diff mbox

drm/atomic: Add WARN_ON when state->acquire_ctx is not set.

Message ID 1462266751-29123-1-git-send-email-maarten.lankhorst@linux.intel.com (mailing list archive)
State New, archived
Headers show

Commit Message

Maarten Lankhorst May 3, 2016, 9:12 a.m. UTC
When I was writing an atomic wrapper for rmfb, I ran into the
following backtrace from lockdep:

Comments

Daniel Vetter May 3, 2016, 12:36 p.m. UTC | #1
On Tue, May 03, 2016 at 11:12:31AM +0200, Maarten Lankhorst wrote:
> When I was writing an atomic wrapper for rmfb, I ran into the
> following backtrace from lockdep:
> 
> =============================================
> [ INFO: possible recursive locking detected ]
> 4.5.0-patser+ #4696 Tainted: G     U
> ---------------------------------------------
> kworker/2:2/2608 is trying to acquire lock:
>  (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c9ddc>] drm_modeset_lock+0x7c/0x120 [drm]
> 
> but task is already holding lock:
>  (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c98cd>] modeset_backoff+0x8d/0x220 [drm]
> 
> other info that might help us debug this:
>  Possible unsafe locking scenario:
> 
>        CPU0
>        ----
>   lock(crtc_ww_class_mutex);
>   lock(crtc_ww_class_mutex);
> 
>  *** DEADLOCK ***
> 
>  May be due to missing lock nesting notation
> 
> 4 locks held by kworker/2:2/2608:
>  #0:  ("events"){.+.+.+}, at: [<ffffffff810a5eea>] process_one_work+0x15a/0x6c0
>  #1:  ((&arg.work)){+.+.+.}, at: [<ffffffff810a5eea>] process_one_work+0x15a/0x6c0
>  #2:  (crtc_ww_class_acquire){+.+.+.}, at: [<ffffffffc004532a>] drm_atomic_helper_remove_fb+0x4a/0x1d0 [drm_kms_helper]
>  #3:  (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c98cd>] modeset_backoff+0x8d/0x220 [drm]
> 
> While lockdep probably catches this bug when it happens, it's better
> to explicitly warn when state->acquire_ctx is not set.
> 
> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>

Yeah, because of the automagic fallback to normal mutex semantics when
acquire_ctx == NULL in ww_mutex_lock this can go boom, and the lockdep
splat is pretty hard to read. Makes sense to have an explicit WARN_ON
about it.

Applied to drm-misc, thanks.
-Daniel

> ---
>  drivers/gpu/drm/drm_atomic.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
> index 0b526423f19f..69adcf3944cc 100644
> --- a/drivers/gpu/drm/drm_atomic.c
> +++ b/drivers/gpu/drm/drm_atomic.c
> @@ -263,6 +263,8 @@ drm_atomic_get_crtc_state(struct drm_atomic_state *state,
>  	int ret, index = drm_crtc_index(crtc);
>  	struct drm_crtc_state *crtc_state;
>  
> +	WARN_ON(!state->acquire_ctx);
> +
>  	crtc_state = drm_atomic_get_existing_crtc_state(state, crtc);
>  	if (crtc_state)
>  		return crtc_state;
> @@ -622,6 +624,8 @@ drm_atomic_get_plane_state(struct drm_atomic_state *state,
>  	int ret, index = drm_plane_index(plane);
>  	struct drm_plane_state *plane_state;
>  
> +	WARN_ON(!state->acquire_ctx);
> +
>  	plane_state = drm_atomic_get_existing_plane_state(state, plane);
>  	if (plane_state)
>  		return plane_state;
> @@ -890,6 +894,8 @@ drm_atomic_get_connector_state(struct drm_atomic_state *state,
>  	struct drm_mode_config *config = &connector->dev->mode_config;
>  	struct drm_connector_state *connector_state;
>  
> +	WARN_ON(!state->acquire_ctx);
> +
>  	ret = drm_modeset_lock(&config->connection_mutex, state->acquire_ctx);
>  	if (ret)
>  		return ERR_PTR(ret);
> -- 
> 2.5.5
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
diff mbox

Patch

=============================================
[ INFO: possible recursive locking detected ]
4.5.0-patser+ #4696 Tainted: G     U
---------------------------------------------
kworker/2:2/2608 is trying to acquire lock:
 (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c9ddc>] drm_modeset_lock+0x7c/0x120 [drm]

but task is already holding lock:
 (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c98cd>] modeset_backoff+0x8d/0x220 [drm]

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(crtc_ww_class_mutex);
  lock(crtc_ww_class_mutex);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

4 locks held by kworker/2:2/2608:
 #0:  ("events"){.+.+.+}, at: [<ffffffff810a5eea>] process_one_work+0x15a/0x6c0
 #1:  ((&arg.work)){+.+.+.}, at: [<ffffffff810a5eea>] process_one_work+0x15a/0x6c0
 #2:  (crtc_ww_class_acquire){+.+.+.}, at: [<ffffffffc004532a>] drm_atomic_helper_remove_fb+0x4a/0x1d0 [drm_kms_helper]
 #3:  (crtc_ww_class_mutex){+.+.+.}, at: [<ffffffffc00c98cd>] modeset_backoff+0x8d/0x220 [drm]

While lockdep probably catches this bug when it happens, it's better
to explicitly warn when state->acquire_ctx is not set.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
---
 drivers/gpu/drm/drm_atomic.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 0b526423f19f..69adcf3944cc 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -263,6 +263,8 @@  drm_atomic_get_crtc_state(struct drm_atomic_state *state,
 	int ret, index = drm_crtc_index(crtc);
 	struct drm_crtc_state *crtc_state;
 
+	WARN_ON(!state->acquire_ctx);
+
 	crtc_state = drm_atomic_get_existing_crtc_state(state, crtc);
 	if (crtc_state)
 		return crtc_state;
@@ -622,6 +624,8 @@  drm_atomic_get_plane_state(struct drm_atomic_state *state,
 	int ret, index = drm_plane_index(plane);
 	struct drm_plane_state *plane_state;
 
+	WARN_ON(!state->acquire_ctx);
+
 	plane_state = drm_atomic_get_existing_plane_state(state, plane);
 	if (plane_state)
 		return plane_state;
@@ -890,6 +894,8 @@  drm_atomic_get_connector_state(struct drm_atomic_state *state,
 	struct drm_mode_config *config = &connector->dev->mode_config;
 	struct drm_connector_state *connector_state;
 
+	WARN_ON(!state->acquire_ctx);
+
 	ret = drm_modeset_lock(&config->connection_mutex, state->acquire_ctx);
 	if (ret)
 		return ERR_PTR(ret);