nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-23 01:50:07 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: bypass smmu for VPR memory access

Browse Source 
SMMU translation should be bypassed for VPR
accesses via GPU.
clear sgt dma address to bypass smmu for VPR.

Bug 1215470

Change-Id: I22df41a9afc447e2502055b7907cc1848a770f26
Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
Reviewed-on: http://git-master/r/696509
(cherry picked from commit a699f55941fa22e90d41a53798956a542b212659)
Reviewed-on: http://git-master/r/707889
This commit is contained in:
Krishna Reddy
2015-02-10 13:56:47 -08:00
committed by Dan Willemsen
parent 0312497ee8
commit 781cbf5c93
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
drivers/gpu/nvgpu/gk20a/platform_gk20a_tegra.c
View File

@@ -121,7 +121,8 @@ static int gk20a_tegra_secure_alloc(struct platform_device *pdev,
}
page = phys_to_page(iova);
sg_set_page(sgt->sgl, page, size, 0);
sg_dma_address(sgt->sgl) = iova;
/* This bypasses SMMU for VPR during gmmu_map. */
sg_dma_address(sgt->sgl) = 0;
desc->sgt = sgt;