Hi Zack,
Please make sure to always cc dri-devel and/or Daniel on these so if I'm away they don't get lost, but also so that they make it into patchwork for processing.
If you have a chance can you resend it, I'll see if we can process this out of band and get it to Linus before release.
Dave.
On Thu, 15 Apr 2021 at 07:14, Zack Rusin zackr@vmware.com wrote:
The following changes since commit d434405aaab7d0ebc516b68a8fc4100922d7f5ef:
Linux 5.12-rc7 (2021-04-11 15:16:13 -0700)
are available in the Git repository at:
git@gitlab.freedesktop.org:zack/vmwgfx.git tags/vmwgfx-fixes-2021-04-14
for you to fetch changes up to 2ef4fb92363c44e8a6f93fd0877b6a7dee6f874d:
drm/vmwgfx: Make sure bo's are unpinned before putting them back (2021-04-14 16:41:31 -0400)
vmwgfx fixes for regressions in 5.12
Here's a set of 3 patches fixing ugly regressions in the vmwgfx driver. We broke lock initialization code and ended up using spinlocks before initialization breaking lockdep. Also there was a bit of a fallout from drm changes which made the core validate that unreferenced buffers have been unpinned. vmwgfx pinning code predates a lot of the core drm and wasn't written to account for those semantics. Fortunately changes required to fix it are not too intrusive. The changes have been validated by our internal ci.
Signed-off-by: Zack Rusin zackr@vmware.com
Zack Rusin (3): drm/vmwgfx: Make sure we unpin no longer needed buffers drm/vmwgfx: Fix the lockdep breakage drm/vmwgfx: Make sure bo's are unpinned before putting them back
drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c | 4 ++++ drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 17 ++++++++--------- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 5 +---- drivers/gpu/drm/vmwgfx/vmwgfx_mob.c | 14 ++++++++++++++ 4 files changed, 27 insertions(+), 13 deletions(-)
On Sun, Apr 18, 2021 at 1:03 AM Dave Airlie airlied@gmail.com wrote:
Hi Zack,
Please make sure to always cc dri-devel and/or Daniel on these so if I'm away they don't get lost, but also so that they make it into patchwork for processing.
If you have a chance can you resend it, I'll see if we can process this out of band and get it to Linus before release.
Or use drm-misc-fixes train, usually it leaves on schedul :-) -Daniel
Dave.
On Thu, 15 Apr 2021 at 07:14, Zack Rusin zackr@vmware.com wrote:
The following changes since commit d434405aaab7d0ebc516b68a8fc4100922d7f5ef:
Linux 5.12-rc7 (2021-04-11 15:16:13 -0700)
are available in the Git repository at:
git@gitlab.freedesktop.org:zack/vmwgfx.git tags/vmwgfx-fixes-2021-04-14
for you to fetch changes up to 2ef4fb92363c44e8a6f93fd0877b6a7dee6f874d:
drm/vmwgfx: Make sure bo's are unpinned before putting them back (2021-04-14 16:41:31 -0400)
vmwgfx fixes for regressions in 5.12
Here's a set of 3 patches fixing ugly regressions in the vmwgfx driver. We broke lock initialization code and ended up using spinlocks before initialization breaking lockdep. Also there was a bit of a fallout from drm changes which made the core validate that unreferenced buffers have been unpinned. vmwgfx pinning code predates a lot of the core drm and wasn't written to account for those semantics. Fortunately changes required to fix it are not too intrusive. The changes have been validated by our internal ci.
Signed-off-by: Zack Rusin zackr@vmware.com
Zack Rusin (3): drm/vmwgfx: Make sure we unpin no longer needed buffers drm/vmwgfx: Fix the lockdep breakage drm/vmwgfx: Make sure bo's are unpinned before putting them back
drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c | 4 ++++ drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 17 ++++++++--------- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 5 +---- drivers/gpu/drm/vmwgfx/vmwgfx_mob.c | 14 ++++++++++++++ 4 files changed, 27 insertions(+), 13 deletions(-)
On 4/17/21 7:02 PM, Dave Airlie wrote:
Hi Zack,
Please make sure to always cc dri-devel and/or Daniel on these so if I'm away they don't get lost, but also so that they make it into patchwork for processing.
If you have a chance can you resend it, I'll see if we can process this out of band and get it to Linus before release.
Hi, Dave.
Thank you! I just noticed that you've sent it out. Would you still like me to resend it through dri-devel? The patches all went through dri-devel already, I just assumed that from dri-misc docs that drm-misc-fixes wouldn't be merged again until after 5.13 was out which is why I went directly.
z
On Wed, Apr 21, 2021 at 7:00 PM Zack Rusin zackr@vmware.com wrote:
On 4/17/21 7:02 PM, Dave Airlie wrote:
Hi Zack,
Please make sure to always cc dri-devel and/or Daniel on these so if I'm away they don't get lost, but also so that they make it into patchwork for processing.
If you have a chance can you resend it, I'll see if we can process this out of band and get it to Linus before release.
Hi, Dave.
Thank you! I just noticed that you've sent it out. Would you still like me to resend it through dri-devel? The patches all went through dri-devel already, I just assumed that from dri-misc docs that drm-misc-fixes wouldn't be merged again until after 5.13 was out which is why I went directly.
drm-misc-fixes will get pulled into the merge window when it misses the last release (which would be 5.12). drm-misc-next-fixes is the special branch which is for fixes during the merge window time, for issues in drm-next (so for 5.13).
I know that ideally maintainters would just cherry-pick bugfixes to the right release branch (like on mesa or xserver or well any other reasonable project), but since drm-misc is volunteer run committers need to pick the right branch. But aside from that drm-misc is always open for bugfixes (and also always open for feature work). -Daniel
dri-devel@lists.freedesktop.org