Re: [petsc-users] Fortran application context, passing a module

It looks to me that you are trying to pass a Fortran 90 derived type as one entry in an array of addresses as an application context (perhaps from some PETSc example). This will not work as written because Fortran doesn't know that PetscFortranAddr is an array of addresses. The

[petsc-users] Fortran application context, passing a module

Dear all I want to pass my own module as an application context in fortran. So I am trying something like this type(mgrid),target :: g PetscFortranAddr :: ctx(6) ctx(1) => g But this gives an error *Error:* Non-POINTER in pointer association context (pointer assignment) at (1) Could you tell

Re: [petsc-users] Understanding PETSc -log_summary

Also this information at the top is often not particularly useful because it includes start up time and any IO etc. We recommend looking that the events below. You can start with the line for TSSolve, or SNESSolve, or KSPSolve depending on if you are using TS, SNES, or KSP. There are

Re: [petsc-users] Question about PETSC

> On Mar 16, 2017, at 4:56 AM, Валентин Егоров > wrote: > > Thank you for answering, I will try. > Sincerely, > Valentin > > Matt is correct. The best way to get into PETSc is by studying the example > codes provided in the source tree. > > As a precursor to

Re: [petsc-users] Understanding PETSc -log_summary

Thanks, Jed. That is helpful. *Ajit Desai* PhD Scholar, Carleton University, Canada On Thu, Mar 16, 2017 at 8:00 AM, Jed Brown wrote: > Ajit Desai writes: > > > Hello Everyone, > > A couple of questions on the *-log_summary* provided by PETSc. > >

Re: [petsc-users] Understanding PETSc -log_summary

Ajit Desai writes: > Hello Everyone, > A couple of questions on the *-log_summary* provided by PETSc. -log_view is the preferred name. > 1. *Avg-Flops & Avg-Flops/sec* are averaged among the participating cores > or averaged over the simulation time or both? Flops on each

[petsc-users] Understanding PETSc -log_summary

Hello Everyone, A couple of questions on the *-log_summary* provided by PETSc. 1. *Avg-Flops & Avg-Flops/sec* are averaged among the participating cores or averaged over the simulation time or both? 2. *Max/Min Flops & Flops/sec* is an indication of load balancing? In simulation-2 these ratios

Re: [petsc-users] Question about PETSC

Thank you for answering, I will try. Sincerely, Valentin 2017-03-16 10:37 GMT+03:00 Dave May : > > > On Thu, 16 Mar 2017 at 07:16, Matthew Knepley wrote: > >> Hi Valentin, >> >> Have you seen this example: https://bitbucket.org/petsc/petsc/src/ >>

Re: [petsc-users] Question about PETSC

On Thu, 16 Mar 2017 at 07:16, Matthew Knepley wrote: > Hi Valentin, > > Have you seen this example: > https://bitbucket.org/petsc/petsc/src/1830d94e4628b31f970259df1d58bc250c9af32a/src/ksp/ksp/examples/tutorials/ex2f.F?at=master=file-view-default > > Would that be enough to

Re: [petsc-users] Question about PETSC

Hi Valentin, Have you seen this example: https://bitbucket.org/petsc/petsc/src/1830d94e4628b31f970259df1d58bc250c9af32a/src/ksp/ksp/examples/tutorials/ex2f.F?at=master=file-view-default Would that be enough to get started? Thanks, Matt On Thu, Mar 16, 2017 at 12:37 AM, Валентин Егоров