gpu: nvgpu: vgpu: change commit_inst sequence in gr_alloc

Since rm-server is going to use gr sources from linux including the
subctx_gv11b.c. commit_inst should be done after global_ctx_buffer
map and commit. gv11b_update_subctx_header is called from rm-server
for alloc_subctx_header which is using global_ctx_buffer_va[PRIV_ACCESS_MAP_VA].

Jira VQRM-2982

Change-Id: Iff953bf0a12db2c6d69d35094969ab9485858025
Signed-off-by: Shashank Singh <shashsingh@nvidia.com>
Signed-off-by: Richard Zhao <rizhao@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1661187
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Sourab Gupta <sourabg@nvidia.com>
Reviewed-by: Thomas Fleury <tfleury@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 7 insertions, 7 deletions

diff --git a/drivers/gpu/nvgpu/vgpu/gr_vgpu.c b/drivers/gpu/nvgpu/vgpu/gr_vgpu.c
index e61996a1..4a3cf075 100644
--- a/drivers/gpu/nvgpu/vgpu/gr_vgpu.c
+++ b/drivers/gpu/nvgpu/vgpu/gr_vgpu.c
@@ -523,13 +523,6 @@ int vgpu_gr_alloc_obj_ctx(struct channel_gk20a  *c, u32 class_num, u32 flags)
                         goto out;
                 }
 
-                /* commit gr ctx buffer */
-                err = g->ops.gr.commit_inst(c, gr_ctx->mem.gpu_va);
-                if (err) {
-                        nvgpu_err(g, "fail to commit gr ctx buffer");
-                        goto out;
-                }
-
                 /* allocate patch buffer */
                 err = vgpu_gr_alloc_channel_patch_ctx(g, c);
                 if (err) {
@@ -550,6 +543,13 @@ int vgpu_gr_alloc_obj_ctx(struct channel_gk20a  *c, u32 class_num, u32 flags)
                         goto out;
                 }
 
+                /* commit gr ctx buffer */
+                err = g->ops.gr.commit_inst(c, gr_ctx->mem.gpu_va);
+                if (err) {
+                        nvgpu_err(g, "fail to commit gr ctx buffer");
+                        goto out;
+                }
+
                 /* load golden image */
                 err = gr_gk20a_elpg_protected_call(g,
                                 vgpu_gr_load_golden_ctx_image(g, c));