to DMGetMatrix,
it would have to put in fake entries before any duplication is possible.
Doesn't it already do this (inserting 0)?
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/4f91c906
--
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/97f1b4b9/attachment.html
by default until
MatSetValues has been called.
But I'm not sure whether either solution will break something else.
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/90b2d398/attachment.html
On Thu, 9 Feb 2012, Jed Brown wrote:
On Thu, Feb 9, 2012 at 23:50, Satish Balay balay at mcs.anl.gov wrote:
On Thu, 9 Feb 2012, Jed Brown wrote:
On Thu, Feb 9, 2012 at 23:36, Satish Balay balay at mcs.anl.gov wrote:
Credential management is simpler because adding/removing keys
On Thu, 9 Feb 2012, Sean Farley wrote:
Now we are back to the original model of user account - and access via
this account [yes - bitbucket makes it easier to crate accounts, and
hopefully better infrastrucutre than 'group permission' for sharing
underneath. [well atleat there is a gui
should do the allocation in
this case (maybe another hook is needed).
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/142bc80c/attachment.html
calling DMSetMatrixPreallocateOnly()?
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/ebb80ce8/attachment.html
On Feb 9, 2012, at 11:15 PM, Sean Farley wrote:
Even if you were right about this specific issue (which you are not) it
doesn't matter. All you've done is removed the need for a releases
subdirectory. What about tutorials subdirectory, externalpackages
subdirectory,
://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/c35efe83/attachment.html
any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/c72fab4e/attachment.html
On Fri, 10 Feb 2012, Satish Balay wrote:
With Bitbucket, GitHub, and all the others, the web account is *not* a unix
account. It's just a way to manage identity/keys. Everyone using ssh has
hg at bitbucket.org (and git at github.com, etc).
sure the implemtation details are different -
On Feb 10, 2012, at 9:13 AM, Matthew Knepley wrote:
The thread has become too deep for me to read, hence the top posting.
Barry's question is the right one: What do we gain by changing?
1) Reliability and Availability
Barry, you should know that this crap about petsc.cs being
their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/f79f6e46/attachment.html
-the-repository
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/d0feadd1/attachment.html
On Fri, 10 Feb 2012, Barry Smith wrote:
On Feb 10, 2012, at 9:13 AM, Matthew Knepley wrote:
The thread has become too deep for me to read, hence the top posting.
Barry's question is the right one: What do we gain by changing?
1) Reliability and Availability
Barry, you
part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/b576c7f2/attachment.html
/20120210/da43f896/attachment.html
Matt,
until we move to bitbucket - can you continue to push to petsc.cs.iit
- and not bitbucket? [for now - pushes to petsc.cs.iit are
automatically pushed to bitbucket]
thanks,
Satish
asterix:/home/balay/spetschg in
running ssh petsc at petsc.cs.iit.edu 'hg -R
lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/8bba15fa/attachment.html
interesting than any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/3eb71f21/attachment.html
Pramod (who did the work) was my student at EPCC last year. We looked at
porting Fluidity (or bits of it) to GPU and as a result decided that working on
PETSc's GPU support was going to be the most promising angle for the time we
had available. We have never got round to benchmarking Fluidity
On Fri, 10 Feb 2012, Matthew Knepley wrote:
until we move to bitbucket - can you continue to push to petsc.cs.iit
- and not bitbucket? [for now - pushes to petsc.cs.iit are
automatically pushed to bitbucket]
I thought we were switched already :) I cannot understand why we would not
On Feb 10, 2012, at 9:34 AM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 09:30, Matthew Knepley knepley at gmail.com wrote:
I decent way of hierarchically organizing our repositories. Tell me how to
do this on bitbucket and you have your slam dunk.
