June 2010 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] [petsc-users] ublas sparse matrix bindings?
by Matthew Knepley 29 Jun '10

29 Jun '10

I am not so sure here. Our problem is not with type parametrization so much I think (since MatScalar is there) but with mixing this with another value type. I suspect that a templated version would have 1G of compiler errors right now. Matt On Tue, Jun 29, 2010 at 1:53 PM, Aron Ahmadia <aron.ahmadia(a)kaust.edu.sa>wrote: > This might be one of the rare cases which makes a strong argument for C++ > template metaprogramming. A well-defined extension could make this > seamless. I know this is already one of the strengths of the Eigen > libraries (freedom to mix types). > > A > > On Tue, Jun 29, 2010 at 2:49 PM, Barry Smith <bsmith(a)mcs.anl.gov> wrote: > >> >> On Jun 29, 2010, at 12:59 PM, Matthew Knepley wrote: >> >> On Tue, Jun 29, 2010 at 12:58 PM, Jed Brown <jed(a)59a2.org> wrote: >> >>> On Tue, 29 Jun 2010 12:43:36 -0500, Matthew Knepley <knepley(a)gmail.com> >>> wrote: >>> > 3) What you really want is to make MatScalar to be float. Then use >>> > doubles for the residual calculation. >>> >>> I have no idea what Luke's usage is, but MatScalar != PetscScalar is not >>> currently supported. >>> >>> http://lists.mcs.anl.gov/pipermail/petsc-dev/2009-June/001417.html >>> >>> I'm intrigued by it because my matrices are only for preconditioning >>> purposes, but haven't caught the bug to try fixing it since the above >>> thread. >> >> >> If someone really needs it, I will fix it. >> >> >> It is a major mother to fix. Unless we can come up with a new clever >> paradigm it will be a big mess of ugly nasty code to "fix". The problem is >> that in many places in the code, values in the arrays of the matrices are >> passed as values through PETSc functions (which are all prototyped as >> PetscScalar) so at each passage one must be write code to convert from >> double to single pass in then free the copy. >> >> Barry >> >> >> >> Matt >> >> >>> >>> Jed >>> >> >> >> >> -- >> 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 >> >> >> > -- 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

1 0

Simplifying PETSc Debugging in Totalview
by Barry Smith 29 Jun '10

29 Jun '10

PETSc users, Jeff Keasler has pointed out that TotalView 8.8 and later support a new feature where libraries can provide code that displays the contents of an object in any useful format (allowing for example, PETSc to provide code to nicely display entries of a sparse matrix in the debugger). I have tried this out by including in petsc-dev a very simple viewer for Vec and Mat; these will just work if you use TotalView 8.8 or higher with PETSc-dev. If any users really love using Totalview with PETSc and would like to see this extended (or extend it themselves :-) the example source code is in src/vec/vec/interface/vector.c called TV_display_type() and please get in touch with us. Thanks Barry

1 0

vacation week of July 5 to 9
by Barry Smith 27 Jun '10

27 Jun '10

Please list the week of July 5 to 9 as vacation for me. I will be traveling and only checking email occasionally. Barry

