[petsc-dev] Exiting with error when using GPUs and non GPU-aware MPI

Colleagues, I did not notice this, but Junchao's MR, "Directly pass root/leafdata to MPI in SF when possible"   https://gitlab.com/petsc/petsc/-/merge_requests/2506 that was merged into master over the weekend causes PETSc to error out if PETSc has been configured with GPU support but the

Re: [petsc-dev] Request for comments: allow C99 internally

I'll just add my vote: +1 to supporting the intersection of C99, C11, C++11, and later standards. Also +1 to sadness that VLAs don't fall into there. (Though I understand the reasons why compiler developers might not want to support them.) --Richard On 3/11/20 8:34 AM, Jed Brown wrote: My

Re: [petsc-dev] Understanding -log_summary with GPUs

Hi Shri, Probably the best way to understand what is going on is to step through things using a debugger, as Junchao suggests. VecAXPY does get used in a lot of places, and maybe it is being called on some vectors that aren't getting their type from the options database? Also, there are

Re: [petsc-dev] [petsc-users] Matrix-free method in PETSc

Well said, Lisandro! Barry, one of the reasons I became a PETSc user (and later developer) was how great you were about patiently answering my (sometimes clueless) questions. You've been doing an amazing job answering user questions for decades. Even the best people (and you are unquestionably

Re: [petsc-dev] PETSc 3.10: September 4, 2018, is the cutoff-date for new features

Hi Satish, Can I add a line about the new AIJSELL Mat type to the changes list before you do that? Meant to do that but didn't get around to it yesterday. --Richard On Wed, Sep 12, 2018, 6:00 AM Balay, Satish wrote: > All release relevant PRs are now merged to master. And I'm ready to tag >

Re: [petsc-dev] GAMG error with MKL

On Mon, Jul 9, 2018 at 10:04 AM, Jed Brown wrote: > Jeff Hammond writes: > > > This is the textbook Wrong Way to write OpenMP and the reason that the > > thread-scalability of DOE applications using MPI+OpenMP sucks. It leads > to > > codes that do fork-join far too often and suffer from death

Re: [petsc-dev] trying to make MKL work with GAMG

e a problem in the near term ... and I don't > understand why there is no equivalent of MatMultSymbolic (setting up > communication maps?). How does if function without this. Please clarify! > > Thanks, > > >> >> --Richard >> >> >&g

Re: [petsc-dev] trying to make MKL work with GAMG

scussions with the MKL team about this, but the current model they have isn't one that I can map all of the PETSc matrix-matrix interface onto. Hopefully I can get them to change this. --Richard > >> >> >>Barry >> >> >> > On Wed, Jul 4, 2018 at 9:44 A

Re: [petsc-dev] GAMG error with MKL

ammond > >>>> wrote: > >>>> > >>>>> > >>>>> > >>>>> On Tue, Jul 3, 2018 at 4:35 PM Mark Adams wrote: > >>>>> > >>>>>> On Tue, Jul 3, 2018 at 1:00 PM Richard Tran Mills

Re: [petsc-dev] trying to make MKL work with GAMG

Hi Mark, I'd never looked at the code for MatCreateMPIAIJWithSeqAIJ(), but it looks like if you are using MPIAIJ matrices but you've set "-mat_seqaij_type seqaijmkl", then the "diagonal" portion of the MPIAIJ matrix ends up as a SEQAIJMKL instance, but the off-diagonal portion ends up as a plain

Re: [petsc-dev] [petsc-maint] Incident INC0122538 MKL on Cori/KNL

Mark, were you trying this in Valgrind with a binary targeting KNL, i.e., built to use AVX-512 instructions? I don't think Valgrind implements all (or any?) of those, so a failure is not a surprise. Indeed, I've had Valgrind choke on some AVX2 instructions, though maybe the most recent versions of

Re: [petsc-dev] GAMG error with MKL

Hi Mark, I'm glad to see you trying out the AIJMKL stuff. I think you are the first person trying to actually use it, so we are probably going to expose some bugs and also some performance issues. My somewhat limited testing has shown that the MKL sparse routines often perform worse than our own

Re: [petsc-dev] testing on Titan

Hi Mark, I've got an OLCF account but don't have access to Titan. Would it be helpful for me to apply to get on whatever allocation you are using so I can poke around at this stuff? (If so, send me a message off list with details of what I should apply for.) Hopefully I could get on there faster

