* [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
@ 2026-07-30 5:36 Nitin Gote
2026-07-30 5:59 ` Raag Jadav
2026-07-30 12:35 ` Matthew Auld
0 siblings, 2 replies; 17+ messages in thread
From: Nitin Gote @ 2026-07-30 5:36 UTC (permalink / raw)
To: intel-xe; +Cc: Nitin Gote, Matthew Brost, Matthew Auld
After hot-unplug, device_del() disables runtime PM. A deferred fd close
can still run later during process exit and destroy remaining exec queues,
which may call xe_pm_runtime_get_noresume().
At this point pm_runtime_get_if_in_use() fails because runtime PM has
already been disabled, so xe_pm_runtime_get_noresume() warns with
"Missing outer runtime PM protection". This is expected after hot-unplug
and does not indicate a missing outer runtime PM guard from the caller.
Skip the warning when the DRM device is already unplugged.
Observed with new IGT core_hotunplug subtests:
igt@core_hotunplug@hotreplug-with-load
igt@core_hotunplug@hotunplug-rescan-with-load
v2: Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
and instead exclude hot-unplug from the WARN in
xe_pm_runtime_get_noresume(). (Matthew Brost)
Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Assisted-by: GitHub-Copilot:claude-sonnet-4.6
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index a5289a9df8d2..a038687d5b2c 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -10,6 +10,7 @@
#include <linux/suspend.h>
#include <linux/dmi.h>
+#include <drm/drm_drv.h>
#include <drm/drm_managed.h>
#include <drm/ttm/ttm_placement.h>
@@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
if (!ref) {
pm_runtime_get_noresume(xe->drm.dev);
- drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
+ drm_WARN(&xe->drm,
+ !drm_dev_is_unplugged(&xe->drm) &&
+ !xe_pm_suspending_or_resuming(xe),
"Missing outer runtime PM protection\n");
}
}
--
2.50.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 5:36 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
@ 2026-07-30 5:59 ` Raag Jadav
2026-07-30 12:05 ` Gote, Nitin R
2026-07-30 12:35 ` Matthew Auld
1 sibling, 1 reply; 17+ messages in thread
From: Raag Jadav @ 2026-07-30 5:59 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe, Matthew Brost, Matthew Auld
On Thu, Jul 30, 2026 at 11:06:36AM +0530, Nitin Gote wrote:
> After hot-unplug, device_del() disables runtime PM. A deferred fd close
> can still run later during process exit and destroy remaining exec queues,
> which may call xe_pm_runtime_get_noresume().
This sounds like could cause many other side effects. Why not make sure
the exec queues are taken care of as part of hot-unplug path?
Raag
^ permalink raw reply [flat|nested] 17+ messages in thread
* RE: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 5:59 ` Raag Jadav
@ 2026-07-30 12:05 ` Gote, Nitin R
2026-07-30 12:31 ` Raag Jadav
0 siblings, 1 reply; 17+ messages in thread
From: Gote, Nitin R @ 2026-07-30 12:05 UTC (permalink / raw)
To: Jadav, Raag
Cc: intel-xe@lists.freedesktop.org, Brost, Matthew, Auld, Matthew
Hi Raag,
> -----Original Message-----
> From: Jadav, Raag <raag.jadav@intel.com>
> Sent: Thursday, July 30, 2026 11:29 AM
> To: Gote, Nitin R <nitin.r.gote@intel.com>
> Cc: intel-xe@lists.freedesktop.org; Brost, Matthew <matthew.brost@intel.com>;
> Auld, Matthew <matthew.auld@intel.com>
> Subject: Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM
> protection after hot-unplug
>
> On Thu, Jul 30, 2026 at 11:06:36AM +0530, Nitin Gote wrote:
> > After hot-unplug, device_del() disables runtime PM. A deferred fd
> > close can still run later during process exit and destroy remaining
> > exec queues, which may call xe_pm_runtime_get_noresume().
>
> This sounds like could cause many other side effects. Why not make sure the exec
> queues are taken care of as part of hot-unplug path?
>
Here In igt@core_hotunplug@hotreplug-with-load and igt@core_hotunplug@hotunplug-rescan-with-load subtests,
the workload keeps the DRM fd open while the device is pci hot-unplugged. Later, fd close reaches
guc_exec_queue_destroy() -> guc_exec_queue_add_msg() -> xe_pm_runtime_get_noresume(),
after pci_remove/device_del() path has already called pm_runtime_disable() during pci hot-unplug.
So the WARN is a false-positive for the already-unplugged device case.
I will update the commit message to make this scope clearer.
Thank you,
Nitin
> Raag
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 12:05 ` Gote, Nitin R
@ 2026-07-30 12:31 ` Raag Jadav
2026-08-06 6:09 ` Gote, Nitin R
0 siblings, 1 reply; 17+ messages in thread
From: Raag Jadav @ 2026-07-30 12:31 UTC (permalink / raw)
To: Gote, Nitin R
Cc: intel-xe@lists.freedesktop.org, Brost, Matthew, Auld, Matthew
On Thu, Jul 30, 2026 at 05:35:26PM +0530, Gote, Nitin R wrote:
> > From: Jadav, Raag <raag.jadav@intel.com>
> > On Thu, Jul 30, 2026 at 11:06:36AM +0530, Nitin Gote wrote:
> > > After hot-unplug, device_del() disables runtime PM. A deferred fd
> > > close can still run later during process exit and destroy remaining
> > > exec queues, which may call xe_pm_runtime_get_noresume().
> >
> > This sounds like could cause many other side effects. Why not make sure the exec
> > queues are taken care of as part of hot-unplug path?
> >
>
> Here In igt@core_hotunplug@hotreplug-with-load and igt@core_hotunplug@hotunplug-rescan-with-load subtests,
> the workload keeps the DRM fd open while the device is pci hot-unplugged. Later, fd close reaches
> guc_exec_queue_destroy() -> guc_exec_queue_add_msg() -> xe_pm_runtime_get_noresume(),
> after pci_remove/device_del() path has already called pm_runtime_disable() during pci hot-unplug.
>
> So the WARN is a false-positive for the already-unplugged device case.
> I will update the commit message to make this scope clearer.
I guess my question was, is it safe have the exec queues lying around until
fd close? Wouldn't it be more appropriate to have them cleaned up as part
of hot-unplug itself?
Raag
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 5:36 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
2026-07-30 5:59 ` Raag Jadav
@ 2026-07-30 12:35 ` Matthew Auld
2026-08-06 6:10 ` Gote, Nitin R
1 sibling, 1 reply; 17+ messages in thread
From: Matthew Auld @ 2026-07-30 12:35 UTC (permalink / raw)
To: Nitin Gote, intel-xe; +Cc: Matthew Brost
On 30/07/2026 06:36, Nitin Gote wrote:
> After hot-unplug, device_del() disables runtime PM. A deferred fd close
> can still run later during process exit and destroy remaining exec queues,
> which may call xe_pm_runtime_get_noresume().
>
> At this point pm_runtime_get_if_in_use() fails because runtime PM has
> already been disabled, so xe_pm_runtime_get_noresume() warns with
> "Missing outer runtime PM protection". This is expected after hot-unplug
> and does not indicate a missing outer runtime PM guard from the caller.
Is the code that calls xe_pm_runtime_get_noresume(), not trying to
protect some kind of hw access to ensure the device stays awake? I think
post unplug hw access is not allowed outside of the devm teardown
sequence. Did you also double check that where this fires, we don't then
touch the hw?
>
> Skip the warning when the DRM device is already unplugged.
>
> Observed with new IGT core_hotunplug subtests:
> igt@core_hotunplug@hotreplug-with-load
> igt@core_hotunplug@hotunplug-rescan-with-load
>
> v2: Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> and instead exclude hot-unplug from the WARN in
> xe_pm_runtime_get_noresume(). (Matthew Brost)
>
> Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
> Cc: Matthew Brost <matthew.brost@intel.com>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Assisted-by: GitHub-Copilot:claude-sonnet-4.6
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
> index a5289a9df8d2..a038687d5b2c 100644
> --- a/drivers/gpu/drm/xe/xe_pm.c
> +++ b/drivers/gpu/drm/xe/xe_pm.c
> @@ -10,6 +10,7 @@
> #include <linux/suspend.h>
> #include <linux/dmi.h>
>
> +#include <drm/drm_drv.h>
> #include <drm/drm_managed.h>
> #include <drm/ttm/ttm_placement.h>
>
> @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
>
> if (!ref) {
> pm_runtime_get_noresume(xe->drm.dev);
> - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> + drm_WARN(&xe->drm,
> + !drm_dev_is_unplugged(&xe->drm) &&
> + !xe_pm_suspending_or_resuming(xe),
> "Missing outer runtime PM protection\n");
> }
> }
^ permalink raw reply [flat|nested] 17+ messages in thread
* RE: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 12:31 ` Raag Jadav
@ 2026-08-06 6:09 ` Gote, Nitin R
2026-08-06 6:27 ` Raag Jadav
0 siblings, 1 reply; 17+ messages in thread
From: Gote, Nitin R @ 2026-08-06 6:09 UTC (permalink / raw)
To: Jadav, Raag
Cc: intel-xe@lists.freedesktop.org, Brost, Matthew, Auld, Matthew
> -----Original Message-----
> From: Jadav, Raag <raag.jadav@intel.com>
> Sent: Thursday, July 30, 2026 6:01 PM
> To: Gote, Nitin R <nitin.r.gote@intel.com>
> Cc: intel-xe@lists.freedesktop.org; Brost, Matthew <matthew.brost@intel.com>;
> Auld, Matthew <matthew.auld@intel.com>
> Subject: Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM
> protection after hot-unplug
>
> On Thu, Jul 30, 2026 at 05:35:26PM +0530, Gote, Nitin R wrote:
> > > From: Jadav, Raag <raag.jadav@intel.com> On Thu, Jul 30, 2026 at
> > > 11:06:36AM +0530, Nitin Gote wrote:
> > > > After hot-unplug, device_del() disables runtime PM. A deferred fd
> > > > close can still run later during process exit and destroy
> > > > remaining exec queues, which may call xe_pm_runtime_get_noresume().
> > >
> > > This sounds like could cause many other side effects. Why not make
> > > sure the exec queues are taken care of as part of hot-unplug path?
> > >
> >
> > Here In igt@core_hotunplug@hotreplug-with-load and
> > igt@core_hotunplug@hotunplug-rescan-with-load subtests, the workload
> > keeps the DRM fd open while the device is pci hot-unplugged. Later, fd
> > close reaches
> > guc_exec_queue_destroy() -> guc_exec_queue_add_msg() ->
> > xe_pm_runtime_get_noresume(), after pci_remove/device_del() path has
> already called pm_runtime_disable() during pci hot-unplug.
> >
> > So the WARN is a false-positive for the already-unplugged device case.
> > I will update the commit message to make this scope clearer.
>
> I guess my question was, is it safe have the exec queues lying around until fd
> close? Wouldn't it be more appropriate to have them cleaned up as part of hot-
> unplug itself?
>
The exec queues are owned by user space through open DRM fds, so they're freed when the fd is closed, not forced at unplug.
drm_dev_unplug() marks the device unplugged but keeps the drm_device alive while fds are open,
we can't destroy a queue while it's still referenced by an open fd.
The hardware side is already cleaned up at unplug; guc_fini_hw() runs in the remove path and sanitizes GuC.
So by the time the fd closes, only the software state of the queue is left, and that's what gets freed here.
And when it does, the cleanup path checks xe_uc_fw_is_running(), sees GuC is gone, and takes the software only teardown.
Nitin
> Raag
^ permalink raw reply [flat|nested] 17+ messages in thread
* RE: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-07-30 12:35 ` Matthew Auld
@ 2026-08-06 6:10 ` Gote, Nitin R
0 siblings, 0 replies; 17+ messages in thread
From: Gote, Nitin R @ 2026-08-06 6:10 UTC (permalink / raw)
To: Auld, Matthew, intel-xe@lists.freedesktop.org; +Cc: Brost, Matthew
> -----Original Message-----
> From: Auld, Matthew <matthew.auld@intel.com>
> Sent: Thursday, July 30, 2026 6:05 PM
> To: Gote, Nitin R <nitin.r.gote@intel.com>; intel-xe@lists.freedesktop.org
> Cc: Brost, Matthew <matthew.brost@intel.com>
> Subject: Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM
> protection after hot-unplug
>
> On 30/07/2026 06:36, Nitin Gote wrote:
> > After hot-unplug, device_del() disables runtime PM. A deferred fd
> > close can still run later during process exit and destroy remaining
> > exec queues, which may call xe_pm_runtime_get_noresume().
> >
> > At this point pm_runtime_get_if_in_use() fails because runtime PM has
> > already been disabled, so xe_pm_runtime_get_noresume() warns with
> > "Missing outer runtime PM protection". This is expected after
> > hot-unplug and does not indicate a missing outer runtime PM guard from the
> caller.
>
> Is the code that calls xe_pm_runtime_get_noresume(), not trying to protect some
> kind of hw access to ensure the device stays awake? I think post unplug hw
> access is not allowed outside of the devm teardown sequence. Did you also
> double check that where this fires, we don't then touch the hw?
Yes, normally xe_pm_runtime_get_noresume() is there to protect queue teardown paths that may touch HW.
I checked again; the path where this WARN fires. After hot-unplug, GuC has already been sanitized by guc_finit_hw(),
So xe_uc_fw_is_running() is false and the cleanup takes the software only destroy path. No queue related hardware cleanup is reached at that point.
The WARN is only because xe_pm_runtime_get_noresume() still takes a PM ref after runtime PM has already been disabled during device removal.
I'll update the commit message in v3 to clarify that the queue destroy path was checked and does not touch actual hw state after unplug.
Nitin
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-06 6:09 ` Gote, Nitin R
@ 2026-08-06 6:27 ` Raag Jadav
0 siblings, 0 replies; 17+ messages in thread
From: Raag Jadav @ 2026-08-06 6:27 UTC (permalink / raw)
To: Gote, Nitin R
Cc: intel-xe@lists.freedesktop.org, Brost, Matthew, Auld, Matthew
On Thu, Aug 06, 2026 at 11:39:50AM +0530, Gote, Nitin R wrote:
> > From: Jadav, Raag <raag.jadav@intel.com>
> > On Thu, Jul 30, 2026 at 05:35:26PM +0530, Gote, Nitin R wrote:
> > > > From: Jadav, Raag <raag.jadav@intel.com> On Thu, Jul 30, 2026 at
> > > > 11:06:36AM +0530, Nitin Gote wrote:
> > > > > After hot-unplug, device_del() disables runtime PM. A deferred fd
> > > > > close can still run later during process exit and destroy
> > > > > remaining exec queues, which may call xe_pm_runtime_get_noresume().
> > > >
> > > > This sounds like could cause many other side effects. Why not make
> > > > sure the exec queues are taken care of as part of hot-unplug path?
> > > >
> > >
> > > Here In igt@core_hotunplug@hotreplug-with-load and
> > > igt@core_hotunplug@hotunplug-rescan-with-load subtests, the workload
> > > keeps the DRM fd open while the device is pci hot-unplugged. Later, fd
> > > close reaches
> > > guc_exec_queue_destroy() -> guc_exec_queue_add_msg() ->
> > > xe_pm_runtime_get_noresume(), after pci_remove/device_del() path has
> > already called pm_runtime_disable() during pci hot-unplug.
> > >
> > > So the WARN is a false-positive for the already-unplugged device case.
> > > I will update the commit message to make this scope clearer.
> >
> > I guess my question was, is it safe have the exec queues lying around until fd
> > close? Wouldn't it be more appropriate to have them cleaned up as part of hot-
> > unplug itself?
> >
>
> The exec queues are owned by user space through open DRM fds, so they're freed when the fd is closed, not forced at unplug.
> drm_dev_unplug() marks the device unplugged but keeps the drm_device alive while fds are open,
> we can't destroy a queue while it's still referenced by an open fd.
>
> The hardware side is already cleaned up at unplug; guc_fini_hw() runs in the remove path and sanitizes GuC.
> So by the time the fd closes, only the software state of the queue is left, and that's what gets freed here.
> And when it does, the cleanup path checks xe_uc_fw_is_running(), sees GuC is gone, and takes the software only teardown.
Makes sense. Although runtime PM is for when you need to interact with the
hardware, so in my understanding we shouldn't be at this point for software
only teardown, but I'll leave it to you all.
Raag
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-06 8:57 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
@ 2026-08-06 8:30 ` Raag Jadav
2026-08-06 9:38 ` Matthew Brost
2026-08-06 8:56 ` ✓ CI.KUnit: success for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2) Patchwork
` (2 subsequent siblings)
3 siblings, 1 reply; 17+ messages in thread
From: Raag Jadav @ 2026-08-06 8:30 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe, Matthew Brost, Matthew Auld
On Thu, Aug 06, 2026 at 02:27:12PM +0530, Nitin Gote wrote:
> Exec queues are owned by user fds and are destroyed when the fd is closed.
> After hot-unplug this can happen from a deferred close during process
> exit, after the device has been removed and runtime PM has been disabled.
> The queue destroy path can call xe_pm_runtime_get_noresume(), and since
> runtime PM is disabled pm_runtime_get_if_in_use() returns no reference,
> so it warns about "Missing outer runtime PM protection".
>
> This is a false positive for the hot-unplug teardown case. The device is
> already unplugged, and the queue destroy path that triggered this warning
> was checked and does not touch hardware state after unplug. GuC has already
> been sanitized by guc_fini_hw(), so no H2G is sent and the hardware
> teardown path is no longer reachable.
>
> Skip the warning when the DRM device is already unplugged.
>
> Observed with new IGT core_hotunplug subtests:
> igt@core_hotunplug@hotreplug-with-load
> igt@core_hotunplug@hotunplug-rescan-with-load
>
> v2:
> - Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> and instead exclude hot-unplug from the WARN in
> xe_pm_runtime_get_noresume(). (Matthew Brost)
>
> v3:
> - Clarify that the queue destroy path was checked and does not touch
> hardware state after unplug. (Matthew Auld)
>
> Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
> Cc: Matthew Brost <matthew.brost@intel.com>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Assisted-by: GitHub-Copilot:claude-opus-4.8
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
> index a5289a9df8d2..a038687d5b2c 100644
> --- a/drivers/gpu/drm/xe/xe_pm.c
> +++ b/drivers/gpu/drm/xe/xe_pm.c
> @@ -10,6 +10,7 @@
> #include <linux/suspend.h>
> #include <linux/dmi.h>
>
> +#include <drm/drm_drv.h>
> #include <drm/drm_managed.h>
> #include <drm/ttm/ttm_placement.h>
>
> @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
From the documentation
* This function should be used in inner places where it is surely already
* protected by outer-bound callers of `xe_pm_runtime_get`.
> if (!ref) {
> pm_runtime_get_noresume(xe->drm.dev);
> - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> + drm_WARN(&xe->drm,
> + !drm_dev_is_unplugged(&xe->drm) &&
> + !xe_pm_suspending_or_resuming(xe),
So IMO this is more of a band-aid.
Raag
> "Missing outer runtime PM protection\n");
> }
> }
> --
> 2.50.1
>
^ permalink raw reply [flat|nested] 17+ messages in thread
* ✓ CI.KUnit: success for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
2026-08-06 8:57 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
2026-08-06 8:30 ` Raag Jadav
@ 2026-08-06 8:56 ` Patchwork
2026-08-06 9:52 ` ✓ Xe.CI.BAT: " Patchwork
2026-08-06 18:39 ` ✗ Xe.CI.FULL: failure " Patchwork
3 siblings, 0 replies; 17+ messages in thread
From: Patchwork @ 2026-08-06 8:56 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
== Series Details ==
Series: drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
URL : https://patchwork.freedesktop.org/series/171320/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[08:55:30] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[08:55:34] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[08:56:06] Starting KUnit Kernel (1/1)...