1 0

Re: [petsc-dev] mecurial emergency
by Barry Smith 26 Jun '10

26 Jun '10

The emergency was only in my repository, never in the master. Barry On Jun 26, 2010, at 6:05 PM, Matthew Knepley wrote: > Is this fixed? Can someone mail when it is over? > > For post mortem, how did this happen? The only way I can > think of is that > > a) Someone forced a push (never never never to this) > > b) Someone pulled into a repo that already had multiple heads > > c) Someone checked in a changset in the middle of a merge > > Matt > > On Sat, Jun 26, 2010 at 11:19 PM, Satish Balay <balay(a)mcs.anl.gov> wrote: > On Sat, 26 Jun 2010, Satish Balay wrote: > > > On Sat, 26 Jun 2010, Barry Smith wrote: > > > > > > > > I need help. FAQ.html has "instructions" to fix this but the instructions are not complete enough: they say > > > > > > "Fixing this is complicated. Basically, you clone the repo > > > before you made head #2, then create the diff for the bad changeset > > > that made head #2. Apply it to the clone and checkin, then pull the > > > master. > > > > > > Now > > > > > > 1) how do I clone the repo before you made head #2? > > > 2) how do you create the diff for the bad changeset that made head # 2 > > > 3) how do you apply it to the clone > > > 4) how do you checkin? Never heard of checkin in mecurial > > > > > > Please let me know this and I will also update the FAQ. > > > > > > > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg merge > > > abort: outstanding uncommitted changes (use 'hg status' to list changes) > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg commit > > > nothing changed > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg status | head -5 > > > ! src/ksp/ksp/examples/tutorials/ex1.c > > > > Looks like some files are deleted. Commit doesn't pick it up > > automatically. [unless 'hg rm' is used for the deletion] > > > > > ? ! > > > ? #configure.log# > > > ? arch-g++.py > > > ? bin/#petscrun# > > > [ > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg heads > > > changeset: 16298:503f0f312436 > > > tag: tip > > > parent: 16297:53f17d50e2f2 > > > parent: 16296:90b6abac8d23 > > > user: Victor Minden victorminden(a)gmail.com > > > date: Fri Jun 25 15:25:05 2010 -0500 > > > description: > > > commit after merge > > > > > > > > > changeset: 16291:47055b889013 > > > user: Barry Smith bsmith(a)mcs.anl.gov > > > date: Sat Jun 26 09:54:02 2010 -0500 > > > files: src/docs/website/documentation/codemanagement.html src/docs/website/documentation/linearsolvertable.html src/docs/website/features/index.html src/docs/website/miscellaneous/acknwldg.html src/docs/website/miscellaneous/funding.html src/sys/makefile src/sys/totalview/makefile src/sys/totalview/tv_data_display.c src/sys/totalview/tv_data_display.h > > > description: > > > minor cleanup of docs > > > added totalview include and source for debugger > > > > > > I would do: > > > > hg update -C 47055b889013 > > hg merge > > hg commit > > I should have added: the above instructions are for the case where > local changes can be discarded. If I have to preserve the local > changes [depending upon how the changes were done] - I might use some > different process. > > The following might work for almost all cases [since the merge is > interfering with the local changes - do the merge in a different/clean > repo] > > i.e: > > hg clone -r 47055b889013 petsc-dev petsc-dev-merge > cd petsc-dev-merge > hg pull [tip] > hg merge > hg commit > hg push [../petsc-dev] > cd ../petsc-dev > > hg status > hg update or 'hg commit/hg merge' or somethig suitable for commiting local changes and merging > local changes > > hg push [petsc.cs.iit.edu] > > Satish > > > > -- > 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

