| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Disable AVX emulation
Set default for streaming stores to "no"
Correct QPX vectorisation (IBM Blue Gene/Q)
Add MIC vectorisation (Intel Xeon Phi)
Convert SSE and AVX vectorisation to using inline functions instead of
macros for code clarity
Define CCTK_BOOLEAN, CCTK_INTEGER and CCTK_BOOLEAN_VEC,
CCTK_INTEGER_VEC to make boolean and integer vectors explicit
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@77 105869f7-3296-0410-a4ea-f4349344b45a
|
|
|
|
| |
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@46 105869f7-3296-0410-a4ea-f4349344b45a
|
|
|
|
|
|
|
|
|
| |
Most of the machines supported by SimFactory that use the Intel
compiler have to set this to "no" to avoid the compiler running out of
memory when compiling ML_BSSN. We choose "no" as the default now to
avoid problems for users of machines not supported in SimFactory.
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@33 105869f7-3296-0410-a4ea-f4349344b45a
|
|
|
|
|
|
| |
So, for the moment treat an empty OPTIMIZE as 'yes', as Cactus itself also does
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@17 105869f7-3296-0410-a4ea-f4349344b45a
|
|
|
|
|
|
| |
OPTIMISE_MODE, which is always provided by Cactus, and consistent with the other two
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@16 105869f7-3296-0410-a4ea-f4349344b45a
|
|
|
|
| |
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@13 105869f7-3296-0410-a4ea-f4349344b45a
|
|
Introduce configuration-time options for vectorisation, including
options to allow architecture-specific choices that may influence
performance.
Introduce "middle" masked stores for large vector sizes and small
loops.
Clean up and simplify some of the implementation code.
git-svn-id: https://svn.cct.lsu.edu/repos/numrel/LSUThorns/Vectors/trunk@10 105869f7-3296-0410-a4ea-f4349344b45a
|