Mailing BB.
results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/94b4ba64/attachment.html
before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/104c6a4b
more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/a68a4ecf/attachment.html
was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/9d6375fa/attachment.html
I have write access to the iit petsc-dev repo so that I can push fixes
to Matt's Sieve stuff. Matt had me switch over to the bitbucket repo and
now I don't have write access. I assume I need a bitbucket login, so I
created one (baagaard) and added my public ssh key. If one of you could
enable
write access to the petsc-dev repo, I would appreciate it.
OK, done.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/e2b6f89f/attachment.html
part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/7ee0feb2/attachment.html
On Feb 10, 2012, at 11:06 AM, Matthew Knepley wrote:
On Fri, Feb 10, 2012 at 11:04 AM, Satish Balay balay at mcs.anl.gov wrote:
sorry about the conflicting messages. We haven't yet completely
switched to bitbucket. So please continue to push to iit repo.
We have not decided that yet.
/pipermail/petsc-dev/attachments/20120210/e49f100e/attachment.html
know we try to fix all bugs and good feature requests within
the hour, but there are topics that take longer. projects.html isn't an
optimal solution.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210
On Feb 10, 2012, at 11:25 AM, Matthew Knepley wrote:
On Fri, Feb 10, 2012 at 11:16 AM, Barry Smith bsmith at mcs.anl.gov wrote:
On Feb 10, 2012, at 11:06 AM, Matthew Knepley wrote:
On Fri, Feb 10, 2012 at 11:04 AM, Satish Balay balay at mcs.anl.gov wrote:
sorry about the conflicting
On Feb 10, 2012, at 11:26 AM, Jed Brown wrote:
One other point is integrated issue tracking. If someone makes a feature
request, we currently manage it purely by not losing the email. With
bitbucket issues, users can follow issues so when we include closes #123
(and other status updates)
.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/517b04c5/attachment.html
/pipermail/petsc-dev/attachments/20120210/5010bac9/attachment.html
On Fri, 10 Feb 2012, Jed Brown wrote:
On Fri, Feb 10, 2012 at 11:30, Barry Smith bsmith at mcs.anl.gov wrote:
On Feb 10, 2012, at 11:26 AM, Jed Brown wrote:
One other point is integrated issue tracking. If someone makes a feature
request, we currently manage it purely by not losing
settings:
https://bitbucket.org/seanfarley/settings/src/d4e93dc190c6/.hgrc
to get this public=yellow, draft=green, secret=red (image attached)
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/b168aad9
to get this public=yellow, draft=green, secret=red (image attached)
Forgot to attach image
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/c08ff9f6/attachment.html
-- next part
, MatType jtype, Mat
*J, MatType ptype, Mat *P),
which mirrors the ultimate use of these matrices in KSPSetOperators.
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/9c2a5c60/attachment.html
we be able to duplicate such unassambled
matrices? Or should they even be call unassembled?
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/9f13be20/attachment.html
://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/6d58b824/attachment.html
.
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/ad82479c/attachment.html
before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/0f08ccf3
-dev/attachments/20120210/f8a80954/attachment.html
://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/07dc79fc/attachment.html
matrices as SNES and KSP would want
anyway?
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/894f92a2/attachment.html
...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/80bb8d62/attachment.html
to the current
code which is already polluted with them.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/37557ff8/attachment.html
was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/d1980ad6/attachment.html
their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/78bd9499/attachment.html
--
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/8b54d621/attachment.html
...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/6b4ab326/attachment.html
/petsc-dev/attachments/20120210/e1b52ffb/attachment.html
:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/25acda6c/attachment.html
to state it.
No uglier than having SNES/KSP take in two matrices, in my view.
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/e50f8b6d/attachment.html
is always the right number of ways to
distinguish matrices.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/16630ac7/attachment.html
Hi All,
I've been developing GPU capabilities for PETSc. The development has
focused mostly on
(1) An efficient multi-GPU SpMV, i.e. MatMult. This is working well.