[08:56:06] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[08:56:07] ================== guc_buf (11 subtests) ===================
[08:56:07] [PASSED] test_smallest
[08:56:07] [PASSED] test_largest
[08:56:07] [PASSED] test_granular
[08:56:07] [PASSED] test_unique
[08:56:07] [PASSED] test_overlap
[08:56:07] [PASSED] test_reusable
[08:56:07] [PASSED] test_too_big
[08:56:07] [PASSED] test_flush
[08:56:07] [PASSED] test_lookup
[08:56:07] [PASSED] test_data
[08:56:07] [PASSED] test_class
[08:56:07] ===================== [PASSED] guc_buf =====================
[08:56:07] =================== guc_dbm (7 subtests) ===================
[08:56:07] [PASSED] test_empty
[08:56:07] [PASSED] test_default
[08:56:07] ======================== test_size ========================
[08:56:07] [PASSED] 4
[08:56:07] [PASSED] 8
[08:56:07] [PASSED] 32
[08:56:07] [PASSED] 256
[08:56:07] ==================== [PASSED] test_size ====================
[08:56:07] ======================= test_reuse ========================
[08:56:07] [PASSED] 4
[08:56:07] [PASSED] 8
[08:56:07] [PASSED] 32
[08:56:07] [PASSED] 256
[08:56:07] =================== [PASSED] test_reuse ====================
[08:56:07] =================== test_range_overlap ====================
[08:56:07] [PASSED] 4
[08:56:07] [PASSED] 8
[08:56:07] [PASSED] 32
[08:56:07] [PASSED] 256
[08:56:07] =============== [PASSED] test_range_overlap ================
[08:56:07] =================== test_range_compact ====================
[08:56:07] [PASSED] 4
[08:56:07] [PASSED] 8
[08:56:07] [PASSED] 32
[08:56:07] [PASSED] 256
[08:56:07] =============== [PASSED] test_range_compact ================
[08:56:07] ==================== test_range_spare =====================
[08:56:07] [PASSED] 4
[08:56:07] [PASSED] 8
[08:56:07] [PASSED] 32
[08:56:07] [PASSED] 256
[08:56:07] ================ [PASSED] test_range_spare =================
[08:56:07] ===================== [PASSED] guc_dbm =====================
[08:56:07] =================== guc_idm (6 subtests) ===================
[08:56:07] [PASSED] bad_init
[08:56:07] [PASSED] no_init
[08:56:07] [PASSED] init_fini
[08:56:07] [PASSED] check_used
[08:56:07] [PASSED] check_quota
[08:56:07] [PASSED] check_all
[08:56:07] ===================== [PASSED] guc_idm =====================
[08:56:07] =============== guc_klv_helpers (9 subtests) ===============
[08:56:07] [PASSED] test_count
[08:56:07] [PASSED] test_encode_u32
[08:56:07] [PASSED] test_encode_u64
[08:56:07] [PASSED] test_encode_string
[08:56:07] [PASSED] test_encode_object_raw
[08:56:07] [PASSED] test_encode_object_klv
[08:56:07] [PASSED] test_encode_object_nested
[08:56:07] [PASSED] test_encode_object_basic
[08:56:07] [PASSED] test_print
[08:56:07] ================= [PASSED] guc_klv_helpers =================
[08:56:07] ================== no_relay (3 subtests) ===================
[08:56:07] [PASSED] xe_drops_guc2pf_if_not_ready
[08:56:07] [PASSED] xe_drops_guc2vf_if_not_ready
[08:56:07] [PASSED] xe_rejects_send_if_not_ready
[08:56:07] ==================== [PASSED] no_relay =====================
[08:56:07] ================== pf_relay (14 subtests) ==================
[08:56:07] [PASSED] pf_rejects_guc2pf_too_short
[08:56:07] [PASSED] pf_rejects_guc2pf_too_long
[08:56:07] [PASSED] pf_rejects_guc2pf_no_payload
[08:56:07] [PASSED] pf_fails_no_payload
[08:56:07] [PASSED] pf_fails_bad_origin
[08:56:07] [PASSED] pf_fails_bad_type
[08:56:07] [PASSED] pf_txn_reports_error
[08:56:07] [PASSED] pf_txn_sends_pf2guc
[08:56:07] [PASSED] pf_sends_pf2guc
[08:56:07] [SKIPPED] pf_loopback_nop (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[08:56:07] [SKIPPED] pf_loopback_echo (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[08:56:07] [SKIPPED] pf_loopback_fail (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[08:56:07] [SKIPPED] pf_loopback_busy (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[08:56:07] [SKIPPED] pf_loopback_retry (requires CONFIG_DRM_XE_DEBUG_SRIOV)
[08:56:07] ==================== [PASSED] pf_relay =====================
[08:56:07] ================== vf_relay (3 subtests) ===================
[08:56:07] [PASSED] vf_rejects_guc2vf_too_short
[08:56:07] [PASSED] vf_rejects_guc2vf_too_long
[08:56:07] [PASSED] vf_rejects_guc2vf_no_payload
[08:56:07] ==================== [PASSED] vf_relay =====================
[08:56:07] ================ pf_gt_config (9 subtests) =================
[08:56:07] [PASSED] fair_contexts_1vf
[08:56:07] [PASSED] fair_doorbells_1vf
[08:56:07] [PASSED] fair_ggtt_1vf
[08:56:07] ====================== fair_vram_1vf ======================
[08:56:07] [PASSED] 3.50 GiB
[08:56:07] [PASSED] 11.5 GiB
[08:56:07] [PASSED] 15.5 GiB
[08:56:07] [PASSED] 31.5 GiB
[08:56:07] [PASSED] 63.5 GiB
[08:56:07] [PASSED] 1.91 GiB
[08:56:07] ================== [PASSED] fair_vram_1vf ==================
[08:56:07] ================ fair_vram_1vf_admin_only =================
[08:56:07] [PASSED] 3.50 GiB
[08:56:07] [PASSED] 11.5 GiB
[08:56:07] [PASSED] 15.5 GiB
[08:56:07] [PASSED] 31.5 GiB
[08:56:07] [PASSED] 63.5 GiB
[08:56:07] [PASSED] 1.91 GiB
[08:56:07] ============ [PASSED] fair_vram_1vf_admin_only =============
[08:56:07] ====================== fair_contexts ======================
[08:56:07] [PASSED] 1 VF
[08:56:07] [PASSED] 2 VFs
[08:56:07] [PASSED] 3 VFs
[08:56:07] [PASSED] 4 VFs
[08:56:07] [PASSED] 5 VFs
[08:56:07] [PASSED] 6 VFs
[08:56:07] [PASSED] 7 VFs
[08:56:07] [PASSED] 8 VFs
[08:56:07] [PASSED] 9 VFs
[08:56:07] [PASSED] 10 VFs
[08:56:07] [PASSED] 11 VFs
[08:56:07] [PASSED] 12 VFs
[08:56:07] [PASSED] 13 VFs
[08:56:07] [PASSED] 14 VFs
[08:56:07] [PASSED] 15 VFs
[08:56:07] [PASSED] 16 VFs
[08:56:07] [PASSED] 17 VFs
[08:56:07] [PASSED] 18 VFs
[08:56:07] [PASSED] 19 VFs
[08:56:07] [PASSED] 20 VFs
[08:56:07] [PASSED] 21 VFs
[08:56:07] [PASSED] 22 VFs
[08:56:07] [PASSED] 23 VFs
[08:56:07] [PASSED] 24 VFs
[08:56:07] [PASSED] 25 VFs
[08:56:07] [PASSED] 26 VFs
[08:56:07] [PASSED] 27 VFs
[08:56:07] [PASSED] 28 VFs
[08:56:07] [PASSED] 29 VFs
[08:56:07] [PASSED] 30 VFs
[08:56:07] [PASSED] 31 VFs
[08:56:07] [PASSED] 32 VFs
[08:56:07] [PASSED] 33 VFs
[08:56:07] [PASSED] 34 VFs
[08:56:07] [PASSED] 35 VFs
[08:56:07] [PASSED] 36 VFs
[08:56:07] [PASSED] 37 VFs
[08:56:07] [PASSED] 38 VFs
[08:56:07] [PASSED] 39 VFs
[08:56:07] [PASSED] 40 VFs
[08:56:07] [PASSED] 41 VFs
[08:56:07] [PASSED] 42 VFs
[08:56:07] [PASSED] 43 VFs
[08:56:07] [PASSED] 44 VFs
[08:56:07] [PASSED] 45 VFs
[08:56:07] [PASSED] 46 VFs
[08:56:07] [PASSED] 47 VFs
[08:56:07] [PASSED] 48 VFs
[08:56:07] [PASSED] 49 VFs
[08:56:07] [PASSED] 50 VFs
[08:56:07] [PASSED] 51 VFs
[08:56:07] [PASSED] 52 VFs
[08:56:07] [PASSED] 53 VFs
[08:56:07] [PASSED] 54 VFs
[08:56:07] [PASSED] 55 VFs
[08:56:07] [PASSED] 56 VFs
[08:56:07] [PASSED] 57 VFs
[08:56:07] [PASSED] 58 VFs
[08:56:07] [PASSED] 59 VFs
[08:56:07] [PASSED] 60 VFs
[08:56:07] [PASSED] 61 VFs
[08:56:07] [PASSED] 62 VFs
[08:56:07] [PASSED] 63 VFs
[08:56:07] ================== [PASSED] fair_contexts ==================
[08:56:07] ===================== fair_doorbells ======================
[08:56:07] [PASSED] 1 VF
[08:56:07] [PASSED] 2 VFs
[08:56:07] [PASSED] 3 VFs
[08:56:07] [PASSED] 4 VFs
[08:56:07] [PASSED] 5 VFs
[08:56:07] [PASSED] 6 VFs
[08:56:07] [PASSED] 7 VFs
[08:56:07] [PASSED] 8 VFs
[08:56:07] [PASSED] 9 VFs
[08:56:07] [PASSED] 10 VFs
[08:56:07] [PASSED] 11 VFs
[08:56:07] [PASSED] 12 VFs
[08:56:07] [PASSED] 13 VFs
[08:56:07] [PASSED] 14 VFs
[08:56:07] [PASSED] 15 VFs
[08:56:07] [PASSED] 16 VFs
[08:56:07] [PASSED] 17 VFs
[08:56:07] [PASSED] 18 VFs
[08:56:07] [PASSED] 19 VFs
[08:56:07] [PASSED] 20 VFs
[08:56:07] [PASSED] 21 VFs
[08:56:07] [PASSED] 22 VFs
[08:56:07] [PASSED] 23 VFs
[08:56:07] [PASSED] 24 VFs
[08:56:07] [PASSED] 25 VFs
[08:56:07] [PASSED] 26 VFs
[08:56:07] [PASSED] 27 VFs
[08:56:07] [PASSED] 28 VFs
[08:56:07] [PASSED] 29 VFs
[08:56:07] [PASSED] 30 VFs
[08:56:07] [PASSED] 31 VFs
[08:56:07] [PASSED] 32 VFs
[08:56:07] [PASSED] 33 VFs
[08:56:07] [PASSED] 34 VFs
[08:56:07] [PASSED] 35 VFs
[08:56:07] [PASSED] 36 VFs
[08:56:07] [PASSED] 37 VFs
[08:56:07] [PASSED] 38 VFs
[08:56:07] [PASSED] 39 VFs
[08:56:07] [PASSED] 40 VFs
[08:56:07] [PASSED] 41 VFs
[08:56:07] [PASSED] 42 VFs
[08:56:07] [PASSED] 43 VFs
[08:56:07] [PASSED] 44 VFs
[08:56:07] [PASSED] 45 VFs
[08:56:07] [PASSED] 46 VFs
[08:56:07] [PASSED] 47 VFs
[08:56:07] [PASSED] 48 VFs
[08:56:07] [PASSED] 49 VFs
[08:56:07] [PASSED] 50 VFs
[08:56:07] [PASSED] 51 VFs
[08:56:07] [PASSED] 52 VFs
[08:56:07] [PASSED] 53 VFs
[08:56:07] [PASSED] 54 VFs
[08:56:07] [PASSED] 55 VFs
[08:56:07] [PASSED] 56 VFs
[08:56:07] [PASSED] 57 VFs
[08:56:07] [PASSED] 58 VFs
[08:56:07] [PASSED] 59 VFs
[08:56:07] [PASSED] 60 VFs
[08:56:07] [PASSED] 61 VFs
[08:56:07] [PASSED] 62 VFs
[08:56:07] [PASSED] 63 VFs
[08:56:07] ================= [PASSED] fair_doorbells ==================
[08:56:07] ======================== fair_ggtt ========================
[08:56:07] [PASSED] 1 VF
[08:56:07] [PASSED] 2 VFs
[08:56:07] [PASSED] 3 VFs
[08:56:07] [PASSED] 4 VFs
[08:56:07] [PASSED] 5 VFs
[08:56:07] [PASSED] 6 VFs
[08:56:07] [PASSED] 7 VFs
[08:56:07] [PASSED] 8 VFs
[08:56:07] [PASSED] 9 VFs
[08:56:07] [PASSED] 10 VFs
[08:56:07] [PASSED] 11 VFs
[08:56:07] [PASSED] 12 VFs
[08:56:07] [PASSED] 13 VFs
[08:56:07] [PASSED] 14 VFs
[08:56:07] [PASSED] 15 VFs
[08:56:07] [PASSED] 16 VFs
[08:56:07] [PASSED] 17 VFs
[08:56:07] [PASSED] 18 VFs
[08:56:07] [PASSED] 19 VFs
[08:56:07] [PASSED] 20 VFs
[08:56:07] [PASSED] 21 VFs
[08:56:07] [PASSED] 22 VFs
[08:56:07] [PASSED] 23 VFs
[08:56:07] [PASSED] 24 VFs
[08:56:07] [PASSED] 25 VFs
[08:56:07] [PASSED] 26 VFs
[08:56:07] [PASSED] 27 VFs
[08:56:07] [PASSED] 28 VFs
[08:56:07] [PASSED] 29 VFs
[08:56:07] [PASSED] 30 VFs
[08:56:07] [PASSED] 31 VFs
[08:56:07] [PASSED] 32 VFs
[08:56:07] [PASSED] 33 VFs
[08:56:07] [PASSED] 34 VFs
[08:56:07] [PASSED] 35 VFs
[08:56:07] [PASSED] 36 VFs
[08:56:07] [PASSED] 37 VFs
[08:56:07] [PASSED] 38 VFs
[08:56:07] [PASSED] 39 VFs
[08:56:07] [PASSED] 40 VFs
[08:56:07] [PASSED] 41 VFs
[08:56:07] [PASSED] 42 VFs
[08:56:07] [PASSED] 43 VFs
[08:56:07] [PASSED] 44 VFs
[08:56:07] [PASSED] 45 VFs
[08:56:07] [PASSED] 46 VFs
[08:56:07] [PASSED] 47 VFs
[08:56:07] [PASSED] 48 VFs
[08:56:07] [PASSED] 49 VFs
[08:56:07] [PASSED] 50 VFs
[08:56:07] [PASSED] 51 VFs
[08:56:07] [PASSED] 52 VFs
[08:56:07] [PASSED] 53 VFs
[08:56:07] [PASSED] 54 VFs
[08:56:07] [PASSED] 55 VFs
[08:56:07] [PASSED] 56 VFs
[08:56:07] [PASSED] 57 VFs
[08:56:07] [PASSED] 58 VFs
[08:56:07] [PASSED] 59 VFs
[08:56:07] [PASSED] 60 VFs
[08:56:07] [PASSED] 61 VFs
[08:56:07] [PASSED] 62 VFs
[08:56:07] [PASSED] 63 VFs
[08:56:07] ==================== [PASSED] fair_ggtt ====================
[08:56:07] ======================== fair_vram ========================
[08:56:07] [PASSED] 1 VF
[08:56:07] [PASSED] 2 VFs
[08:56:07] [PASSED] 3 VFs
[08:56:07] [PASSED] 4 VFs
[08:56:07] [PASSED] 5 VFs
[08:56:07] [PASSED] 6 VFs
[08:56:07] [PASSED] 7 VFs
[08:56:07] [PASSED] 8 VFs
[08:56:07] [PASSED] 9 VFs
[08:56:07] [PASSED] 10 VFs
[08:56:07] [PASSED] 11 VFs
[08:56:07] [PASSED] 12 VFs
[08:56:07] [PASSED] 13 VFs
[08:56:07] [PASSED] 14 VFs
[08:56:07] [PASSED] 15 VFs
[08:56:07] [PASSED] 16 VFs
[08:56:07] [PASSED] 17 VFs
[08:56:07] [PASSED] 18 VFs
[08:56:07] [PASSED] 19 VFs
[08:56:07] [PASSED] 20 VFs
[08:56:07] [PASSED] 21 VFs
[08:56:07] [PASSED] 22 VFs
[08:56:07] [PASSED] 23 VFs
[08:56:07] [PASSED] 24 VFs
[08:56:07] [PASSED] 25 VFs
[08:56:07] [PASSED] 26 VFs
[08:56:07] [PASSED] 27 VFs
[08:56:07] [PASSED] 28 VFs
[08:56:07] [PASSED] 29 VFs
[08:56:07] [PASSED] 30 VFs
[08:56:07] [PASSED] 31 VFs
[08:56:07] [PASSED] 32 VFs
[08:56:07] [PASSED] 33 VFs
[08:56:07] [PASSED] 34 VFs
[08:56:07] [PASSED] 35 VFs
[08:56:07] [PASSED] 36 VFs
[08:56:07] [PASSED] 37 VFs
[08:56:07] [PASSED] 38 VFs
[08:56:07] [PASSED] 39 VFs
[08:56:07] [PASSED] 40 VFs
[08:56:07] [PASSED] 41 VFs
[08:56:07] [PASSED] 42 VFs
[08:56:07] [PASSED] 43 VFs
[08:56:07] [PASSED] 44 VFs
[08:56:07] [PASSED] 45 VFs
[08:56:07] [PASSED] 46 VFs
[08:56:07] [PASSED] 47 VFs
[08:56:07] [PASSED] 48 VFs
[08:56:07] [PASSED] 49 VFs
[08:56:07] [PASSED] 50 VFs
[08:56:07] [PASSED] 51 VFs
[08:56:07] [PASSED] 52 VFs
[08:56:07] [PASSED] 53 VFs
[08:56:07] [PASSED] 54 VFs
[08:56:07] [PASSED] 55 VFs
[08:56:07] [PASSED] 56 VFs
[08:56:07] [PASSED] 57 VFs
[08:56:07] [PASSED] 58 VFs
[08:56:07] [PASSED] 59 VFs
[08:56:07] [PASSED] 60 VFs
[08:56:07] [PASSED] 61 VFs
[08:56:07] [PASSED] 62 VFs
[08:56:07] [PASSED] 63 VFs
[08:56:07] ==================== [PASSED] fair_vram ====================
[08:56:07] ================== [PASSED] pf_gt_config ===================
[08:56:07] ===================== lmtt (1 subtest) =====================
[08:56:07] ======================== test_ops =========================
[08:56:07] [PASSED] 2-level
[08:56:07] [PASSED] multi-level
[08:56:07] ==================== [PASSED] test_ops =====================
[08:56:07] ====================== [PASSED] lmtt =======================
[08:56:07] ================= sriov_packet (1 subtest) =================
[08:56:07] [PASSED] test_descriptor_init
[08:56:07] ================== [PASSED] sriov_packet ===================
[08:56:07] ================= pf_service (11 subtests) =================
[08:56:07] [PASSED] pf_negotiate_any
[08:56:07] [PASSED] pf_negotiate_base_match
[08:56:07] [PASSED] pf_negotiate_base_newer
[08:56:07] [PASSED] pf_negotiate_base_next
[08:56:07] [SKIPPED] pf_negotiate_base_older (no older minor)
[08:56:07] [PASSED] pf_negotiate_base_prev
[08:56:07] [PASSED] pf_negotiate_latest_match
[08:56:07] [PASSED] pf_negotiate_latest_newer
[08:56:07] [PASSED] pf_negotiate_latest_next
[08:56:07] [SKIPPED] pf_negotiate_latest_older (no older minor)
[08:56:07] [SKIPPED] pf_negotiate_latest_prev (no prev major)
[08:56:07] =================== [PASSED] pf_service ====================
[08:56:07] ================= xe_guc_g2g (2 subtests) ==================
[08:56:07] ============== xe_live_guc_g2g_kunit_default ==============
[08:56:07] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[08:56:07] ============== xe_live_guc_g2g_kunit_allmem ===============
[08:56:07] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[08:56:07] =================== [SKIPPED] xe_guc_g2g ===================
[08:56:07] =================== xe_mocs (2 subtests) ===================
[08:56:07] ================ xe_live_mocs_kernel_kunit ================
[08:56:07] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[08:56:07] ================ xe_live_mocs_reset_kunit =================
[08:56:07] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[08:56:07] ==================== [SKIPPED] xe_mocs =====================
[08:56:07] ================= xe_migrate (2 subtests) ==================
[08:56:07] ================= xe_migrate_sanity_kunit =================
[08:56:07] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[08:56:07] ================== xe_validate_ccs_kunit ==================
[08:56:07] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[08:56:07] =================== [SKIPPED] xe_migrate ===================
[08:56:07] ================== xe_dma_buf (1 subtest) ==================
[08:56:07] ==================== xe_dma_buf_kunit =====================
[08:56:07] ================ [SKIPPED] xe_dma_buf_kunit ================
[08:56:07] =================== [SKIPPED] xe_dma_buf ===================
[08:56:07] ================= xe_bo_shrink (1 subtest) =================
[08:56:07] =================== xe_bo_shrink_kunit ====================
[08:56:07] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[08:56:07] ================== [SKIPPED] xe_bo_shrink ==================
[08:56:07] ==================== xe_bo (2 subtests) ====================
[08:56:07] ================== xe_ccs_migrate_kunit ===================
[08:56:07] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[08:56:07] ==================== xe_bo_evict_kunit ====================
[08:56:07] =============== [SKIPPED] xe_bo_evict_kunit ================
[08:56:07] ===================== [SKIPPED] xe_bo ======================
[08:56:07] ==================== args (13 subtests) ====================
[08:56:07] [PASSED] count_args_test
[08:56:07] [PASSED] call_args_example
[08:56:07] [PASSED] call_args_test
[08:56:07] [PASSED] drop_first_arg_example
[08:56:07] [PASSED] drop_first_arg_test
[08:56:07] [PASSED] first_arg_example
[08:56:07] [PASSED] first_arg_test
[08:56:07] [PASSED] last_arg_example
[08:56:07] [PASSED] last_arg_test
[08:56:07] [PASSED] pick_arg_example
[08:56:07] [PASSED] if_args_example
[08:56:07] [PASSED] if_args_test
[08:56:07] [PASSED] sep_comma_example
[08:56:07] ====================== [PASSED] args =======================
[08:56:07] =================== xe_pci (3 subtests) ====================
[08:56:07] ==================== check_graphics_ip ====================
[08:56:07] [PASSED] 12.00 Xe_LP
[08:56:07] [PASSED] 12.10 Xe_LP+
[08:56:07] [PASSED] 12.55 Xe_HPG
[08:56:07] [PASSED] 12.60 Xe_HPC
[08:56:07] [PASSED] 12.70 Xe_LPG
[08:56:07] [PASSED] 12.71 Xe_LPG
[08:56:07] [PASSED] 12.74 Xe_LPG+
[08:56:07] [PASSED] 20.01 Xe2_HPG
[08:56:07] [PASSED] 20.02 Xe2_HPG
[08:56:07] [PASSED] 20.04 Xe2_LPG
[08:56:07] [PASSED] 30.00 Xe3_LPG
[08:56:07] [PASSED] 30.01 Xe3_LPG
[08:56:07] [PASSED] 30.03 Xe3_LPG
[08:56:07] [PASSED] 30.04 Xe3_LPG
[08:56:07] [PASSED] 30.05 Xe3_LPG
[08:56:07] [PASSED] 35.10 Xe3p_LPG
[08:56:07] [PASSED] 35.11 Xe3p_XPC
[08:56:07] ================ [PASSED] check_graphics_ip ================
[08:56:07] ===================== check_media_ip ======================
[08:56:07] [PASSED] 12.00 Xe_M
[08:56:07] [PASSED] 12.55 Xe_HPM
[08:56:07] [PASSED] 13.00 Xe_LPM+
[08:56:07] [PASSED] 13.01 Xe2_HPM
[08:56:07] [PASSED] 20.00 Xe2_LPM
[08:56:07] [PASSED] 30.00 Xe3_LPM
[08:56:07] [PASSED] 30.02 Xe3_LPM
[08:56:07] [PASSED] 35.00 Xe3p_LPM
