[petsc-dev] MatMatMult gives different results

[Jed Brown]

On Thu, Feb 9, 2012 at 08:30, Matthew Knepley <knepley at gmail.com> wrote:

> On Thu, Feb 9, 2012 at 2:48 AM, Alexander Grayver <agrayver at gfz-potsdam.de
> > wrote:
>
>> **
>> lib/petsc-dev> hg pull -u
>> pulling from http://petsc.cs.iit.edu/petsc/petsc-dev
>> searching for changes
>> no changes found
>>
>> Does it take some time till it's there?
>>
>
> You have to remake Fortran stubs: make allfortranstubs
>
> and then make.
>

So we have some confusion right now because Matt is pushing to bitbucket
first and Barry is pushing to petsc.cs.iit.edu. I would be happy to
experiment with using bitbucket as the primary for petsc-dev and have
petsc.cs.iit.edu be the mirror (so users can keep tracking petsc-dev
without changing URLs). The simple convenience of having https:// and
ssh:// trivially push/pull the same repository is enough to make me prefer
interacting with bitbucket (screwy networks filter ssh).

But "two-way mirror" does not work, so we have to agree about where we are
pushing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120209/d77c4a6e/attachment.html>