(2) Triangular Solve used in ILU preconditioners; i.e. MatSolve. The
performance of this ... is what it is :|
This code is in beta
.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/1703d2eb/attachment.html
/20120210/44e57d01/attachment.html
One other point, my code supports
(1) CUDA 4.1, Thrust 1.5, CUSP 0.3
or
(2) CUDA 4.0, Thurst 1.4, CUSP 0.2
While it may work with mercurial versions of Thrust and CUSP, I make no
guarantees.
-Paul
Hi All,
I've been developing GPU capabilities for PETSc. The development has
focused mostly
compare with the CUSPARSE
implementation?
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/39fc772d/attachment.html
-dev/attachments/20120210/ee64f4b1/attachment.html
--
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/ae810e80/attachment.html
transparent name for that option --
mf_operator seems rather obscure.
Dmitry.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/2f022244/attachment.html
:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/01b6d884/attachment.html
On Feb 10, 2012, at 3:40 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 15:24, Dmitry Karpeev karpeev at mcs.anl.gov wrote:
So, as I understand, this behavior is reasonable for the use with SNES, since
the matrix is assembled with a call to SNES(DM)ComputeJacobian. DMGetMatrix
is also used
part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/feda4216/attachment.html
On Feb 10, 2012, at 4:54 PM, Sean Farley wrote:
What if I don't have an openid?
Everything uses OpenID now and bitbucket uses it too. I went over this with
you last time. Your gmail, facebook, yahoo, etc. are all automatically
OpenIDs.
I sure don't want to use my facebook account to
On Feb 10, 2012, at 5:00 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 16:56, Barry Smith bsmith at mcs.anl.gov wrote:
I realize that since the DMs may often share the grid information and only
be different in the stencil width or box type or something Jed (or maybe it
is Matt) is going to
On Fri, 10 Feb 2012, Barry Smith wrote:
On Feb 10, 2012, at 4:54 PM, Sean Farley wrote:
What if I don't have an openid?
Everything uses OpenID now and bitbucket uses it too. I went over this with
you last time. Your gmail, facebook, yahoo, etc. are all automatically
OpenIDs.
://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/dc2fcc1c/attachment.html
however it pleases
you.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/1bc82829/attachment.html
On Feb 10, 2012, at 5:14 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 17:08, Dmitry Karpeev karpeev at mcs.anl.gov wrote:
I think it's completely natural for a DM to assemble two operators -- the
discretizations for the two are likely to be related anyway -- as soon as we
decide that it's
On Feb 10, 2012, at 5:17 PM, Sean Farley wrote:
I sure don't want to use my facebook account to access work related stuff,
that is absurd.
Your bsmith at mcs.anl.gov is already used as your OpenID, I've seen this
already. You don't have to use it. I'm just saying it's the easiest way to
On Feb 10, 2012, at 5:09 PM, Satish Balay wrote:
On Fri, 10 Feb 2012, Barry Smith wrote:
On Feb 10, 2012, at 4:54 PM, Sean Farley wrote:
What if I don't have an openid?
Everything uses OpenID now and bitbucket uses it too. I went over this with
you last time. Your gmail,
On Feb 10, 2012, at 5:09 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 17:01, Barry Smith bsmith at mcs.anl.gov wrote:
What if I don't have an openid?
Everything uses OpenID now and bitbucket uses it too. I went over this with
you last time. Your gmail, facebook, yahoo, etc. are all
/petsc-dev/attachments/20120210/6f37877d/attachment.html
:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/aa19d5d0/attachment.html
On Fri, 10 Feb 2012, Barry Smith wrote:
I am totally confused. Why can I use my BarryFSmith account to admin
repositories people make and push and pull repositories but suddenly when I
want to have admin at the petsc level I need to use this other openid
thing?? Why, why why?
Hmm
.
Dmitry.
