gpu: nvgpu: fix race condition with poweroff

When gpu rail-gating is enabled, it is possible that
both rail gating code and system shudown can start
executing gk20a_pm_prepare_poweroff() in parallel.
To synchronize this execution, protect gk20a_pm_prepare_poweroff()
with a mutex lock.

Bug 200168805

Change-Id: I19536a43ed20c3e82b32c316922dc3e19e3f59bb
Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-on: http://git-master/r/999548
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

2 files changed, 11 insertions, 3 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/gk20a.c b/drivers/gpu/nvgpu/gk20a/gk20a.c
index 0982ecdf..1000eba2 100644
--- a/drivers/gpu/nvgpu/gk20a/gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/gk20a.c
@@ -669,6 +669,7 @@ static int gk20a_init_support(struct platform_device *dev)
         mutex_init(&g->dbg_sessions_lock);
         mutex_init(&g->client_lock);
         mutex_init(&g->ch_wdt_lock);
+        mutex_init(&g->poweroff_lock);
 
         mutex_init(&g->interleave_lock);
         g->num_interleaved_channels = 0;
@@ -689,17 +690,19 @@ static int gk20a_pm_prepare_poweroff(struct device *dev)
 
         gk20a_dbg_fn("");
 
-        gk20a_scale_suspend(pdev);
+        mutex_lock(&g->poweroff_lock);
 
         if (!g->power_on)
-                return 0;
+                goto done;
+
+        gk20a_scale_suspend(pdev);
 
         /* cancel any pending cde work */
         gk20a_cde_suspend(g);
 
         ret = gk20a_channel_suspend(g);
         if (ret)
-                return ret;
+                goto done;
 
         /* disable elpg before gr or fifo suspend */
         ret |= gk20a_pmu_destroy(g);
@@ -723,6 +726,9 @@ static int gk20a_pm_prepare_poweroff(struct device *dev)
         /* Stop CPU from accessing the GPU registers. */
         gk20a_lockout_registers(g);
 
+done:
+        mutex_unlock(&g->poweroff_lock);
+
         return ret;
 }
 
diff --git a/drivers/gpu/nvgpu/gk20a/gk20a.h b/drivers/gpu/nvgpu/gk20a/gk20a.h
index b02d6111..0207588f 100644
--- a/drivers/gpu/nvgpu/gk20a/gk20a.h
+++ b/drivers/gpu/nvgpu/gk20a/gk20a.h
@@ -521,6 +521,8 @@ struct gk20a {
 
         struct mutex ch_wdt_lock;
 
+        struct mutex poweroff_lock;
+
         /* Channel priorities */
         u32 timeslice_low_priority_us;
         u32 timeslice_medium_priority_us;