[08:56:07] [PASSED] 35.03 Xe3p_HPM
[08:56:07] ================= [PASSED] check_media_ip ==================
[08:56:07] =================== check_platform_desc ===================
[08:56:07] [PASSED] 0x9A60 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A68 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A70 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A40 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A49 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A59 (TIGERLAKE)
[08:56:07] [PASSED] 0x9A78 (TIGERLAKE)
[08:56:07] [PASSED] 0x9AC0 (TIGERLAKE)
[08:56:07] [PASSED] 0x9AC9 (TIGERLAKE)
[08:56:07] [PASSED] 0x9AD9 (TIGERLAKE)
[08:56:07] [PASSED] 0x9AF8 (TIGERLAKE)
[08:56:07] [PASSED] 0x4C80 (ROCKETLAKE)
[08:56:07] [PASSED] 0x4C8A (ROCKETLAKE)
[08:56:07] [PASSED] 0x4C8B (ROCKETLAKE)
[08:56:07] [PASSED] 0x4C8C (ROCKETLAKE)
[08:56:07] [PASSED] 0x4C90 (ROCKETLAKE)
[08:56:07] [PASSED] 0x4C9A (ROCKETLAKE)
[08:56:07] [PASSED] 0x4680 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4682 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4688 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x468A (ALDERLAKE_S)
[08:56:07] [PASSED] 0x468B (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4690 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4692 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4693 (ALDERLAKE_S)
[08:56:07] [PASSED] 0x46A0 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46A1 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46A2 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46A3 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46A6 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46A8 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46AA (ALDERLAKE_P)
[08:56:07] [PASSED] 0x462A (ALDERLAKE_P)
[08:56:07] [PASSED] 0x4626 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x4628 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46B0 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46B1 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46B2 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46B3 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46C0 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46C1 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46C2 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46C3 (ALDERLAKE_P)
[08:56:07] [PASSED] 0x46D0 (ALDERLAKE_N)
[08:56:07] [PASSED] 0x46D1 (ALDERLAKE_N)
[08:56:07] [PASSED] 0x46D2 (ALDERLAKE_N)
[08:56:07] [PASSED] 0x46D3 (ALDERLAKE_N)
[08:56:07] [PASSED] 0x46D4 (ALDERLAKE_N)
[08:56:07] [PASSED] 0xA721 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7A1 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7A9 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7AC (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7AD (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA720 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7A0 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7A8 (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7AA (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA7AB (ALDERLAKE_P)
[08:56:07] [PASSED] 0xA780 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA781 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA782 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA783 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA788 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA789 (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA78A (ALDERLAKE_S)
[08:56:07] [PASSED] 0xA78B (ALDERLAKE_S)
[08:56:07] [PASSED] 0x4905 (DG1)
[08:56:07] [PASSED] 0x4906 (DG1)
[08:56:07] [PASSED] 0x4907 (DG1)
[08:56:07] [PASSED] 0x4908 (DG1)
[08:56:07] [PASSED] 0x4909 (DG1)
[08:56:07] [PASSED] 0x56C0 (DG2)
[08:56:07] [PASSED] 0x56C2 (DG2)
[08:56:07] [PASSED] 0x56C1 (DG2)
[08:56:07] [PASSED] 0x7D51 (METEORLAKE)
[08:56:07] [PASSED] 0x7DD1 (METEORLAKE)
[08:56:07] [PASSED] 0x7D41 (METEORLAKE)
[08:56:07] [PASSED] 0x7D67 (METEORLAKE)
[08:56:07] [PASSED] 0xB640 (METEORLAKE)
[08:56:07] [PASSED] 0x56A0 (DG2)
[08:56:07] [PASSED] 0x56A1 (DG2)
[08:56:07] [PASSED] 0x56A2 (DG2)
[08:56:07] [PASSED] 0x56BE (DG2)
[08:56:07] [PASSED] 0x56BF (DG2)
[08:56:07] [PASSED] 0x5690 (DG2)
[08:56:07] [PASSED] 0x5691 (DG2)
[08:56:07] [PASSED] 0x5692 (DG2)
[08:56:07] [PASSED] 0x56A5 (DG2)
[08:56:07] [PASSED] 0x56A6 (DG2)
[08:56:07] [PASSED] 0x56B0 (DG2)
[08:56:07] [PASSED] 0x56B1 (DG2)
[08:56:07] [PASSED] 0x56BA (DG2)
[08:56:07] [PASSED] 0x56BB (DG2)
[08:56:07] [PASSED] 0x56BC (DG2)
[08:56:07] [PASSED] 0x56BD (DG2)
[08:56:07] [PASSED] 0x5693 (DG2)
[08:56:07] [PASSED] 0x5694 (DG2)
[08:56:07] [PASSED] 0x5695 (DG2)
[08:56:07] [PASSED] 0x56A3 (DG2)
[08:56:07] [PASSED] 0x56A4 (DG2)
[08:56:07] [PASSED] 0x56B2 (DG2)
[08:56:07] [PASSED] 0x56B3 (DG2)
[08:56:07] [PASSED] 0x5696 (DG2)
[08:56:07] [PASSED] 0x5697 (DG2)
[08:56:07] [PASSED] 0xB69 (PVC)
[08:56:07] [PASSED] 0xB6E (PVC)
[08:56:07] [PASSED] 0xBD4 (PVC)
[08:56:07] [PASSED] 0xBD5 (PVC)
[08:56:07] [PASSED] 0xBD6 (PVC)
[08:56:07] [PASSED] 0xBD7 (PVC)
[08:56:07] [PASSED] 0xBD8 (PVC)
[08:56:07] [PASSED] 0xBD9 (PVC)
[08:56:07] [PASSED] 0xBDA (PVC)
[08:56:07] [PASSED] 0xBDB (PVC)
[08:56:07] [PASSED] 0xBE0 (PVC)
[08:56:07] [PASSED] 0xBE1 (PVC)
[08:56:07] [PASSED] 0xBE5 (PVC)
[08:56:07] [PASSED] 0x7D40 (METEORLAKE)
[08:56:07] [PASSED] 0x7D45 (METEORLAKE)
[08:56:07] [PASSED] 0x7D55 (METEORLAKE)
[08:56:07] [PASSED] 0x7D60 (METEORLAKE)
[08:56:07] [PASSED] 0x7DD5 (METEORLAKE)
[08:56:07] [PASSED] 0x6420 (LUNARLAKE)
[08:56:07] [PASSED] 0x64A0 (LUNARLAKE)
[08:56:07] [PASSED] 0x64B0 (LUNARLAKE)
[08:56:07] [PASSED] 0xE202 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE209 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE20B (BATTLEMAGE)
[08:56:07] [PASSED] 0xE20C (BATTLEMAGE)
[08:56:07] [PASSED] 0xE20D (BATTLEMAGE)
[08:56:07] [PASSED] 0xE210 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE211 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE212 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE216 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE220 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE221 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE222 (BATTLEMAGE)
[08:56:07] [PASSED] 0xE223 (BATTLEMAGE)
[08:56:07] [PASSED] 0xB080 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB081 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB082 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB083 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB084 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB085 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB086 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB087 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB08F (PANTHERLAKE)
[08:56:07] [PASSED] 0xB090 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB0A0 (PANTHERLAKE)
[08:56:07] [PASSED] 0xB0B0 (PANTHERLAKE)
[08:56:07] [PASSED] 0xFD80 (PANTHERLAKE)
[08:56:07] [PASSED] 0xFD81 (PANTHERLAKE)
[08:56:07] [PASSED] 0xD740 (NOVALAKE_S)
[08:56:07] [PASSED] 0xD741 (NOVALAKE_S)
[08:56:07] [PASSED] 0xD742 (NOVALAKE_S)
[08:56:07] [PASSED] 0xD743 (NOVALAKE_S)
[08:56:07] [PASSED] 0xD745 (NOVALAKE_S)
[08:56:07] [PASSED] 0xD74A (NOVALAKE_S)
[08:56:07] [PASSED] 0xD74B (NOVALAKE_S)
[08:56:07] [PASSED] 0x674C (CRESCENTISLAND)
[08:56:07] [PASSED] 0x674D (CRESCENTISLAND)
[08:56:07] [PASSED] 0x674E (CRESCENTISLAND)
[08:56:07] [PASSED] 0x674F (CRESCENTISLAND)
[08:56:07] [PASSED] 0x6750 (CRESCENTISLAND)
[08:56:07] [PASSED] 0xD750 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD751 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD752 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD753 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD754 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD755 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD756 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD757 (NOVALAKE_P)
[08:56:07] [PASSED] 0xD75F (NOVALAKE_P)
[08:56:07] =============== [PASSED] check_platform_desc ===============
[08:56:07] ===================== [PASSED] xe_pci ======================
[08:56:07] ============= xe_rtp_tables_test (5 subtests) ==============
[08:56:07] ================== xe_rtp_table_gt_test ===================
[08:56:07] [PASSED] gt_was/14011060649
[08:56:07] [PASSED] gt_was/14011059788
[08:56:07] [PASSED] gt_was/14015795083
[08:56:07] [PASSED] gt_was/16021867713
[08:56:07] [PASSED] gt_was/14019449301
[08:56:07] [PASSED] gt_was/16028005424
[08:56:07] [PASSED] gt_was/14026578760
[08:56:07] [PASSED] gt_was/1409420604
[08:56:07] [PASSED] gt_was/1408615072
[08:56:07] [PASSED] gt_was/22010523718
[08:56:07] [PASSED] gt_was/14011006942
[08:56:07] [PASSED] gt_was/14014830051
[08:56:07] [PASSED] gt_was/18018781329
[08:56:07] [PASSED] gt_was/1509235366
[08:56:07] [PASSED] gt_was/18018781329
[08:56:07] [PASSED] gt_was/16016694945
[08:56:07] [PASSED] gt_was/14018575942
[08:56:07] [PASSED] gt_was/22016670082
[08:56:07] [PASSED] gt_was/22016670082
[08:56:07] [PASSED] gt_was/14017421178
[08:56:07] [PASSED] gt_was/16025250150
[08:56:07] [PASSED] gt_was/14021871409
[08:56:07] [PASSED] gt_was/16021865536
[08:56:07] [PASSED] gt_was/14021486841
[08:56:07] [PASSED] gt_was/14025160223
[08:56:07] [PASSED] gt_was/14026144927, 16029437861, 14026127056
[08:56:07] [PASSED] gt_was/14025635424
[08:56:07] [PASSED] gt_was/16028005424
[08:56:07] ============== [PASSED] xe_rtp_table_gt_test ===============
[08:56:07] ================== xe_rtp_table_gt_test ===================
[08:56:07] [PASSED] gt_tunings/Tuning: Blend Fill Caching Optimization Disable
[08:56:07] [PASSED] gt_tunings/Tuning: 32B Access Enable
[08:56:07] [PASSED] gt_tunings/Tuning: L3 cache
[08:56:07] [PASSED] gt_tunings/Tuning: L3 cache - media
[08:56:07] [PASSED] gt_tunings/Tuning: Compression Overfetch
[08:56:07] [PASSED] gt_tunings/Tuning: Compression Overfetch - media
[08:56:07] [PASSED] gt_tunings/Tuning: Enable compressible partial write overfetch in L3
[08:56:07] [PASSED] gt_tunings/Tuning: Enable compressible partial write overfetch in L3 - media
[08:56:07] [PASSED] gt_tunings/Tuning: L2 Overfetch Compressible Only
[08:56:07] [PASSED] gt_tunings/Tuning: L2 Overfetch Compressible Only - media
[08:56:07] [PASSED] gt_tunings/Tuning: Stateless compression control
[08:56:07] [PASSED] gt_tunings/Tuning: Stateless compression control - media
[08:56:07] [PASSED] gt_tunings/Tuning: L3 RW flush all Cache
[08:56:07] [PASSED] gt_tunings/Tuning: L3 RW flush all cache - media
[08:56:07] [PASSED] gt_tunings/Tuning: Set STLB Bank Hash Mode to 4KB
[08:56:07] ============== [PASSED] xe_rtp_table_gt_test ===============
[08:56:07] ================== xe_rtp_table_oob_test ==================
[08:56:07] [PASSED] oob_was/1607983814
[08:56:07] [PASSED] oob_was/16010904313
[08:56:07] [PASSED] oob_was/18022495364
[08:56:07] [PASSED] oob_was/22012773006
[08:56:07] [PASSED] oob_was/14014475959
[08:56:07] [PASSED] oob_was/22011391025
[08:56:07] [PASSED] oob_was/22012727170
[08:56:07] [PASSED] oob_was/22012727685
[08:56:07] [PASSED] oob_was/22016596838
[08:56:07] [PASSED] oob_was/18020744125
[08:56:07] [PASSED] oob_was/1409600907
[08:56:07] [PASSED] oob_was/22014953428
[08:56:07] [PASSED] oob_was/16017236439
[08:56:07] [PASSED] oob_was/14019821291
[08:56:07] [PASSED] oob_was/14015076503
[08:56:07] [PASSED] oob_was/14018913170
[08:56:07] [PASSED] oob_was/14018094691
[08:56:07] [PASSED] oob_was/18024947630
[08:56:07] [PASSED] oob_was/16022287689
[08:56:07] [PASSED] oob_was/13011645652
[08:56:07] [PASSED] oob_was/14022293748
[08:56:07] [PASSED] oob_was/22019794406
[08:56:07] [PASSED] oob_was/22019338487
[08:56:07] [PASSED] oob_was/16023588340
[08:56:07] [PASSED] oob_was/14019789679
[08:56:07] [PASSED] oob_was/14022866841
[08:56:07] [PASSED] oob_was/16021333562
[08:56:07] [PASSED] oob_was/14016712196
[08:56:07] [PASSED] oob_was/14015568240
[08:56:07] [PASSED] oob_was/18013179988
[08:56:07] [PASSED] oob_was/1508761755
[08:56:07] [PASSED] oob_was/16023105232
[08:56:07] [PASSED] oob_was/16026508708
[08:56:07] [PASSED] oob_was/14020001231
[08:56:07] [PASSED] oob_was/16023683509
[08:56:07] [PASSED] oob_was/14025515070
[08:56:07] [PASSED] oob_was/15015404425_disable
[08:56:07] [PASSED] oob_was/16026007364
[08:56:07] [PASSED] oob_was/14020316580
[08:56:07] [PASSED] oob_was/14025883347
[08:56:07] [PASSED] oob_was/16029380221
[08:56:07] [PASSED] oob_was/22022079272
[08:56:07] [PASSED] oob_was/16029897822
[08:56:07] [PASSED] oob_was/14027054324
[08:56:07] ============== [PASSED] xe_rtp_table_oob_test ==============
[08:56:07] ================ xe_rtp_table_dev_oob_test ================
[08:56:07] [PASSED] device_oob_was/22010954014
[08:56:07] [PASSED] device_oob_was/15015404425
[08:56:07] [PASSED] device_oob_was/22019338487_display
[08:56:07] [PASSED] device_oob_was/14022085890
[08:56:07] [PASSED] device_oob_was/14026539277
[08:56:07] [PASSED] device_oob_was/14026633728
[08:56:07] [PASSED] device_oob_was/14026746987
[08:56:07] [PASSED] device_oob_was/14026779378
[08:56:07] ============ [PASSED] xe_rtp_table_dev_oob_test ============
[08:56:07] ========== xe_rtp_table_missing_upper_bound_test ==========
[08:56:07] [PASSED] register_whitelist/WaAllowPMDepthAndInvocationCountAccessFromUMD, 1408556865
[08:56:07] [PASSED] register_whitelist/1508744258, 14012131227, 1808121037
[08:56:07] [PASSED] register_whitelist/1806527549
[08:56:07] [PASSED] register_whitelist/allow_read_ctx_timestamp
[08:56:07] [PASSED] register_whitelist/allow_read_queue_timestamp
[08:56:07] [PASSED] register_whitelist/16014440446
[08:56:07] [PASSED] register_whitelist/16017236439
[08:56:07] [PASSED] register_whitelist/16020183090
[08:56:07] [PASSED] register_whitelist/14024997852
[08:56:07] [PASSED] register_whitelist/14024997852
[08:56:07] ====== [PASSED] xe_rtp_table_missing_upper_bound_test ======
[08:56:07] =============== [PASSED] xe_rtp_tables_test ================
[08:56:07] =================== xe_rtp (3 subtests) ====================
[08:56:07] =================== xe_rtp_rules_tests ====================
[08:56:07] [PASSED] no
[08:56:07] [PASSED] yes
[08:56:07] [PASSED] no-and-no
[08:56:07] [PASSED] no-and-yes
[08:56:07] [PASSED] yes-and-no
[08:56:07] [PASSED] yes-and-yes
[08:56:07] [PASSED] no-or-no
[08:56:07] [PASSED] no-or-yes
[08:56:07] [PASSED] yes-or-no
[08:56:07] [PASSED] yes-or-yes
[08:56:07] [PASSED] no-yes-or-yes-no
[08:56:07] [PASSED] no-yes-or-yes-yes
[08:56:07] [PASSED] yes-yes-or-no-yes
[08:56:07] [PASSED] yes-yes-or-yes-yes
[08:56:07] [PASSED] no-no-or-yes-or-no
[08:56:07] [PASSED] or
[08:56:07] [PASSED] or-yes
[08:56:07] [PASSED] or-no
[08:56:07] [PASSED] yes-or
[08:56:07] [PASSED] no-or
[08:56:07] [PASSED] no-or-or-yes
[08:56:07] [PASSED] yes-or-or-no
[08:56:07] [PASSED] no-or-or-no
[08:56:07] [PASSED] missing-context-engine-class
[08:56:07] [PASSED] missing-context-engine-class-or-yes
[08:56:07] [PASSED] missing-context-engine-class-or-or-yes
[08:56:07] =============== [PASSED] xe_rtp_rules_tests ================
[08:56:07] =============== xe_rtp_process_to_sr_tests ================
[08:56:07] [PASSED] coalesce-same-reg
[08:56:07] [PASSED] coalesce-same-reg-literal-and-func
[08:56:07] [PASSED] no-match-no-add
[08:56:07] [PASSED] two-regs-two-entries
[08:56:07] [PASSED] clr-one-set-other
[08:56:07] [PASSED] set-field
[08:56:07] [PASSED] conflict-duplicate
[08:56:07] [PASSED] conflict-not-disjoint
[08:56:07] [PASSED] conflict-not-disjoint-literal-and-func
[08:56:07] [PASSED] conflict-reg-type
[08:56:07] [PASSED] bad-mcr-reg-forced-to-regular
[08:56:07] [PASSED] bad-regular-reg-forced-to-mcr
[08:56:07] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[08:56:07] ================== xe_rtp_process_tests ===================
[08:56:07] [PASSED] active1
[08:56:07] [PASSED] active2
[08:56:07] [PASSED] active-inactive
[08:56:07] [PASSED] inactive-active
[08:56:07] [PASSED] inactive-active-inactive
[08:56:07] [PASSED] inactive-inactive-inactive
[08:56:07] ============== [PASSED] xe_rtp_process_tests ===============
[08:56:07] ===================== [PASSED] xe_rtp ======================
[08:56:07] ==================== xe_wa (1 subtest) =====================
[08:56:07] ======================== xe_wa_gt =========================
[08:56:07] [PASSED] TIGERLAKE B0
[08:56:07] [PASSED] DG1 A0
[08:56:07] [PASSED] DG1 B0
[08:56:07] [PASSED] ALDERLAKE_S A0
[08:56:07] [PASSED] ALDERLAKE_S B0
[08:56:07] [PASSED] ALDERLAKE_S C0
[08:56:07] [PASSED] ALDERLAKE_S D0
[08:56:07] [PASSED] ALDERLAKE_P A0
[08:56:07] [PASSED] ALDERLAKE_P B0
[08:56:07] [PASSED] ALDERLAKE_P C0
[08:56:07] [PASSED] ALDERLAKE_S RPLS D0
[08:56:07] [PASSED] ALDERLAKE_P RPLU E0
[08:56:07] [PASSED] DG2 G10 C0
[08:56:07] [PASSED] DG2 G11 B1
[08:56:07] [PASSED] DG2 G12 A1
[08:56:07] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[08:56:07] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[08:56:07] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[08:56:07] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[08:56:07] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[08:56:07] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[08:56:07] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[08:56:07] ==================== [PASSED] xe_wa_gt =====================
[08:56:07] ====================== [PASSED] xe_wa ======================
[08:56:07] ============================================================
[08:56:07] Testing complete. Ran 742 tests: passed: 724, skipped: 18
[08:56:07] Elapsed time: 36.942s total, 4.324s configuring, 31.952s building, 0.636s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[08:56:07] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[08:56:09] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[08:56:34] Starting KUnit Kernel (1/1)...
