There is a new IB flag that enables this new behavior.
Full invalidation is unnecessary for RELEASE_MEM and doesn't make sense
when draw calls from two adjacent gfx IBs run in parallel. This will be
the new default for Mesa.

v2: bump the version
Signed-off-by: NMarek Olšák <marek.olsak@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This callback writes a value to a register and then reads
back another register and waits for a value in a single
operation.

Provide a helper function using two operations for engines
that don't support this opertion.
Reviewed-by: NHuang Rui <ray.huang@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Stuffing the PASID mapping into the VM flush isn't flexible enough since
the PASID mapping changes not as often as we need a VM flush.

v2: add missing use of gmc_v7_0_emit_pasid_mapping
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Beneficial when a lot of processes are waiting for VMIDs.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Allows us to wait for a register value/mask on a ring.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

All HDP invalidation and most flush can now be replaced by the generic
ASIC function.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Acked-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

rd the pasid from the VM code to the emit_vm_flush function and update
all implementations with the new parameter.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

sed -i "s/vm_id/vmid/g" drivers/gpu/drm/amd/amdgpu/*.c
sed -i "s/vm_id/vmid/g" drivers/gpu/drm/amd/amdgpu/*.h
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

On Tonga VF, there're 2 sources updating wptr registers for
sdma3: 1) polling mem and 2) doorbell. When doorbell and polling
mem are both enabled on sdma3, there will be collision hit in
occasion between those two sources when ucode and h/w are doing
the updating on wptr register in parallel. Issue doesn't happen
on CP GFX/Compute since CP drops all doorbell writes when VF is
inactive. So enable polling mem and don't use doorbell for SDMA3.
Signed-off-by: NPixel Ding <Pixel.Ding@amd.com>
Reviewed-by: NMonk Liu <monk.liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This fence is used by KFD to keep memory resident while user mode
queues are enabled. Trying to evict memory will trigger the
enable_signaling callback, which starts a KFD eviction, which
involves preempting user mode queues before signaling the fence.
There is one such fence per process.

v2:
* Grab a reference to mm_struct
* Dereference fence after NULL check
* Simplify fence release, no need to signal without anyone waiting
* Added signed-off-by Harish, who is the original author of this code

v3:
* update MAINTAINERS file
* change amd_kfd_ prefix to amdkfd_
* remove useless initialization of variable to NULL

v4:
* set amdkfd_fence_ops to be static
* Suggested by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NHarish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NOded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: NOded Gabbay <oded.gabbay@gmail.com>

Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NOded Gabbay <oded.gabbay@gmail.com>

This moves and renames the AMDGPU scheduler to a common location in DRM
in order to facilitate re-use by other drivers. This is mostly a straight
forward rename with no code changes.

One notable exception is the function to_drm_sched_fence(), which is no
longer a inline header function to avoid the need to export the
drm_sched_fence_ops_scheduled and drm_sched_fence_ops_finished structures.
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Tested-by: NDieter Nützel <Dieter@nuetzel-hh.de>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NLucas Stach <l.stach@pengutronix.de>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

cleanups, now only operate on the given ring
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Register accessing is performed when IRQ is disabled. Never sleep in
this function.

Known issue: dead sleep in many use cases of index/data registers.

v2:
 - wrap polling fence functions.
 - don't trigger IRQ for polling in case of wrongly fence signal.

v3:
 - handle wrap round gracefully.
 - add comments for polling function

v4:
 - don't return negative timeout confused with error code
Signed-off-by: Npding <Pixel.Ding@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add an initial framework for changing the HW priorities of rings. The
framework allows requesting priority changes for the lifetime of an
amdgpu_job. After the job completes the priority will decay to the next
lowest priority for which a request is still valid.

A new ring function set_priority() can now be populated to take care of
the HW specific programming sequence for priority changes.

v2: set priority before emitting IB, and take a ref on amdgpu_job
v3: use AMD_SCHED_PRIORITY_* instead of AMDGPU_CTX_PRIORITY_*
v4: plug amdgpu_ring_restore_priority_cb into amdgpu_job_free_cb
v5: use atomic for tracking job priorities instead of last_job
v6: rename amdgpu_ring_priority_[get/put]() and align parameters
v7: replace spinlocks with mutexes for KIQ compatibility
v8: raise ring priority during cs_ioctl, instead of job_run
v9: priority_get() before push_job()
Reviewed-by: NChristian König <christian.koenig@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This patch provides a guarantee that the first n queues allocated by
an application will be on different pipes. Where n is the number of
pipes available from the hardware.

