Re: [PATCH,RFC] Introduce RUN_UNDER_VALGRIND in test-suite

["Martin Liška" <mliska@suse.cz>]

On 12/03/2015 03:15 PM, Bernd Schmidt wrote:
> On 11/23/2015 10:34 AM, Martin Liška wrote:
>> On 11/21/2015 05:26 AM, Hans-Peter Nilsson wrote:
>>> IIRC you can replace the actual dg-runtest proc with your own
>>> (implementing a wrapper).  Grep aroung, I think we do that
>>> already.  That's certainly preferable instead of touching all
>>> callers.
>>
>> You are right, the suggested patch was over-kill, wrapper should be fine for that.
>> Currently I've been playing with a bit different approach (suggested by Markus),
>> where I would like to enable valgrind in gcc.c using an environmental variable.
>>
>> Question is if it should replace existing ENABLE_VALGRIND_CHECKING and how to
>> integrate it with a valgrind suppressions file?
> 
> This patch still seems to be in the queue. I've been looking at it every now and then, without really forming an opinion. In any case, I think we'll need to postpone this to stage1 at this point.
> 
> Wouldn't it be better to fix issues first and only then enable running the testsuite with valgrind, rather than make a suppression file?
> 
> Your latest patch seems to add the option of running the compiler without ENABLE_CHECKING_VALGRIND being defined. Doesn't this run into problems when the support in ggc isn't compiled in?
> 
> 
> Bernd

Hi.

Right, the patch is in queue and can wait for next stage1. I must agree with Hans-Peter Nilsson that we should
mainly focus on removal of memory leaks (and other invalid operations) rather that maintaining a list of suppressions.
After that, integration with existing configure machine should be easily doable, I guess.

I've just run the test-suite (with default languages) and report file was post-processed with my script [1] that
groups same back-traces together.

Currently we have ~200000 errors, in ~4000 different back-traces.

Majority of them (~2600 BTs) are in fortran FE (BT contains 'gfc_'): [2].
The rest contains some issues in CP FE, many GGC invalid read/write operations ([4]) and many
memory leaks in gcc.c (for instance option handling).

My question is if a bug should be created for all fortran issues and whether it's realistic that
they can be eventually fixed in next stage1?

Thanks,
Martin

[1] https://github.com/marxin/script-misc/blob/master/valgrind-grep.py
[2] https://drive.google.com/file/d/0B0pisUJ80pO1ZjdCVlZoeGZQNjg/view?usp=sharing
[3] https://drive.google.com/file/d/0B0pisUJ80pO1aFZTWk5sVTBlcHc/view?usp=sharing
[4] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68758