[08:56:34] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[08:56:34] ============ drm_test_pick_cmdline (2 subtests) ============
[08:56:34] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[08:56:34] =============== drm_test_pick_cmdline_named ===============
[08:56:34] [PASSED] NTSC
[08:56:34] [PASSED] NTSC-J
[08:56:34] [PASSED] PAL
[08:56:34] [PASSED] PAL-M
[08:56:34] =========== [PASSED] drm_test_pick_cmdline_named ===========
[08:56:34] ============== [PASSED] drm_test_pick_cmdline ==============
[08:56:34] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[08:56:34] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[08:56:34] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[08:56:34] =========== drm_validate_clone_mode (2 subtests) ===========
[08:56:34] ============== drm_test_check_in_clone_mode ===============
[08:56:34] [PASSED] in_clone_mode
[08:56:34] [PASSED] not_in_clone_mode
[08:56:34] ========== [PASSED] drm_test_check_in_clone_mode ===========
[08:56:34] =============== drm_test_check_valid_clones ===============
[08:56:34] [PASSED] not_in_clone_mode
[08:56:34] [PASSED] valid_clone
[08:56:34] [PASSED] invalid_clone
[08:56:34] =========== [PASSED] drm_test_check_valid_clones ===========
[08:56:34] ============= [PASSED] drm_validate_clone_mode =============
[08:56:34] ============= drm_validate_modeset (1 subtest) =============
[08:56:34] [PASSED] drm_test_check_connector_changed_modeset
[08:56:34] ============== [PASSED] drm_validate_modeset ===============
[08:56:34] ====== drm_test_bridge_get_current_state (1 subtest) =======
[08:56:34] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[08:56:34] ======== [PASSED] drm_test_bridge_get_current_state ========
[08:56:34] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[08:56:34] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[08:56:34] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[08:56:34] [PASSED] drm_test_drm_bridge_helper_hdmi_output_bus_fmts
[08:56:34] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[08:56:34] ============== drm_bridge_alloc (2 subtests) ===============
[08:56:34] [PASSED] drm_test_drm_bridge_alloc_basic
[08:56:34] [PASSED] drm_test_drm_bridge_alloc_get_put
[08:56:34] ================ [PASSED] drm_bridge_alloc =================
[08:56:34] ============= drm_bridge_bus_fmt (5 subtests) ==============
[08:56:34] [PASSED] drm_test_bridge_rgb_yuv_rgb
[08:56:34] [PASSED] drm_test_bridge_must_convert_to_yuv444
[08:56:34] [PASSED] drm_test_bridge_hdmi_auto_rgb
[08:56:34] [PASSED] drm_test_bridge_auto_first
[08:56:34] [PASSED] drm_test_bridge_rgb_yuv_no_path
[08:56:34] =============== [PASSED] drm_bridge_bus_fmt ================
[08:56:34] ============= drm_cmdline_parser (40 subtests) =============
[08:56:34] [PASSED] drm_test_cmdline_force_d_only
[08:56:34] [PASSED] drm_test_cmdline_force_D_only_dvi
[08:56:34] [PASSED] drm_test_cmdline_force_D_only_hdmi
[08:56:34] [PASSED] drm_test_cmdline_force_D_only_not_digital
[08:56:34] [PASSED] drm_test_cmdline_force_e_only
[08:56:34] [PASSED] drm_test_cmdline_res
[08:56:34] [PASSED] drm_test_cmdline_res_vesa
[08:56:34] [PASSED] drm_test_cmdline_res_vesa_rblank
[08:56:34] [PASSED] drm_test_cmdline_res_rblank
[08:56:34] [PASSED] drm_test_cmdline_res_bpp
[08:56:34] [PASSED] drm_test_cmdline_res_refresh
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[08:56:34] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[08:56:34] [PASSED] drm_test_cmdline_res_margins_force_on
[08:56:34] [PASSED] drm_test_cmdline_res_vesa_margins
[08:56:34] [PASSED] drm_test_cmdline_name
[08:56:34] [PASSED] drm_test_cmdline_name_bpp
[08:56:34] [PASSED] drm_test_cmdline_name_option
[08:56:34] [PASSED] drm_test_cmdline_name_bpp_option
[08:56:34] [PASSED] drm_test_cmdline_rotate_0
[08:56:34] [PASSED] drm_test_cmdline_rotate_90
[08:56:34] [PASSED] drm_test_cmdline_rotate_180
[08:56:34] [PASSED] drm_test_cmdline_rotate_270
[08:56:34] [PASSED] drm_test_cmdline_hmirror
[08:56:34] [PASSED] drm_test_cmdline_vmirror
[08:56:34] [PASSED] drm_test_cmdline_margin_options
[08:56:34] [PASSED] drm_test_cmdline_multiple_options
[08:56:34] [PASSED] drm_test_cmdline_bpp_extra_and_option
[08:56:34] [PASSED] drm_test_cmdline_extra_and_option
[08:56:34] [PASSED] drm_test_cmdline_freestanding_options
[08:56:34] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[08:56:34] [PASSED] drm_test_cmdline_panel_orientation
[08:56:34] ================ drm_test_cmdline_invalid =================
[08:56:34] [PASSED] margin_only
[08:56:34] [PASSED] interlace_only
[08:56:34] [PASSED] res_missing_x
[08:56:34] [PASSED] res_missing_y
[08:56:34] [PASSED] res_bad_y
[08:56:34] [PASSED] res_missing_y_bpp
[08:56:34] [PASSED] res_bad_bpp
[08:56:34] [PASSED] res_bad_refresh
[08:56:34] [PASSED] res_bpp_refresh_force_on_off
[08:56:34] [PASSED] res_invalid_mode
[08:56:34] [PASSED] res_bpp_wrong_place_mode
[08:56:34] [PASSED] name_bpp_refresh
[08:56:34] [PASSED] name_refresh
[08:56:34] [PASSED] name_refresh_wrong_mode
[08:56:34] [PASSED] name_refresh_invalid_mode
[08:56:34] [PASSED] rotate_multiple
[08:56:34] [PASSED] rotate_invalid_val
[08:56:34] [PASSED] rotate_truncated
[08:56:34] [PASSED] invalid_option
[08:56:34] [PASSED] invalid_tv_option
[08:56:34] [PASSED] truncated_tv_option
[08:56:34] ============ [PASSED] drm_test_cmdline_invalid =============
[08:56:34] =============== drm_test_cmdline_tv_options ===============
[08:56:34] [PASSED] NTSC
[08:56:34] [PASSED] NTSC_443
[08:56:34] [PASSED] NTSC_J
[08:56:34] [PASSED] PAL
[08:56:34] [PASSED] PAL_M
[08:56:34] [PASSED] PAL_N
[08:56:34] [PASSED] SECAM
[08:56:34] [PASSED] MONO_525
[08:56:34] [PASSED] MONO_625
[08:56:34] =========== [PASSED] drm_test_cmdline_tv_options ===========
[08:56:34] =============== [PASSED] drm_cmdline_parser ================
[08:56:34] ========== drmm_connector_hdmi_init (20 subtests) ==========
[08:56:34] [PASSED] drm_test_connector_hdmi_init_valid
[08:56:34] [PASSED] drm_test_connector_hdmi_init_bpc_8
[08:56:34] [PASSED] drm_test_connector_hdmi_init_bpc_10
[08:56:34] [PASSED] drm_test_connector_hdmi_init_bpc_12
[08:56:34] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[08:56:34] [PASSED] drm_test_connector_hdmi_init_bpc_null
[08:56:34] [PASSED] drm_test_connector_hdmi_init_formats_empty
[08:56:34] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[08:56:34] === drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[08:56:34] [PASSED] supported_formats=0x9 yuv420_allowed=1
[08:56:34] [PASSED] supported_formats=0x9 yuv420_allowed=0
[08:56:34] [PASSED] supported_formats=0x5 yuv420_allowed=1
[08:56:34] [PASSED] supported_formats=0x5 yuv420_allowed=0
[08:56:34] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[08:56:34] [PASSED] drm_test_connector_hdmi_init_null_ddc
[08:56:34] [PASSED] drm_test_connector_hdmi_init_null_product
[08:56:34] [PASSED] drm_test_connector_hdmi_init_null_vendor
[08:56:34] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[08:56:34] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[08:56:34] [PASSED] drm_test_connector_hdmi_init_product_valid
[08:56:34] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[08:56:34] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[08:56:34] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[08:56:34] ========= drm_test_connector_hdmi_init_type_valid =========
[08:56:34] [PASSED] HDMI-A
[08:56:34] [PASSED] HDMI-B
[08:56:34] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[08:56:34] ======== drm_test_connector_hdmi_init_type_invalid ========
[08:56:34] [PASSED] Unknown
[08:56:34] [PASSED] VGA
[08:56:34] [PASSED] DVI-I
[08:56:34] [PASSED] DVI-D
[08:56:34] [PASSED] DVI-A
[08:56:34] [PASSED] Composite
[08:56:34] [PASSED] SVIDEO
[08:56:34] [PASSED] LVDS
[08:56:34] [PASSED] Component
[08:56:34] [PASSED] DIN
[08:56:34] [PASSED] DP
[08:56:34] [PASSED] TV
[08:56:34] [PASSED] eDP
[08:56:34] [PASSED] Virtual
[08:56:34] [PASSED] DSI
[08:56:34] [PASSED] DPI
[08:56:34] [PASSED] Writeback
[08:56:34] [PASSED] SPI
[08:56:34] [PASSED] USB
[08:56:34] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[08:56:34] ============ [PASSED] drmm_connector_hdmi_init =============
[08:56:34] ============= drmm_connector_init (3 subtests) =============
[08:56:34] [PASSED] drm_test_drmm_connector_init
[08:56:34] [PASSED] drm_test_drmm_connector_init_null_ddc
[08:56:34] ========= drm_test_drmm_connector_init_type_valid =========
[08:56:34] [PASSED] Unknown
[08:56:34] [PASSED] VGA
[08:56:34] [PASSED] DVI-I
[08:56:34] [PASSED] DVI-D
[08:56:34] [PASSED] DVI-A
[08:56:34] [PASSED] Composite
[08:56:34] [PASSED] SVIDEO
[08:56:34] [PASSED] LVDS
[08:56:34] [PASSED] Component
[08:56:34] [PASSED] DIN
[08:56:34] [PASSED] DP
[08:56:34] [PASSED] HDMI-A
[08:56:34] [PASSED] HDMI-B
[08:56:34] [PASSED] TV
[08:56:34] [PASSED] eDP
[08:56:34] [PASSED] Virtual
[08:56:34] [PASSED] DSI
[08:56:34] [PASSED] DPI
[08:56:34] [PASSED] Writeback
[08:56:34] [PASSED] SPI
[08:56:34] [PASSED] USB
[08:56:34] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[08:56:34] =============== [PASSED] drmm_connector_init ===============
[08:56:34] ========= drm_connector_dynamic_init (6 subtests) ==========
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_init
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_init_properties
[08:56:34] ===== drm_test_drm_connector_dynamic_init_type_valid ======
[08:56:34] [PASSED] Unknown
[08:56:34] [PASSED] VGA
[08:56:34] [PASSED] DVI-I
[08:56:34] [PASSED] DVI-D
[08:56:34] [PASSED] DVI-A
[08:56:34] [PASSED] Composite
[08:56:34] [PASSED] SVIDEO
[08:56:34] [PASSED] LVDS
[08:56:34] [PASSED] Component
[08:56:34] [PASSED] DIN
[08:56:34] [PASSED] DP
[08:56:34] [PASSED] HDMI-A
[08:56:34] [PASSED] HDMI-B
[08:56:34] [PASSED] TV
[08:56:34] [PASSED] eDP
[08:56:34] [PASSED] Virtual
[08:56:34] [PASSED] DSI
[08:56:34] [PASSED] DPI
[08:56:34] [PASSED] Writeback
[08:56:34] [PASSED] SPI
[08:56:34] [PASSED] USB
[08:56:34] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[08:56:34] ======== drm_test_drm_connector_dynamic_init_name =========
[08:56:34] [PASSED] Unknown
[08:56:34] [PASSED] VGA
[08:56:34] [PASSED] DVI-I
[08:56:34] [PASSED] DVI-D
[08:56:34] [PASSED] DVI-A
[08:56:34] [PASSED] Composite
[08:56:34] [PASSED] SVIDEO
[08:56:34] [PASSED] LVDS
[08:56:34] [PASSED] Component
[08:56:34] [PASSED] DIN
[08:56:34] [PASSED] DP
[08:56:34] [PASSED] HDMI-A
[08:56:34] [PASSED] HDMI-B
[08:56:34] [PASSED] TV
[08:56:34] [PASSED] eDP
[08:56:34] [PASSED] Virtual
[08:56:34] [PASSED] DSI
[08:56:34] [PASSED] DPI
[08:56:34] [PASSED] Writeback
[08:56:34] [PASSED] SPI
[08:56:34] [PASSED] USB
[08:56:34] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[08:56:34] =========== [PASSED] drm_connector_dynamic_init ============
[08:56:34] ==== drm_connector_dynamic_register_early (4 subtests) =====
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[08:56:34] ====== [PASSED] drm_connector_dynamic_register_early =======
[08:56:34] ======= drm_connector_dynamic_register (7 subtests) ========
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[08:56:34] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[08:56:34] ========= [PASSED] drm_connector_dynamic_register ==========
[08:56:34] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[08:56:34] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[08:56:34] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[08:56:34] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[08:56:34] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[08:56:34] ========== drm_test_get_tv_mode_from_name_valid ===========
[08:56:34] [PASSED] NTSC
[08:56:34] [PASSED] NTSC-443
[08:56:34] [PASSED] NTSC-J
[08:56:34] [PASSED] PAL
[08:56:34] [PASSED] PAL-M
[08:56:34] [PASSED] PAL-N
[08:56:34] [PASSED] SECAM
[08:56:34] [PASSED] Mono
[08:56:34] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[08:56:34] [PASSED] drm_test_get_tv_mode_from_name_truncated
[08:56:34] ============ [PASSED] drm_get_tv_mode_from_name ============
[08:56:34] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[08:56:34] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[08:56:34] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid =
[08:56:34] [PASSED] VIC 96
[08:56:34] [PASSED] VIC 97
[08:56:34] [PASSED] VIC 101
[08:56:34] [PASSED] VIC 102
[08:56:34] [PASSED] VIC 106
[08:56:34] [PASSED] VIC 107
[08:56:34] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[08:56:34] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[08:56:34] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[08:56:34] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[08:56:34] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[08:56:34] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[08:56:34] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[08:56:34] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[08:56:34] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ====
[08:56:34] [PASSED] Automatic
[08:56:34] [PASSED] Full
[08:56:34] [PASSED] Limited 16:235
[08:56:34] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[08:56:34] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[08:56:34] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[08:56:34] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[08:56:34] === drm_test_drm_hdmi_connector_get_output_format_name ====
[08:56:34] [PASSED] RGB
[08:56:34] [PASSED] YUV 4:2:0
[08:56:34] [PASSED] YUV 4:2:2
[08:56:34] [PASSED] YUV 4:4:4
[08:56:34] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[08:56:34] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[08:56:34] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[08:56:34] ============= drm_damage_helper (21 subtests) ==============
[08:56:34] [PASSED] drm_test_damage_iter_no_damage
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_src_moved
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_not_visible
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[08:56:34] [PASSED] drm_test_damage_iter_no_damage_no_fb
[08:56:34] [PASSED] drm_test_damage_iter_simple_damage
[08:56:34] [PASSED] drm_test_damage_iter_single_damage
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_outside_src
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_src_moved
[08:56:34] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[08:56:34] [PASSED] drm_test_damage_iter_damage
[08:56:34] [PASSED] drm_test_damage_iter_damage_one_intersect
[08:56:34] [PASSED] drm_test_damage_iter_damage_one_outside
[08:56:34] [PASSED] drm_test_damage_iter_damage_src_moved
[08:56:34] [PASSED] drm_test_damage_iter_damage_not_visible
[08:56:34] ================ [PASSED] drm_damage_helper ================
[08:56:34] ============== drm_dp_mst_helper (3 subtests) ==============
[08:56:34] ============== drm_test_dp_mst_calc_pbn_mode ==============
[08:56:34] [PASSED] Clock 154000 BPP 30 DSC disabled
[08:56:34] [PASSED] Clock 234000 BPP 30 DSC disabled
[08:56:34] [PASSED] Clock 297000 BPP 24 DSC disabled
[08:56:34] [PASSED] Clock 332880 BPP 24 DSC enabled
[08:56:34] [PASSED] Clock 324540 BPP 24 DSC enabled
[08:56:34] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[08:56:34] ============== drm_test_dp_mst_calc_pbn_div ===============
[08:56:34] [PASSED] Link rate 2000000 lane count 4
[08:56:34] [PASSED] Link rate 2000000 lane count 2
[08:56:34] [PASSED] Link rate 2000000 lane count 1
[08:56:34] [PASSED] Link rate 1350000 lane count 4
[08:56:34] [PASSED] Link rate 1350000 lane count 2
[08:56:34] [PASSED] Link rate 1350000 lane count 1
[08:56:34] [PASSED] Link rate 1000000 lane count 4
[08:56:34] [PASSED] Link rate 1000000 lane count 2
[08:56:34] [PASSED] Link rate 1000000 lane count 1
[08:56:34] [PASSED] Link rate 810000 lane count 4
[08:56:34] [PASSED] Link rate 810000 lane count 2
[08:56:34] [PASSED] Link rate 810000 lane count 1
[08:56:34] [PASSED] Link rate 540000 lane count 4
[08:56:34] [PASSED] Link rate 540000 lane count 2
[08:56:34] [PASSED] Link rate 540000 lane count 1
[08:56:34] [PASSED] Link rate 270000 lane count 4
[08:56:34] [PASSED] Link rate 270000 lane count 2
[08:56:34] [PASSED] Link rate 270000 lane count 1
[08:56:34] [PASSED] Link rate 162000 lane count 4
[08:56:34] [PASSED] Link rate 162000 lane count 2
[08:56:34] [PASSED] Link rate 162000 lane count 1
[08:56:34] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[08:56:34] ========= drm_test_dp_mst_sideband_msg_req_decode =========
[08:56:34] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[08:56:34] [PASSED] DP_POWER_UP_PHY with port number
[08:56:34] [PASSED] DP_POWER_DOWN_PHY with port number
[08:56:34] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[08:56:34] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[08:56:34] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[08:56:34] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[08:56:34] [PASSED] DP_QUERY_PAYLOAD with port number
[08:56:34] [PASSED] DP_QUERY_PAYLOAD with VCPI
[08:56:34] [PASSED] DP_REMOTE_DPCD_READ with port number
[08:56:34] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[08:56:34] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[08:56:34] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[08:56:34] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[08:56:34] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[08:56:34] [PASSED] DP_REMOTE_I2C_READ with port number
[08:56:34] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[08:56:34] [PASSED] DP_REMOTE_I2C_READ with transactions array
[08:56:34] [PASSED] DP_REMOTE_I2C_WRITE with port number
[08:56:34] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[08:56:34] [PASSED] DP_REMOTE_I2C_WRITE with data array
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[08:56:34] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[08:56:34] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[08:56:34] ================ [PASSED] drm_dp_mst_helper ================
[08:56:34] ================== drm_exec (7 subtests) ===================
[08:56:34] [PASSED] sanitycheck
[08:56:34] [PASSED] test_lock
[08:56:34] [PASSED] test_lock_unlock
[08:56:34] [PASSED] test_duplicates
[08:56:34] [PASSED] test_prepare
[08:56:34] [PASSED] test_prepare_array
[08:56:34] [PASSED] test_multiple_loops
[08:56:34] ==================== [PASSED] drm_exec =====================
[08:56:34] =========== drm_format_helper_test (17 subtests) ===========
[08:56:34] ============== drm_test_fb_xrgb8888_to_gray8 ==============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[08:56:34] ============= drm_test_fb_xrgb8888_to_rgb332 ==============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[08:56:34] ============= drm_test_fb_xrgb8888_to_rgb565 ==============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[08:56:34] ============ drm_test_fb_xrgb8888_to_xrgb1555 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[08:56:34] ============ drm_test_fb_xrgb8888_to_argb1555 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[08:56:34] ============ drm_test_fb_xrgb8888_to_rgba5551 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[08:56:34] ============= drm_test_fb_xrgb8888_to_rgb888 ==============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[08:56:34] ============= drm_test_fb_xrgb8888_to_bgr888 ==============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[08:56:34] ============ drm_test_fb_xrgb8888_to_argb8888 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[08:56:34] =========== drm_test_fb_xrgb8888_to_xrgb2101010 ===========
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[08:56:34] =========== drm_test_fb_xrgb8888_to_argb2101010 ===========
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[08:56:34] ============== drm_test_fb_xrgb8888_to_mono ===============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[08:56:34] ==================== drm_test_fb_swab =====================
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ================ [PASSED] drm_test_fb_swab =================
[08:56:34] ============ drm_test_fb_xrgb8888_to_xbgr8888 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[08:56:34] ============ drm_test_fb_xrgb8888_to_abgr8888 =============
[08:56:34] [PASSED] single_pixel_source_buffer
[08:56:34] [PASSED] single_pixel_clip_rectangle
[08:56:34] [PASSED] well_known_colors
[08:56:34] [PASSED] destination_pitch
[08:56:34] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[08:56:34] ================= drm_test_fb_clip_offset =================
[08:56:34] [PASSED] pass through
[08:56:34] [PASSED] horizontal offset
[08:56:34] [PASSED] vertical offset
[08:56:34] [PASSED] horizontal and vertical offset
[08:56:34] [PASSED] horizontal offset (custom pitch)
[08:56:34] [PASSED] vertical offset (custom pitch)
[08:56:34] [PASSED] horizontal and vertical offset (custom pitch)
[08:56:34] ============= [PASSED] drm_test_fb_clip_offset =============
[08:56:34] =================== drm_test_fb_memcpy ====================
[08:56:34] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[08:56:34] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[08:56:34] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[08:56:34] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[08:56:34] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[08:56:34] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[08:56:34] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[08:56:34] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[08:56:34] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[08:56:34] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[08:56:34] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[08:56:34] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[08:56:34] =============== [PASSED] drm_test_fb_memcpy ================
[08:56:34] ============= [PASSED] drm_format_helper_test ==============
[08:56:34] ================= drm_format (18 subtests) =================
[08:56:34] [PASSED] drm_test_format_block_width_invalid
[08:56:34] [PASSED] drm_test_format_block_width_one_plane
[08:56:34] [PASSED] drm_test_format_block_width_two_plane
[08:56:34] [PASSED] drm_test_format_block_width_three_plane
[08:56:34] [PASSED] drm_test_format_block_width_tiled
[08:56:34] [PASSED] drm_test_format_block_height_invalid
[08:56:34] [PASSED] drm_test_format_block_height_one_plane
[08:56:34] [PASSED] drm_test_format_block_height_two_plane
[08:56:34] [PASSED] drm_test_format_block_height_three_plane
[08:56:34] [PASSED] drm_test_format_block_height_tiled
[08:56:34] [PASSED] drm_test_format_min_pitch_invalid
[08:56:34] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[08:56:34] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[08:56:34] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[08:56:34] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[08:56:34] [PASSED] drm_test_format_min_pitch_two_plane
[08:56:34] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[08:56:34] [PASSED] drm_test_format_min_pitch_tiled
[08:56:34] =================== [PASSED] drm_format ====================
[08:56:34] ============== drm_framebuffer (10 subtests) ===============
[08:56:34] ========== drm_test_framebuffer_check_src_coords ==========
[08:56:34] [PASSED] Success: source fits into fb
[08:56:34] [PASSED] Fail: overflowing fb with x-axis coordinate
[08:56:34] [PASSED] Fail: overflowing fb with y-axis coordinate
[08:56:34] [PASSED] Fail: overflowing fb with source width
[08:56:34] [PASSED] Fail: overflowing fb with source height
[08:56:34] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[08:56:34] [PASSED] drm_test_framebuffer_cleanup
[08:56:34] =============== drm_test_framebuffer_create ===============
[08:56:34] [PASSED] ABGR8888 normal sizes
[08:56:34] [PASSED] ABGR8888 max sizes
[08:56:34] [PASSED] ABGR8888 pitch greater than min required
[08:56:34] [PASSED] ABGR8888 pitch less than min required
[08:56:34] [PASSED] ABGR8888 Invalid width
[08:56:34] [PASSED] ABGR8888 Invalid buffer handle
[08:56:34] [PASSED] No pixel format
[08:56:34] [PASSED] ABGR8888 Width 0
[08:56:34] [PASSED] ABGR8888 Height 0
[08:56:34] [PASSED] ABGR8888 Out of bound height * pitch combination
[08:56:34] [PASSED] ABGR8888 Large buffer offset