This helps avoid ring aliasing which can result in work executing in
time-sliced mode instead of truly parallel mode.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Overwriting still used ring content has a low probability to cause
problems, not writing at all has 100% probability to cause problems.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>

Keep them where they belong.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>

  In review, Christian would like to keep the logic
  inside amdgpu_vm.c with a cost of slightly slower.
  The loop is still optimized out with this patch.

v2: remove the if statement. Now it is not slower.
Signed-off-by: NAlex Xie <AlexBin.Xie@amd.com>
Reviewed-by: NChristian König <christian.koeng@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Depending on usage patterns, the current LRU policy may create a
non-injective mapping between userspace ring ids and kernel rings.

This behaviour is undesired as apps that attempt to fill all HW blocks
would be unable to reach some of them.

This change forces the LRU policy to create bijective mappings only.

v2: compress ring_blacklist
v3: simplify amdgpu_ring_is_blacklisted() logic
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Reviewed-by: NNicolai Hähnle <nicolai.haehnle@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Use an LRU policy to map usermode rings to HW compute queues.

Most compute clients use one queue, and usually the first queue
available. This results in poor pipe/queue work distribution when
multiple compute apps are running. In most cases pipe 0 queue 0 is
the only queue that gets used.

In order to better distribute work across multiple HW queues, we adopt
a policy to map the usermode ring ids to the LRU HW queue.

This fixes a large majority of multi-app compute workloads sharing the
same HW queue, even though 7 other queues are available.

v2: use ring->funcs->type instead of ring->hw_ip
v3: remove amdgpu_queue_mapper_funcs
v4: change ring_lru_list_lock to spinlock, grab only once in lru_get()
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

  Move several if statements and a loop statment from
  run time to initialization time.
Signed-off-by: NAlex Xie <AlexBin.Xie@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add the ring function callbacks for the encode rings.
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Needed for the proper command sequence for VCN.
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add the ring function callbacks for the decode ring.
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Acked-by: NChunming Zhou <david1.zhou@amd.com>
Acked-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

1,TDR will kickout guilty job if it hang exceed the threshold
of the given one from kernel paramter "job_hang_limit", that
way a bad command stream will not infinitly cause GPU hang.

by default this threshold is 1 so a job will be kicked out
after it hang.

2,if a job timeout TDR routine will not reset all sched/ring,
instead if will only reset on the givn one which is indicated
by @job of amdgpu_sriov_gpu_reset, that way we don't need to
reset and recover each sched/ring if we already know which job
cause GPU hang.

3,unblock sriov_gpu_reset for AI family.

V2:
1:put kickout guilty job after sched parked.
2:since parking scheduler prior to kickout already occupies a
while, we can do last check on the in question job before
doing hw_reset.

TODO:
1:when a job is considered as guilty, we should mark some flag
in its fence status flag, and let UMD side aware that this
fence signaling is not due to job complete but job hang.

2:if gpu reset cause all video memory lost, we need introduce
a new policy to implement TDR, like drop all jobs not yet
signaled, and all IOCTL on this device will return ERROR
DEVICE_LOST.
this will be implemented later.
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

AI affected:

CP/HW team requires KMD insert FRAME_CONTROL(end) after
the last IB and before the fence of this DMAframe.

this is to make sure the cache are flushed, and it's a must
change no matter MCBP/SR-IOV or bare-metal case because new
CP hw won't do the cache flush for each IB anymore, it just
leaves it to KMD now.

with this patch, certain MCBP hang issue when rendering
vulkan/chained-ib are resolved.

v2: drop gfx8 changes.  gfx8 is not affected (Alex)
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For Vega10 we have 18 VM invalidation engines for each VMHUB.

Start to assign them manually to the rings.

v2: add a BUG_ON if we use to many engines
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add the info which ring belonging to which VMHUB.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

ring->buf_mask need be set prior to ring_clear_ring invoke
and fix ring_clear_ring as well which should use buf_mask
instead of ptr_mask
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Avoids passing around additional parameters during setup.
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

And initialize them
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NLeo Liu <leo.liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Newer asics need 64 bit wptrs.  If the wptr is now
smaller than the rptr that doesn't indicate a wrap-around
anymore.

v2: integrate Christian's comments.
Signed-off-by: NKen Wang <Qingqing.Wang@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

vi_mqd is only used by VI family but mqd_ptr and mqd_backup is
common for all ASIC, so change the pointer to void.
Signed-off-by: NXiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMonk Liu <Monk.Liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

we can use it clear ring buffer instead of fullfill
0, which is not correct for engine
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

introduce a new mqd member in ring is for later usage.
we need keep a clean version of MQD for the purpose
of recovering compute rings from hang.
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

During virtual runtime, need to send command to kiq ring to
read/write GPU registers. Add two interface to support the two
actions.
Signed-off-by: NXiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: NMonk Linu <Monk.Liu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add a new ring type definition for KIQ. KIQ is used for interaction
between driver and CP.
Signed-off-by: NXiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: NTrigger Huang <trigger.huang@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>