1 0

Re: [petsc-dev] mecurial emergency
by Matthew Knepley 26 Jun '10

26 Jun '10

Is this fixed? Can someone mail when it is over? For post mortem, how did this happen? The only way I can think of is that a) Someone forced a push (never never never to this) b) Someone pulled into a repo that already had multiple heads c) Someone checked in a changset in the middle of a merge Matt On Sat, Jun 26, 2010 at 11:19 PM, Satish Balay <balay(a)mcs.anl.gov> wrote: > On Sat, 26 Jun 2010, Satish Balay wrote: > > > On Sat, 26 Jun 2010, Barry Smith wrote: > > > > > > > > I need help. FAQ.html has "instructions" to fix this but the > instructions are not complete enough: they say > > > > > > "Fixing this is complicated. Basically, you clone the repo > > > before you made head #2, then create the diff for the bad changeset > > > that made head #2. Apply it to the clone and checkin, then pull the > > > master. > > > > > > Now > > > > > > 1) how do I clone the repo before you made head #2? > > > 2) how do you create the diff for the bad changeset that made head # 2 > > > 3) how do you apply it to the clone > > > 4) how do you checkin? Never heard of checkin in mecurial > > > > > > Please let me know this and I will also update the FAQ. > > > > > > > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg merge > > > abort: outstanding uncommitted changes (use 'hg status' to list > changes) > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg commit > > > nothing changed > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg status | > head -5 > > > ! src/ksp/ksp/examples/tutorials/ex1.c > > > > Looks like some files are deleted. Commit doesn't pick it up > > automatically. [unless 'hg rm' is used for the deletion] > > > > > ? ! > > > ? #configure.log# > > > ? arch-g++.py > > > ? bin/#petscrun# > > > [ > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg heads > > > changeset: 16298:503f0f312436 > > > tag: tip > > > parent: 16297:53f17d50e2f2 > > > parent: 16296:90b6abac8d23 > > > user: Victor Minden victorminden(a)gmail.com > > > date: Fri Jun 25 15:25:05 2010 -0500 > > > description: > > > commit after merge > > > > > > > > > changeset: 16291:47055b889013 > > > user: Barry Smith bsmith(a)mcs.anl.gov > > > date: Sat Jun 26 09:54:02 2010 -0500 > > > files: src/docs/website/documentation/codemanagement.html > src/docs/website/documentation/linearsolvertable.html > src/docs/website/features/index.html > src/docs/website/miscellaneous/acknwldg.html > src/docs/website/miscellaneous/funding.html src/sys/makefile > src/sys/totalview/makefile src/sys/totalview/tv_data_display.c > src/sys/totalview/tv_data_display.h > > > description: > > > minor cleanup of docs > > > added totalview include and source for debugger > > > > > > I would do: > > > > hg update -C 47055b889013 > > hg merge > > hg commit > > I should have added: the above instructions are for the case where > local changes can be discarded. If I have to preserve the local > changes [depending upon how the changes were done] - I might use some > different process. > > The following might work for almost all cases [since the merge is > interfering with the local changes - do the merge in a different/clean > repo] > > i.e: > > hg clone -r 47055b889013 petsc-dev petsc-dev-merge > cd petsc-dev-merge > hg pull [tip] > hg merge > hg commit > hg push [../petsc-dev] > cd ../petsc-dev > > hg status > hg update or 'hg commit/hg merge' or somethig suitable for commiting local > changes and merging > local changes > > hg push [petsc.cs.iit.edu] > > Satish > -- 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

1 0

Re: [petsc-dev] mecurial emergency
by Satish Balay 26 Jun '10

26 Jun '10

On Sat, 26 Jun 2010, Satish Balay wrote: > On Sat, 26 Jun 2010, Barry Smith wrote: > > > > > I need help. FAQ.html has "instructions" to fix this but the instructions are not complete enough: they say > > > > "Fixing this is complicated. Basically, you clone the repo > > before you made head #2, then create the diff for the bad changeset > > that made head #2. Apply it to the clone and checkin, then pull the > > master. > > > > Now > > > > 1) how do I clone the repo before you made head #2? > > 2) how do you create the diff for the bad changeset that made head # 2 > > 3) how do you apply it to the clone > > 4) how do you checkin? Never heard of checkin in mecurial > > > > Please let me know this and I will also update the FAQ. > > > > > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg merge > > abort: outstanding uncommitted changes (use 'hg status' to list changes) > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg commit > > nothing changed > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg status | head -5 > > ! src/ksp/ksp/examples/tutorials/ex1.c > > Looks like some files are deleted. Commit doesn't pick it up > automatically. [unless 'hg rm' is used for the deletion] > > > ? ! > > ? #configure.log# > > ? arch-g++.py > > ? bin/#petscrun# > > [ > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg heads > > changeset: 16298:503f0f312436 > > tag: tip > > parent: 16297:53f17d50e2f2 > > parent: 16296:90b6abac8d23 > > user: Victor Minden victorminden(a)gmail.com > > date: Fri Jun 25 15:25:05 2010 -0500 > > description: > > commit after merge > > > > > > changeset: 16291:47055b889013 > > user: Barry Smith bsmith(a)mcs.anl.gov > > date: Sat Jun 26 09:54:02 2010 -0500 > > files: src/docs/website/documentation/codemanagement.html src/docs/website/documentation/linearsolvertable.html src/docs/website/features/index.html src/docs/website/miscellaneous/acknwldg.html src/docs/website/miscellaneous/funding.html src/sys/makefile src/sys/totalview/makefile src/sys/totalview/tv_data_display.c src/sys/totalview/tv_data_display.h > > description: > > minor cleanup of docs > > added totalview include and source for debugger > > > I would do: > > hg update -C 47055b889013 > hg merge > hg commit I should have added: the above instructions are for the case where local changes can be discarded. If I have to preserve the local changes [depending upon how the changes were done] - I might use some different process. The following might work for almost all cases [since the merge is interfering with the local changes - do the merge in a different/clean repo] i.e: hg clone -r 47055b889013 petsc-dev petsc-dev-merge cd petsc-dev-merge hg pull [tip] hg merge hg commit hg push [../petsc-dev] cd ../petsc-dev hg status hg update or 'hg commit/hg merge' or somethig suitable for commiting local changes and merging local changes hg push [petsc.cs.iit.edu] Satish