[08:56:34] [PASSED] ABGR8888 Buffer offset for inexistent plane
[08:56:34] [PASSED] ABGR8888 Invalid flag
[08:56:34] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[08:56:34] [PASSED] ABGR8888 Valid buffer modifier
[08:56:34] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[08:56:34] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] NV12 Normal sizes
[08:56:34] [PASSED] NV12 Max sizes
[08:56:34] [PASSED] NV12 Invalid pitch
[08:56:34] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[08:56:34] [PASSED] NV12 different modifier per-plane
[08:56:34] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[08:56:34] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] NV12 Modifier for inexistent plane
[08:56:34] [PASSED] NV12 Handle for inexistent plane
[08:56:34] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[08:56:34] [PASSED] YVU420 Normal sizes
[08:56:34] [PASSED] YVU420 Max sizes
[08:56:34] [PASSED] YVU420 Invalid pitch
[08:56:34] [PASSED] YVU420 Different pitches
[08:56:34] [PASSED] YVU420 Different buffer offsets/pitches
[08:56:34] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[08:56:34] [PASSED] YVU420 Valid modifier
[08:56:34] [PASSED] YVU420 Different modifiers per plane
[08:56:34] [PASSED] YVU420 Modifier for inexistent plane
[08:56:34] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[08:56:34] [PASSED] X0L2 Normal sizes
[08:56:34] [PASSED] X0L2 Max sizes
[08:56:34] [PASSED] X0L2 Invalid pitch
[08:56:34] [PASSED] X0L2 Pitch greater than minimum required
[08:56:34] [PASSED] X0L2 Handle for inexistent plane
[08:56:34] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[08:56:34] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[08:56:34] [PASSED] X0L2 Valid modifier
[08:56:34] [PASSED] X0L2 Modifier for inexistent plane
[08:56:34] =========== [PASSED] drm_test_framebuffer_create ===========
[08:56:34] [PASSED] drm_test_framebuffer_free
[08:56:34] [PASSED] drm_test_framebuffer_init
[08:56:34] [PASSED] drm_test_framebuffer_init_bad_format
[08:56:34] [PASSED] drm_test_framebuffer_init_dev_mismatch
[08:56:34] [PASSED] drm_test_framebuffer_lookup
[08:56:34] [PASSED] drm_test_framebuffer_lookup_inexistent
[08:56:34] [PASSED] drm_test_framebuffer_modifiers_not_supported
[08:56:34] ================= [PASSED] drm_framebuffer =================
[08:56:34] ================ drm_gem_shmem (8 subtests) ================
[08:56:34] [PASSED] drm_gem_shmem_test_obj_create
[08:56:34] [PASSED] drm_gem_shmem_test_obj_create_private
[08:56:34] [PASSED] drm_gem_shmem_test_pin_pages
[08:56:34] [PASSED] drm_gem_shmem_test_vmap
[08:56:34] [PASSED] drm_gem_shmem_test_get_sg_table
[08:56:34] [PASSED] drm_gem_shmem_test_get_pages_sgt
[08:56:34] [PASSED] drm_gem_shmem_test_madvise
[08:56:34] [PASSED] drm_gem_shmem_test_purge
[08:56:34] ================== [PASSED] drm_gem_shmem ==================
[08:56:34] === drm_atomic_helper_connector_hdmi_check (29 subtests) ===
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[08:56:34] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 =======
[08:56:34] [PASSED] Automatic
[08:56:34] [PASSED] Full
[08:56:34] [PASSED] Limited 16:235
[08:56:34] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[08:56:34] [PASSED] drm_test_check_disable_connector
[08:56:34] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[08:56:34] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[08:56:34] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[08:56:34] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[08:56:34] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[08:56:34] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[08:56:34] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[08:56:34] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[08:56:34] [PASSED] drm_test_check_output_bpc_dvi
[08:56:34] [PASSED] drm_test_check_output_bpc_format_vic_1
[08:56:34] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[08:56:34] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[08:56:34] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[08:56:34] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[08:56:34] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[08:56:34] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[08:56:34] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[08:56:34] ============ drm_test_check_hdmi_color_format =============
[08:56:34] [PASSED] AUTO -> RGB
[08:56:34] [PASSED] YCBCR422 -> YUV422
[08:56:34] [PASSED] YCBCR420 -> YUV420
[08:56:34] [PASSED] YCBCR444 -> YUV444
[08:56:34] [PASSED] RGB -> RGB
[08:56:34] ======== [PASSED] drm_test_check_hdmi_color_format =========
[08:56:34] ======== drm_test_check_hdmi_color_format_420_only ========
[08:56:34] [PASSED] RGB should fail
[08:56:34] [PASSED] YUV444 should fail
[08:56:34] [PASSED] YUV422 should fail
[08:56:34] [PASSED] YUV420 should work
[08:56:34] ==== [PASSED] drm_test_check_hdmi_color_format_420_only ====
[08:56:34] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[08:56:34] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[08:56:34] [PASSED] drm_test_check_broadcast_rgb_value
[08:56:34] [PASSED] drm_test_check_bpc_8_value
[08:56:34] [PASSED] drm_test_check_bpc_10_value
[08:56:34] [PASSED] drm_test_check_bpc_12_value
[08:56:34] [PASSED] drm_test_check_format_value
[08:56:34] [PASSED] drm_test_check_tmds_char_value
[08:56:34] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[08:56:34] = drm_atomic_helper_connector_hdmi_mode_valid (7 subtests) =
[08:56:34] [PASSED] drm_test_check_mode_valid
[08:56:34] [PASSED] drm_test_check_mode_valid_reject
[08:56:34] [PASSED] drm_test_check_mode_valid_reject_rate
[08:56:34] [PASSED] drm_test_check_mode_valid_reject_max_clock
[08:56:34] [PASSED] drm_test_check_mode_valid_yuv420_only_max_clock
[08:56:34] [PASSED] drm_test_check_mode_valid_reject_yuv420_only_connector
[08:56:34] [PASSED] drm_test_check_mode_valid_accept_yuv420_also_connector_rgb
[08:56:34] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[08:56:34] = drm_atomic_helper_connector_hdmi_infoframes (5 subtests) =
[08:56:34] [PASSED] drm_test_check_infoframes
[08:56:34] [PASSED] drm_test_check_reject_avi_infoframe
[08:56:34] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_8
[08:56:34] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_10
[08:56:34] [PASSED] drm_test_check_reject_audio_infoframe
[08:56:34] === [PASSED] drm_atomic_helper_connector_hdmi_infoframes ===
[08:56:34] ================= drm_managed (2 subtests) =================
[08:56:34] [PASSED] drm_test_managed_release_action
[08:56:34] [PASSED] drm_test_managed_run_action
[08:56:34] =================== [PASSED] drm_managed ===================
[08:56:34] =================== drm_mm (6 subtests) ====================
[08:56:34] [PASSED] drm_test_mm_init
[08:56:34] [PASSED] drm_test_mm_debug
[08:56:34] [PASSED] drm_test_mm_align32
[08:56:34] [PASSED] drm_test_mm_align64
[08:56:34] [PASSED] drm_test_mm_lowest
[08:56:34] [PASSED] drm_test_mm_highest
[08:56:34] ===================== [PASSED] drm_mm ======================
[08:56:34] ============= drm_modes_analog_tv (5 subtests) =============
[08:56:34] [PASSED] drm_test_modes_analog_tv_mono_576i
[08:56:34] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[08:56:34] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[08:56:34] [PASSED] drm_test_modes_analog_tv_pal_576i
[08:56:34] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[08:56:34] =============== [PASSED] drm_modes_analog_tv ===============
[08:56:34] ============== drm_plane_helper (2 subtests) ===============
[08:56:34] =============== drm_test_check_plane_state ================
[08:56:34] [PASSED] clipping_simple
[08:56:34] [PASSED] clipping_rotate_reflect
[08:56:34] [PASSED] positioning_simple
[08:56:34] [PASSED] upscaling
[08:56:34] [PASSED] downscaling
[08:56:34] [PASSED] rounding1
[08:56:34] [PASSED] rounding2
[08:56:34] [PASSED] rounding3
[08:56:34] [PASSED] rounding4
[08:56:34] =========== [PASSED] drm_test_check_plane_state ============
[08:56:34] =========== drm_test_check_invalid_plane_state ============
[08:56:34] [PASSED] positioning_invalid
[08:56:34] [PASSED] upscaling_invalid
[08:56:34] [PASSED] downscaling_invalid
[08:56:34] ======= [PASSED] drm_test_check_invalid_plane_state ========
[08:56:34] ================ [PASSED] drm_plane_helper =================
[08:56:34] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[08:56:34] ====== drm_test_connector_helper_tv_get_modes_check =======
[08:56:34] [PASSED] None
[08:56:34] [PASSED] PAL
[08:56:34] [PASSED] NTSC
[08:56:34] [PASSED] Both, NTSC Default
[08:56:34] [PASSED] Both, PAL Default
[08:56:34] [PASSED] Both, NTSC Default, with PAL on command-line
[08:56:34] [PASSED] Both, PAL Default, with NTSC on command-line
[08:56:34] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[08:56:34] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[08:56:34] ================== drm_rect (9 subtests) ===================
[08:56:34] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[08:56:34] [PASSED] drm_test_rect_clip_scaled_not_clipped
[08:56:34] [PASSED] drm_test_rect_clip_scaled_clipped
[08:56:34] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[08:56:34] ================= drm_test_rect_intersect =================
[08:56:34] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[08:56:34] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[08:56:34] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[08:56:34] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[08:56:34] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[08:56:34] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[08:56:34] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[08:56:34] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[08:56:34] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[08:56:34] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[08:56:34] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[08:56:34] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[08:56:34] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[08:56:34] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[08:56:34] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
[08:56:34] ============= [PASSED] drm_test_rect_intersect =============
[08:56:34] ================ drm_test_rect_calc_hscale ================
[08:56:34] [PASSED] normal use
[08:56:34] [PASSED] out of max range
[08:56:34] [PASSED] out of min range
[08:56:34] [PASSED] zero dst
[08:56:34] [PASSED] negative src
[08:56:34] [PASSED] negative dst
[08:56:34] ============ [PASSED] drm_test_rect_calc_hscale ============
[08:56:34] ================ drm_test_rect_calc_vscale ================
[08:56:34] [PASSED] normal use
[08:56:34] [PASSED] out of max range
[08:56:34] [PASSED] out of min range
[08:56:34] [PASSED] zero dst
[08:56:34] [PASSED] negative src
[08:56:34] [PASSED] negative dst
[08:56:34] ============ [PASSED] drm_test_rect_calc_vscale ============
[08:56:34] ================== drm_test_rect_rotate ===================
[08:56:34] [PASSED] reflect-x
[08:56:34] [PASSED] reflect-y
[08:56:34] [PASSED] rotate-0
[08:56:34] [PASSED] rotate-90
[08:56:34] [PASSED] rotate-180
[08:56:34] [PASSED] rotate-270
[08:56:34] ============== [PASSED] drm_test_rect_rotate ===============
[08:56:34] ================ drm_test_rect_rotate_inv =================
[08:56:34] [PASSED] reflect-x
[08:56:34] [PASSED] reflect-y
[08:56:34] [PASSED] rotate-0
[08:56:34] [PASSED] rotate-90
[08:56:34] [PASSED] rotate-180
[08:56:34] [PASSED] rotate-270
[08:56:34] ============ [PASSED] drm_test_rect_rotate_inv =============
[08:56:34] ==================== [PASSED] drm_rect =====================
[08:56:34] ============ drm_sysfb_modeset_test (1 subtest) ============
[08:56:34] ============ drm_test_sysfb_build_fourcc_list =============
[08:56:34] [PASSED] no native formats
[08:56:34] [PASSED] XRGB8888 as native format
[08:56:34] [PASSED] remove duplicates
[08:56:34] [PASSED] convert alpha formats
[08:56:34] [PASSED] random formats
[08:56:34] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[08:56:34] ============= [PASSED] drm_sysfb_modeset_test ==============
[08:56:34] ================== drm_fixp (2 subtests) ===================
[08:56:34] [PASSED] drm_test_int2fixp
[08:56:34] [PASSED] drm_test_sm2fixp
[08:56:34] ==================== [PASSED] drm_fixp =====================
[08:56:34] ============================================================
[08:56:34] Testing complete. Ran 637 tests: passed: 637
[08:56:34] Elapsed time: 26.899s total, 1.802s configuring, 24.932s building, 0.138s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[08:56:34] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[08:56:36] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[08:56:46] Starting KUnit Kernel (1/1)...
[08:56:46] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[08:56:46] ================= ttm_device (5 subtests) ==================
[08:56:46] [PASSED] ttm_device_init_basic
[08:56:46] [PASSED] ttm_device_init_multiple
[08:56:46] [PASSED] ttm_device_fini_basic
[08:56:46] [PASSED] ttm_device_init_no_vma_man
[08:56:46] ================== ttm_device_init_pools ==================
[08:56:46] [PASSED] No DMA allocations, no DMA32 required
[08:56:46] [PASSED] DMA allocations, DMA32 required
[08:56:46] [PASSED] No DMA allocations, DMA32 required
[08:56:46] [PASSED] DMA allocations, no DMA32 required
[08:56:46] ============== [PASSED] ttm_device_init_pools ==============
[08:56:46] =================== [PASSED] ttm_device ====================
[08:56:46] ================== ttm_pool (8 subtests) ===================
[08:56:46] ================== ttm_pool_alloc_basic ===================
[08:56:46] [PASSED] One page
[08:56:46] [PASSED] More than one page
[08:56:46] [PASSED] Above the allocation limit
[08:56:46] [PASSED] One page, with coherent DMA mappings enabled
[08:56:46] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[08:56:46] ============== [PASSED] ttm_pool_alloc_basic ===============
[08:56:46] ============== ttm_pool_alloc_basic_dma_addr ==============
[08:56:46] [PASSED] One page
[08:56:46] [PASSED] More than one page
[08:56:46] [PASSED] Above the allocation limit
[08:56:46] [PASSED] One page, with coherent DMA mappings enabled
[08:56:46] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[08:56:46] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[08:56:46] [PASSED] ttm_pool_alloc_order_caching_match
[08:56:46] [PASSED] ttm_pool_alloc_caching_mismatch
[08:56:46] [PASSED] ttm_pool_alloc_order_mismatch
[08:56:46] [PASSED] ttm_pool_free_dma_alloc
[08:56:46] [PASSED] ttm_pool_free_no_dma_alloc
[08:56:46] [PASSED] ttm_pool_fini_basic
[08:56:46] ==================== [PASSED] ttm_pool =====================
[08:56:46] ================ ttm_resource (8 subtests) =================
[08:56:46] ================= ttm_resource_init_basic =================
[08:56:46] [PASSED] Init resource in TTM_PL_SYSTEM
[08:56:46] [PASSED] Init resource in TTM_PL_VRAM
[08:56:46] [PASSED] Init resource in a private placement
[08:56:46] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[08:56:46] ============= [PASSED] ttm_resource_init_basic =============
[08:56:46] [PASSED] ttm_resource_init_pinned
[08:56:46] [PASSED] ttm_resource_fini_basic
[08:56:46] [PASSED] ttm_resource_manager_init_basic
[08:56:46] [PASSED] ttm_resource_manager_usage_basic
[08:56:46] [PASSED] ttm_resource_manager_set_used_basic
[08:56:46] [PASSED] ttm_sys_man_alloc_basic
[08:56:46] [PASSED] ttm_sys_man_free_basic
[08:56:46] ================== [PASSED] ttm_resource ===================
[08:56:46] =================== ttm_tt (15 subtests) ===================
[08:56:46] ==================== ttm_tt_init_basic ====================
[08:56:46] [PASSED] Page-aligned size
[08:56:46] [PASSED] Extra pages requested
[08:56:46] ================ [PASSED] ttm_tt_init_basic ================
[08:56:46] [PASSED] ttm_tt_init_misaligned
[08:56:46] [PASSED] ttm_tt_fini_basic
[08:56:46] [PASSED] ttm_tt_fini_sg
[08:56:46] [PASSED] ttm_tt_fini_shmem
[08:56:46] [PASSED] ttm_tt_create_basic
[08:56:46] [PASSED] ttm_tt_create_invalid_bo_type
[08:56:46] [PASSED] ttm_tt_create_ttm_exists
[08:56:46] [PASSED] ttm_tt_create_failed
[08:56:46] [PASSED] ttm_tt_destroy_basic
[08:56:46] [PASSED] ttm_tt_populate_null_ttm
[08:56:46] [PASSED] ttm_tt_populate_populated_ttm
[08:56:46] [PASSED] ttm_tt_unpopulate_basic
[08:56:46] [PASSED] ttm_tt_unpopulate_empty_ttm
[08:56:46] [PASSED] ttm_tt_swapin_basic
[08:56:46] ===================== [PASSED] ttm_tt ======================
[08:56:46] =================== ttm_bo (14 subtests) ===================
[08:56:46] =========== ttm_bo_reserve_optimistic_no_ticket ===========
[08:56:46] [PASSED] Cannot be interrupted and sleeps
[08:56:46] [PASSED] Cannot be interrupted, locks straight away
[08:56:46] [PASSED] Can be interrupted, sleeps
[08:56:46] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[08:56:46] [PASSED] ttm_bo_reserve_locked_no_sleep
[08:56:46] [PASSED] ttm_bo_reserve_no_wait_ticket
[08:56:46] [PASSED] ttm_bo_reserve_double_resv
[08:56:46] [PASSED] ttm_bo_reserve_interrupted
[08:56:46] [PASSED] ttm_bo_reserve_deadlock
[08:56:46] [PASSED] ttm_bo_unreserve_basic
[08:56:46] [PASSED] ttm_bo_unreserve_pinned
[08:56:46] [PASSED] ttm_bo_unreserve_bulk
[08:56:46] [PASSED] ttm_bo_fini_basic
[08:56:46] [PASSED] ttm_bo_fini_shared_resv
[08:56:46] [PASSED] ttm_bo_pin_basic
[08:56:46] [PASSED] ttm_bo_pin_unpin_resource
[08:56:46] [PASSED] ttm_bo_multiple_pin_one_unpin
[08:56:46] ===================== [PASSED] ttm_bo ======================
[08:56:46] ============== ttm_bo_validate (22 subtests) ===============
[08:56:46] ============== ttm_bo_init_reserved_sys_man ===============
[08:56:46] [PASSED] Buffer object for userspace
[08:56:46] [PASSED] Kernel buffer object
[08:56:46] [PASSED] Shared buffer object
[08:56:46] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[08:56:46] ============== ttm_bo_init_reserved_mock_man ==============
[08:56:46] [PASSED] Buffer object for userspace
[08:56:46] [PASSED] Kernel buffer object
[08:56:46] [PASSED] Shared buffer object
[08:56:46] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[08:56:46] [PASSED] ttm_bo_init_reserved_resv
[08:56:46] ================== ttm_bo_validate_basic ==================
[08:56:46] [PASSED] Buffer object for userspace
[08:56:46] [PASSED] Kernel buffer object
[08:56:46] [PASSED] Shared buffer object
[08:56:46] ============== [PASSED] ttm_bo_validate_basic ==============
[08:56:46] [PASSED] ttm_bo_validate_invalid_placement
[08:56:46] ============= ttm_bo_validate_same_placement ==============
[08:56:46] [PASSED] System manager
[08:56:46] [PASSED] VRAM manager
[08:56:46] ========= [PASSED] ttm_bo_validate_same_placement ==========
[08:56:46] [PASSED] ttm_bo_validate_failed_alloc
[08:56:46] [PASSED] ttm_bo_validate_pinned
[08:56:46] [PASSED] ttm_bo_validate_busy_placement
[08:56:46] ================ ttm_bo_validate_multihop =================
[08:56:46] [PASSED] Buffer object for userspace
[08:56:46] [PASSED] Kernel buffer object
[08:56:46] [PASSED] Shared buffer object
[08:56:46] ============ [PASSED] ttm_bo_validate_multihop =============
[08:56:46] ========== ttm_bo_validate_no_placement_signaled ==========
[08:56:46] [PASSED] Buffer object in system domain, no page vector
[08:56:46] [PASSED] Buffer object in system domain with an existing page vector
[08:56:46] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[08:56:46] ======== ttm_bo_validate_no_placement_not_signaled ========
[08:56:46] [PASSED] Buffer object for userspace
[08:56:46] [PASSED] Kernel buffer object
[08:56:46] [PASSED] Shared buffer object
[08:56:46] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[08:56:46] [PASSED] ttm_bo_validate_move_fence_signaled
[08:56:46] ========= ttm_bo_validate_move_fence_not_signaled =========
[08:56:46] [PASSED] Waits for GPU
[08:56:46] [PASSED] Tries to lock straight away
[08:56:46] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[08:56:46] [PASSED] ttm_bo_validate_swapout
[08:56:46] [PASSED] ttm_bo_validate_happy_evict
[08:56:46] [PASSED] ttm_bo_validate_all_pinned_evict
[08:56:46] [PASSED] ttm_bo_validate_allowed_only_evict
[08:56:46] [PASSED] ttm_bo_validate_deleted_evict
[08:56:46] [PASSED] ttm_bo_validate_busy_domain_evict
[08:56:46] [PASSED] ttm_bo_validate_evict_gutting
[08:56:46] [PASSED] ttm_bo_validate_recrusive_evict
[08:56:46] ================= [PASSED] ttm_bo_validate =================
[08:56:46] ============================================================
[08:56:46] Testing complete. Ran 102 tests: passed: 102
[08:56:46] Elapsed time: 11.841s total, 1.819s configuring, 9.807s building, 0.175s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
@ 2026-08-06 8:57 Nitin Gote
2026-08-06 8:30 ` Raag Jadav
` (3 more replies)
0 siblings, 4 replies; 17+ messages in thread
From: Nitin Gote @ 2026-08-06 8:57 UTC (permalink / raw)
To: intel-xe; +Cc: Nitin Gote, Matthew Brost, Matthew Auld
Exec queues are owned by user fds and are destroyed when the fd is closed.
After hot-unplug this can happen from a deferred close during process
exit, after the device has been removed and runtime PM has been disabled.
The queue destroy path can call xe_pm_runtime_get_noresume(), and since
runtime PM is disabled pm_runtime_get_if_in_use() returns no reference,
so it warns about "Missing outer runtime PM protection".
This is a false positive for the hot-unplug teardown case. The device is
already unplugged, and the queue destroy path that triggered this warning
was checked and does not touch hardware state after unplug. GuC has already
been sanitized by guc_fini_hw(), so no H2G is sent and the hardware
teardown path is no longer reachable.
Skip the warning when the DRM device is already unplugged.
Observed with new IGT core_hotunplug subtests:
igt@core_hotunplug@hotreplug-with-load
igt@core_hotunplug@hotunplug-rescan-with-load
v2:
- Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
and instead exclude hot-unplug from the WARN in
xe_pm_runtime_get_noresume(). (Matthew Brost)
v3:
- Clarify that the queue destroy path was checked and does not touch
hardware state after unplug. (Matthew Auld)
Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Assisted-by: GitHub-Copilot:claude-opus-4.8
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index a5289a9df8d2..a038687d5b2c 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -10,6 +10,7 @@
#include <linux/suspend.h>
#include <linux/dmi.h>
+#include <drm/drm_drv.h>
#include <drm/drm_managed.h>
#include <drm/ttm/ttm_placement.h>
@@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
if (!ref) {
pm_runtime_get_noresume(xe->drm.dev);
- drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
+ drm_WARN(&xe->drm,
+ !drm_dev_is_unplugged(&xe->drm) &&
+ !xe_pm_suspending_or_resuming(xe),
"Missing outer runtime PM protection\n");
}
}
--
2.50.1
^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-06 8:30 ` Raag Jadav
@ 2026-08-06 9:38 ` Matthew Brost
2026-08-06 19:01 ` Matthew Brost
0 siblings, 1 reply; 17+ messages in thread
From: Matthew Brost @ 2026-08-06 9:38 UTC (permalink / raw)
To: Raag Jadav; +Cc: Nitin Gote, intel-xe, Matthew Auld
On Thu, Aug 06, 2026 at 10:30:20AM +0200, Raag Jadav wrote:
> On Thu, Aug 06, 2026 at 02:27:12PM +0530, Nitin Gote wrote:
> > Exec queues are owned by user fds and are destroyed when the fd is closed.
> > After hot-unplug this can happen from a deferred close during process
> > exit, after the device has been removed and runtime PM has been disabled.
> > The queue destroy path can call xe_pm_runtime_get_noresume(), and since
> > runtime PM is disabled pm_runtime_get_if_in_use() returns no reference,
> > so it warns about "Missing outer runtime PM protection".
> >
> > This is a false positive for the hot-unplug teardown case. The device is
> > already unplugged, and the queue destroy path that triggered this warning
> > was checked and does not touch hardware state after unplug. GuC has already
> > been sanitized by guc_fini_hw(), so no H2G is sent and the hardware
> > teardown path is no longer reachable.
> >
> > Skip the warning when the DRM device is already unplugged.
