projects / platform / kernel / linux-rpi.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a22f0ec) | patch
habanalabs: use mmu cache range invalidation
authorAlon Mizrahi <amizrahi@habana.ai>
Thu, 8 Apr 2021 12:30:59 +0000 (15:30 +0300)
committerOded Gabbay <ogabbay@kernel.org>
Fri, 18 Jun 2021 12:23:38 +0000 (15:23 +0300)
commit08c03a19662fd628e8866d89769d594c1d8c8093
treeadcfeb7d61b0c87f61db3d855285d11030ad31bftree | snapshot
parenta22f0ec0731ddf6355e6bf93ded189fca2542cdfcommit | diff
habanalabs: use mmu cache range invalidation

Use mmu cache range invalidation instead of entire cache invalidation
because it yields better performance.

In GOYA and GAUDI, always use entire cache invalidation because these
ASICs don't support range invalidation.

Signed-off-by: Alon Mizrahi <amizrahi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
drivers/misc/habanalabs/common/habanalabs.h diff | blob | history
drivers/misc/habanalabs/common/memory.c diff | blob | history
drivers/misc/habanalabs/gaudi/gaudi.c diff | blob | history
drivers/misc/habanalabs/goya/goya.c diff | blob | history
Domain: System / Kernel;