Skip to contentSkip to conversation

drm: fix double drm_put_minor() in fail paths

PATCH In review Rob Clark 2 participants 3 messages Dec 12, 2013

Subsystems (2)
lore
Share & tools
lore

Rob Clark

author
#0 lore raw

PATCH 1/1 drm: fix double drm_put_minor() in fail paths

If driver failed to load (for example, -EPROBE_DEFER), we'd end up doing drm_put_minor() both from drm_dev_register() and drm_dev_free(), the second time with a bogus pointer.

---
 drivers/gpu/drm/drm_stub.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/drm_stub.c +3 -0
index f53d524..f245d99 100644
@@ -567,10 +567,13 @@ err_unload:
 		dev->driver->unload(dev);
 err_primary_node:
 	drm_put_minor(dev->primary);
+	dev->primary = NULL;
 err_render_node:
 	drm_put_minor(dev->render);
+	dev->render = NULL;
 err_control_node:
 	drm_put_minor(dev->control);
+	dev->control = NULL;
 err_agp:
 	if (dev->driver->bus->agp_destroy)
 		dev->driver->bus->agp_destroy(dev);
Signature · 1 line
1.8.4.2
#1 lore raw
Quoted 3 lines · Rob Clark

If driver failed to load (for example, -EPROBE_DEFER), we'd end up doing drm_put_minor() both from drm_dev_register() and drm_dev_free(), the second time with a bogus pointer.

FYI, I sent a similar patch ~week ago that changed put_minor into unplug_minor per David Herrmann's advice.

Quoted 31 lines
---
 drivers/gpu/drm/drm_stub.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/drm_stub.c +3 -1
index f53d524..f245d99 100644
@@ -567,10 +567,13 @@ err_unload:
                dev->driver->unload(dev);
 err_primary_node:
        drm_put_minor(dev->primary);
+       dev->primary = NULL;
 err_render_node:
        drm_put_minor(dev->render);
+       dev->render = NULL;
 err_control_node:
        drm_put_minor(dev->control);
+       dev->control = NULL;
 err_agp:
        if (dev->driver->bus->agp_destroy)
                dev->driver->bus->agp_destroy(dev);
--

1.8.4.2

_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel

author
#2 lore raw
Quoted 7 lines · Ilia Mirkin
Quoted 3 lines · Rob Clark

If driver failed to load (for example, -EPROBE_DEFER), we'd end up doing drm_put_minor() both from drm_dev_register() and drm_dev_free(), the second time with a bogus pointer.

FYI, I sent a similar patch ~week ago that changed put_minor into unplug_minor per David Herrmann's advice.

ahh, ok.. I didn't see it, but if it fixes the same issue we can drop my patch..

BR,
-R

Quoted 31 lines
Quoted 31 lines
---
 drivers/gpu/drm/drm_stub.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/drm_stub.c +3 -1
index f53d524..f245d99 100644
@@ -567,10 +567,13 @@ err_unload:
                dev->driver->unload(dev);
 err_primary_node:
        drm_put_minor(dev->primary);
+       dev->primary = NULL;
 err_render_node:
        drm_put_minor(dev->render);
+       dev->render = NULL;
 err_control_node:
        drm_put_minor(dev->control);
+       dev->control = NULL;
 err_agp:
        if (dev->driver->bus->agp_destroy)
                dev->driver->bus->agp_destroy(dev);
--

1.8.4.2

_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel