 34m3.215s
sys 12m38.778s
Not a huge difference.
Jerry
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Aw: Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-26 1:24 ` Jerry D
@ 2024-03-26 7:49 ` Richard Biener
0 siblings, 0 replies; 13+ messages in thread
From: Richard Biener @ 2024-03-26 7:49 UTC (permalink / raw)
To: Jerry D; +Cc: Harald Anlauf, gfortran
On Tue, Mar 26, 2024 at 2:24 AM Jerry D <jvdelisle2@gmail.com> wrote:
>
> On 3/25/24 3:30 PM, Harald Anlauf wrote:
> >> On 3/25/24 12:53 PM, Harald Anlauf wrote:
> >>>>>>> I noticed by chance that we have quite a lot of improperly specified do-do
> >>>>>>> directives in the testsuite.
> >>>>>>>
> >>>>>>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> >>>>>>> 83
> >>>>>>>
> >>>>>
> >>>>> I think this is on purpose.
> >>>>> The idea to use a "feature" in dejagnu to only iterate once and not
> >>>>> over all possible options. So execution time can be lowered a bit.
> >>>>>
> >>>>> But I don't know if this hack still works, it definitely did work some years ago.
> >>>>>
> >>>>> Cheers,
> >>>>> Manfred
> >>>
> >>> Is this "feature" documented somewhere? I don't see it on
> >>>
> >>> https://gcc.gnu.org/onlinedocs/gcc-13.2.0/gccint/Directives.html
> >>>
> >>> Given that the dg-directives are important and possibly fragile,
> >>> and since we had issues in the past, can we check that a test
> >>> that was added works the way intended?
> >>>
> >>>>>>> Note that with two blanks instead of just one a testcase does not get executed.
> >>>>>>>
> >>>>>>> Does anybody want to earn the honors to change the directives and
> >>>>>>> check for "fallout" in the testsuite?
> >>>>>>>
> >>>>>>> Cheers,
> >>>>>>> Harald
> >>>>>>>
> >>>>>>
> >>>>>> One failure after fixing all the spaces ( sed is our friend ).
> >>>>>>
> >>>>>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
> >>>
> >>> This does actually point to an issue with the testcase:
> >>> it only works properly with optimization enabled.
> >>>
> >>> Manual inspection of this test and the expected dump suggests
> >>> that e.g. -O1 could have been added to the dg-options directive.
> >>>
> >>> Shouldn't we fix at least the dg-options of that testcase?
> >>>
> >>> Cheers,
> >>> Harald
> >>>
> >>
> >> I restored the one test that appeared to fail so that it had the two
> >> spaces 'trick'. When run in the test suite, it is compiled with -O
> >> which does invoke the optimization. I manually checked the tree dump
> >> with this option and it indeed has all the _gfortran_matmul calls removed.
> >>
> >> I am inclined to leave these all untouched with the two spaces in place.
> >>
> >> From the test log:
> >>
> >> PASS: gfortran.dg/inline_matmul_1.f90 -O execution test
> >> PASS: gfortran.dg/inline_matmul_1.f90 -O scan-tree-dump-times
> >> optimized "_gfortran_matmul" 0
> >
> > Alright, then leave it that way.
> >
> > I find it somewhat unsatisfactory though, to have a behavior of the
> > testsuite harness that is so intranparent.
> >
> > If it is a simple oversight that the behavior of double space was
> > never documented, it could simply be fixed, for the sake of everybody.
> >
> > The take-home message for me is - whenever I write a testcase that
> > relies on this behavior - to add a comment in the header that this
> > is intended behavior, and set all compiler flags appropriately...
> >
> > Cheers,
> > Harald
> >
> >
>
> With two spaces there are 9 tests run at -O, with one space there are 54
> variations executed. Specifying the optimization option in the
> dg-options also runs 54 variations. It appears that the "feature" was
> done on purpose, but it is frustrating that this is not documented anywhere.
It would be nice to document it in sourcebuild.texi and to put a comment before
the dg-do with two spaces indicating that two spaces are on purpose.
I also don't remember this being implemented ...
Richard.
> I do remember discussions way back about some tests taking a long time
> to run and needing a way to speed up the overall testing. I will look
> over the documentation and see if we can get this addressed.
>
> For my own test cases, I have always specified the options I want and
> for front-end work, optimizations don't usually matter except in this
> case. I don't have time to go through all 83 tests.
>
> From my system here:
>
> Testing of trunk complete..... one space
>
> real 4m56.294s
> user 35m51.082s
> sys 12m51.283s
>
>
> Testing of trunk complete..... two spaces
>
> real 4m44.421s
> user 34m3.215s
> sys 12m38.778s
>
> Not a huge difference.
