Hi
On Wed, Jul 10, 2013 at 2:11 PM, Daniel Vetter daniel.vetter@ffwll.ch wrote:
We kzalloc the driver node at init time, so no need to do this again.
Signed-off-by: Daniel Vetter daniel.vetter@ffwll.ch
drivers/gpu/drm/drm_fops.c | 9 --------- 1 file changed, 9 deletions(-)
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c index 57e3014..9610997 100644 --- a/drivers/gpu/drm/drm_fops.c +++ b/drivers/gpu/drm/drm_fops.c @@ -73,15 +73,6 @@ static int drm_setup(struct drm_device * dev) for (i = 0; i < ARRAY_SIZE(dev->counts); i++) atomic_set(&dev->counts[i], 0);
dev->sigdata.lock = NULL;dev->context_flag = 0;dev->last_context = 0;dev->last_checked = 0;dev->if_version = 0;dev->buf_async = NULL;
drm_setup() is called on every first open. I digged through core DRM and noticed that we have a lot of code that cleans up on lastclose and reinitializes on firstopen which is skipped for non-UMS drivers to preserve KMS state if no user keeps DRM node open. I would like to hear a comment in the commit-message which explains why we don't reset "if_version", anymore. Are you sure UMS user-space does not depend on this logic?
I have no idea of UMS, but if that reinitialization on lastclose+firstopen is not relied on, I guess we can remove a lot more code. Just wanted to make sure that this is right.
Cheers David
DRM_DEBUG("\n"); /*-- 1.8.3.2
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel