nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-25 02:52:51 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: vmarea: Fix sparse documentation

Browse Source 
Sparse mappings are a mechanism to allow oversubscription of the GPU's
accessible memory to a given texture or other buffer object.

Change-Id: I675a73468cc8c1b1e333502be871404a49a025c4
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2284837
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
This commit is contained in:
Alex Waterman
2020-01-24 11:59:33 -06:00
parent 21f8b366cd
commit 1a204f6c7a
1 changed files with 10 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
drivers/gpu/nvgpu/include/nvgpu/vm_area.h
View File

@@ -61,8 +61,10 @@ struct nvgpu_vm_area {
*/
u64 size;
/**
* The GPU unmapping needed before using the vm_area.
*/
* Mark the vm area as sparse.
*
* @sa NVGPU_VM_AREA_ALLOC_SPARSE
*/
bool sparse;
};
@@ -80,7 +82,12 @@ nvgpu_vm_area_from_vm_area_list(struct nvgpu_list_node *node)
#define NVGPU_VM_AREA_ALLOC_FIXED_OFFSET BIT32(0)
/**
* Allocation of vm_area at random address.
* Mark the vmarea as sparse: this means that the vmarea's entire range of PTEs
* is mapped as sparse. Sparse mappings are mappings in which the valid bit is
* set to 0, but the volatile (cached) bit is set to 1.
*
* The purpose here is to allow an oversubscription of physical memory
* for a particular texture or other object.
*/
#define NVGPU_VM_AREA_ALLOC_SPARSE BIT32(1)