>
> Jerry
>
>
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Aw: Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-25 22:30 ` Aw: " Harald Anlauf
2024-03-26 1:24 ` Jerry D
@ 2024-03-26 8:53 ` Mikael Morin
1 sibling, 0 replies; 13+ messages in thread
From: Mikael Morin @ 2024-03-26 8:53 UTC (permalink / raw)
To: Harald Anlauf, Jerry D; +Cc: Manfred Schwarb, fortran
Le 25/03/2024 à 23:30, Harald Anlauf a écrit :
>> On 3/25/24 12:53 PM, Harald Anlauf wrote:
>>>>>>>
(...)
>>>>>>
>>>>>> One failure after fixing all the spaces ( sed is our friend ).
>>>>>>
>>>>>> FAIL: gfortran.dg/inline_matmul_1.f90 -O0 scan-tree-dump-times optimized "_gfortran_matmul" 0
>>>
>>> This does actually point to an issue with the testcase:
>>> it only works properly with optimization enabled.
>>>
>>> Manual inspection of this test and the expected dump suggests
>>> that e.g. -O1 could have been added to the dg-options directive.
>>>
>>> Shouldn't we fix at least the dg-options of that testcase?
>>>
For what it's worth, the scan-tree-dump-times directive can be
conditionalized on target __OPTIMIZE__, so that it's active with
optimization and inactive with -O0. This would make the test pass
regardless of compiler flags.
Of course if the goal is lowering the test execution time, it wouldn't help.
(...)
>
> I find it somewhat unsatisfactory though, to have a behavior of the
> testsuite harness that is so intranparent.
>
Agreed.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-03-24 20:19 Invalid "dg-do run" in the testsuite (with 2 blanks) Harald Anlauf
2024-03-24 23:26 ` Jerry D
2024-03-25 4:53 ` Jerry D
@ 2024-04-02 6:46 ` Andrew Pinski
2024-04-02 17:44 ` Jerry D
2 siblings, 1 reply; 13+ messages in thread
From: Andrew Pinski @ 2024-04-02 6:46 UTC (permalink / raw)
To: Harald Anlauf; +Cc: fortran
On Sun, Mar 24, 2024 at 1:19 PM Harald Anlauf <anlauf@gmx.de> wrote:
>
> Dear all,
>
> I noticed by chance that we have quite a lot of improperly specified do-do
> directives in the testsuite.
>
> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
> 83
>
> Note that with two blanks instead of just one a testcase does not get executed.
>
> Does anybody want to earn the honors to change the directives and
> check for "fallout" in the testsuite?
Note this seems related to
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=28032 (or maybe this is
the "workaround" for not doing torture options on the testcases).
If this gets documented to avoid doing the torture options, then the
testsuite should be auditted for dg-options with `-ON` and add the
extra space for dg-do to get the non torture way.
Thanks,
Andrew
>
> Cheers,
> Harald
>
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: Invalid "dg-do run" in the testsuite (with 2 blanks)
2024-04-02 6:46 ` Andrew Pinski
@ 2024-04-02 17:44 ` Jerry D
0 siblings, 0 replies; 13+ messages in thread
From: Jerry D @ 2024-04-02 17:44 UTC (permalink / raw)
To: Andrew Pinski, Harald Anlauf; +Cc: fortran, Jerry D
On 4/1/24 11:46 PM, Andrew Pinski wrote:
> On Sun, Mar 24, 2024 at 1:19 PM Harald Anlauf <anlauf@gmx.de> wrote:
>>
>> Dear all,
>>
>> I noticed by chance that we have quite a lot of improperly specified do-do
>> directives in the testsuite.
>>
>> % grep "dg-do run" gcc/testsuite/gfortran.dg/ -rl|wc -l
>> 83
>>
>> Note that with two blanks instead of just one a testcase does not get executed.
>>
>> Does anybody want to earn the honors to change the directives and
>> check for "fallout" in the testsuite?
>
> Note this seems related to
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=28032 (or maybe this is
> the "workaround" for not doing torture options on the testcases).
>
> If this gets documented to avoid doing the torture options, then the
> testsuite should be auditted for dg-options with `-ON` and add the
> extra space for dg-do to get the non torture way.
>
> Thanks,
> Andrew
>
>>
>> Cheers,
>> Harald
>>
Thanks for this additional info. I have not settled on exactly what to say in the documentation.
I noticed that when I specified a specific -ON it still did them all regardless. I did not know if this was expected or a feature.
Jerry -
^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2024-04-02 17:44 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-03-24 20:19 Invalid "dg-do run" in the testsuite (with 2 blanks) Harald Anlauf
2024-03-24 23:26 ` Jerry D
2024-03-25 4:53 ` Jerry D
2024-03-25 8:27 ` Manfred Schwarb
2024-03-25 17:38 ` Jerry D
2024-03-25 19:53 ` Harald Anlauf
2024-03-25 20:52 ` Jerry D
2024-03-25 22:30 ` Aw: " Harald Anlauf
2024-03-26 1:24 ` Jerry D
2024-03-26 7:49 ` Richard Biener
2024-03-26 8:53 ` Mikael Morin
2024-04-02 6:46 ` Andrew Pinski
2024-04-02 17:44 ` Jerry D
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).