projects / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5b6a1ae) | patch
nvc0: bump the amount of shared memory per MP on Maxwell
authorSamuel Pitoiset <samuel.pitoiset@gmail.com>
Mon, 25 Apr 2016 22:15:25 +0000 (00:15 +0200)
committerSamuel Pitoiset <samuel.pitoiset@gmail.com>
Mon, 25 Apr 2016 22:32:25 +0000 (00:32 +0200)
commitd12c3b02ffdcb54fd3c919ab1d13603a793d20ef
treefa50c0a29bacf8ed7a760e4d503cad27fa6a156btree
parent5b6a1aee4674d24d57fbbaac86619e429bd031e5commit | diff
nvc0: bump the amount of shared memory per MP on Maxwell

According to the CUDA compute capability version, GM10x can expose
64KB of shared memory while GM20x can use 96KB.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
src/gallium/drivers/nouveau/nvc0/nvc0_screen.c diff | blob | history