nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-22 09:12:24 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: make user vma start,end pde aligned

Browse Source 
Any PDE can allocate memory with a specific page size. That means memory
allocation with page size 4K and 64K will be realized by different PDEs
with page size (or PTE size) 4K and 64K respectively. To accomplish this
user vma is required to be pde aligned.
Currently, user vma is aligned by (big_page_size << 10) carried over
from when pde size was equivalent to (big_page_size << 10).

Modify user vma alignment check to use pde size.

JIRA NVGPU-5302

Change-Id: I2c6599fe50ce9fb081dd1f5a8cd6aa48b17b33b4
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2428327
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Sami Kiminki <skiminki@nvidia.com>
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: Sami Kiminki <skiminki@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Vedashree Vidwans
2020-10-14 21:59:16 -07:00
committed by Alex Waterman
parent 00d1e10ff2
commit 69948919b7
4 changed files with 25 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
userspace/units/mm/vm/vm.c
View File

@@ -2058,7 +2058,7 @@ int test_vm_pde_coverage_bit_count(struct unit_module *m, struct gk20a *g,
int ret = UNIT_FAIL;
struct vm_gk20a *vm = create_test_vm(m, g);
bit_count = nvgpu_vm_pde_coverage_bit_count(vm);
bit_count = nvgpu_vm_pde_coverage_bit_count(g, vm->big_page_size);
if (bit_count != GP10B_PDE_BIT_COUNT) {
unit_err(m, "invalid PDE bit count\n");