drm/nouveau: set irq_enabled manually
authorIlia Mirkin <imirkin@alum.mit.edu>
Thu, 30 Jan 2014 00:53:00 +0000 (19:53 -0500)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 7 Mar 2014 05:30:00 +0000 (21:30 -0800)
commit 7d3428cd4b2ad51af86fdbdf8284ca38fa95e601 upstream.

Since commit 0fa9061ae8c ("drm/nouveau/mc: handle irq-related setup
ourselves"), drm_device->irq_enabled remained unset. This is needed in
order to properly wait for a vblank event in the generic drm code.

See https://bugs.freedesktop.org/show_bug.cgi?id=74195

Reported-by: Jan Janecek <janjanjanx@gmail.com>
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/gpu/drm/nouveau/nouveau_drm.c

index 383f4e6ea9d164c59509c7788666cf89fb6abbe0..4598a6afea1e4945f690013ad2c746680906e56f 100644 (file)
@@ -339,6 +339,8 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
        if (ret)
                goto fail_device;
 
+       dev->irq_enabled = true;
+
        /* workaround an odd issue on nvc1 by disabling the device's
         * nosnoop capability.  hopefully won't cause issues until a
         * better fix is found - assuming there is one...
@@ -426,6 +428,7 @@ nouveau_drm_remove(struct pci_dev *pdev)
        struct nouveau_drm *drm = nouveau_drm(dev);
        struct nouveau_object *device;
 
+       dev->irq_enabled = false;
        device = drm->client.base.device;
        drm_put_dev(dev);