Skip Menu |
Report information
Id: 131879
Status: new
Priority: 0/
Queue: perl6

Owner: Nobody
Requestors: alex.jakimenko [at] gmail.com
Cc:
AdminCc:

Severity: (no value)
Tag: (no value)
Platform: (no value)
Patch Status: (no value)
VM: (no value)



Subject: [REGRESSION] Memory “leak”
After recent rakudo updates whateverables started having some problems. At first I thought that they simply became slower, but upon further investigation it turned out that their memory usage was increasing during the operation (which is most likely the reason for them to slow down during the execution). MasterDuke++ did a lot of investigation and measurements, so I'm hoping to see his comments here. I don't have much to say, so I'll just give a bunch of links. Please bear with me. Initial conversation and relatively golfed down code: https://irclog.perlgeek.de/moarvm/2017-08-05#i_14973432 https://gist.github.com/MasterDuke17/685b627a6a2749483dc5ec09c6a777a4 dogbert11++ reproducing the issue https://irclog.perlgeek.de/moarvm/2017-08-06#i_14976177 (note that it seems like there's no memory leak, but the memory usage is growing) More: https://irclog.perlgeek.de/perl6/2017-08-08#i_14987204 I don't know if this affects anybody else, but jdv79++ was talking about some memory issues recently: https://irclog.perlgeek.de/perl6-dev/2017-08-05#i_14973488
There was some progress on this. jnthn++

I guess the most promising commit regarding this issue is https://github.com/MoarVM/MoarVM/commit/0d2e5f8cc0264eb495ea4f20f3a0c90d05d55710 , but there's at least one other commit that could've improved the situation ( https://github.com/rakudo/rakudo/commit/c59b986785fe17f53e1664af1aa60e785ff31624 ).

I will test this again after we have MoarVM bumped (or sooner if I have time).

On 2017-08-10 10:43:42, alex.jakimenko@gmail.com wrote:
Show quoted text
> After recent rakudo updates whateverables started having some
> problems. At first I thought that they simply became slower, but upon
> further investigation it turned out that their memory usage was
> increasing during the operation (which is most likely the reason for
> them to slow down during the execution).
>
> MasterDuke++ did a lot of investigation and measurements, so I'm
> hoping to see his comments here. I don't have much to say, so I'll
> just give a bunch of links. Please bear with me.
>
> Initial conversation and relatively golfed down code:
> https://irclog.perlgeek.de/moarvm/2017-08-05#i_14973432
> https://gist.github.com/MasterDuke17/685b627a6a2749483dc5ec09c6a777a4
>
> dogbert11++ reproducing the issue
> https://irclog.perlgeek.de/moarvm/2017-08-06#i_14976177
>
> (note that it seems like there's no memory leak, but the memory usage
> is growing)
>
> More: https://irclog.perlgeek.de/perl6/2017-08-08#i_14987204
>
>
> I don't know if this affects anybody else, but jdv79++ was talking
> about some memory issues recently: https://irclog.perlgeek.de/perl6-
> dev/2017-08-05#i_14973488


OK, this is still not resolved, but it got noticeably better.

The issue is reproducible with the snippet provided in the original report.
On 2017-08-11 10:01:42, alex.jakimenko@gmail.com wrote:
Show quoted text
> There was some progress on this. jnthn++
>
> I guess the most promising commit regarding this issue is
> https://github.com/MoarVM/MoarVM/commit/0d2e5f8cc0264eb495ea4f20f3a0c90d05d55710
> , but there's at least one other commit that could've improved the
> situation (
> https://github.com/rakudo/rakudo/commit/c59b986785fe17f53e1664af1aa60e785ff31624
> ).
>
> I will test this again after we have MoarVM bumped (or sooner if I
> have time).
>
> On 2017-08-10 10:43:42, alex.jakimenko@gmail.com wrote:
> > After recent rakudo updates whateverables started having some
> > problems. At first I thought that they simply became slower, but upon
> > further investigation it turned out that their memory usage was
> > increasing during the operation (which is most likely the reason for
> > them to slow down during the execution).
> >
> > MasterDuke++ did a lot of investigation and measurements, so I'm
> > hoping to see his comments here. I don't have much to say, so I'll
> > just give a bunch of links. Please bear with me.
> >
> > Initial conversation and relatively golfed down code:
> > https://irclog.perlgeek.de/moarvm/2017-08-05#i_14973432
> > https://gist.github.com/MasterDuke17/685b627a6a2749483dc5ec09c6a777a4
> >
> > dogbert11++ reproducing the issue
> > https://irclog.perlgeek.de/moarvm/2017-08-06#i_14976177
> >
> > (note that it seems like there's no memory leak, but the memory usage
> > is growing)
> >
> > More: https://irclog.perlgeek.de/perl6/2017-08-08#i_14987204
> >
> >
> > I don't know if this affects anybody else, but jdv79++ was talking
> > about some memory issues recently: https://irclog.perlgeek.de/perl6-
> > dev/2017-08-05#i_14973488




This service is sponsored and maintained by Best Practical Solutions and runs on Perl.org infrastructure.

For issues related to this RT instance (aka "perlbug"), please contact perlbug-admin at perl.org