Re: [petsc-dev] plans for preconditioners for SeqSELL

On Mon, Jun 25, 2018 at 6:12 PM, Richard Tran Mills wrote: > Hi Hong, > > On Mon, Jun 25, 2018 at 2:52 PM, Zhang, Hong wrote: > >> >> >> On Jun 25, 2018, at 4:09 PM, Mills, Richard Tran wrote: >> >> Hi Hong, >> >> Thank

Re: [petsc-dev] plans for preconditioners for SeqSELL

petsc/src/master/src/mat/impls/sell/seq/sell.c?fileview >> er=file-view-default%20#sell.c-251 >> >> You were probably mislead by the code >> >> if (A->rmap->bs > 1) { >> ierr = MatConvert_Basic(A,newtype,reuse,newmat);CHKERRQ(ierr); >> PetscFunctionReturn(0);

Re: [petsc-dev] plans for preconditioners for SeqSELL

ierr = MatConvert_Basic(A,newtype,reuse,newmat);CHKERRQ(ierr); > PetscFunctionReturn(0); > } > which should be removed. > > Thanks, > Hong (Mr.) > > > On Jun 25, 2018, at 3:06 PM, Richard Tran Mills wrote: > > Hi Everyone (especially Herr Doktor Hong Zhang), &g

Re: [petsc-dev] upcoming release and testing [v2]

Satish: Oops, I noticed that the order of the developer photos got messed up (Hong and Stefano's photos switched) when I updated the website. Have pushed a fix to master. --Richard On Thu, Apr 5, 2018 at 7:07 PM, Balay, Satish <ba...@mcs.anl.gov> wrote: > On Fri, 6 Apr 2018, Richard T

Re: [petsc-dev] upcoming release and testing [v2]

wrote: > Thanks! I pushed some fixes to rmills/doc-add-rmills-to-user-manual. > > Satish > > On Fri, 6 Apr 2018, Richard Tran Mills wrote: > > > Hi Satish, > > > > I decided that I don't have time to polish things as much as I'd like, so > > I'll

Re: [petsc-dev] upcoming release and testing [v2]

, Satish <ba...@mcs.anl.gov> wrote: > Sure - we can push doc changes without going through the test cycle [just > a doc build cycle] > > thanks, > Satish > > On Thu, 5 Apr 2018, Richard Tran Mills wrote: > > > Hi Satish, > > > > I'm working on rig

Re: [petsc-dev] upcoming release and testing [v2]

Hi Satish, I'm working on right now on some edits to part of the "Performance Turning" part of the manual. If I can get these edits to a state I'm happy with, I'd like them to go in the release, but I'm not sure if I will have them ready before this evening. --Richard On Thu, Apr 5, 2018 at

Re: [petsc-dev] Target next PETSc/TAO release for March 30

asked for. =) --Richard On Fri, Mar 23, 2018 at 10:21 AM, Jed Brown <j...@jedbrown.org> wrote: > I think it's feasible, but are we ever going to have more than a week's > notice so that we can have a known merge window followed by a few days > of feature freeze? > > Ric

Re: [petsc-dev] plans for preconditioners for SeqSELL

"SELL has to implement everything and usually this means to manually convert back to AIJ". If there are no objections I'd like to clean this up. (Subclassing AIJ was unfortunately not available at the time Hong started his great work on SELL) Best regards, Karli On 03/03/2018 07:52 AM

Re: [petsc-dev] Handling pull requests in a better way

Hi Karl, Since I know this is in many ways a thankless task, let me thank you for taking this on! I'm a bit late to the discussion, but I want to point out one of the issues I've encountered with pull requests: Often a pull request is submitted with multiple reviewers listed, and it's

[petsc-dev] Rename csrperm files to aijperm?

Is there any good reason behind keeping the 'csrperm' in file and directory names when MATCSRPERM was changed to MATAIJPERM? If not, I'm going to rename things to 'aijperm' to stay consistent with the type name. --Richard

Re: [petsc-dev] plans for preconditioners for SeqSELL

On Wed, Feb 14, 2018 at 6:15 PM, Jed Brown <j...@jedbrown.org> wrote: > Richard Tran Mills <rtmi...@anl.gov> writes: > > > On Tue, Feb 13, 2018 at 8:48 PM, Jed Brown <j...@jedbrown.org> wrote: > > > >> Richard Tran Mills <rtmi...@anl.gov> writes:

Re: [petsc-dev] plans for preconditioners for SeqSELL

it easier for users in a lot other cases as well, and might make the use of SELL much more likely. If no one has already done some work on this, I'll take a stab at it. --Richard On Mon, Feb 12, 2018 at 10:04 AM, Richard Tran Mills <rtmi...@anl.gov> wrote: > On Mon, Feb 12, 2018 at 8:47

Re: [petsc-dev] plans for preconditioners for SeqSELL

On Tue, Feb 13, 2018 at 8:48 PM, Jed Brown <j...@jedbrown.org> wrote: > Richard Tran Mills <rtmi...@anl.gov> writes: > > > I haven't experimented very thoroughly with it (hmm... should probably do > > such experiments), but I believe that, once matrix rows

Re: [petsc-dev] plans for preconditioners for SeqSELL

s AIJ is faster since it uses the inodes routines. > > Is this expected? > These are the flags I used to compile PETSc on > KNL --COPTFLAGS=-xMIC-AVX512 -O3 -mP2OPT_hpo_vec_remainder=F; > > > 2018-02-12 21:04 GMT+03:00 Richard Tran Mills <rtmi...@anl.gov>: > &g

Re: [petsc-dev] plans for preconditioners for SeqSELL

On Mon, Feb 12, 2018 at 8:47 AM, Smith, Barry F. wrote: > > > > On Feb 12, 2018, at 10:25 AM, Stefano Zampini > wrote: > > > > Barry, > > > > for sure Amat,Pmat is the right approach; however, with complicated user > codes, we are not always in

Re: [petsc-dev] PETSc Quarterly Telecon

Karl, Thanks for organizing this. Unfortunately, I'll be on a plane at that time so I won't be able to join. I would be grateful if someone could write some brief meeting minutes for those of us who are interested but cannot make it. Thanks, Richard On Mon, Dec 18, 2017 at 2:51 AM, Karl Rupp

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

On Sun, Nov 12, 2017 at 9:54 AM, Smith, Barry F. wrote: > > > > On Nov 12, 2017, at 11:21 AM, Matthew Knepley wrote: > > > > On Sun, Nov 12, 2017 at 12:17 PM, Satish Balay > wrote: > > On Sun, 12 Nov 2017, Matthew Knepley wrote: > > > >

Re: [petsc-dev] broken nightlybuilds (next vs next-tmp)

Hi Satish, Thanks for taking the initiative to switch to testing next-tmp to help clear up the constipation with moving things into master. It looks like there hasn't been any graduation of the branches you've been putting into next-tmp in a few days, though. Is this just because you haven't had

Re: [petsc-dev] development version for MATAIJMKL type mat

On Tue, Oct 10, 2017 at 1:03 PM, Mark Adams wrote: > putting this back on the list. > > On Tue, Oct 10, 2017 at 3:21 PM, Bakytzhan Kallemov > wrote: > >> >> >> >> Forwarded Message >> Subject: Re: [petsc-dev] development version for

Re: [petsc-dev] Getting credit for Contributors graph on github?

ers who want to get credit for their PETSc > commits. > > > Begin forwarded message: > > *From: *Jed Brown <j...@jedbrown.org> > *Subject: **Re: [petsc-maint] Contributors graph on github?* > *Date: *September 27, 2017 at 11:01:48 AM CDT > *To: *Richard Tran Mills <

Re: [petsc-dev] So PETSc's function names are restricted to 31 characters?

gov> wrote: > Does PETSc build faster at 38,000 feet? > > On Sep 22, 2017, at 8:47 PM, Richard Tran Mills <rtmi...@anl.gov> wrote: > > Certainly! Being 38,000 feet in the air is no barrier to PETSc development! > > --Richard > > On Fri, Sep 22, 2017 at 6:44 PM, Smith

Re: [petsc-dev] So PETSc's function names are restricted to 31 characters?

Certainly! Being 38,000 feet in the air is no barrier to PETSc development! --Richard On Fri, Sep 22, 2017 at 6:44 PM, Smith, Barry F. <bsm...@mcs.anl.gov> wrote: > > Are you on the airplane? > > > > On Sep 22, 2017, at 8:43 PM > > , Richard Tran Mills <rtm

Re: [petsc-dev] So PETSc's function names are restricted to 31 characters?

