gpu: nvgpu: SLAB allocation for page allocator

Add the ability to do "SLAB" allocation in the page allocator. This
is generally useful since the allocator manages 64K pages but often
we only need 4k chunks (for example when allocating memory for page
table entries).

Bug 1799159
JIRA DNVGPU-100

Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: http://git-master/r/1225322
(cherry picked from commit 299a5639243e44be504391d9155b4ae17d914aa2)
Change-Id: Ib3a8558d40ba16bd3a413f4fd38b146beaa3c66b
Reviewed-on: http://git-master/r/1227924
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 5 insertions, 2 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
index 1bacb70a..df58feb2 100644
--- a/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/mm_gk20a.c
@@ -925,8 +925,11 @@ static int gk20a_init_vidmem(struct mm_gk20a *mm)
                                         bootstrap_base, bootstrap_size,
                                         SZ_4K, 0);
 
-        err = gk20a_page_allocator_init(&g->mm.vidmem.allocator, "vidmem",
-                                base, size - base, default_page_size, 0);
+        err = gk20a_page_allocator_init(&g->mm.vidmem.allocator,
+                                        "vidmem",
+                                        base, size - base,
+                                        default_page_size,
+                                        GPU_ALLOC_4K_VIDMEM_PAGES);
         if (err) {
                 gk20a_err(d, "Failed to register vidmem for size %zu: %d",
                                 size, err);