Signed-off-by: NKarol Herbst <karolherbst@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NKarol Herbst <karolherbst@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

v2: Set entry to 0xff if not found
    Add cap entry for ver 0x30 tables
    Rework to fix memory leak
v3: More error checks
    Simplify check for invalid entries
v4: disable for ver 0x10 for now
    move assignments after the second last return
Signed-off-by: NKarol Herbst <karolherbst@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Greatly improves the chances of recovering the GPU from a CTXSW_TIMEOUT.

Tested with piglit's arb_shader_image_load_store-atomicity, which causes
GR to hang in such a way that recovery failed (CTXSW_TIMEOUT continually
re-triggers).
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This will serve as a basis for implementing some improvements to how
we recover the GPU from channel errors.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The previous commit simply changes the interface, but should result in
the same behaviour as previously.  This commit has been split out from
it as it can result in a different channel being selected.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

We'll be wanting to reuse this logic in more places.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This will be used by callers outside of fifo interrupt handlers.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

FIFO gives us load/save/switch status, and we need to be able to determine
which direction a "switch" is failing during channel recovery.

In order to do this, we apparently need to query the engine itself.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This prevents *very* long waits while attempting to destroy channels
after a fault has occurred.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

These will be used to improve error recovery behaviour.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

There are instances (such as non-recoverable GPU page faults) where
NVKM decides that a channel's context is no longer viable, and will
be removed from the runlist.

This commit notifies the owner of the channel when this happens, so
it has the opportunity to take some kind of recovery action instead
of hanging.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

If the led class registration fails, we free drm->led but do not reset
it to NULL, which means that the suspend/resume/fini function will act
as if everything went well in init() and will likely crash the kernel.

This patch adds the missing drm->led = NULL.
Reported-by: NEmmanuel Pescosta <emmanuelpescosta099@gmail.com>
Signed-off-by: NMartin Peres <martin.peres@free.fr>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Tested on a G92, seems to work. Confirmed by 8 mmiotraces.
Signed-off-by: NKarol Herbst <karolherbst@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NKarol Herbst <karolherbst@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The issue was recorded in fd.o bug 27501, not 25701.
Signed-off-by: NIlia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This will allow the DRM to share memory objects between clients later
down the track.

For the moment, the only immediate benefit is less logic required to
handle suspend/resume.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

These were a little insane.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

We never have any need for a double-linked list here, and as there's
generally a large number of these objects, replace it with a single-
linked list in order to save some memory.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

We want a supervisor client of NVKM (such as the DRM) to be able to
allow sharing of resources (such as memory objects) between clients.

To allow this, the supervisor creates all its clients as children of
itself, and will use an upcoming ioctl to permit sharing.

Currently it's not possible for indirect clients to use subclients.
Supporting this will require an additional field in the main ioctl.
This isn't important currently, but will need to be fixed for virt.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Preparation for supporting subclients.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Preparation for supporting subclients, and also good for consistency.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>