nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-22 17:36:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: fall back to sysmem for generic allocs

Browse Source 
In gk20a_gmmu_alloc_attr(), which is used for in-kernel allocations,
fall back to attempting to allocate sysmem when vidmem allocation fails.

Bug 1809939

Change-Id: I0397026fd1b3bc803f6d8bb7409e05ab31ec961d
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1215447
(cherry picked from commit 3ec37992b830cee917e8ad35ede50e048907014a)
Reviewed-on: http://git-master/r/1217687
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Konsta Holtta
2016-09-06 11:18:51 +03:00
committed by mobile promotions
parent eee2744d49
commit 1a63ca3a65
1 changed files with 10 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
drivers/gpu/nvgpu/gk20a/mm_gk20a.c
View File

@@ -2793,8 +2793,16 @@ int gk20a_gmmu_alloc(struct gk20a *g, size_t size, struct mem_desc *mem)
int gk20a_gmmu_alloc_attr(struct gk20a *g, enum dma_attr attr, size_t size,
struct mem_desc *mem)
{
if (g->mm.vidmem_is_vidmem)
return gk20a_gmmu_alloc_attr_vid(g, attr, size, mem);
if (g->mm.vidmem_is_vidmem) {
int err = gk20a_gmmu_alloc_attr_vid(g, attr, size, mem);
if (!err)
return 0;
/*
* Fall back to sysmem (which may then also fail) in case
* vidmem is exhausted.
*/
}
return gk20a_gmmu_alloc_attr_sys(g, attr, size, mem);
}