| Age | Commit message (Collapse) | Author |
|---|
|
We want to make sure jobs that trigger unhandled page fault get killed
immediately, and jobs queued to the same context after the pagefault
occured get executed properly.
v3:
* New patch
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
|
|
We had use-after-free faults in the past because the MMU context
while released while jobs were still in flight, and the job cleanup
path assumed the context was still present. Add a test to make sure this
won't happen again.
v2:
* Add R-b
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
|
|
Extend the pan-reset test to make sure jobs from other context get
resumed after a reset.
v2:
* New patch
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
|
|
Use NULL jobs and WRITE_VALUE job loops to test job submission. This
way we get tests that pass on both Midgard and Bifrost without pulling
the pan_pack() infrastructure from mesa. Note that we'll have to do that
at some point if we want to add complex tests, but we're not there yet.
v2:
* Move of the pan-reset changes out of this commit
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
|
|
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
|
|
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Petri Latvala <petri.latvala@intel.com>
|
