- changed status to open
- removed comment
Improve performance of CarpetInterp2 interpolation
The branch https://bitbucket.org/eschnett/carpet/branch/ianhinder/fasterp_opt#diff contains an optimisation to the CarpetInterp2 interpolation routine which improved the performance of Llama interpatch interpolation in my test by a factor of 5. I did this a while ago, and haven't looked at it recently. Before merging, the following should be done: 1. Check that is applies cleanly to the current version of Carpet 2. Decide whether the vectorisation pragmas need to be protected by Cactus preprocessor guards Or any other changes which people think might be necessary. This should wait until after the upcoming (May 2015) release of the ET.
Keyword:
Comments (6)
-
reporter -
reporter - removed comment
This optimisation applies cleanly to the release, but not to master, due to the code reformat. Erik, would you mind making a version of this patch that applies to master?
-
- removed comment
Converting a patch is not possible. Instead, I will need to convert the whole file both before and after applying the patch. Has this patch been applied? Do you have a branch name / commit id / commit message?
-
reporter - removed comment
It's in the original ticket description: https://bitbucket.org/eschnett/carpet/branch/ianhinder/fasterp_opt. The branch name is "fasterp_opt". Thanks!
-
- changed status to resolved
- removed comment
The code in that branch was applied to Carpet in 2c896aecd8f7bc973c282ea80c42dc21de5e9c65 "CarpetInterp2: Vectorize interpolation" in 2016-04-27.
-
- edited description
- changed status to closed
- Log in to comment