I've never encountered a recent Fortran compiler that doesn't allow maximum identifier length to exceed the 31 characters specified in Fortran 95. Since Fortran 2003 extended this to 63, I vote for that. --Richard On Fri, Sep 22, 2017 at 6:36 PM, Smith, Barry F. wrote: > >

Re: [petsc-dev] (no subject)

Thanks for sharing this, Barry. I haven't had time to read their paper, but it looks worth a read. Hong, since many machine-learning or data-mining problems can be cast as linear algebra problems (several examples involving eigenproblems come to mind), I'm guessing that there must be several

[petsc-dev] Why separate PCASMGetSubKSP/PCGASMGetSubKSP/PCBJacobiGetSubKSP?

Folks, I see that we have separate interface routines PCASMGetSubKSP(), PCGASMGetSubKSP(), and PCBJacobiGetSubKSP(), all of which have an identical interface and do the same thing. It seems like there ought to instead be just one routine called something like PCGetSubKSP(). Is there a good reason

Re: [petsc-dev] hypre + OMP and MKL-AIJ

Hi Mark, Regarding MatMatMult() support, I think it should be fairly trivial for me to add that to AIJMKL. However, MKL doesn't currently have a matrix triple product routine (though the possibility of adding this has come up before in discussions with Alex Kalinkin from the MKL sparse team), so

[petsc-dev] How are we deciding the memory alignment that BuildSystem should choose?

Folks, I am wondering how PETSc's BuildSystem currently chooses the memory alignment to use. In the example file I provide in the repo for the Cori KNL nodes, I specify '--with-memalign=64' to match the cache line width. If I don't do this, then configure chooses a 16 byte alignment. Do people

Re: [petsc-dev] Small fix associating MatPtAP_seqaijperm_seqmaij_C

--Richard On Wed, Aug 16, 2017 at 3:40 PM, Richard Tran Mills <rtmi...@anl.gov> wrote: > Satish et al., > > I noticed a failure when testing with SeqAIJPERM and have made a small fix > that is on branch rmills/fix-seqaijperm_seqmaij. If such a small fix needs > to be tested in

[petsc-dev] Small fix associating MatPtAP_seqaijperm_seqmaij_C

, for convenience I have appended the git commit. --Richard rmills@Richards-MBP:tests(rmills/fix-seqaijperm_seqmaij=)arch-osx-clang-gfortran7-debug$ git show commit 1fde5765381347da5003c8b2045311701bd352b4 (HEAD -> rmills/fix-seqaijperm_seqmaij, origin/rmills/fix-seqaijperm_seqmaij) Author: Richard Tran Mi

Re: [petsc-dev] PETSc release?

words?) I tested on my Mac with clang and gfortran with real/complex and single/double-precision and the tests look fairly clean. Would like to merge this to next soon so that it can be tested further. Thanks, Richard On Mon, Aug 14, 2017 at 1:04 PM, Richard Tran Mills <rtmi...@anl.gov>

Re: [petsc-dev] PETSc release?

My estimate of "a few days" ended up being off (discovered many problems when running the test suite with it using SeqAIJMKL as the AIJ subtype), but I think my initial AIJMKL class is now close to completion and have submitted a pull request. --Richard On Mon, Jul 24, 2017 at 12:42 PM, Karl

Re: [petsc-dev] Test failures in PetscCommBuildTwoSided_Ibarrier?

ugs). >> >> Barry >> >> > On Aug 1, 2017, at 6:58 PM, Dave May <dave.mayhe...@gmail.com> wrote: >> > >> > >> > >> > On 2 August 2017 at 00:39, Richard Tran Mills <rtmi...@anl.gov> wrote: >>

Re: [petsc-dev] Test failures in PetscCommBuildTwoSided_Ibarrier?

from brew. > > And the nightlybuilds are fine [will have to check if something broke > today] > > There was previously an issue with gcc from brew - but don't remember what > it was. > > Satish > > On Tue, 1 Aug 2017, Richard Tran Mills wrote: > > > Hi Folks, > > &g

[petsc-dev] Test failures in PetscCommBuildTwoSided_Ibarrier?

Hi Folks, I am seeing a lot of errors when I run the full test suite (with latest master) on my MacBook, related to SEGVs inside (or thereabouts) PetscCommBuildTwoSided_Ibarrier. Do I have something mis-configured or ... ? Sample error message below: > [2]PETSC ERROR:

[petsc-dev] Running just a part of the test suite?

