gpu: nvgpu: Increase GPFIFO priv cmd queue size

We're doing two sync point increments, so we need to allocate space
for it.

Change-Id: I663abb18a930eb3955379d5f8dd1b08c8fa56897
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/719884

1 files changed, 1 insertions, 1 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/channel_gk20a.c b/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
index dc44d4b1..7fa8a552 100644
--- a/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/channel_gk20a.c
@@ -866,7 +866,7 @@ static int channel_gk20a_alloc_priv_cmdbuf(struct channel_gk20a *c)
            max size of priv_cmdbuf is :
            (gpfifo entry number * (2 / 3) * (4 + 6) * 4 bytes */
         size = roundup_pow_of_two(
-                c->gpfifo.entry_num * 2 * 10 * sizeof(u32) / 3);
+                c->gpfifo.entry_num * 2 * 12 * sizeof(u32) / 3);
 
         err = gk20a_gmmu_alloc_map(ch_vm, size, &q->mem);
         if (err) {