gpu: nvgpu: make cbc alloc os specific

CBC base needs to be aligned to 64KB. On Linux this is
achieved making compbit backing size multiple of 64KB.
However QNX nvmap alloc function does not allocate
memory aligned to requested size and needs to overallocate
to satisfy alignment requirement. Make cbc alloc function OS
specific to be able to modify QNX code.

Also align cbc base address to 64KB before writing to CBC BASE
register.

Bug 200426427

Change-Id: Ic867501403f2e2a4ba41ad5a8ed6f9c5c8ffa3f4
Signed-off-by: Aparna Das <aparnad@nvidia.com>
(cherry picked from commit 3f1e1133a46ebfc9763c649d7b839d069cae5a36)
Reviewed-on: https://git-master.nvidia.com/r/1786046
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 2 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/common/fb/fb_gv11b.c b/drivers/gpu/nvgpu/common/fb/fb_gv11b.c
index 664b2dd1..6973e04d 100644
--- a/drivers/gpu/nvgpu/common/fb/fb_gv11b.c
+++ b/drivers/gpu/nvgpu/common/fb/fb_gv11b.c
@@ -102,6 +102,8 @@ void gv11b_fb_init_cbc(struct gk20a *g, struct gr_gk20a *gr)
         else
                 compbit_store_iova = nvgpu_mem_get_addr(g,
                                                         &gr->compbit_store.mem);
+        /* must be aligned to 64 KB */
+        compbit_store_iova = roundup(compbit_store_iova, (u64)SZ_64K);
 
         compbit_base_post_divide64 = compbit_store_iova >>
                 fb_mmu_cbc_base_address_alignment_shift_v();