Hi Folks, I'm not sure how to easily run only portions of the PETSc test suite. The tests for 'Mat' have been very helpful to the debugging work I am currently doing; I'd like to be able to run only these instead of the entire suite. I think (?) these particular tests are still all in the old

[petsc-dev] Encountering SeqAIJ matrices seem to have no actual matrix; intended behavior?

Hi Folks, I am still slogging away at debugging my AIJMKL matrix class using the PETSc test suite. In some cases, such as mat/examples/tests/ex160.c, I am seeing a failure of the mkl_sparse_optimize() call inside MatAssembly. Digging into this, it appears that the problem is that the A array (the

Re: [petsc-dev] Routines that change matrix entries without requiring another MatAssemblyBegin/End?

On Sun, Jul 30, 2017 at 5:42 AM, Karl Rupp wrote: > >> > >> Well, I thought of AIJMKL keeping track of the state for which > mkl_sparse_optimize() was called. If the matrix state changes, the next > call to MatMult() > > > > is MatMult the only operation that needs

Re: [petsc-dev] Routines that change matrix entries without requiring another MatAssemblyBegin/End?

On Fri, Jul 28, 2017 at 12:32 PM, Smith, Barry F. <bsm...@mcs.anl.gov> wrote: > > > On Jul 28, 2017, at 2:13 PM, Richard Tran Mills <rtmi...@anl.gov> wrote: > > > > On Fri, Jul 28, 2017 at 12:02 PM, Stefano Zampini < > stefano.zamp...@gmail.c

Re: [petsc-dev] Routines that change matrix entries without requiring another MatAssemblyBegin/End?

On Fri, Jul 28, 2017 at 12:12 PM, Dave May <dave.mayhe...@gmail.com> wrote: > > On Fri, 28 Jul 2017 at 20:58, Dave May <dave.mayhe...@gmail.com> wrote: > >> Hi Richard, >> >> On Fri, 28 Jul 2017 at 19:51, Richard Tran Mills <rtmi...@anl.gov> wrote:

Re: [petsc-dev] Routines that change matrix entries without requiring another MatAssemblyBegin/End?

at PETSc has after these are used by mkl_sparse_optimize(). --Richard > > > > On Jul 28, 2017, at 9:58 PM, Barry Smith <bsm...@mcs.anl.gov> wrote: > > > > > >> On Jul 28, 2017, at 12:50 PM, Richard Tran Mills <rtmi...@anl.gov> > wrote: > >> > >>

[petsc-dev] Routines that change matrix entries without requiring another MatAssemblyBegin/End?

Folks, I've been debugging my rmills/add-aijmkl branch, and I found that one bug I had was related to what happens when I am using the MKL sparse inspector-executor routines inside an AIJMKL matrix and then MatScale is called on the matrix. MatScale modifies the entries of the matrix, thereby

Re: [petsc-dev] PETSc release?

> take long for this monster, so it is fine if it cannot go in the release. > > > > Thanks, > > Hong (Mr.) > > > >> On Jul 24, 2017, at 11:23 AM, Richard Tran Mills <rtmi...@anl.gov> > wrote: > >> > >> On Sun, Jul 23, 2017 at 3:08 PM,

Re: [petsc-dev] PETSc release?

On Sun, Jul 23, 2017 at 3:08 PM, Smith, Barry F. wrote: > >Anything stopping us from making a PETSc release? > >Barry > > I'd really like to get my rmills/add-aijmkl branch in a form that can go in. Should be able to have this ready for everyone's review in a day or

[petsc-dev] MatDuplicate documentation needs clarification

Folks, Looking at the MatDuplicate documentation, I think we could use some clarification. For the second argument, what this currently says is "*op *- either MAT_DO_NOT_COPY_VALUES or MAT_COPY_VALUES, cause it to copy the numerical values in the matrix MAT_SHARE_NONZERO_PATTERN to share the

[petsc-dev] What should MatDuplicate_SeqAIJMKL() do?