> >
> > Observed with new IGT core_hotunplug subtests:
> > igt@core_hotunplug@hotreplug-with-load
> > igt@core_hotunplug@hotunplug-rescan-with-load
> >
> > v2:
> > - Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> > and instead exclude hot-unplug from the WARN in
> > xe_pm_runtime_get_noresume(). (Matthew Brost)
> >
I don't think this was suggestion but also I believe I misunderstood the
problem, but maybe this works.
> > v3:
> > - Clarify that the queue destroy path was checked and does not touch
> > hardware state after unplug. (Matthew Auld)
> >
> > Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
> > Cc: Matthew Brost <matthew.brost@intel.com>
> > Cc: Matthew Auld <matthew.auld@intel.com>
> > Assisted-by: GitHub-Copilot:claude-opus-4.8
> > Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
> > index a5289a9df8d2..a038687d5b2c 100644
> > --- a/drivers/gpu/drm/xe/xe_pm.c
> > +++ b/drivers/gpu/drm/xe/xe_pm.c
> > @@ -10,6 +10,7 @@
> > #include <linux/suspend.h>
> > #include <linux/dmi.h>
> >
> > +#include <drm/drm_drv.h>
> > #include <drm/drm_managed.h>
> > #include <drm/ttm/ttm_placement.h>
> >
> > @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
>
> From the documentation
>
> * This function should be used in inner places where it is surely already
> * protected by outer-bound callers of `xe_pm_runtime_get`.
>
We should have those.
> > if (!ref) {
> > pm_runtime_get_noresume(xe->drm.dev);
> > - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> > + drm_WARN(&xe->drm,
> > + !drm_dev_is_unplugged(&xe->drm) &&
> > + !xe_pm_suspending_or_resuming(xe),
>
> So IMO this is more of a band-aid.
>
So if I'm understanding the problem correctly,
pm_runtime_get_if_in_use() (via guc_exec_queue_add_msg()) fails when the
device is unplugged? I don't see how the PM reference count could be
zero, regardless of whether the device is plugged in or unplugged.
My suggestion was (and still is) that guc_exec_queue_destroy() should
always add a message and determine whether the device/firmware is still
alive while processing that message, then either kick the destory or
issue H2G. Other messages likely are also missing proper device alive
checks too.
If, for some reason, obtaining a PM reference via
xe_pm_runtime_get_noresume() doesn't work when the device is unplugged,
then that logic should be handled in the message submission layer. For
example, use drm_dev_enter()/drm_dev_exit() for hot-unplug protection,
only take a PM reference if the device is still bound, encode that
information into the message, and then determine the appropriate next
steps when the message is processed.
Sorry a lot of this stuff in xe_guc_submit.c around corner cases /
teardwons is a mess from early Xe work I did, we should aim to clean
this up with clear semantics.
Matt
> Raag
>
> > "Missing outer runtime PM protection\n");
> > }
> > }
> > --
> > 2.50.1
> >
^ permalink raw reply [flat|nested] 17+ messages in thread
* ✓ Xe.CI.BAT: success for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
2026-08-06 8:57 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
2026-08-06 8:30 ` Raag Jadav
2026-08-06 8:56 ` ✓ CI.KUnit: success for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2) Patchwork
@ 2026-08-06 9:52 ` Patchwork
2026-08-06 18:39 ` ✗ Xe.CI.FULL: failure " Patchwork
3 siblings, 0 replies; 17+ messages in thread
From: Patchwork @ 2026-08-06 9:52 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 908 bytes --]
== Series Details ==
Series: drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
URL : https://patchwork.freedesktop.org/series/171320/
State : success
== Summary ==
CI Bug Log - changes from xe-5552-8d11ccdc98daac5e969243b42907fd060a650091_BAT -> xe-pw-171320v2_BAT
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (15 -> 13)
------------------------------
Missing (2): bat-bmg-2 bat-bmg-3
Changes
-------
No changes found
Build changes
-------------
* Linux: xe-5552-8d11ccdc98daac5e969243b42907fd060a650091 -> xe-pw-171320v2
IGT_9041: 9041
xe-5552-8d11ccdc98daac5e969243b42907fd060a650091: 8d11ccdc98daac5e969243b42907fd060a650091
xe-pw-171320v2: 171320v2
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/index.html
[-- Attachment #2: Type: text/html, Size: 1456 bytes --]
^ permalink raw reply [flat|nested] 17+ messages in thread
* ✗ Xe.CI.FULL: failure for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
2026-08-06 8:57 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
` (2 preceding siblings ...)
2026-08-06 9:52 ` ✓ Xe.CI.BAT: " Patchwork
@ 2026-08-06 18:39 ` Patchwork
3 siblings, 0 replies; 17+ messages in thread
From: Patchwork @ 2026-08-06 18:39 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 21296 bytes --]
== Series Details ==
Series: drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2)
URL : https://patchwork.freedesktop.org/series/171320/
State : failure
== Summary ==
CI Bug Log - changes from xe-5552-8d11ccdc98daac5e969243b42907fd060a650091_FULL -> xe-pw-171320v2_FULL
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with xe-pw-171320v2_FULL absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in xe-pw-171320v2_FULL, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (2 -> 2)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in xe-pw-171320v2_FULL:
### IGT changes ###
#### Possible regressions ####
* igt@kms_cursor_crc@cursor-random-256x256:
- shard-bmg: [PASS][1] -> [FAIL][2]
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-1/igt@kms_cursor_crc@cursor-random-256x256.html
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-5/igt@kms_cursor_crc@cursor-random-256x256.html
Known issues
------------
Here are the changes found in xe-pw-171320v2_FULL that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip:
- shard-lnl: NOTRUN -> [SKIP][3] ([Intel XE#3658] / [Intel XE#7360])
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip.html
* igt@kms_big_fb@y-tiled-32bpp-rotate-0:
- shard-lnl: NOTRUN -> [SKIP][4] ([Intel XE#1124]) +2 other tests skip
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_big_fb@y-tiled-32bpp-rotate-0.html
* igt@kms_bw@linear-tiling-3-displays-target-2160x1440p:
- shard-lnl: NOTRUN -> [SKIP][5] ([Intel XE#367])
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_bw@linear-tiling-3-displays-target-2160x1440p.html
* igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs@pipe-c-dp-2:
- shard-bmg: [PASS][6] -> [INCOMPLETE][7] ([Intel XE#7084] / [Intel XE#8150]) +1 other test incomplete
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-7/igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs@pipe-c-dp-2.html
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-6/igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs@pipe-c-dp-2.html
* igt@kms_ccs@missing-ccs-buffer-y-tiled-gen12-rc-ccs-cc:
- shard-lnl: NOTRUN -> [SKIP][8] ([Intel XE#2887]) +3 other tests skip
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_ccs@missing-ccs-buffer-y-tiled-gen12-rc-ccs-cc.html
* igt@kms_chamelium_hpd@dp-hpd-storm:
- shard-lnl: NOTRUN -> [SKIP][9] ([Intel XE#373])
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_chamelium_hpd@dp-hpd-storm.html
* igt@kms_content_protection@dp-mst-lic-type-0:
- shard-lnl: NOTRUN -> [SKIP][10] ([Intel XE#307] / [Intel XE#6974])
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_content_protection@dp-mst-lic-type-0.html
* igt@kms_cursor_crc@cursor-random-256x256@pipe-a-dp-2:
- shard-bmg: [PASS][11] -> [FAIL][12] ([Intel XE#8117])
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-1/igt@kms_cursor_crc@cursor-random-256x256@pipe-a-dp-2.html
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-5/igt@kms_cursor_crc@cursor-random-256x256@pipe-a-dp-2.html
* igt@kms_cursor_crc@cursor-sliding-128x42:
- shard-lnl: NOTRUN -> [SKIP][13] ([Intel XE#1424]) +1 other test skip
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_cursor_crc@cursor-sliding-128x42.html
* igt@kms_cursor_legacy@cursora-vs-flipb-varying-size:
- shard-lnl: NOTRUN -> [SKIP][14] ([Intel XE#309] / [Intel XE#7343])
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_cursor_legacy@cursora-vs-flipb-varying-size.html
* igt@kms_cursor_legacy@flip-vs-cursor-legacy:
- shard-bmg: [PASS][15] -> [FAIL][16] ([Intel XE#7571])
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-1/igt@kms_cursor_legacy@flip-vs-cursor-legacy.html
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-5/igt@kms_cursor_legacy@flip-vs-cursor-legacy.html
* igt@kms_dsc@dsc-fractional-bpp:
- shard-lnl: NOTRUN -> [SKIP][17] ([Intel XE#8265])
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_dsc@dsc-fractional-bpp.html
* igt@kms_flip@2x-dpms-vs-vblank-race:
- shard-lnl: NOTRUN -> [SKIP][18] ([Intel XE#1421])
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_flip@2x-dpms-vs-vblank-race.html
* igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1:
- shard-lnl: [PASS][19] -> [FAIL][20] ([Intel XE#301])
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-2/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-8/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
* igt@kms_flip@flip-vs-expired-vblank@c-edp1:
- shard-lnl: [PASS][21] -> [FAIL][22] ([Intel XE#301] / [Intel XE#3149])
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-4/igt@kms_flip@flip-vs-expired-vblank@c-edp1.html
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_flip@flip-vs-expired-vblank@c-edp1.html
* igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-downscaling:
- shard-lnl: NOTRUN -> [SKIP][23] ([Intel XE#7178] / [Intel XE#7351])
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-downscaling.html
* igt@kms_frontbuffer_tracking@drrs-1p-primscrn-shrfb-msflip-blt:
- shard-lnl: NOTRUN -> [SKIP][24] ([Intel XE#6312] / [Intel XE#651]) +2 other tests skip
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@drrs-1p-primscrn-shrfb-msflip-blt.html
* igt@kms_frontbuffer_tracking@drrshdr-1p-rte:
- shard-lnl: NOTRUN -> [SKIP][25] ([Intel XE#6312]) +3 other tests skip
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@drrshdr-1p-rte.html
* igt@kms_frontbuffer_tracking@fbcdrrshdr-2p-primscrn-spr-indfb-draw-render:
- shard-lnl: NOTRUN -> [SKIP][26] ([Intel XE#7905]) +6 other tests skip
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@fbcdrrshdr-2p-primscrn-spr-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@fbchdr-1p-primscrn-pri-shrfb-draw-render:
- shard-lnl: NOTRUN -> [SKIP][27] ([Intel XE#7865]) +2 other tests skip
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@fbchdr-1p-primscrn-pri-shrfb-draw-render.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-fullscreen:
- shard-lnl: NOTRUN -> [SKIP][28] ([Intel XE#656] / [Intel XE#7905]) +6 other tests skip
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-fullscreen.html
* igt@kms_frontbuffer_tracking@fbcpsrhdr-abgr161616f-draw-mmap-wc:
- shard-lnl: NOTRUN -> [SKIP][29] ([Intel XE#7061])
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_frontbuffer_tracking@fbcpsrhdr-abgr161616f-draw-mmap-wc.html
* igt@kms_joiner@invalid-modeset-force-big-joiner:
- shard-lnl: NOTRUN -> [SKIP][30] ([Intel XE#7086] / [Intel XE#7390])
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_joiner@invalid-modeset-force-big-joiner.html
* igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf:
- shard-lnl: NOTRUN -> [SKIP][31] ([Intel XE#2893] / [Intel XE#4608] / [Intel XE#7304])
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf.html
* igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf@pipe-a-edp-1:
- shard-lnl: NOTRUN -> [SKIP][32] ([Intel XE#4608])
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf@pipe-a-edp-1.html
* igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf@pipe-b-edp-1:
- shard-lnl: NOTRUN -> [SKIP][33] ([Intel XE#4608] / [Intel XE#7304])
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_psr2_sf@fbc-psr2-cursor-plane-move-continuous-exceed-sf@pipe-b-edp-1.html
* igt@kms_psr@pr-primary-render:
- shard-lnl: NOTRUN -> [SKIP][34] ([Intel XE#1406])
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_psr@pr-primary-render.html
* igt@kms_rotation_crc@bad-pixel-format:
- shard-lnl: NOTRUN -> [SKIP][35] ([Intel XE#3414] / [Intel XE#3904] / [Intel XE#7342])
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_rotation_crc@bad-pixel-format.html
* igt@xe_configfs@ctx-restore-mid-bb-invalid:
- shard-bmg: [PASS][36] -> [ABORT][37] ([Intel XE#8007])
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-10/igt@xe_configfs@ctx-restore-mid-bb-invalid.html
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-2/igt@xe_configfs@ctx-restore-mid-bb-invalid.html
* igt@xe_evict@evict-threads-small-multi-vm:
- shard-lnl: NOTRUN -> [SKIP][38] ([Intel XE#6540] / [Intel XE#688]) +2 other tests skip
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_evict@evict-threads-small-multi-vm.html
* igt@xe_exec_balancer@no-exec-virtual-userptr-invalidate-race:
- shard-lnl: NOTRUN -> [SKIP][39] ([Intel XE#7482]) +2 other tests skip
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_exec_balancer@no-exec-virtual-userptr-invalidate-race.html
* igt@xe_exec_basic@multigpu-many-execqueues-many-vm-basic-defer-mmap:
- shard-lnl: NOTRUN -> [SKIP][40] ([Intel XE#1392]) +1 other test skip
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_exec_basic@multigpu-many-execqueues-many-vm-basic-defer-mmap.html
* igt@xe_exec_fault_mode@many-multi-queue-invalid-fault:
- shard-lnl: NOTRUN -> [SKIP][41] ([Intel XE#8374])
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_exec_fault_mode@many-multi-queue-invalid-fault.html
* igt@xe_exec_multi_queue@two-queues-preempt-mode-fault-userptr-invalidate:
- shard-lnl: NOTRUN -> [SKIP][42] ([Intel XE#8364]) +6 other tests skip
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_exec_multi_queue@two-queues-preempt-mode-fault-userptr-invalidate.html
* igt@xe_exec_threads@threads-multi-queue-cm-shared-vm-basic:
- shard-lnl: NOTRUN -> [SKIP][43] ([Intel XE#8378]) +2 other tests skip
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_exec_threads@threads-multi-queue-cm-shared-vm-basic.html
* igt@xe_pat@pat-index-xelp:
- shard-lnl: NOTRUN -> [SKIP][44] ([Intel XE#7590] / [Intel XE#977])
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_pat@pat-index-xelp.html
* igt@xe_pm@s4-d3cold-basic-exec:
- shard-lnl: NOTRUN -> [SKIP][45] ([Intel XE#2284] / [Intel XE#366] / [Intel XE#7370])
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_pm@s4-d3cold-basic-exec.html
* igt@xe_survivability@i2c-functionality:
- shard-lnl: NOTRUN -> [SKIP][46] ([Intel XE#8415])
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_survivability@i2c-functionality.html
* igt@xe_vm@overcommit-nonfault-vram-lr-external-nodefer:
- shard-lnl: NOTRUN -> [SKIP][47] ([Intel XE#7892])
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@xe_vm@overcommit-nonfault-vram-lr-external-nodefer.html
* igt@xe_wedged@basic-wedged:
- shard-bmg: [PASS][48] -> [ABORT][49] ([Intel XE#8591])
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-5/igt@xe_wedged@basic-wedged.html
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-6/igt@xe_wedged@basic-wedged.html
#### Possible fixes ####
* igt@kms_flip@flip-vs-expired-vblank-interruptible@a-edp1:
- shard-lnl: [FAIL][50] ([Intel XE#301]) -> [PASS][51]
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-2/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-edp1.html
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-8/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-edp1.html
* igt@kms_setmode@basic@pipe-b-edp-1:
- shard-lnl: [FAIL][52] ([Intel XE#8618]) -> [PASS][53] +2 other tests pass
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-2/igt@kms_setmode@basic@pipe-b-edp-1.html
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-8/igt@kms_setmode@basic@pipe-b-edp-1.html
#### Warnings ####
* igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions:
- shard-lnl: [SKIP][54] ([Intel XE#309] / [Intel XE#7343]) -> [SKIP][55] ([Intel XE#309] / [Intel XE#7343] / [Intel XE#7935])
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-2/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions.html
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-8/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions.html
* igt@kms_flip@flip-vs-expired-vblank:
- shard-lnl: [FAIL][56] ([Intel XE#301]) -> [FAIL][57] ([Intel XE#301] / [Intel XE#3149])
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-lnl-4/igt@kms_flip@flip-vs-expired-vblank.html
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-lnl-3/igt@kms_flip@flip-vs-expired-vblank.html
* igt@kms_hdr@brightness-with-hdr:
- shard-bmg: [SKIP][58] ([Intel XE#3374] / [Intel XE#3544]) -> [SKIP][59] ([Intel XE#3544])
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-2/igt@kms_hdr@brightness-with-hdr.html
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-10/igt@kms_hdr@brightness-with-hdr.html
* igt@kms_tiled_display@basic-test-pattern-with-chamelium:
- shard-bmg: [SKIP][60] ([Intel XE#2426] / [Intel XE#5848]) -> [SKIP][61] ([Intel XE#2509] / [Intel XE#7437])
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-5552-8d11ccdc98daac5e969243b42907fd060a650091/shard-bmg-7/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/shard-bmg-3/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
[Intel XE#1124]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1124
[Intel XE#1392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1392
[Intel XE#1406]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1406
[Intel XE#1421]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1421
[Intel XE#1424]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1424
[Intel XE#2284]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2284
[Intel XE#2426]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2426
[Intel XE#2509]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2509
[Intel XE#2887]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2887
[Intel XE#2893]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2893
[Intel XE#301]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/301
[Intel XE#307]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/307
[Intel XE#309]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/309
[Intel XE#3149]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3149
[Intel XE#3374]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3374
[Intel XE#3414]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3414
[Intel XE#3544]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3544
[Intel XE#3658]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3658
[Intel XE#366]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/366
[Intel XE#367]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/367
[Intel XE#373]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/373
[Intel XE#3904]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3904
[Intel XE#4608]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4608
[Intel XE#5848]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5848
[Intel XE#6312]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6312
[Intel XE#651]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/651
[Intel XE#6540]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6540
[Intel XE#656]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/656
[Intel XE#688]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/688
[Intel XE#6974]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6974
[Intel XE#7061]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7061
[Intel XE#7084]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7084
[Intel XE#7086]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7086
[Intel XE#7178]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7178
[Intel XE#7304]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7304
[Intel XE#7342]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7342
[Intel XE#7343]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7343
[Intel XE#7351]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7351
[Intel XE#7360]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7360
[Intel XE#7370]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7370
[Intel XE#7390]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7390
[Intel XE#7437]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7437
[Intel XE#7482]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7482
[Intel XE#7571]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7571
[Intel XE#7590]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7590
[Intel XE#7865]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7865
[Intel XE#7892]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7892
[Intel XE#7905]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7905
[Intel XE#7935]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/7935
[Intel XE#8007]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8007
[Intel XE#8117]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8117
[Intel XE#8150]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8150
[Intel XE#8265]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8265
[Intel XE#8364]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8364
[Intel XE#8374]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8374
[Intel XE#8378]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8378
[Intel XE#8415]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8415
[Intel XE#8591]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8591
[Intel XE#8618]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/8618
[Intel XE#977]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/977
Build changes
-------------
* Linux: xe-5552-8d11ccdc98daac5e969243b42907fd060a650091 -> xe-pw-171320v2
IGT_9041: 9041
xe-5552-8d11ccdc98daac5e969243b42907fd060a650091: 8d11ccdc98daac5e969243b42907fd060a650091
xe-pw-171320v2: 171320v2
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-171320v2/index.html
[-- Attachment #2: Type: text/html, Size: 23852 bytes --]
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-06 9:38 ` Matthew Brost
@ 2026-08-06 19:01 ` Matthew Brost
2026-08-07 13:38 ` Gote, Nitin R
0 siblings, 1 reply; 17+ messages in thread
From: Matthew Brost @ 2026-08-06 19:01 UTC (permalink / raw)
To: Raag Jadav; +Cc: Nitin Gote, intel-xe, Matthew Auld
On Thu, Aug 06, 2026 at 02:38:03AM -0700, Matthew Brost wrote:
> On Thu, Aug 06, 2026 at 10:30:20AM +0200, Raag Jadav wrote:
> > On Thu, Aug 06, 2026 at 02:27:12PM +0530, Nitin Gote wrote:
> > > Exec queues are owned by user fds and are destroyed when the fd is closed.
> > > After hot-unplug this can happen from a deferred close during process
> > > exit, after the device has been removed and runtime PM has been disabled.
> > > The queue destroy path can call xe_pm_runtime_get_noresume(), and since
> > > runtime PM is disabled pm_runtime_get_if_in_use() returns no reference,
> > > so it warns about "Missing outer runtime PM protection".
> > >
> > > This is a false positive for the hot-unplug teardown case. The device is
> > > already unplugged, and the queue destroy path that triggered this warning
> > > was checked and does not touch hardware state after unplug. GuC has already
> > > been sanitized by guc_fini_hw(), so no H2G is sent and the hardware
> > > teardown path is no longer reachable.
> > >
> > > Skip the warning when the DRM device is already unplugged.
> > >
> > > Observed with new IGT core_hotunplug subtests:
> > > igt@core_hotunplug@hotreplug-with-load
> > > igt@core_hotunplug@hotunplug-rescan-with-load
> > >
> > > v2:
> > > - Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> > > and instead exclude hot-unplug from the WARN in
> > > xe_pm_runtime_get_noresume(). (Matthew Brost)
> > >
>
> I don't think this was suggestion but also I believe I misunderstood the
> problem, but maybe this works.
>
> > > v3:
> > > - Clarify that the queue destroy path was checked and does not touch
> > > hardware state after unplug. (Matthew Auld)
> > >
> > > Link: https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=4
> > > Cc: Matthew Brost <matthew.brost@intel.com>
> > > Cc: Matthew Auld <matthew.auld@intel.com>
> > > Assisted-by: GitHub-Copilot:claude-opus-4.8
> > > Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> > > ---
> > > drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> > > 1 file changed, 4 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
> > > index a5289a9df8d2..a038687d5b2c 100644
> > > --- a/drivers/gpu/drm/xe/xe_pm.c
> > > +++ b/drivers/gpu/drm/xe/xe_pm.c
> > > @@ -10,6 +10,7 @@
> > > #include <linux/suspend.h>
> > > #include <linux/dmi.h>
> > >
> > > +#include <drm/drm_drv.h>
> > > #include <drm/drm_managed.h>
> > > #include <drm/ttm/ttm_placement.h>
> > >
> > > @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct xe_device *xe)
> >
> > From the documentation
> >
> > * This function should be used in inner places where it is surely already
> > * protected by outer-bound callers of `xe_pm_runtime_get`.
> >
>
> We should have those.
>
> > > if (!ref) {
> > > pm_runtime_get_noresume(xe->drm.dev);
> > > - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> > > + drm_WARN(&xe->drm,
> > > + !drm_dev_is_unplugged(&xe->drm) &&
> > > + !xe_pm_suspending_or_resuming(xe),
> >
> > So IMO this is more of a band-aid.
> >
>
> So if I'm understanding the problem correctly,
> pm_runtime_get_if_in_use() (via guc_exec_queue_add_msg()) fails when the
> device is unplugged? I don't see how the PM reference count could be
> zero, regardless of whether the device is plugged in or unplugged.
