gk20a: Unlock mutex upon queue full.

Unlock mutex upon queue full then return -EAGAIN.

Bug 1551294

Change-Id: I2a06a99028f98e6602a4cc27d15015bfe1db4eda
Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com>
Reviewed-on: http://git-master/r/501235
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Konsta Holtta <kholtta@nvidia.com>
Tested-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>

1 files changed, 1 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/pmu_gk20a.c b/drivers/gpu/nvgpu/gk20a/pmu_gk20a.c
index 9d52c943..f0ce3c9a 100644
--- a/drivers/gpu/nvgpu/gk20a/pmu_gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/pmu_gk20a.c
@@ -1760,6 +1760,7 @@ static int pmu_queue_open_write(struct pmu_gk20a *pmu,
 
         if (!pmu_queue_has_room(pmu, queue, size, &rewind)) {
                 gk20a_err(dev_from_gk20a(gk20a_from_pmu(pmu)), "queue full");
+                pmu_queue_unlock(pmu, queue);
                 return -EAGAIN;
         }