[petsc-dev] Recommended Petsc Build Procedure
At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? Thanks, Dave
[petsc-dev] Recommended Petsc Build Procedure
On Sun, Apr 22, 2012 at 10:42 AM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? You do not have CMake, and thus did not get the production printout. The make is indeed legacy, but just as fast as CMake (Aron), as I believe the Python is as well. The Python is experimental, but I have been using it for a year and it works fine. Matt Thanks, Dave -- 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 -- next part -- An HTML attachment was scrubbed... URL: http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120422/aa7527c0/attachment.html
[petsc-dev] Recommended Petsc Build Procedure
You can do parallel builds with CMake... A On Sun, Apr 22, 2012 at 5:57 PM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: I have cmake on my system but configure.log does not document that approach. What is your build command to build petsc with cmake? Thanks, Dave Aron Ahmadia writes: I use the CMake build because it's the fastest, though it requires you to have CMake installed on your system. A On Sun, Apr 22, 2012 at 5:42 PM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? Thanks, Dave -- next part -- An HTML attachment was scrubbed... URL: http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120422/c3f86817/attachment.html
[petsc-dev] Recommended Petsc Build Procedure
The recommend usage is make If cmake was found with ./configure it will use it automatically if you do make, if it was not found it will use the legacy make Barry On Apr 22, 2012, at 10:19 AM, Dave Nystrom wrote: OK. I'll mail the configure.log to petsc-maint. But this did not start out as a configure problem - I was just asking for a recommendation on usage. And the response you cite below was to Aron before I received your response. Matthew Knepley writes: On Sun, Apr 22, 2012 at 10:57 AM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: I have cmake on my system but configure.log does not document that approach. We did not find it. This is a configure question, and should be mailed to petsc-maint, not the dev list. What is your build command to build petsc with cmake? make Matt Thanks, Dave Aron Ahmadia writes: I use the CMake build because it's the fastest, though it requires you to have CMake installed on your system. A On Sun, Apr 22, 2012 at 5:42 PM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? Thanks, Dave -- 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
[petsc-dev] Recommended Petsc Build Procedure
Thanks. So far, I have been using make. Barry Smith writes: The recommend usage is make If cmake was found with ./configure it will use it automatically if you do make, if it was not found it will use the legacy make Barry On Apr 22, 2012, at 10:19 AM, Dave Nystrom wrote: OK. I'll mail the configure.log to petsc-maint. But this did not start out as a configure problem - I was just asking for a recommendation on usage. And the response you cite below was to Aron before I received your response. Matthew Knepley writes: On Sun, Apr 22, 2012 at 10:57 AM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: I have cmake on my system but configure.log does not document that approach. We did not find it. This is a configure question, and should be mailed to petsc-maint, not the dev list. What is your build command to build petsc with cmake? make Matt Thanks, Dave Aron Ahmadia writes: I use the CMake build because it's the fastest, though it requires you to have CMake installed on your system. A On Sun, Apr 22, 2012 at 5:42 PM, Dave Nystrom Dave.Nystrom at tachyonlogic.com wrote: At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? Thanks, Dave -- 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
[petsc-dev] Recommended Petsc Build Procedure
On Sun, 22 Apr 2012, Dave Nystrom wrote: At the end of configure.log, there are two possible ways to build petsc-dev that are specified. Which is the recommended way to build - using make or using python? I have been using make. Also, one is labeled as legacy and one is labeled as experimental. That gives the impression of having a choice between an old, archaic method or a new, experimental approach. Should one just be labeled as production? Currently the first choice printed by configure is the recommended choice. [this is either cmake or legacy depending on what configure figured out]. Python is labled experimental as most of us [except Matt] haven't actively used it yet. As more folks use it actively - we can remove the 'experimental' tag off it. Satish