>
> My suggestion was (and still is) that guc_exec_queue_destroy() should
> always add a message and determine whether the device/firmware is still
> alive while processing that message, then either kick the destory or
> issue H2G. Other messages likely are also missing proper device alive
> checks too.
>
> If, for some reason, obtaining a PM reference via
> xe_pm_runtime_get_noresume() doesn't work when the device is unplugged,
> then that logic should be handled in the message submission layer. For
> example, use drm_dev_enter()/drm_dev_exit() for hot-unplug protection,
> only take a PM reference if the device is still bound, encode that
> information into the message, and then determine the appropriate next
> steps when the message is processed.
>
Below is rough sketch of what I was thinking. We can also probably rip
out EXEC_QUEUE_FLAG_PERMANENT then too. We'd also need solid
explaination how why xe_pm_runtime_get_noresume doesn't work if device
is unplugged, as that part it unclear to me too.
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 9036f89dff7d..08000f8db28a 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1812,32 +1812,15 @@ static void __guc_exec_queue_destroy_async(struct work_struct *w)
static void guc_exec_queue_destroy_async(struct xe_exec_queue *q)
{
INIT_WORK(&q->guc->destroy_async, __guc_exec_queue_destroy_async);
-
- /* We must block on kernel engines so slabs are empty on driver unload */
- if (q->flags & EXEC_QUEUE_FLAG_PERMANENT || exec_queue_wedged(q))
- guc_exec_queue_do_destroy(q);
- else
- xe_destroy_wq_queue(&q->guc->destroy_async);
+ xe_destroy_wq_queue(&q->guc->destroy_async);
}
-static void __guc_exec_queue_destroy(struct xe_guc *guc, struct xe_exec_queue *q)
-{
- /*
- * Might be done from within the GPU scheduler, need to do async as we
- * fini the scheduler when the engine is fini'd, the scheduler can't
- * complete fini within itself (circular dependency). Async resolves
- * this we and don't really care when everything is fini'd, just that it
- * is.
- */
- guc_exec_queue_destroy_async(q);
-}
-
-static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
+static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
struct xe_guc *guc = exec_queue_to_guc(q);
- xe_gt_assert(guc_to_gt(guc), !(q->flags & EXEC_QUEUE_FLAG_PERMANENT));
trace_xe_exec_queue_cleanup_entity(q);
/*
@@ -1850,10 +1833,10 @@ static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
* it is safe to directly destroy the exec queue on driver side, as the GuC
* will not process further requests and all resources must be cleaned up locally.
*/
- if (exec_queue_registered(q) && xe_uc_fw_is_running(&guc->fw))
+ if (bound && exec_queue_registered(q) && xe_uc_fw_is_running(&guc->fw))
disable_scheduling_deregister(guc, q);
else
- __guc_exec_queue_destroy(guc, q);
+ guc_exec_queue_destroy_async(q);
}
static bool guc_exec_queue_allowed_to_change_state(struct xe_exec_queue *q)
@@ -1861,12 +1844,13 @@ static bool guc_exec_queue_allowed_to_change_state(struct xe_exec_queue *q)
return !exec_queue_killed_or_banned_or_wedged(q) && exec_queue_registered(q);
}
-static void __guc_exec_queue_process_msg_set_sched_props(struct xe_sched_msg *msg)
+static void __guc_exec_queue_process_msg_set_sched_props(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
struct xe_guc *guc = exec_queue_to_guc(q);
- if (guc_exec_queue_allowed_to_change_state(q))
+ if (guc_exec_queue_allowed_to_change_state(q) && bound)
init_policies(guc, q);
kfree(msg);
}
@@ -1904,13 +1888,14 @@ static void suspend_fence_signal(struct xe_exec_queue *q)
__suspend_fence_signal(q);
}
-static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
+static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
struct xe_guc *guc = exec_queue_to_guc(q);
if (guc_exec_queue_allowed_to_change_state(q) && !exec_queue_suspended(q) &&
- exec_queue_enabled(q)) {
+ exec_queue_enabled(q) && bound) {
wait_event(guc->ct.wq, vf_recovery(guc) ||
((q->guc->resume_time != RESUME_PENDING ||
xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q)));
@@ -1934,11 +1919,12 @@ static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
}
}
-static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
+static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
- if (guc_exec_queue_allowed_to_change_state(q)) {
+ if (guc_exec_queue_allowed_to_change_state(q) && bound) {
clear_exec_queue_suspended(q);
if (!exec_queue_enabled(q)) {
q->guc->resume_time = RESUME_PENDING;
@@ -1950,17 +1936,20 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
}
}
-static void __guc_exec_queue_process_msg_set_multi_queue_priority(struct xe_sched_msg *msg)
+static void
+__guc_exec_queue_process_msg_set_multi_queue_priority(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
- if (guc_exec_queue_allowed_to_change_state(q))
+ if (guc_exec_queue_allowed_to_change_state(q) && bound)
guc_exec_queue_send_cgp_sync(q, 0);
kfree(msg);
}
-static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
+static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg,
+ bool bound)
{
struct xe_exec_queue *q = msg->private_data;
@@ -1969,7 +1958,7 @@ static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
* CGP update + CGP_SYNC (re-applies the current priority from
* q->multi_queue.priority).
*/
- if (guc_exec_queue_allowed_to_change_state(q))
+ if (guc_exec_queue_allowed_to_change_state(q) && bound)
guc_exec_queue_send_cgp_sync(q, 0);
}
@@ -1982,37 +1971,46 @@ static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
#define OPCODE_MASK 0xf
#define MSG_LOCKED BIT(8)
#define MSG_HEAD BIT(9)
+#define MSG_PM_REF BIT(10)
static void guc_exec_queue_process_msg(struct xe_sched_msg *msg)
{
struct xe_device *xe = guc_to_xe(exec_queue_to_guc(msg->private_data));
+ int idx;
+ bool pm_ref = !!(msg->opcode & MSG_PM_REF),
+ bound = drm_dev_enter(&xe->drm, &idx);
trace_xe_sched_msg_recv(msg);
- switch (msg->opcode) {
+ switch (msg->opcode & OPCODE_MASK) {
case CLEANUP:
- __guc_exec_queue_process_msg_cleanup(msg);
+ __guc_exec_queue_process_msg_cleanup(msg, bound);
break;
case SET_SCHED_PROPS:
- __guc_exec_queue_process_msg_set_sched_props(msg);
+ __guc_exec_queue_process_msg_set_sched_props(msg, bound);
break;
case SUSPEND:
- __guc_exec_queue_process_msg_suspend(msg);
+ __guc_exec_queue_process_msg_suspend(msg, bound);
break;
case RESUME:
- __guc_exec_queue_process_msg_resume(msg);
+ __guc_exec_queue_process_msg_resume(msg, bound);
break;
case SET_MULTI_QUEUE_PRIORITY:
- __guc_exec_queue_process_msg_set_multi_queue_priority(msg);
+ __guc_exec_queue_process_msg_set_multi_queue_priority(msg,
+ bound);
break;
case CGP_SYNC_MSG:
- __guc_exec_queue_process_msg_cgp_sync(msg);
+ __guc_exec_queue_process_msg_cgp_sync(msg, bound);
break;
default:
XE_WARN_ON("Unknown message type");
}
- xe_pm_runtime_put(xe);
+ if (pm_ref)
+ xe_pm_runtime_put(xe);
+
+ if (bound)
+ drm_dev_exit(idx);
}
static const struct drm_sched_backend_ops drm_sched_ops = {
@@ -2137,10 +2135,17 @@ static void guc_exec_queue_kill(struct xe_exec_queue *q)
static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg *msg,
u32 opcode)
{
- xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
+ struct xe_guc *guc = exec_queue_to_guc(q);
+ struct xe_device *xe = guc_to_xe(guc);
+ int idx;
+ bool bound = drm_dev_enter(&xe->drm, &idx);
INIT_LIST_HEAD(&msg->link);
msg->opcode = opcode & OPCODE_MASK;
+ if (bound) {
+ xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
+ msg->opcode |= MSG_PM_REF;
+ }
msg->private_data = q;
trace_xe_sched_msg_add(msg);
@@ -2150,6 +2155,9 @@ static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
xe_sched_add_msg_locked(&q->guc->sched, msg);
else
xe_sched_add_msg(&q->guc->sched, msg);
+
+ if (bound)
+ drm_dev_exit(idx);
}
static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q,
@@ -2182,10 +2190,7 @@ static void guc_exec_queue_destroy(struct xe_exec_queue *q)
{
struct xe_sched_msg *msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
- if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) && !exec_queue_wedged(q))
- guc_exec_queue_add_msg(q, msg, CLEANUP);
- else
- __guc_exec_queue_destroy(exec_queue_to_guc(q), q);
+ guc_exec_queue_add_msg(q, msg, CLEANUP);
}
static int guc_exec_queue_set_priority(struct xe_exec_queue *q,
@@ -2650,7 +2655,7 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
}
if (do_destroy)
- __guc_exec_queue_destroy(guc, q);
+ guc_exec_queue_destroy_async(q);
}
static int guc_submit_reset_prepare(struct xe_guc *guc)
@@ -3296,7 +3301,7 @@ static void handle_deregister_done(struct xe_guc *guc, struct xe_exec_queue *q)
trace_xe_exec_queue_deregister_done(q);
clear_exec_queue_registered(q);
- __guc_exec_queue_destroy(guc, q);
+ guc_exec_queue_destroy_async(q);
}
> Sorry a lot of this stuff in xe_guc_submit.c around corner cases /
> teardwons is a mess from early Xe work I did, we should aim to clean
> this up with clear semantics.
>
> Matt
>
> > Raag
> >
> > > "Missing outer runtime PM protection\n");
> > > }
> > > }
> > > --
> > > 2.50.1
> > >
^ permalink raw reply related [flat|nested] 17+ messages in thread
* RE: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-06 19:01 ` Matthew Brost
@ 2026-08-07 13:38 ` Gote, Nitin R
2026-08-07 18:39 ` Matthew Brost
0 siblings, 1 reply; 17+ messages in thread
From: Gote, Nitin R @ 2026-08-07 13:38 UTC (permalink / raw)
To: Brost, Matthew
Cc: intel-xe@lists.freedesktop.org, Auld, Matthew, Jadav, Raag
Hi Matt,
> -----Original Message-----
> From: Brost, Matthew <matthew.brost@intel.com>
> Sent: Friday, August 7, 2026 12:32 AM
> To: Jadav, Raag <raag.jadav@intel.com>
> Cc: Gote, Nitin R <nitin.r.gote@intel.com>; intel-xe@lists.freedesktop.org; Auld,
> Matthew <matthew.auld@intel.com>
> Subject: Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM
> protection after hot-unplug
>
> On Thu, Aug 06, 2026 at 02:38:03AM -0700, Matthew Brost wrote:
> > On Thu, Aug 06, 2026 at 10:30:20AM +0200, Raag Jadav wrote:
> > > On Thu, Aug 06, 2026 at 02:27:12PM +0530, Nitin Gote wrote:
> > > > Exec queues are owned by user fds and are destroyed when the fd is closed.
> > > > After hot-unplug this can happen from a deferred close during
> > > > process exit, after the device has been removed and runtime PM has been
> disabled.
> > > > The queue destroy path can call xe_pm_runtime_get_noresume(), and
> > > > since runtime PM is disabled pm_runtime_get_if_in_use() returns no
> > > > reference, so it warns about "Missing outer runtime PM protection".
> > > >
> > > > This is a false positive for the hot-unplug teardown case. The
> > > > device is already unplugged, and the queue destroy path that
> > > > triggered this warning was checked and does not touch hardware
> > > > state after unplug. GuC has already been sanitized by
> > > > guc_fini_hw(), so no H2G is sent and the hardware teardown path is no
> longer reachable.
> > > >
> > > > Skip the warning when the DRM device is already unplugged.
> > > >
> > > > Observed with new IGT core_hotunplug subtests:
> > > > igt@core_hotunplug@hotreplug-with-load
> > > > igt@core_hotunplug@hotunplug-rescan-with-load
> > > >
> > > > v2:
> > > > - Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> > > > and instead exclude hot-unplug from the WARN in
> > > > xe_pm_runtime_get_noresume(). (Matthew Brost)
> > > >
> >
> > I don't think this was suggestion but also I believe I misunderstood
> > the problem, but maybe this works.
> >
> > > > v3:
> > > > - Clarify that the queue destroy path was checked and does not touch
> > > > hardware state after unplug. (Matthew Auld)
> > > >
> > > > Link:
> > > > https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=
> > > > 4
> > > > Cc: Matthew Brost <matthew.brost@intel.com>
> > > > Cc: Matthew Auld <matthew.auld@intel.com>
> > > > Assisted-by: GitHub-Copilot:claude-opus-4.8
> > > > Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> > > > ---
> > > > drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> > > > 1 file changed, 4 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/drivers/gpu/drm/xe/xe_pm.c
> > > > b/drivers/gpu/drm/xe/xe_pm.c index a5289a9df8d2..a038687d5b2c
> > > > 100644
> > > > --- a/drivers/gpu/drm/xe/xe_pm.c
> > > > +++ b/drivers/gpu/drm/xe/xe_pm.c
> > > > @@ -10,6 +10,7 @@
> > > > #include <linux/suspend.h>
> > > > #include <linux/dmi.h>
> > > >
> > > > +#include <drm/drm_drv.h>
> > > > #include <drm/drm_managed.h>
> > > > #include <drm/ttm/ttm_placement.h>
> > > >
> > > > @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct
> > > > xe_device *xe)
> > >
> > > From the documentation
> > >
> > > * This function should be used in inner places where it is surely
> > > already
> > > * protected by outer-bound callers of `xe_pm_runtime_get`.
> > >
> >
> > We should have those.
> >
> > > > if (!ref) {
> > > > pm_runtime_get_noresume(xe->drm.dev);
> > > > - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> > > > + drm_WARN(&xe->drm,
> > > > + !drm_dev_is_unplugged(&xe->drm) &&
> > > > + !xe_pm_suspending_or_resuming(xe),
> > >
> > > So IMO this is more of a band-aid.
> > >
> >
> > So if I'm understanding the problem correctly,
> > pm_runtime_get_if_in_use() (via guc_exec_queue_add_msg()) fails when
> > the device is unplugged? I don't see how the PM reference count could
> > be zero, regardless of whether the device is plugged in or unplugged.
> >
> > My suggestion was (and still is) that guc_exec_queue_destroy() should
> > always add a message and determine whether the device/firmware is
> > still alive while processing that message, then either kick the
> > destory or issue H2G. Other messages likely are also missing proper
> > device alive checks too.
> >
Yes, it was. Initially, I thought to send this as a separate patch once this issue was fixed.
However, I think it's better to include it in this patch itself.
> > If, for some reason, obtaining a PM reference via
> > xe_pm_runtime_get_noresume() doesn't work when the device is
> > unplugged, then that logic should be handled in the message submission
> > layer. For example, use drm_dev_enter()/drm_dev_exit() for hot-unplug
> > protection, only take a PM reference if the device is still bound,
> > encode that information into the message, and then determine the
> > appropriate next steps when the message is processed.
> >
>
> Below is rough sketch of what I was thinking. We can also probably rip out
> EXEC_QUEUE_FLAG_PERMANENT then too. We'd also need solid explaination
> how why xe_pm_runtime_get_noresume doesn't work if device is unplugged, as
> that part it unclear to me too.
Regarding why xe_pm_runtime_get_noresume() doesn't work after hot-unplug, I did some debugging.
You were right that this is not a zero refcount case. At the warning observed: usage_count=3, status=2(RPM_SUSPENDED) and disable_depth=1.
During hot-unplug, the PCI remove path goes through: device_del() -> pm_runtime_remove() -> __pm_runtime_disable(), which disables runtime pm
and increments disable_depth from 0 to 1. The old struct device stays alive because the old drm_device is still referenced by an open fd.
When that fd is finally closed, guc_exec_queue_destroy() runs on the old device where runtime PM is already disabled (disable_depth=1).
In that state, pm_runtime_get_if_in_use() returns -EINVAL regardless of the non-zero usage count, which causes xe_pm_runtime_get_noresume()
to hit the "Missing outer runtime PM protection" WARN.
So this does not appear to be an actual missing runtime PM reference. The failure is caused by runtime PM already being disabled during device removal.
I'll move the handling to the message submission layer as you suggested, taking the PM reference only while drm_dev_enter() indicates the device is still bound,
instead of trying to gate the warning.
To confirm this, I added some debug log and saw:
disable_depth=1, status=2(RPM_SUSPENDED), usage_count=3
Thank you for the rough sketch, Matt. I'll work on that approach.
Nitin
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c
> b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 9036f89dff7d..08000f8db28a 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -1812,32 +1812,15 @@ static void __guc_exec_queue_destroy_async(struct
> work_struct *w) static void guc_exec_queue_destroy_async(struct
> xe_exec_queue *q) {
> INIT_WORK(&q->guc->destroy_async, __guc_exec_queue_destroy_async);
> -
> - /* We must block on kernel engines so slabs are empty on driver unload */
> - if (q->flags & EXEC_QUEUE_FLAG_PERMANENT || exec_queue_wedged(q))
> - guc_exec_queue_do_destroy(q);
> - else
> - xe_destroy_wq_queue(&q->guc->destroy_async);
> + xe_destroy_wq_queue(&q->guc->destroy_async);
> }
>
> -static void __guc_exec_queue_destroy(struct xe_guc *guc, struct
> xe_exec_queue *q) -{
> - /*
> - * Might be done from within the GPU scheduler, need to do async as we
> - * fini the scheduler when the engine is fini'd, the scheduler can't
> - * complete fini within itself (circular dependency). Async resolves
> - * this we and don't really care when everything is fini'd, just that it
> - * is.
> - */
> - guc_exec_queue_destroy_async(q);
> -}
> -
> -static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg
> *msg)
> +static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg
> *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
> struct xe_guc *guc = exec_queue_to_guc(q);
>
> - xe_gt_assert(guc_to_gt(guc), !(q->flags &
> EXEC_QUEUE_FLAG_PERMANENT));
> trace_xe_exec_queue_cleanup_entity(q);
>
> /*
> @@ -1850,10 +1833,10 @@ static void
> __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
> * it is safe to directly destroy the exec queue on driver side, as the GuC
> * will not process further requests and all resources must be cleaned up
> locally.
> */
> - if (exec_queue_registered(q) && xe_uc_fw_is_running(&guc->fw))
> + if (bound && exec_queue_registered(q) &&
> + xe_uc_fw_is_running(&guc->fw))
> disable_scheduling_deregister(guc, q);
> else
> - __guc_exec_queue_destroy(guc, q);
> + guc_exec_queue_destroy_async(q);
> }
>
> static bool guc_exec_queue_allowed_to_change_state(struct xe_exec_queue
> *q) @@ -1861,12 +1844,13 @@ static bool
> guc_exec_queue_allowed_to_change_state(struct xe_exec_queue *q)
> return !exec_queue_killed_or_banned_or_wedged(q) &&
> exec_queue_registered(q); }
>
> -static void __guc_exec_queue_process_msg_set_sched_props(struct
> xe_sched_msg *msg)
> +static void __guc_exec_queue_process_msg_set_sched_props(struct
> xe_sched_msg *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
> struct xe_guc *guc = exec_queue_to_guc(q);
>
> - if (guc_exec_queue_allowed_to_change_state(q))
> + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> init_policies(guc, q);
> kfree(msg);
> }
> @@ -1904,13 +1888,14 @@ static void suspend_fence_signal(struct
> xe_exec_queue *q)
> __suspend_fence_signal(q);
> }
>
> -static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg
> *msg)
> +static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg
> *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
> struct xe_guc *guc = exec_queue_to_guc(q);
>
> if (guc_exec_queue_allowed_to_change_state(q) &&
> !exec_queue_suspended(q) &&
> - exec_queue_enabled(q)) {
> + exec_queue_enabled(q) && bound) {
> wait_event(guc->ct.wq, vf_recovery(guc) ||
> ((q->guc->resume_time != RESUME_PENDING ||
> xe_guc_read_stopped(guc)) &&
> !exec_queue_pending_disable(q))); @@ -1934,11 +1919,12 @@ static void
> __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
> }
> }
>
> -static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg
> *msg)
> +static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg
> *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
>
> - if (guc_exec_queue_allowed_to_change_state(q)) {
> + if (guc_exec_queue_allowed_to_change_state(q) && bound) {
> clear_exec_queue_suspended(q);
> if (!exec_queue_enabled(q)) {
> q->guc->resume_time = RESUME_PENDING; @@ -1950,17
> +1936,20 @@ static void __guc_exec_queue_process_msg_resume(struct
> xe_sched_msg *msg)
> }
> }
>
> -static void __guc_exec_queue_process_msg_set_multi_queue_priority(struct
> xe_sched_msg *msg)
> +static void
> +__guc_exec_queue_process_msg_set_multi_queue_priority(struct
> xe_sched_msg *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
>
> - if (guc_exec_queue_allowed_to_change_state(q))
> + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> guc_exec_queue_send_cgp_sync(q, 0);
>
> kfree(msg);
> }
>
> -static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg
> *msg)
> +static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg
> *msg,
> + bool bound)
> {
> struct xe_exec_queue *q = msg->private_data;
>
> @@ -1969,7 +1958,7 @@ static void
> __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
> * CGP update + CGP_SYNC (re-applies the current priority from
> * q->multi_queue.priority).
