[gmx-developers] Wallcycle bug?

Mark Abraham mark.j.abraham at gmail.com
Tue Sep 15 17:27:03 CEST 2015


Hi,

Jenkins mostly runs non-pinned mdrun, and under heavy load we probably see
more thread migration, and thus such de-sync events. I've only seen the
issue on slaves hosted under VM, so I think it's not clear that this is an
x86, rather than VM+x86 property.

Load is heavy lately because we're running multiple master-branch configs
while getting the new releng machinery stable. We envisaged a few days
that's now been over a week, and barring this unforeseen issue, I think
we're more or less happy with things now. Teemu is there value in dealing
with more of the releng patches in gerrit before we disable the old
master-branch configs?

Mark

On Tue, Sep 15, 2015 at 5:17 PM Berk Hess <hess at kth.se> wrote:

> Hi,
>
> We didn't realize that modern x86 cpus no longer synchronize clocks
> between cores. We should get the fix for this merged asap:
> https://gerrit.gromacs.org/#/c/5091/
>
> Cheers,
>
> Berk
>
> On 09/15/2015 05:06 PM, David van der Spoel wrote:
> > Hi,
> >
> > my patch stumbled over this issue in Jenkins:
> >
> > Program:     gmx mdrun, VERSION 5.2-dev-20150915-ef00eb4-local
> > Source file: src/gromacs/timing/wallcycle.cpp (line 398)
> > Function:    void subtract_cycles(wallcc_t *, int, int)
> >
> > Assertion failed:
> > Condition: wcc[ewc_main].c >= wcc[ewc_sub].c
> > Subcounter cannot have more ticks than parent
> >
> > For more information and tips for troubleshooting, please check the
> > GROMACS
> > website at http://www.gromacs.org/Documentation/Errors
> >
> > Any clue? Or is this assertion just overly pedantic?
>
> --
> Gromacs Developers mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers
> or send a mail to gmx-developers-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150915/46180993/attachment.html>


More information about the gromacs.org_gmx-developers mailing list