All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] drm/vmwgfx: Fix a destoy-while-held mutex problem.
@ 2018-03-21  9:53 Thomas Hellstrom
  2018-03-21  9:53   ` Thomas Hellstrom
  0 siblings, 1 reply; 3+ messages in thread
From: Thomas Hellstrom @ 2018-03-21  9:53 UTC (permalink / raw)
  To: dri-devel; +Cc: linux-graphics-maintainer, Thomas Hellstrom, stable

When validating legacy surfaces, the backup bo might be destroyed at
surface validate time. However, the kms resource validation code may have
the bo reserved, so we will destroy a locked mutex. While there shouldn't
be any other users of that mutex when it is destroyed, it causes a lock
leak and thus throws a lockdep error.

Fix this by having the kms resource validation code hold a reference to
the bo while we have it reserved. We do this by introducing a validation
context which might come in handy when the kms code is extended to validate
multiple resources or buffers.

Cc: <stable@vger.kernel.org>
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>
---
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.c  | 28 +++++++++++++++++++---------
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.h  | 12 +++++++++---
 drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c |  5 +++--
 drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c |  5 +++--
 4 files changed, 34 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index b3d62aa01a9b..3c824fd7cbf3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -31,7 +31,6 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_rect.h>
 
-
 /* Might need a hrtimer here? */
 #define VMWGFX_PRESENT_RATE ((HZ / 60 > 0) ? HZ / 60 : 1)
 
@@ -2517,9 +2516,12 @@ void vmw_kms_helper_buffer_finish(struct vmw_private *dev_priv,
  * Helper to be used if an error forces the caller to undo the actions of
  * vmw_kms_helper_resource_prepare.
  */
-void vmw_kms_helper_resource_revert(struct vmw_resource *res)
+void vmw_kms_helper_resource_revert(struct vmw_validation_ctx *ctx)
 {
-	vmw_kms_helper_buffer_revert(res->backup);
+	struct vmw_resource *res = ctx->res;
+
+	vmw_kms_helper_buffer_revert(ctx->buf);
+	vmw_dmabuf_unreference(&ctx->buf);
 	vmw_resource_unreserve(res, false, NULL, 0);
 	mutex_unlock(&res->dev_priv->cmdbuf_mutex);
 }
@@ -2536,10 +2538,14 @@ void vmw_kms_helper_resource_revert(struct vmw_resource *res)
  * interrupted by a signal.
  */
 int vmw_kms_helper_resource_prepare(struct vmw_resource *res,
-				    bool interruptible)
+				    bool interruptible,
+				    struct vmw_validation_ctx *ctx)
 {
 	int ret = 0;
 
+	ctx->buf = NULL;
+	ctx->res = res;
+
 	if (interruptible)
 		ret = mutex_lock_interruptible(&res->dev_priv->cmdbuf_mutex);
 	else
@@ -2558,6 +2564,8 @@ int vmw_kms_helper_resource_prepare(struct vmw_resource *res,
 						    res->dev_priv->has_mob);
 		if (ret)
 			goto out_unreserve;
+
+		ctx->buf = vmw_dmabuf_reference(res->backup);
 	}
 	ret = vmw_resource_validate(res);
 	if (ret)
@@ -2565,7 +2573,7 @@ int vmw_kms_helper_resource_prepare(struct vmw_resource *res,
 	return 0;
 
 out_revert:
-	vmw_kms_helper_buffer_revert(res->backup);
+	vmw_kms_helper_buffer_revert(ctx->buf);
 out_unreserve:
 	vmw_resource_unreserve(res, false, NULL, 0);
 out_unlock:
@@ -2581,11 +2589,13 @@ int vmw_kms_helper_resource_prepare(struct vmw_resource *res,
  * @out_fence: Optional pointer to a fence pointer. If non-NULL, a
  * ref-counted fence pointer is returned here.
  */
-void vmw_kms_helper_resource_finish(struct vmw_resource *res,
-			     struct vmw_fence_obj **out_fence)
+void vmw_kms_helper_resource_finish(struct vmw_validation_ctx *ctx,
+				    struct vmw_fence_obj **out_fence)
 {
-	if (res->backup || out_fence)
-		vmw_kms_helper_buffer_finish(res->dev_priv, NULL, res->backup,
+	struct vmw_resource *res = ctx->res;
+
+	if (ctx->buf || out_fence)
+		vmw_kms_helper_buffer_finish(res->dev_priv, NULL, ctx->buf,
 					     out_fence, NULL);
 
 	vmw_resource_unreserve(res, false, NULL, 0);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index 948362c43c73..3d2ca280eaa7 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -240,6 +240,11 @@ struct vmw_display_unit {
 	int set_gui_y;
 };
 
+struct vmw_validation_ctx {
+	struct vmw_resource *res;
+	struct vmw_dma_buffer *buf;
+};
+
 #define vmw_crtc_to_du(x) \
 	container_of(x, struct vmw_display_unit, crtc)
 #define vmw_connector_to_du(x) \
@@ -296,9 +301,10 @@ void vmw_kms_helper_buffer_finish(struct vmw_private *dev_priv,
 				  struct drm_vmw_fence_rep __user *
 				  user_fence_rep);
 int vmw_kms_helper_resource_prepare(struct vmw_resource *res,
-				    bool interruptible);
-void vmw_kms_helper_resource_revert(struct vmw_resource *res);
-void vmw_kms_helper_resource_finish(struct vmw_resource *res,
+				    bool interruptible,
+				    struct vmw_validation_ctx *ctx);
+void vmw_kms_helper_resource_revert(struct vmw_validation_ctx *ctx);
+void vmw_kms_helper_resource_finish(struct vmw_validation_ctx *ctx,
 				    struct vmw_fence_obj **out_fence);
 int vmw_kms_readback(struct vmw_private *dev_priv,
 		     struct drm_file *file_priv,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index 63a4cd794b73..3ec9eae831b8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -909,12 +909,13 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv,
 	struct vmw_framebuffer_surface *vfbs =
 		container_of(framebuffer, typeof(*vfbs), base);
 	struct vmw_kms_sou_surface_dirty sdirty;
+	struct vmw_validation_ctx ctx;
 	int ret;
 
 	if (!srf)
 		srf = &vfbs->surface->res;
 
-	ret = vmw_kms_helper_resource_prepare(srf, true);
+	ret = vmw_kms_helper_resource_prepare(srf, true, &ctx);
 	if (ret)
 		return ret;
 
@@ -933,7 +934,7 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv,
 	ret = vmw_kms_helper_dirty(dev_priv, framebuffer, clips, vclips,
 				   dest_x, dest_y, num_clips, inc,
 				   &sdirty.base);
-	vmw_kms_helper_resource_finish(srf, out_fence);
+	vmw_kms_helper_resource_finish(&ctx, out_fence);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index b68d74888ab1..6b969e5dea2a 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -980,12 +980,13 @@ int vmw_kms_stdu_surface_dirty(struct vmw_private *dev_priv,
 	struct vmw_framebuffer_surface *vfbs =
 		container_of(framebuffer, typeof(*vfbs), base);
 	struct vmw_stdu_dirty sdirty;
+	struct vmw_validation_ctx ctx;
 	int ret;
 
 	if (!srf)
 		srf = &vfbs->surface->res;
 
-	ret = vmw_kms_helper_resource_prepare(srf, true);
+	ret = vmw_kms_helper_resource_prepare(srf, true, &ctx);
 	if (ret)
 		return ret;
 
@@ -1008,7 +1009,7 @@ int vmw_kms_stdu_surface_dirty(struct vmw_private *dev_priv,
 				   dest_x, dest_y, num_clips, inc,
 				   &sdirty.base);
 out_finish:
-	vmw_kms_helper_resource_finish(srf, out_fence);
+	vmw_kms_helper_resource_finish(&ctx, out_fence);
 
 	return ret;
 }
-- 
2.14.3

^ permalink raw reply related	[flat|nested] 3+ messages in thread
* [PATCH] drm/vmwgfx: Fix black screen and device errors when running without fbdev
  2018-03-21  9:53 [PATCH] drm/vmwgfx: Fix a destoy-while-held mutex problem Thomas Hellstrom
@ 2018-03-21  9:53   ` Thomas Hellstrom
  0 siblings, 0 replies; 3+ messages in thread
From: Thomas Hellstrom @ 2018-03-21  9:53 UTC (permalink / raw)
  To: dri-devel; +Cc: linux-graphics-maintainer, Thomas Hellstrom, stable

When we are running without fbdev, transitioning from the login screen to
X or gnome-shell/wayland will cause a vt switch and the driver will disable
svga mode, losing all modesetting resources. However, the kms atomic state
does not reflect that and may think that a crtc is still turned on, which
will cause device errors when we try to bind an fb to the crtc, and the
screen will remain black.

Fix this by turning off all kms resources before disabling svga mode.

Cc: <stable@vger.kernel.org>
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>
---
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 13 +++++++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.h |  1 +
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 11 +++++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.h |  1 -
 4 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 184340d486c3..86d25f18aa99 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -1337,6 +1337,19 @@ static void __vmw_svga_disable(struct vmw_private *dev_priv)
  */
 void vmw_svga_disable(struct vmw_private *dev_priv)
 {
+	/*
+	 * Disabling SVGA will turn off device modesetting capabilities, so
+	 * notify KMS about that so that it doesn't cache atomic state that
+	 * isn't valid anymore, for example crtcs turned on.
+	 * Strictly we'd want to do this under the SVGA lock (or an SVGA mutex),
+	 * but vmw_kms_lost_device() takes the reservation sem and thus we'll
+	 * end up with lock order reversal. Thus, a master may actually perform
+	 * a new modeset just after we call vmw_kms_lost_device() and race with
+	 * vmw_svga_disable(), but that should at worst cause atomic KMS state
+	 * to be inconsistent with the device, causing modesetting problems.
+	 *
+	 */
+	vmw_kms_lost_device(dev_priv->dev);
 	ttm_write_lock(&dev_priv->reservation_sem, false);
 	spin_lock(&dev_priv->svga_lock);
 	if (dev_priv->bdev.man[TTM_PL_VRAM].use_type) {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index d08753e8fd94..9116fe8baebc 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -938,6 +938,7 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 int vmw_kms_update_layout_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 void vmw_kms_legacy_hotspot_clear(struct vmw_private *dev_priv);
+void vmw_kms_lost_device(struct drm_device *dev);
 
 int vmw_dumb_create(struct drm_file *file_priv,
 		    struct drm_device *dev,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index ead61015cd79..b3d62aa01a9b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -2851,3 +2851,14 @@ int vmw_kms_set_config(struct drm_mode_set *set,
 
 	return drm_atomic_helper_set_config(set, ctx);
 }
+
+
+/**
+ * vmw_kms_lost_device - Notify kms that modesetting capabilities will be lost
+ *
+ * @dev: Pointer to the drm device
+ */
+void vmw_kms_lost_device(struct drm_device *dev)
+{
+	drm_atomic_helper_shutdown(dev);
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index cd9da2dd79af..948362c43c73 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -439,5 +439,4 @@ int vmw_kms_stdu_dma(struct vmw_private *dev_priv,
 
 int vmw_kms_set_config(struct drm_mode_set *set,
 		       struct drm_modeset_acquire_ctx *ctx);
-
 #endif
-- 
2.14.3

^ permalink raw reply related	[flat|nested] 3+ messages in thread
* [PATCH] drm/vmwgfx: Fix black screen and device errors when running without fbdev
@ 2018-03-21  9:53   ` Thomas Hellstrom
  0 siblings, 0 replies; 3+ messages in thread
From: Thomas Hellstrom @ 2018-03-21  9:53 UTC (permalink / raw)
  To: dri-devel; +Cc: Thomas Hellstrom, linux-graphics-maintainer, stable

When we are running without fbdev, transitioning from the login screen to
X or gnome-shell/wayland will cause a vt switch and the driver will disable
svga mode, losing all modesetting resources. However, the kms atomic state
does not reflect that and may think that a crtc is still turned on, which
will cause device errors when we try to bind an fb to the crtc, and the
screen will remain black.

Fix this by turning off all kms resources before disabling svga mode.

Cc: <stable@vger.kernel.org>
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>
---
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 13 +++++++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.h |  1 +
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 11 +++++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.h |  1 -
 4 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 184340d486c3..86d25f18aa99 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -1337,6 +1337,19 @@ static void __vmw_svga_disable(struct vmw_private *dev_priv)
  */
 void vmw_svga_disable(struct vmw_private *dev_priv)
 {
+	/*
+	 * Disabling SVGA will turn off device modesetting capabilities, so
+	 * notify KMS about that so that it doesn't cache atomic state that
+	 * isn't valid anymore, for example crtcs turned on.
+	 * Strictly we'd want to do this under the SVGA lock (or an SVGA mutex),
+	 * but vmw_kms_lost_device() takes the reservation sem and thus we'll
+	 * end up with lock order reversal. Thus, a master may actually perform
+	 * a new modeset just after we call vmw_kms_lost_device() and race with
+	 * vmw_svga_disable(), but that should at worst cause atomic KMS state
+	 * to be inconsistent with the device, causing modesetting problems.
+	 *
+	 */
+	vmw_kms_lost_device(dev_priv->dev);
 	ttm_write_lock(&dev_priv->reservation_sem, false);
 	spin_lock(&dev_priv->svga_lock);
 	if (dev_priv->bdev.man[TTM_PL_VRAM].use_type) {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index d08753e8fd94..9116fe8baebc 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -938,6 +938,7 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 int vmw_kms_update_layout_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 void vmw_kms_legacy_hotspot_clear(struct vmw_private *dev_priv);
+void vmw_kms_lost_device(struct drm_device *dev);
 
 int vmw_dumb_create(struct drm_file *file_priv,
 		    struct drm_device *dev,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index ead61015cd79..b3d62aa01a9b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -2851,3 +2851,14 @@ int vmw_kms_set_config(struct drm_mode_set *set,
 
 	return drm_atomic_helper_set_config(set, ctx);
 }
+
+
+/**
+ * vmw_kms_lost_device - Notify kms that modesetting capabilities will be lost
+ *
+ * @dev: Pointer to the drm device
+ */
+void vmw_kms_lost_device(struct drm_device *dev)
+{
+	drm_atomic_helper_shutdown(dev);
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index cd9da2dd79af..948362c43c73 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -439,5 +439,4 @@ int vmw_kms_stdu_dma(struct vmw_private *dev_priv,
 
 int vmw_kms_set_config(struct drm_mode_set *set,
 		       struct drm_modeset_acquire_ctx *ctx);
-
 #endif
-- 
2.14.3

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

^ permalink raw reply related	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-03-21  9:54 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-03-21  9:53 [PATCH] drm/vmwgfx: Fix a destoy-while-held mutex problem Thomas Hellstrom
2018-03-21  9:53 ` [PATCH] drm/vmwgfx: Fix black screen and device errors when running without fbdev Thomas Hellstrom
2018-03-21  9:53   ` Thomas Hellstrom

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.