Fellow PETSc developers, I'm trying to figure out what my MatDuplicate() implementation for AIJMKL should do when the sparse inspector-executor model is being used. I think that the behavior should be * If MAT_COPY_VALUES has been specified, then mkl_sparse_optimize() should be executed for the

Re: [petsc-dev] error configuring on KNL

On Fri, Jun 30, 2017 at 3:27 PM, Balay, Satish <ba...@mcs.anl.gov> wrote: > On Fri, 30 Jun 2017, Richard Tran Mills wrote: > > > Hmm. I get the same error when adding "--download-hypre=0" to an > otherwise > > functional configure script for Cori KNL. Hypre's

Re: [petsc-dev] error configuring on KNL

Hmm. I get the same error when adding "--download-hypre=0" to an otherwise functional configure script for Cori KNL. Hypre's configure doesn't like something about cross-compiling for KNL. It might be possible to make this work by building on the compute nodes, though this is, unfortunately, not

Re: [petsc-dev] error configuring on KNL

Hi Mark, Take a look at the example template I've put together on Cori and see if it is helpful to you: /global/homes/r/rmills/proj/petsc/config/arch-cray-xc40-knl-opt.py (I note that the optimization levels for the different compilers were chosen after a fair amount of testing.) I'm about

[petsc-dev] How to check MKL requirements for a matrix class

PETSc developers, I have been working on an AIJMKL matrix class that enables use of MKL sparse BLAS routines. I need to keep this from breaking the build when a user does not have MKL. What should I check for? I see when I build on a machine where I have specified that MKL be used for BLAS and

Re: [petsc-dev] Supporting --download-hdf5 in Cygwin?

there. --Richard On 10/4/13 5:45 PM, Matthew Knepley wrote: On Fri, Oct 4, 2013 at 4:42 PM, Richard Tran Mills r...@eecs.utk.edu mailto:r...@eecs.utk.edu wrote: Hi Satish, I got this message attempting to use the Gnu compilers; that's all that I've got on this system. Should I send

Re: [petsc-dev] Supporting --download-hdf5 in Cygwin?

://self.name+' does not support --download-'+self.downloadname.lower()+' on Microsoft Windows') Matt If thats not the case - the bug is somewere else.. satish On Mon, 7 Oct 2013, Richard Tran Mills wrote: Matt, I tried

Re: [petsc-dev] Disabling a header in configure based on a bad link?

experience with *disabling* functionality in BuildSystem? Should we be less optimistic about having the header and only enable it if the test passes? Where would it go? Cheers, Aron -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics

[petsc-dev] Supporting --download-hdf5 in Cygwin?

. --Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University of Tennessee, Knoxville E-mail: rmi...@ornl.gov V: 865-241-3198 http

Re: [petsc-dev] Supporting --download-hdf5 in Cygwin?

Hi Satish, I got this message attempting to use the Gnu compilers; that's all that I've got on this system. Should I send the configure.log to petsc-maint? --Richard On 10/4/13 4:34 PM, Satish Balay wrote: On Fri, 4 Oct 2013, Richard Tran Mills wrote: Hi Folks, If I tell configure.py

[petsc-dev] Deprecated PETSc functions and Fortran codes?

as with the C code? Not a huge issue of course, but it would be nice to have the consistency. --Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory

Re: [petsc-dev] Adding DMLocalToLocalBegin/End: What about PetscSF?

construct some other default for local-to-local, or is there a good way to remap the global vector layout onto a local one? --Richard On 6/12/13 11:25 AM, Jed Brown wrote: Richard Tran Mills r...@eecs.utk.edu writes: PETSc developers, I am finally getting around to adding the DMLocalToLocalBegin

[petsc-dev] Adding DMLocalToLocalBegin/End: What about PetscSF?

as for a global-to-local. Could someone who is familiar with PetscSF comment on this? Thanks, Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory

[petsc-dev] Remove or deprecate DMDALocalToLocalBegin/End when adding DMLocalToLocalBegin/End

-- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University of Tennessee, Knoxville E-mail: rmi...@ornl.gov V: 865-241-3198 http://climate.ornl.gov

[petsc-dev] What to do with DMDALocalToLocal routines when adding DMLocalToLocalBegin()/End()?

Barry, Yes, you are right. I had erroneously thought that it was a user routine with a manual page; I was thrown off by the nice comment at the beginning. =) --Richard On 3/27/13 11:14 PM, Barry Smith wrote: On Mar 27, 2013, at 2:23 PM, Richard Tran Mills rtm at eecs.utk.edu wrote

[petsc-dev] What to do with DMDALocalToLocal routines when adding DMLocalToLocalBegin()/End()?

DMDALocalToLocalCreate() (which is called lazily by DMDALocalToLocalBegin()) as-is, however. Does this sound OK? --Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge

[petsc-dev] Add natural-to-global and global-to-natural operations for DM?

On 3/19/13 11:38 PM, Jed Brown wrote: On Tue, Mar 19, 2013 at 9:54 PM, Richard Tran Mills rtm at eecs.utk.edu mailto:rtm at eecs.utk.edu wrote: [...] Obviously, there may be subtypes of DM for which no natural to global operation makes sense. At the type level, adding

[petsc-dev] Add natural-to-global and global-to-natural operations for DM?

not need need a universal DMGlobalToNatural(); I do not object to a DMLocalToLocalBegin/End(), though local is not universal we do have the local concept at the DM level already and it will remain there. Barry On Mar 19, 2013, at 9:54 PM, Richard Tran Mills rtm at eecs.utk.edu

[petsc-dev] Increasing timeout for external package builds?

-- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University of Tennessee, Knoxville E-mail: rmills at ornl.gov V: 865-241-3198 http://climate.ornl.gov

[petsc-dev] PETSc repositories have been moved

requests to it. -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University of Tennessee, Knoxville E-mail: rmills at ornl.gov V: 865-241-3198

[petsc-dev] Some questions about adding default global to local, etc., functions for DMShell

without first setting the scatter context via DMShellSetGlobalToLocalVecScatter()? Thanks, Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory

[petsc-dev] Need mechanism to set LocalToGlobal mapping in a DMShell

I think that adding DMLocalToLocalBegin/End makes sense from a consistency standpoint. Should this be done by adding localtolocalbegin/end pointers to dm-ops? --Richard On 2/21/13 4:35 PM, Barry Smith wrote: On Feb 21, 2013, at 3:31 PM, Richard Tran Mills rtm at eecs.utk.edu wrote: I

[petsc-dev] Need mechanism to set LocalToGlobal mapping in a DMShell

you need it? You're in charge of making the matrices and vectors so you can give them local-to-global mappings. The only solver that calls DMGetLocalToGlobalMapping is PCEXOTIC, which only works with DMDA anyway. On Thu, Feb 21, 2013 at 12:35 PM, Richard Tran Mills rtm at eecs.utk.edu

[petsc-dev] Need mechanism to set LocalToGlobal mapping in a DMShell

-- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University of Tennessee, Knoxville E-mail: rmills at ornl.gov V: 865-241-3198 http://climate.ornl.gov

[petsc-dev] Uncrustify has fucked up my file

You know... I really regret that I don't have the time to regularly follow this list, because I'm sure I miss out on a ton of gems like this exchange! Clearly I need to make some time somehow. I think Matt might look good with the ridiculous hat and glasses. --Richard On 1/31/13 8:35 PM,

[petsc-dev] [mpich-discuss] MPICH migration to git

decided that Emacs was too bloated. Now *there's* some good trolling material! =) --Richard -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory

[petsc-dev] Counting Fortran function pointers

, 2013 at 5:48 PM, Jed Brown jedbrown at mcs.anl.gov mailto:jedbrown at mcs.anl.gov wrote: Richard, it's better to mail the list. No, we haven't reworked subtype Fortran pointers yet, but I think we know what needs to be done. On Mon, Jan 7, 2013 at 1:42 PM, Richard Tran Mills

[petsc-dev] [mpich-discuss] MPICH migration to git

feature with modern git (to be fair, it's with a develop version of mercurial). -- Richard Tran Mills, Ph.D. Computational Earth Scientist | Joint Assistant Professor Hydrogeochemical Dynamics Team | EECS and Earth Planetary Sciences Oak Ridge National Laboratory | University

[petsc-dev] [mpich-discuss] MPICH migration to git

On 1/9/13 11:37 PM, Barry Smith wrote: On Jan 9, 2013, at 10:35 PM, Richard Tran Mills rtm at eecs.utk.edu wrote: Git does some very cool stuff, but I have to agree with Sean's assessment of the user interface, and that's the reason I prefer Mercurial. This is not so much an issue

[petsc-dev] Is src/snes/examples/tutorials/ex50.c now ex19.c?

Hi Folks, src/ts/examples/tutorials/ex26.c references the file src/snes/examples/tutorials/ex50.c, which doesn't exist. Looking through the commit logs, it looks like this has been replaced with ex19.c? Or is that a somewhat different example? --Richard -- Richard Tran Mills, Ph.D

[petsc-dev] Is src/snes/examples/tutorials/ex50.c now ex19.c?

Thanks, Jed. I'll update the the ex26.c TS example to say ex19 instead of ex50. --Richard On 1/8/13 3:56 PM, Jed Brown wrote: Yes, it was moved to ex19.c once it was complete. On Tue, Jan 8, 2013 at 2:46 PM, Richard Tran Mills rtm at eecs.utk.edu mailto:rtm at eecs.utk.edu wrote

[petsc-dev] Updating the downloaded PLAPACK?

Hi Folks, I have corrected a few problems in the downloaded PLAPACK tarball (ftp://ftp.mcs.anl.gov/pub/petsc/externalpackages/PLAPACKR32-hg.tar.gz) due to missing function prototypes. The name of this tarball leads me to believe that we keep a Mercurial repo of this maintained version of

[petsc-dev] Stupid BuildSystem question: How do I force rebuild of a downloaded package?

Hi Folks, I am trying to test the changes I have made to the PLAPACKR32-hg package before pushing them back up and asking Satish to rebuild the tarball. I guessed that I could do this by deleting $PETSC_DIR/externalpackages/PLAPACKR32-hg and $PETSC_DIR/$PETSC_ARCH/lib/libPLAPACK.a, pointing

[petsc-dev] Stupid BuildSystem question: How do I force rebuild of a downloaded package?

the tarball on the FTP site? Thanks, Richard On 11/7/2011 3:26 PM, Satish Balay wrote: On Mon, 7 Nov 2011, Richard Tran Mills wrote: Hi Folks, I am trying to test the changes I have made to the PLAPACKR32-hg package before pushing them back up and asking Satish to rebuild the tarball. I

[petsc-dev] Trying to get the CMake instructions to work, unsuccessfully

need to reconfigure for a specific IDE instead of the default Makefiles (and some IDEs, such as QtCreator and KDevelop4, will also do this for you). -- Richard Tran Mills, Ph.D.| E-mail: rmills at climate.ornl.gov Computational Scientist | Phone: (865) 241-3198

[petsc-dev] Trying to get the CMake instructions to work, unsuccessfully

? Thanks, Richard On 2/2/2011 11:21 AM, Jed Brown wrote: On Wed, Feb 2, 2011 at 13:12, Richard Tran Mills rmills at climate.ornl.gov mailto:rmills at climate.ornl.gov wrote: Thanks for your responses. However, that is the first thing that I tried: rmills at ubuntu:~/proj/petsc-dev

[petsc-dev] Trying to get the CMake instructions to work, unsuccessfully

-strict-aliasing -Wno-unknown-pragmas -g', '-DCMAKE_C_COMPILER=/home/rmills/proj/petsc-dev/ubuntu-gnu_g/bin/mpicc', '-DCMAKE_Fortran_FLAGS= -Wall -Wno-unused-variable -g', '-DCMAKE_Fortran_COMPILER=/home/rmills/proj/petsc-dev/ubuntu-gnu_g/bin/mpif90' -- Richard Tran Mills, Ph.D

[petsc-dev] Trying to get the CMake instructions to work, unsuccessfully

--download-hypre=1 --download-ml=1 --download-blacs=1 --download-scalapack=1 --download-parmetis=1 --download-plapack=1 --with-cmake=/usr/bin/cmake PETSC_ARCH=ubuntu-gnu_g --Richard On 2/2/2011 6:22 PM, Richard Tran Mills wrote: Hi Jed, I get the following: - rmills at ubuntu:~/proj

[petsc-dev] PETSc programming model for multi-core systems

. Thus one could do an upfront static load balancing of most of the data then when the first cores run out of their static work they do the rest of the work with the dynamic balancing. Barry Jed -- Richard Tran Mills, Ph.D.| E-mail: rmills at climate.ornl.gov Computational

[petsc-dev] [GPU] Performance on Fermi

your log_summary, then PETSC_USE_DEBUG was definitely defined when plog.c was compiled. It's really much easier to have two separate builds and always use the optimized one when profiling. Jed -- Richard Tran Mills, Ph.D.| E-mail: rmills at climate.ornl.gov Computational Scientist

[petsc-dev] improvement to ./configure

-- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener -- Richard Tran Mills, Ph.D.| E-mail: rmills at climate.ornl.gov Computational Scientist

[petsc-dev] improvement to ./configure

of a filename. Barry On Jan 8, 2010, at 4:13 PM, Richard Tran Mills wrote: Matt, I now get something like the following when executing a configuration script with a '_' in it: Traceback (most recent call last): File config/ubuntu-gnu_g.py, line 18, in module

  1   2   >