[Nouveau] [PATCH RFC 02/20] pm: fix a potential race condition when creating an engine context

[Samuel Pitoiset]

There is always the possiblity that the ppm->context pointer would get
partially updated and accidentally would equal ctx. This would allow two
contexts to co-exist, which is not acceptable. Moving the test to the
critical section takes care of this problem.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Signed-off-by: Martin Peres <martin.peres at free.fr>
---
 drm/nouveau/nvkm/engine/pm/base.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drm/nouveau/nvkm/engine/pm/base.c b/drm/nouveau/nvkm/engine/pm/base.c
index 274457c..4cf36a3 100644
--- a/drm/nouveau/nvkm/engine/pm/base.c
+++ b/drm/nouveau/nvkm/engine/pm/base.c
@@ -358,12 +358,11 @@ nvkm_perfctx_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	mutex_lock(&nv_subdev(ppm)->mutex);
 	if (ppm->context == NULL)
 		ppm->context = ctx;
-	mutex_unlock(&nv_subdev(ppm)->mutex);
-
 	if (ctx != ppm->context)
-		return -EBUSY;
+		ret = -EBUSY;
+	mutex_unlock(&nv_subdev(ppm)->mutex);
 
-	return 0;
+	return ret;
 }
 
 struct nvkm_oclass
-- 
2.4.2