| commit | e534a221c1df74d3366b0fbaf1e43fe06686cea1 | |
| author | Sven Verdoolaege <skimo@kotnet.org> | |
| Tue, 22 Apr 2014 12:43:11 +0000 (22 14:43 +0200) | ||
| committer | Sven Verdoolaege <skimo@kotnet.org> | |
| Wed, 15 Apr 2015 20:50:21 +0000 (15 22:50 +0200) | ||
| tree | dcce8fc14134d5f2e534da69363b011f329457de | treesnapshot (tar.gz zip) |
| parent | 3f56bd1e0586085555234dbb9acf8f46555b3815 | commitdiff |
gpu: only add synchronization after copies from shared to global that require it
If the data written back from shared memory to global memory
is not needed within the same kernel invocation and is also not overwritten,
or if it is only needed inside the same iteration of the bands where
synchronization is inserted, then there is no need to insert synchronization
after that copy.
Signed-off-by: Sven Verdoolaege <skimo@kotnet.org>
If the data written back from shared memory to global memory
is not needed within the same kernel invocation and is also not overwritten,
or if it is only needed inside the same iteration of the bands where
synchronization is inserted, then there is no need to insert synchronization
after that copy.
Signed-off-by: Sven Verdoolaege <skimo@kotnet.org>
| gpu.c | diffblobblamehistory |