> */
> - if (guc_exec_queue_allowed_to_change_state(q))
> + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> guc_exec_queue_send_cgp_sync(q, 0); }
>
> @@ -1982,37 +1971,46 @@ static void
> __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
> #define OPCODE_MASK 0xf
> #define MSG_LOCKED BIT(8)
> #define MSG_HEAD BIT(9)
> +#define MSG_PM_REF BIT(10)
>
> static void guc_exec_queue_process_msg(struct xe_sched_msg *msg) {
> struct xe_device *xe = guc_to_xe(exec_queue_to_guc(msg->private_data));
> + int idx;
> + bool pm_ref = !!(msg->opcode & MSG_PM_REF),
> + bound = drm_dev_enter(&xe->drm, &idx);
>
> trace_xe_sched_msg_recv(msg);
>
> - switch (msg->opcode) {
> + switch (msg->opcode & OPCODE_MASK) {
> case CLEANUP:
> - __guc_exec_queue_process_msg_cleanup(msg);
> + __guc_exec_queue_process_msg_cleanup(msg, bound);
> break;
> case SET_SCHED_PROPS:
> - __guc_exec_queue_process_msg_set_sched_props(msg);
> + __guc_exec_queue_process_msg_set_sched_props(msg,
> + bound);
> break;
> case SUSPEND:
> - __guc_exec_queue_process_msg_suspend(msg);
> + __guc_exec_queue_process_msg_suspend(msg, bound);
> break;
> case RESUME:
> - __guc_exec_queue_process_msg_resume(msg);
> + __guc_exec_queue_process_msg_resume(msg, bound);
> break;
> case SET_MULTI_QUEUE_PRIORITY:
> - __guc_exec_queue_process_msg_set_multi_queue_priority(msg);
> + __guc_exec_queue_process_msg_set_multi_queue_priority(msg,
> +
> + bound);
> break;
> case CGP_SYNC_MSG:
> - __guc_exec_queue_process_msg_cgp_sync(msg);
> + __guc_exec_queue_process_msg_cgp_sync(msg, bound);
> break;
> default:
> XE_WARN_ON("Unknown message type");
> }
>
> - xe_pm_runtime_put(xe);
> + if (pm_ref)
> + xe_pm_runtime_put(xe);
> +
> + if (bound)
> + drm_dev_exit(idx);
> }
>
> static const struct drm_sched_backend_ops drm_sched_ops = { @@ -2137,10
> +2135,17 @@ static void guc_exec_queue_kill(struct xe_exec_queue *q) static
> void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
> *msg,
> u32 opcode) {
> - xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
> + struct xe_guc *guc = exec_queue_to_guc(q);
> + struct xe_device *xe = guc_to_xe(guc);
> + int idx;
> + bool bound = drm_dev_enter(&xe->drm, &idx);
>
> INIT_LIST_HEAD(&msg->link);
> msg->opcode = opcode & OPCODE_MASK;
> + if (bound) {
> + xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
> + msg->opcode |= MSG_PM_REF;
> + }
> msg->private_data = q;
>
> trace_xe_sched_msg_add(msg);
> @@ -2150,6 +2155,9 @@ static void guc_exec_queue_add_msg(struct
> xe_exec_queue *q, struct xe_sched_msg
> xe_sched_add_msg_locked(&q->guc->sched, msg);
> else
> xe_sched_add_msg(&q->guc->sched, msg);
> +
> + if (bound)
> + drm_dev_exit(idx);
> }
>
> static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q, @@ -
> 2182,10 +2190,7 @@ static void guc_exec_queue_destroy(struct xe_exec_queue
> *q) {
> struct xe_sched_msg *msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
>
> - if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) &&
> !exec_queue_wedged(q))
> - guc_exec_queue_add_msg(q, msg, CLEANUP);
> - else
> - __guc_exec_queue_destroy(exec_queue_to_guc(q), q);
> + guc_exec_queue_add_msg(q, msg, CLEANUP);
> }
>
> static int guc_exec_queue_set_priority(struct xe_exec_queue *q, @@ -2650,7
> +2655,7 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct
> xe_exec_queue *q)
> }
>
> if (do_destroy)
> - __guc_exec_queue_destroy(guc, q);
> + guc_exec_queue_destroy_async(q);
> }
>
> static int guc_submit_reset_prepare(struct xe_guc *guc) @@ -3296,7 +3301,7
> @@ static void handle_deregister_done(struct xe_guc *guc, struct
> xe_exec_queue *q)
> trace_xe_exec_queue_deregister_done(q);
>
> clear_exec_queue_registered(q);
> - __guc_exec_queue_destroy(guc, q);
> + guc_exec_queue_destroy_async(q);
> }
>
> > Sorry a lot of this stuff in xe_guc_submit.c around corner cases /
> > teardwons is a mess from early Xe work I did, we should aim to clean
> > this up with clear semantics.
> >
> > Matt
> >
> > > Raag
> > >
> > > > "Missing outer runtime PM protection\n");
> > > > }
> > > > }
> > > > --
> > > > 2.50.1
> > > >
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug
2026-08-07 13:38 ` Gote, Nitin R
@ 2026-08-07 18:39 ` Matthew Brost
0 siblings, 0 replies; 17+ messages in thread
From: Matthew Brost @ 2026-08-07 18:39 UTC (permalink / raw)
To: Gote, Nitin R; +Cc: intel-xe@lists.freedesktop.org, Auld, Matthew, Jadav, Raag
On Fri, Aug 07, 2026 at 07:38:40AM -0600, Gote, Nitin R wrote:
> Hi Matt,
>
> > -----Original Message-----
> > From: Brost, Matthew <matthew.brost@intel.com>
> > Sent: Friday, August 7, 2026 12:32 AM
> > To: Jadav, Raag <raag.jadav@intel.com>
> > Cc: Gote, Nitin R <nitin.r.gote@intel.com>; intel-xe@lists.freedesktop.org; Auld,
> > Matthew <matthew.auld@intel.com>
> > Subject: Re: [PATCH] drm/xe/pm: do not warn about missing runtime PM
> > protection after hot-unplug
> >
> > On Thu, Aug 06, 2026 at 02:38:03AM -0700, Matthew Brost wrote:
> > > On Thu, Aug 06, 2026 at 10:30:20AM +0200, Raag Jadav wrote:
> > > > On Thu, Aug 06, 2026 at 02:27:12PM +0530, Nitin Gote wrote:
> > > > > Exec queues are owned by user fds and are destroyed when the fd is closed.
> > > > > After hot-unplug this can happen from a deferred close during
> > > > > process exit, after the device has been removed and runtime PM has been
> > disabled.
> > > > > The queue destroy path can call xe_pm_runtime_get_noresume(), and
> > > > > since runtime PM is disabled pm_runtime_get_if_in_use() returns no
> > > > > reference, so it warns about "Missing outer runtime PM protection".
> > > > >
> > > > > This is a false positive for the hot-unplug teardown case. The
> > > > > device is already unplugged, and the queue destroy path that
> > > > > triggered this warning was checked and does not touch hardware
> > > > > state after unplug. GuC has already been sanitized by
> > > > > guc_fini_hw(), so no H2G is sent and the hardware teardown path is no
> > longer reachable.
> > > > >
> > > > > Skip the warning when the DRM device is already unplugged.
> > > > >
> > > > > Observed with new IGT core_hotunplug subtests:
> > > > > igt@core_hotunplug@hotreplug-with-load
> > > > > igt@core_hotunplug@hotunplug-rescan-with-load
> > > > >
> > > > > v2:
> > > > > - Drop the drm_dev_is_unplugged() bypass from guc_exec_queue_destroy()
> > > > > and instead exclude hot-unplug from the WARN in
> > > > > xe_pm_runtime_get_noresume(). (Matthew Brost)
> > > > >
> > >
> > > I don't think this was suggestion but also I believe I misunderstood
> > > the problem, but maybe this works.
> > >
> > > > > v3:
> > > > > - Clarify that the queue destroy path was checked and does not touch
> > > > > hardware state after unplug. (Matthew Auld)
> > > > >
> > > > > Link:
> > > > > https://patchwork.freedesktop.org/patch/725773/?series=166744&rev=
> > > > > 4
> > > > > Cc: Matthew Brost <matthew.brost@intel.com>
> > > > > Cc: Matthew Auld <matthew.auld@intel.com>
> > > > > Assisted-by: GitHub-Copilot:claude-opus-4.8
> > > > > Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> > > > > ---
> > > > > drivers/gpu/drm/xe/xe_pm.c | 5 ++++-
> > > > > 1 file changed, 4 insertions(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/drivers/gpu/drm/xe/xe_pm.c
> > > > > b/drivers/gpu/drm/xe/xe_pm.c index a5289a9df8d2..a038687d5b2c
> > > > > 100644
> > > > > --- a/drivers/gpu/drm/xe/xe_pm.c
> > > > > +++ b/drivers/gpu/drm/xe/xe_pm.c
> > > > > @@ -10,6 +10,7 @@
> > > > > #include <linux/suspend.h>
> > > > > #include <linux/dmi.h>
> > > > >
> > > > > +#include <drm/drm_drv.h>
> > > > > #include <drm/drm_managed.h>
> > > > > #include <drm/ttm/ttm_placement.h>
> > > > >
> > > > > @@ -914,7 +915,9 @@ void xe_pm_runtime_get_noresume(struct
> > > > > xe_device *xe)
> > > >
> > > > From the documentation
> > > >
> > > > * This function should be used in inner places where it is surely
> > > > already
> > > > * protected by outer-bound callers of `xe_pm_runtime_get`.
> > > >
> > >
> > > We should have those.
> > >
> > > > > if (!ref) {
> > > > > pm_runtime_get_noresume(xe->drm.dev);
> > > > > - drm_WARN(&xe->drm, !xe_pm_suspending_or_resuming(xe),
> > > > > + drm_WARN(&xe->drm,
> > > > > + !drm_dev_is_unplugged(&xe->drm) &&
> > > > > + !xe_pm_suspending_or_resuming(xe),
> > > >
> > > > So IMO this is more of a band-aid.
> > > >
> > >
> > > So if I'm understanding the problem correctly,
> > > pm_runtime_get_if_in_use() (via guc_exec_queue_add_msg()) fails when
> > > the device is unplugged? I don't see how the PM reference count could
> > > be zero, regardless of whether the device is plugged in or unplugged.
> > >
> > > My suggestion was (and still is) that guc_exec_queue_destroy() should
> > > always add a message and determine whether the device/firmware is
> > > still alive while processing that message, then either kick the
> > > destory or issue H2G. Other messages likely are also missing proper
> > > device alive checks too.
> > >
>
> Yes, it was. Initially, I thought to send this as a separate patch once this issue was fixed.
> However, I think it's better to include it in this patch itself.
>
> > > If, for some reason, obtaining a PM reference via
> > > xe_pm_runtime_get_noresume() doesn't work when the device is
> > > unplugged, then that logic should be handled in the message submission
> > > layer. For example, use drm_dev_enter()/drm_dev_exit() for hot-unplug
> > > protection, only take a PM reference if the device is still bound,
> > > encode that information into the message, and then determine the
> > > appropriate next steps when the message is processed.
> > >
> >
> > Below is rough sketch of what I was thinking. We can also probably rip out
> > EXEC_QUEUE_FLAG_PERMANENT then too. We'd also need solid explaination
> > how why xe_pm_runtime_get_noresume doesn't work if device is unplugged, as
> > that part it unclear to me too.
>
> Regarding why xe_pm_runtime_get_noresume() doesn't work after hot-unplug, I did some debugging.
> You were right that this is not a zero refcount case. At the warning observed: usage_count=3, status=2(RPM_SUSPENDED) and disable_depth=1.
>
> During hot-unplug, the PCI remove path goes through: device_del() -> pm_runtime_remove() -> __pm_runtime_disable(), which disables runtime pm
> and increments disable_depth from 0 to 1. The old struct device stays alive because the old drm_device is still referenced by an open fd.
> When that fd is finally closed, guc_exec_queue_destroy() runs on the old device where runtime PM is already disabled (disable_depth=1).
> In that state, pm_runtime_get_if_in_use() returns -EINVAL regardless of the non-zero usage count, which causes xe_pm_runtime_get_noresume()
> to hit the "Missing outer runtime PM protection" WARN.
>
Thanks for looking ino this part. Make sure to put this into the commit
and bonus points for updating our xe_pm.c doc around hazard of PM refs
after the device is hotplugged. I suspect this isn't the only case in Xe
where we have missing drm_dev_enter/exits.
Matt
> So this does not appear to be an actual missing runtime PM reference. The failure is caused by runtime PM already being disabled during device removal.
> I'll move the handling to the message submission layer as you suggested, taking the PM reference only while drm_dev_enter() indicates the device is still bound,
> instead of trying to gate the warning.
>
> To confirm this, I added some debug log and saw:
> disable_depth=1, status=2(RPM_SUSPENDED), usage_count=3
>
> Thank you for the rough sketch, Matt. I'll work on that approach.
>
> Nitin
>
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c
> > b/drivers/gpu/drm/xe/xe_guc_submit.c
> > index 9036f89dff7d..08000f8db28a 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> > @@ -1812,32 +1812,15 @@ static void __guc_exec_queue_destroy_async(struct
> > work_struct *w) static void guc_exec_queue_destroy_async(struct
> > xe_exec_queue *q) {
> > INIT_WORK(&q->guc->destroy_async, __guc_exec_queue_destroy_async);
> > -
> > - /* We must block on kernel engines so slabs are empty on driver unload */
> > - if (q->flags & EXEC_QUEUE_FLAG_PERMANENT || exec_queue_wedged(q))
> > - guc_exec_queue_do_destroy(q);
> > - else
> > - xe_destroy_wq_queue(&q->guc->destroy_async);
> > + xe_destroy_wq_queue(&q->guc->destroy_async);
> > }
> >
> > -static void __guc_exec_queue_destroy(struct xe_guc *guc, struct
> > xe_exec_queue *q) -{
> > - /*
> > - * Might be done from within the GPU scheduler, need to do async as we
> > - * fini the scheduler when the engine is fini'd, the scheduler can't
> > - * complete fini within itself (circular dependency). Async resolves
> > - * this we and don't really care when everything is fini'd, just that it
> > - * is.
> > - */
> > - guc_exec_queue_destroy_async(q);
> > -}
> > -
> > -static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg
> > *msg)
> > +static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg
> > *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> > struct xe_guc *guc = exec_queue_to_guc(q);
> >
> > - xe_gt_assert(guc_to_gt(guc), !(q->flags &
> > EXEC_QUEUE_FLAG_PERMANENT));
> > trace_xe_exec_queue_cleanup_entity(q);
> >
> > /*
> > @@ -1850,10 +1833,10 @@ static void
> > __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
> > * it is safe to directly destroy the exec queue on driver side, as the GuC
> > * will not process further requests and all resources must be cleaned up
> > locally.
> > */
> > - if (exec_queue_registered(q) && xe_uc_fw_is_running(&guc->fw))
> > + if (bound && exec_queue_registered(q) &&
> > + xe_uc_fw_is_running(&guc->fw))
> > disable_scheduling_deregister(guc, q);
> > else
> > - __guc_exec_queue_destroy(guc, q);
> > + guc_exec_queue_destroy_async(q);
> > }
> >
> > static bool guc_exec_queue_allowed_to_change_state(struct xe_exec_queue
> > *q) @@ -1861,12 +1844,13 @@ static bool
> > guc_exec_queue_allowed_to_change_state(struct xe_exec_queue *q)
> > return !exec_queue_killed_or_banned_or_wedged(q) &&
> > exec_queue_registered(q); }
> >
> > -static void __guc_exec_queue_process_msg_set_sched_props(struct
> > xe_sched_msg *msg)
> > +static void __guc_exec_queue_process_msg_set_sched_props(struct
> > xe_sched_msg *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> > struct xe_guc *guc = exec_queue_to_guc(q);
> >
> > - if (guc_exec_queue_allowed_to_change_state(q))
> > + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> > init_policies(guc, q);
> > kfree(msg);
> > }
> > @@ -1904,13 +1888,14 @@ static void suspend_fence_signal(struct
> > xe_exec_queue *q)
> > __suspend_fence_signal(q);
> > }
> >
> > -static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg
> > *msg)
> > +static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg
> > *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> > struct xe_guc *guc = exec_queue_to_guc(q);
> >
> > if (guc_exec_queue_allowed_to_change_state(q) &&
> > !exec_queue_suspended(q) &&
> > - exec_queue_enabled(q)) {
> > + exec_queue_enabled(q) && bound) {
> > wait_event(guc->ct.wq, vf_recovery(guc) ||
> > ((q->guc->resume_time != RESUME_PENDING ||
> > xe_guc_read_stopped(guc)) &&
> > !exec_queue_pending_disable(q))); @@ -1934,11 +1919,12 @@ static void
> > __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
> > }
> > }
> >
> > -static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg
> > *msg)
> > +static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg
> > *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> >
> > - if (guc_exec_queue_allowed_to_change_state(q)) {
> > + if (guc_exec_queue_allowed_to_change_state(q) && bound) {
> > clear_exec_queue_suspended(q);
> > if (!exec_queue_enabled(q)) {
> > q->guc->resume_time = RESUME_PENDING; @@ -1950,17
> > +1936,20 @@ static void __guc_exec_queue_process_msg_resume(struct
> > xe_sched_msg *msg)
> > }
> > }
> >
> > -static void __guc_exec_queue_process_msg_set_multi_queue_priority(struct
> > xe_sched_msg *msg)
> > +static void
> > +__guc_exec_queue_process_msg_set_multi_queue_priority(struct
> > xe_sched_msg *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> >
> > - if (guc_exec_queue_allowed_to_change_state(q))
> > + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> > guc_exec_queue_send_cgp_sync(q, 0);
> >
> > kfree(msg);
> > }
> >
> > -static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg
> > *msg)
> > +static void __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg
> > *msg,
> > + bool bound)
> > {
> > struct xe_exec_queue *q = msg->private_data;
> >
> > @@ -1969,7 +1958,7 @@ static void
> > __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
> > * CGP update + CGP_SYNC (re-applies the current priority from
> > * q->multi_queue.priority).
> > */
> > - if (guc_exec_queue_allowed_to_change_state(q))
> > + if (guc_exec_queue_allowed_to_change_state(q) && bound)
> > guc_exec_queue_send_cgp_sync(q, 0); }
> >
> > @@ -1982,37 +1971,46 @@ static void
> > __guc_exec_queue_process_msg_cgp_sync(struct xe_sched_msg *msg)
> > #define OPCODE_MASK 0xf
> > #define MSG_LOCKED BIT(8)
> > #define MSG_HEAD BIT(9)
> > +#define MSG_PM_REF BIT(10)
> >
> > static void guc_exec_queue_process_msg(struct xe_sched_msg *msg) {
> > struct xe_device *xe = guc_to_xe(exec_queue_to_guc(msg->private_data));
> > + int idx;
> > + bool pm_ref = !!(msg->opcode & MSG_PM_REF),
> > + bound = drm_dev_enter(&xe->drm, &idx);
> >
> > trace_xe_sched_msg_recv(msg);
> >
> > - switch (msg->opcode) {
> > + switch (msg->opcode & OPCODE_MASK) {
> > case CLEANUP:
> > - __guc_exec_queue_process_msg_cleanup(msg);
> > + __guc_exec_queue_process_msg_cleanup(msg, bound);
> > break;
> > case SET_SCHED_PROPS:
> > - __guc_exec_queue_process_msg_set_sched_props(msg);
> > + __guc_exec_queue_process_msg_set_sched_props(msg,
> > + bound);
> > break;
> > case SUSPEND:
> > - __guc_exec_queue_process_msg_suspend(msg);
> > + __guc_exec_queue_process_msg_suspend(msg, bound);
> > break;
> > case RESUME:
> > - __guc_exec_queue_process_msg_resume(msg);
> > + __guc_exec_queue_process_msg_resume(msg, bound);
> > break;
> > case SET_MULTI_QUEUE_PRIORITY:
> > - __guc_exec_queue_process_msg_set_multi_queue_priority(msg);
> > + __guc_exec_queue_process_msg_set_multi_queue_priority(msg,
> > +
> > + bound);
> > break;
> > case CGP_SYNC_MSG:
> > - __guc_exec_queue_process_msg_cgp_sync(msg);
> > + __guc_exec_queue_process_msg_cgp_sync(msg, bound);
> > break;
> > default:
> > XE_WARN_ON("Unknown message type");
> > }
> >
> > - xe_pm_runtime_put(xe);
> > + if (pm_ref)
> > + xe_pm_runtime_put(xe);
> > +
> > + if (bound)
> > + drm_dev_exit(idx);
> > }
> >
> > static const struct drm_sched_backend_ops drm_sched_ops = { @@ -2137,10
> > +2135,17 @@ static void guc_exec_queue_kill(struct xe_exec_queue *q) static
> > void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
> > *msg,
> > u32 opcode) {
> > - xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
> > + struct xe_guc *guc = exec_queue_to_guc(q);
> > + struct xe_device *xe = guc_to_xe(guc);
> > + int idx;
> > + bool bound = drm_dev_enter(&xe->drm, &idx);
> >
> > INIT_LIST_HEAD(&msg->link);
> > msg->opcode = opcode & OPCODE_MASK;
> > + if (bound) {
> > + xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q)));
> > + msg->opcode |= MSG_PM_REF;
> > + }
> > msg->private_data = q;
> >
> > trace_xe_sched_msg_add(msg);
> > @@ -2150,6 +2155,9 @@ static void guc_exec_queue_add_msg(struct
> > xe_exec_queue *q, struct xe_sched_msg
> > xe_sched_add_msg_locked(&q->guc->sched, msg);
> > else
> > xe_sched_add_msg(&q->guc->sched, msg);
> > +
> > + if (bound)
> > + drm_dev_exit(idx);
> > }
> >
> > static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q, @@ -
> > 2182,10 +2190,7 @@ static void guc_exec_queue_destroy(struct xe_exec_queue
> > *q) {
> > struct xe_sched_msg *msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
> >
> > - if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) &&
> > !exec_queue_wedged(q))
> > - guc_exec_queue_add_msg(q, msg, CLEANUP);
> > - else
> > - __guc_exec_queue_destroy(exec_queue_to_guc(q), q);
> > + guc_exec_queue_add_msg(q, msg, CLEANUP);
> > }
> >
> > static int guc_exec_queue_set_priority(struct xe_exec_queue *q, @@ -2650,7
> > +2655,7 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct
> > xe_exec_queue *q)
> > }
> >
> > if (do_destroy)
> > - __guc_exec_queue_destroy(guc, q);
> > + guc_exec_queue_destroy_async(q);
> > }
> >
> > static int guc_submit_reset_prepare(struct xe_guc *guc) @@ -3296,7 +3301,7
> > @@ static void handle_deregister_done(struct xe_guc *guc, struct
> > xe_exec_queue *q)
> > trace_xe_exec_queue_deregister_done(q);
> >
> > clear_exec_queue_registered(q);
> > - __guc_exec_queue_destroy(guc, q);
> > + guc_exec_queue_destroy_async(q);
> > }
> >
> > > Sorry a lot of this stuff in xe_guc_submit.c around corner cases /
> > > teardwons is a mess from early Xe work I did, we should aim to clean
> > > this up with clear semantics.
> > >
> > > Matt
> > >
> > > > Raag
> > > >
> > > > > "Missing outer runtime PM protection\n");
> > > > > }
> > > > > }
> > > > > --
> > > > > 2.50.1
> > > > >
^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2026-08-07 18:39 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-06 8:57 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
2026-08-06 8:30 ` Raag Jadav
2026-08-06 9:38 ` Matthew Brost
2026-08-06 19:01 ` Matthew Brost
2026-08-07 13:38 ` Gote, Nitin R
2026-08-07 18:39 ` Matthew Brost
2026-08-06 8:56 ` ✓ CI.KUnit: success for drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug (rev2) Patchwork
2026-08-06 9:52 ` ✓ Xe.CI.BAT: " Patchwork
2026-08-06 18:39 ` ✗ Xe.CI.FULL: failure " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2026-07-30 5:36 [PATCH] drm/xe/pm: do not warn about missing runtime PM protection after hot-unplug Nitin Gote
2026-07-30 5:59 ` Raag Jadav
2026-07-30 12:05 ` Gote, Nitin R
2026-07-30 12:31 ` Raag Jadav
2026-08-06 6:09 ` Gote, Nitin R
2026-08-06 6:27 ` Raag Jadav
2026-07-30 12:35 ` Matthew Auld
2026-08-06 6:10 ` Gote, Nitin R
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox