Add documentation for device wedged event along with its consumer expectations. For now it is amended to 'Device reset' chapter, but with extended functionality in the future it can be refactored into its own chapter. Signed-off-by: Raag Jadav <raag.jadav@xxxxxxxxx> --- Documentation/gpu/drm-uapi.rst | 42 ++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst index 370d820be248..c1186dfd283d 100644 --- a/Documentation/gpu/drm-uapi.rst +++ b/Documentation/gpu/drm-uapi.rst @@ -313,6 +313,22 @@ driver separately, with no common DRM interface. Ideally this should be properly integrated at DRM scheduler to provide a common ground for all drivers. After a reset, KMD should reject new command submissions for affected contexts. +Drivers can optionally make use of device wedged event (implemented as +drm_dev_wedged_event() in DRM subsystem) which notifies userspace of wedged +(hanged/unusable) state of the DRM device through a uevent. This is useful +especially in cases where the device is no longer operating as expected even +after a hardware reset and has become unrecoverable from driver context. +Purpose of this implementation is to provide drivers a generic way to recover +with the help of userspace intervention, and hence the vendor agnostic nature +of the event. + +Different drivers may have different ideas of a "wedged device" depending on +their hardware implementation. It is up to the drivers to decide when they see +the need for recovery and how they want to recover from the available methods. +Current implementation defines three recovery methods, out of which, drivers +can choose to support any one or multiple of them. Preferred recovery method +will be sent in the uevent environment as WEDGED=<method>. + User Mode Driver ---------------- @@ -323,6 +339,32 @@ if the UMD requires it. After detecting a reset, UMD will then proceed to report it to the application using the appropriate API error code, as explained in the section below about robustness. +On device wedged scenario, userspace will receive a uevent from KMD with +its preferred recovery method in the uevent environment as WEDGED=<method>. +Userspace consumers (sysadmin) can define udev rules to parse this event +and take respective action to recover the device. + +.. table:: Wedged Device Recovery + + =============== ================================== + Recovery method Consumer expectations + =============== ================================== + rebind unbind + rebind driver + bus-reset unbind + reset bus device + rebind + reboot reboot system + =============== ================================== + +Userspace consumers can optionally read the recovery methods supported by the +device via ``wedge_recovery`` sysfs attribute:: + + $ cat /sys/class/drm/card<N>/wedge_recovery + rebind + bus-reset + reboot + +This is useful in cases where the device supports multiple recovery methods +which can be used as fallbacks. + Robustness ---------- -- 2.34.1
