linux-stable.git/drivers/gpu, branch linux-5.12.y

drm/amdkfd: fix sysfs kobj leak

2021-07-20T14:02:16+00:00

[ Upstream commit dcdb4d904b4bd3078fe8d4d24b1658560d6078ef ]

3 cases of kobj leak, which causes memory leak:

kobj_type must have release() method to free memory from release
callback. Don't need NULL default_attrs to init kobj.

sysfs files created under kobj_status should be removed with kobj_status
as parent kobject.

Remove queue sysfs files when releasing queue from process MMU notifier
release callback.

Signed-off-by: Philip Yang 
Reviewed-by: Felix Kuehling 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu: fix Navi1x tcp power gating hang when issuing lightweight invalidaiton

2021-07-20T14:02:16+00:00

[ Upstream commit 9c26ddb1c5b6e30c6bca48b8ad9205d96efe93d0 ]

Fix TCP hang when a lightweight invalidation happens on Navi1x.

Signed-off-by: Evan Quan 
Reviewed-by: Lijo Lazar 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/gma500: Add the missed drm_gem_object_put() in psb_user_framebuffer_create()

2021-07-20T14:02:15+00:00

[ Upstream commit cd8f318fbd266b127ffc93cc4c1eaf9a5196fafb ]

psb_user_framebuffer_create() misses to call drm_gem_object_put() in an
error path. Add the missed function call to fix it.

Signed-off-by: Jing Xiangfeng 
Signed-off-by: Daniel Vetter 
Link: https://patchwork.freedesktop.org/patch/msgid/20210629115956.15160-1-jingxiangfeng@huawei.com
Signed-off-by: Sasha Levin

drm/dp_mst: Add missing drm parameters to recently added call to drm_dbg_kms()

2021-07-20T14:02:02+00:00

commit 24ff3dc18b99c4b912ab1746e803ddb3be5ced4c upstream.

Commit 3769e4c0af5b ("drm/dp_mst: Avoid to mess up payload table by
ports in stale topology") added to calls to drm_dbg_kms() but it
missed the first parameter, the drm device breaking the build.

Fixes: 3769e4c0af5b ("drm/dp_mst: Avoid to mess up payload table by ports in stale topology")
Cc: Wayne Lin 
Cc: Lyude Paul 
Cc: dri-devel@lists.freedesktop.org
Cc: stable@vger.kernel.org
Signed-off-by: José Roberto de Souza 
Reviewed-by: Lyude Paul 
Signed-off-by: Lyude Paul 
Link: https://patchwork.freedesktop.org/patch/msgid/20210616194415.36926-1-jose.souza@intel.com
Signed-off-by: Greg Kroah-Hartman

drm/dp_mst: Avoid to mess up payload table by ports in stale topology

2021-07-20T14:02:02+00:00

commit 3769e4c0af5b82c8ea21d037013cb9564dfaa51f upstream.

[Why]
After unplug/hotplug hub from the system, userspace might start to
clear stale payloads gradually. If we call drm_dp_mst_deallocate_vcpi()
to release stale VCPI of those ports which are not relating to current
topology, we have chane to wrongly clear active payload table entry for
current topology.

E.g.
We have allocated VCPI 1 in current payload table and we call
drm_dp_mst_deallocate_vcpi() to clean VCPI 1 in stale topology. In
drm_dp_mst_deallocate_vcpi(), it will call drm_dp_mst_put_payload_id()
tp put VCPI 1 and which means ID 1 is available again. Thereafter, if we
want to allocate a new payload stream, it will find ID 1 is available by
drm_dp_mst_assign_payload_id(). However, ID 1 is being used

[How]
Check target sink is relating to current topology or not before doing
any payload table update.
Searching upward to find the target sink's relevant root branch device.
If the found root branch device is not the same root of current
topology, don't update payload table.

Changes since v1:
* Change debug macro to use drm_dbg_kms() instead
* Amend the commit message to add Cc tag.

Signed-off-by: Wayne Lin 
Cc: stable@vger.kernel.org
Signed-off-by: Lyude Paul 
Link: https://patchwork.freedesktop.org/patch/msgid/20210616035501.3776-3-Wayne.Lin@amd.com
Reviewed-by: Lyude Paul 
Signed-off-by: Greg Kroah-Hartman

drm/dp_mst: Do not set proposed vcpi directly

2021-07-20T14:02:02+00:00

commit 35d3e8cb35e75450f87f87e3d314e2d418b6954b upstream.

[Why]
When we receive CSN message to notify one port is disconnected, we will
implicitly set its corresponding num_slots to 0. Later on, we will
eventually call drm_dp_update_payload_part1() to arrange down streams.

In drm_dp_update_payload_part1(), we iterate over all proposed_vcpis[]
to do the update. Not specific to a target sink only. For example, if we
light up 2 monitors, Monitor_A and Monitor_B, and then we unplug
Monitor_B. Later on, when we call drm_dp_update_payload_part1() to try
to update payload for Monitor_A, we'll also implicitly clean payload for
Monitor_B at the same time. And finally, when we try to call
drm_dp_update_payload_part1() to clean payload for Monitor_B, we will do
nothing at this time since payload for Monitor_B has been cleaned up
previously.

For StarTech 1to3 DP hub, it seems like if we didn't update DPCD payload
ID table then polling for "ACT Handled"(BIT_1 of DPCD 002C0h) will fail
and this polling will last for 3 seconds.

Therefore, guess the best way is we don't set the proposed_vcpi[]
diretly. Let user of these herlper functions to set the proposed_vcpi
directly.

[How]
1. Revert commit 7617e9621bf2 ("drm/dp_mst: clear time slots for ports
invalid")
2. Tackle the issue in previous commit by skipping those trasient
proposed VCPIs. These stale VCPIs shoulde be explicitly cleared by
user later on.

Changes since v1:
* Change debug macro to use drm_dbg_kms() instead
* Amend the commit message to add Fixed & Cc tags

Signed-off-by: Wayne Lin 
Fixes: 7617e9621bf2 ("drm/dp_mst: clear time slots for ports invalid")
Cc: Lyude Paul 
Cc: Wayne Lin 
Cc: Maarten Lankhorst 
Cc: Maxime Ripard 
Cc: Thomas Zimmermann 
Cc: dri-devel@lists.freedesktop.org
Cc:  # v5.5+
Signed-off-by: Lyude Paul 
Link: https://patchwork.freedesktop.org/patch/msgid/20210616035501.3776-2-Wayne.Lin@amd.com
Reviewed-by: Lyude Paul 
Signed-off-by: Greg Kroah-Hartman

drm/i915/gt: Fix -EDEADLK handling regression

2021-07-20T14:02:01+00:00

commit 2feeb52859fc1ab94cd35b61ada3a6ac4ff24243 upstream.

The conversion to ww mutexes failed to address the fence code which
already returns -EDEADLK when we run out of fences. Ww mutexes on
the other hand treat -EDEADLK as an internal errno value indicating
a need to restart the operation due to a deadlock. So now when the
fence code returns -EDEADLK the higher level code erroneously
restarts everything instead of returning the error to userspace
as is expected.

To remedy this let's switch the fence code to use a different errno
value for this. -ENOBUFS seems like a semi-reasonable unique choice.
Apart from igt the only user of this I could find is sna, and even
there all we do is dump the current fence registers from debugfs
into the X server log. So no user visible functionality is affected.
If we really cared about preserving this we could of course convert
back to -EDEADLK higher up, but doesn't seem like that's worth
the hassle here.

Not quite sure which commit specifically broke this, but I'll
just attribute it to the general gem ww mutex work.

Cc: stable@vger.kernel.org
Cc: Maarten Lankhorst 
Cc: Thomas Hellström 
Testcase: igt/gem_pread/exhaustion
Testcase: igt/gem_pwrite/basic-exhaustion
Testcase: igt/gem_fenced_exec_thrash/too-many-fences
Fixes: 80f0b679d6f0 ("drm/i915: Add an implementation for i915_gem_ww_ctx locking, v2.")
Signed-off-by: Ville Syrjälä 
Link: https://patchwork.freedesktop.org/patch/msgid/20210630164413.25481-1-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst 
(cherry picked from commit 78d2ad7eb4e1f0e9cd5d79788446b6092c21d3e0)
Signed-off-by: Rodrigo Vivi 
Signed-off-by: Greg Kroah-Hartman

drm/i915/gtt: drop the page table optimisation

2021-07-20T14:02:01+00:00

commit 0abb33bfca0fb74df76aac03e90ce685016ef7be upstream.

We skip filling out the pt with scratch entries if the va range covers
the entire pt, since we later have to fill it with the PTEs for the
object pages anyway. However this might leave open a small window where
the PTEs don't point to anything valid for the HW to consume.

When for example using 2M GTT pages this fill_px() showed up as being
quite significant in perf measurements, and ends up being completely
wasted since we ignore the pt and just use the pde directly.

Anyway, currently we have our PTE construction split between alloc and
insert, which is probably slightly iffy nowadays, since the alloc
doesn't actually allocate anything anymore, instead it just sets up the
page directories and points the PTEs at the scratch page. Later when we
do the insert step we re-program the PTEs again. Better might be to
squash the alloc and insert into a single step, then bringing back this
optimisation(along with some others) should be possible.

Fixes: 14826673247e ("drm/i915: Only initialize partially filled pagetables")
Signed-off-by: Matthew Auld 
Cc: Jon Bloomfield 
Cc: Chris Wilson 
Cc: Daniel Vetter 
Cc:  # v4.15+
Reviewed-by: Daniel Vetter 
Link: https://patchwork.freedesktop.org/patch/msgid/20210713130431.2392740-1-matthew.auld@intel.com
(cherry picked from commit 8f88ca76b3942d82e2c1cea8735ec368d89ecc15)
Signed-off-by: Rodrigo Vivi 
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: add another Renoir DID

2021-07-20T14:02:01+00:00

commit 775da83005cb61d4c213c636df9337da05714ff1 upstream.

Add new PCI device id.

Signed-off-by: Jinzhou Su 
Reviewed-by: Huang Rui 
Reviewed-by: Alex Deucher 
Signed-off-by: Alex Deucher 
Cc: stable@vger.kernel.org # 5.11.x
Signed-off-by: Greg Kroah-Hartman

drm/ast: Remove reference to struct drm_device.pdev

2021-07-19T08:01:27+00:00

commit 0ecb51824e838372e01330752503ddf9c0430ef7 upstream.

Using struct drm_device.pdev is deprecated. Upcast with to_pci_dev()
from struct drm_device.dev to get the PCI device structure.

v9:
	* fix remaining pdev references

Signed-off-by: Thomas Zimmermann 
Reviewed-by: Michael J. Ruhl 
Fixes: ba4e0339a6a3 ("drm/ast: Fixed CVE for DP501")
Cc: KuoHsiang Chou 
Cc: kernel test robot 
Cc: Thomas Zimmermann 
Cc: Dave Airlie 
Cc: dri-devel@lists.freedesktop.org
Link: https://patchwork.freedesktop.org/patch/msgid/20210429105101.25667-2-tzimmermann@suse.de
Signed-off-by: Greg Kroah-Hartman