gpu: nvgpu: add gops.fifo.ch_abort_clean_up

Channel abort clean up is only needed by native and vgpu driver but not
RM server. RM server expects guest will clean up itself. RM server
should not set the callback.

Jira VQRM-3058

Change-Id: I11b49b6f2d51c871e31de16955d487dca82609cb
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1679705
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 1 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/gm20b/hal_gm20b.c b/drivers/gpu/nvgpu/gm20b/hal_gm20b.c
index b91795fa..f2ae97cf 100644
--- a/drivers/gpu/nvgpu/gm20b/hal_gm20b.c
+++ b/drivers/gpu/nvgpu/gm20b/hal_gm20b.c
@@ -438,6 +438,7 @@ static const struct gpu_ops gm20b_ops = {
                 .tsg_bind_channel = gk20a_tsg_bind_channel,
                 .tsg_unbind_channel = gk20a_fifo_tsg_unbind_channel,
                 .post_event_id = gk20a_tsg_event_id_post_event,
+                .ch_abort_clean_up = gk20a_channel_abort_clean_up,
 #ifdef CONFIG_TEGRA_GK20A_NVHOST
                 .alloc_syncpt_buf = gk20a_fifo_alloc_syncpt_buf,
                 .free_syncpt_buf = gk20a_fifo_free_syncpt_buf,