1 0

Re: [petsc-dev] mecurial emergency
by Satish Balay 26 Jun '10

26 Jun '10

On Sat, 26 Jun 2010, Barry Smith wrote: > > I need help. FAQ.html has "instructions" to fix this but the instructions are not complete enough: they say > > "Fixing this is complicated. Basically, you clone the repo > before you made head #2, then create the diff for the bad changeset > that made head #2. Apply it to the clone and checkin, then pull the > master. > > Now > > 1) how do I clone the repo before you made head #2? > 2) how do you create the diff for the bad changeset that made head # 2 > 3) how do you apply it to the clone > 4) how do you checkin? Never heard of checkin in mecurial > > Please let me know this and I will also update the FAQ. > > > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg merge > abort: outstanding uncommitted changes (use 'hg status' to list changes) > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg commit > nothing changed > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg status | head -5 > ! src/ksp/ksp/examples/tutorials/ex1.c Looks like some files are deleted. Commit doesn't pick it up automatically. [unless 'hg rm' is used for the deletion] > ? ! > ? #configure.log# > ? arch-g++.py > ? bin/#petscrun# > [ > > [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg heads > changeset: 16298:503f0f312436 > tag: tip > parent: 16297:53f17d50e2f2 > parent: 16296:90b6abac8d23 > user: Victor Minden victorminden(a)gmail.com > date: Fri Jun 25 15:25:05 2010 -0500 > description: > commit after merge > > > changeset: 16291:47055b889013 > user: Barry Smith bsmith(a)mcs.anl.gov > date: Sat Jun 26 09:54:02 2010 -0500 > files: src/docs/website/documentation/codemanagement.html src/docs/website/documentation/linearsolvertable.html src/docs/website/features/index.html src/docs/website/miscellaneous/acknwldg.html src/docs/website/miscellaneous/funding.html src/sys/makefile src/sys/totalview/makefile src/sys/totalview/tv_data_display.c src/sys/totalview/tv_data_display.h > description: > minor cleanup of docs > added totalview include and source for debugger I would do: hg update -C 47055b889013 hg merge hg commit Satish

1 0

mecurial emergency
by Barry Smith 26 Jun '10

26 Jun '10

I need help. FAQ.html has "instructions" to fix this but the instructions are not complete enough: they say "Fixing this is complicated. Basically, you clone the repo before you made head #2, then create the diff for the bad changeset that made head #2. Apply it to the clone and checkin, then pull the master. Now 1) how do I clone the repo before you made head #2? 2) how do you create the diff for the bad changeset that made head # 2 3) how do you apply it to the clone 4) how do you checkin? Never heard of checkin in mecurial Please let me know this and I will also update the FAQ. [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg merge abort: outstanding uncommitted changes (use 'hg status' to list changes) [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg commit nothing changed [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg status | head -5 ! src/ksp/ksp/examples/tutorials/ex1.c ? ! ? #configure.log# ? arch-g++.py ? bin/#petscrun# [ [barry-smiths-macbook-pro:src/sys/totalview] barrysmith% hg heads changeset: 16298:503f0f312436 tag: tip parent: 16297:53f17d50e2f2 parent: 16296:90b6abac8d23 user: Victor Minden victorminden(a)gmail.com date: Fri Jun 25 15:25:05 2010 -0500 description: commit after merge changeset: 16291:47055b889013 user: Barry Smith bsmith(a)mcs.anl.gov date: Sat Jun 26 09:54:02 2010 -0500 files: src/docs/website/documentation/codemanagement.html src/docs/website/documentation/linearsolvertable.html src/docs/website/features/index.html src/docs/website/miscellaneous/acknwldg.html src/docs/website/miscellaneous/funding.html src/sys/makefile src/sys/totalview/makefile src/sys/totalview/tv_data_display.c src/sys/totalview/tv_data_display.h description: minor cleanup of docs added totalview include and source for debugger

1 0

Re: [petsc-dev] g95 with PETSc and somefort.F
by Satish Balay 25 Jun '10

25 Jun '10

I think its a g95 bug on Mac. I see this on the Mac - but on on linux. Will report the bug. moving 'block data PetscCommInit' into a different sourcefile is a workarround. Satish On Wed, 23 Jun 2010, Barry Smith wrote: > > Anybody know what this problem is and how to fix it? > > Barry > > > make PETSC_DIR=/Users/barrysmith/Src/petsc-dev PETSC_ARCH=arch-g95 > /Users/barrysmith/Src/petsc-dev/arch-g95/bin/mpif90 -c -Wall -Wno-unused-variable -g -I/Users/barrysmith/Src/petsc-dev/arch-g95/include -I/Users/barrysmith/Src/petsc-dev/include -I/opt/local/include -I/Users/barrysmith/Src/petsc-dev/arch-g95/include -I/Users/barrysmith/Src/petsc-dev/arch-g95/include -I/opt/local/include -I/Users/barrysmith/Src/petsc-dev/arch-g95/include -o somefort.o somefort.F > Warning (115): Line 17 of somefort.F is being truncated > Warning (115): Line 18 of somefort.F is being truncated > Warning (115): Line 19 of somefort.F is being truncated > Warning (115): Line 17 of somefort.F is being truncated > Warning (115): Line 18 of somefort.F is being truncated > Warning (115): Line 19 of somefort.F is being truncated > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:321:non-relocatable subtraction expression, "_petscfortran9_" minus "L00000000002$pb" > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:321:symbol: "_petscfortran9_" can't be undefined in a subtraction expression > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:118:non-relocatable subtraction expression, "_petscfortran10_" minus "L00000000001$pb" > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:118:symbol: "_petscfortran10_" can't be undefined in a subtraction expression > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:113:non-relocatable subtraction expression, "_petscfortran9_" minus "L00000000001$pb" > /var/folders/M4/M4PTePd3FM4rLBeN0lPCkE+++TI/-Tmp-//ccQR27B9.s:113:symbol: "_petscfortran9_" can't be undefined in a subtraction expression > make[1]: [somefort.o] Error 1 (ignored) > /usr/bin/ar cr /Users/barrysmith/Src/petsc-dev/arch-g95/lib/libpetsc.a somefort.o > ar: somefort.o: No such file or directory > make[1]: [libf] Error 1 (ignored) > /usr/bin/ranlib: file: /Users/barrysmith/Src/petsc-dev/arch-g95/lib/libpetsc.a(mhyp.o) has no symbols > > >

1 0

Re: [petsc-dev] KSPAddOptionsChecker() needed?
by Jed Brown 25 Jun '10

25 Jun '10

On Fri, 25 Jun 2010 18:53:48 +0800, Matthew Knepley <knepley(a)gmail.com> wrote: > No cast needed in straight C :) Assuming you like warnings. Of course you can make a macro that does the cast and forwards to a real function, but that's not how the PetscObject* API currently works. Jed

1 0