Barry
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/aaa3901c/attachment.html
On Feb 10, 2012, at 6:22 PM, Matthew Knepley wrote:
On Fri, Feb 10, 2012 at 5:29 PM, Barry Smith bsmith at mcs.anl.gov wrote:
On Feb 10, 2012, at 5:09 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 17:01, Barry Smith bsmith at mcs.anl.gov wrote:
What if I don't have an openid?
--
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/ed6a7561/attachment.html
On Feb 10, 2012, at 6:36 PM, Satish Balay wrote:
On Fri, 10 Feb 2012, Barry Smith wrote:
I am totally confused. Why can I use my BarryFSmith account to admin
repositories people make and push and pull repositories but suddenly when I
want to have admin at the petsc level I need to use
was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/20ead33c/attachment.html
On Fri, 10 Feb 2012, Satish Balay wrote:
On Fri, 10 Feb 2012, Barry Smith wrote:
I am totally confused. Why can I use my BarryFSmith account to admin
repositories people make and push and pull repositories but suddenly when I
want to have admin at the petsc level I need to use this
On Feb 10, 2012, at 6:40 PM, Sean Farley wrote:
Hmm openid is orthongonal. As I understand things - its just a
machanism to not maintain yet-another-passwd-for-yet-another-site.
Yes, Satish is right about this.
[i.e setup openid with any openid provider. and then user this
credential to
/pipermail/petsc-dev/attachments/20120210/9437fc93/attachment.html
seems very natural.
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/4a60359d/attachment.html
part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/b420867f/attachment.html
about which DM on which to
PetscObjectCompose() things needed by certain callbacks (e.g. in the FAS
with TS stuff I'm doing).
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/6c802a2f
On Feb 10, 2012, at 6:41 PM, Matthew Knepley wrote:
On Fri, Feb 10, 2012 at 6:39 PM, Barry Smith bsmith at mcs.anl.gov wrote:
Not true. Make a clone of bitbucket.org/petsc/petsc.bitbucket.org make a
change to it and try to push, you cannot. You need to send email to Sean
asking him
Satish,
This is totally irrelevant to the discussion.
Barry
On Feb 10, 2012, at 6:43 PM, Satish Balay wrote:
On Fri, 10 Feb 2012, Satish Balay wrote:
On Fri, 10 Feb 2012, Barry Smith wrote:
I am totally confused. Why can I use my BarryFSmith account to admin
repositories
On Feb 10, 2012, at 6:48 PM, Sean Farley wrote:
Not true. Make a clone of bitbucket.org/petsc/petsc.bitbucket.org make a
change to it and try to push, you cannot. You need to send email to Sean
asking him for write permission to that repository. So cannot access
repositories in petsc
On Fri, 10 Feb 2012, Barry Smith wrote:
The model is:
you have an 'account' - and can create repos there. Now you are the
owner of the repos.
Then you can give 'admin' privilidges to other 'account' users on any
given repo. And also create groups to give read/write permissions
was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/e2fcda64/attachment.html
On Feb 10, 2012, at 6:45 PM, Dmitry Karpeev wrote:
On Fri, Feb 10, 2012 at 5:24 PM, Barry Smith bsmith at mcs.anl.gov wrote:
On Feb 10, 2012, at 5:14 PM, Jed Brown wrote:
On Fri, Feb 10, 2012 at 17:08, Dmitry Karpeev karpeev at mcs.anl.gov
wrote:
I think it's completely natural
which DM on which to
PetscObjectCompose() things needed by certain callbacks (e.g. in the FAS
with TS stuff I'm doing).
-- next part --
An HTML attachment was scrubbed...
URL:
http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120210/01c2a20c/attachment.html
ok - didn't realise the complete context.
satish
On Fri, 10 Feb 2012, Barry Smith wrote:
Satish,
This is totally irrelevant to the discussion.
Barry
On Feb 10, 2012, at 6:43 PM, Satish Balay wrote:
On Fri, 10 Feb 2012, Satish Balay wrote:
On Fri, 10 Feb 2012, Barry
1 - 100 of 116 matches
Mail list logo
