Do ring clear before ring test, otherwise compute ring test will
fail after gpu resetting. Still can't find the root cause, just
workaround it.
Signed-off-by: NXiangliang.Yu <Xiangliang.Yu@amd.com>
Acked-by: NMonk Liu <Monk.Liu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

since now gpu reset is unified with gpu_recover
for both bare-metal and SR-IOV:

1)rename in_sriov_reset to in_gpu_reset
2)move lock_reset from adev->virt to adev
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>

Normally all waiting get timeout if there's one.
Release the lock and return immediately when timeout happens.

v2:
 - set the se_sh to broadcase before return
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: Npding <Pixel.Ding@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

When this VF stays in exclusive mode for long, other VFs will be
impacted.

The redundant messages causes exclusive mode timeout when they're
redirected. That is a normal use case for cloud service to redirect
guest log to virtual serial port.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: Npding <Pixel.Ding@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

those RLC used buffers are not cleared in GFX's sw_fini
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>

Using the ARRAY_SIZE macro improves the readability of the code.

Found with Coccinelle with the following semantic patch:
@r depends on (org || report)@
type T;
T[] E;
position p;
@@
(
 (sizeof(E)@p /sizeof(*E))
|
 (sizeof(E)@p /sizeof(E[...]))
|
 (sizeof(E)@p /sizeof(T))
)
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NJérémy Lefaure <jeremy.lefaure@lse.epita.fr>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Programming CP_HQD_QUEUE_PRIORITY enables a queue to take priority over
other queues on the same pipe. Multiple queues on a pipe are timesliced
so this gives us full precedence over other queues.

Programming CP_HQD_PIPE_PRIORITY changes the SPI_ARB_PRIORITY of the
wave as follows:
        0x2: CS_H
        0x1: CS_M
        0x0: CS_L

The SPI block will then dispatch work according to the policy set by
SPI_ARB_PRIORITY. In the current policy CS_H is higher priority than
gfx.

In order to prevent getting stuck in loops of resources bouncing between
GFX and high priority compute and introducing further latency, we
statically reserve a portion of the pipe.

v2: fix srbm_select to ring->queue and use ring->funcs->type
v3: use AMD_SCHED_PRIORITY_* instead of AMDGPU_CTX_PRIORITY_*
v4: switch int to enum amd_sched_priority
v5: corresponding changes for srbm_lock
v6: change CU reservation to PIPE_PERCENT allocation
v7: use kiq instead of MMIO
v8: back to MMIO, and make the implementation sleep safe.
v9: corresponding changes for splitting HIGH into _HW/_SW
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>

Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NRex Zhu <Rex.Zhu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This isn't safe if we have multiple GPUs plugged in, since
there is only one copy of this struct in the bss, just allocate
on stack, it's 40/108 bytes which should be safe.
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NRex Zhu <Rex.Zhu@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

v2:
move kcq_disable out of SRIOV, make it genearal
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>

bo_free on csa is too late to put in amdgpu_fini because that
time ttm is already finished,
Move it earlier to avoid the page fault.
Signed-off-by: NMonk Liu <Monk.Liu@amd.com>
Signed-off-by: NHorace Chen <horace.chen@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

currently in_reset is only used in sriov gpu reset, and it
will be used for other non-gfx hw component later, like
PSP, so move it from gfx to adev and rename to in_sriov_reset
make more sense.
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>

Signed-off-by: NEvan Quan <evan.quan@amd.com>
Reviewed-by: NFlora Cui <Flora.Cui@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Confirmed with the hw team.  It's the same for all asics.
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Newer versions of the CP firmware require changes in how the driver
initializes the hw block.
Change the firmware name for new firmware to maintain compatibility with
older kernels.
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NEvan Quan <evan.quan@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This register only has a single instance in the hardware. Its value
applies to all VMIDS.
Signed-off-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NOded Gabbay <oded.gabbay@gmail.com>

Saves us even more loc.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Saves us quite a bunch of loc.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Signed-off-by: NHuang Rui <ray.huang@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This is just future proofing code, not something that can be triggered
in real life.  We're testing to make sure we don't shift wrap when we
do "1ull << i" so "i" has to be in the 0-63 range.  If it's 64 then we
have gone too far.
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

With previous golden settings, compute task can't use
reserved LDS (32K) on CU0 and CU1. On 64K LDS system,
if compute work group allocate more than 32K LDS, then
it can't be dispatched to CU0 and CU1 because of the
reservation. This enables compute task to use reserved
LDS on CU0 and CU1.
Signed-off-by: NOak Zeng <Oak.Zeng@amd.com>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Drop the deactivation in KIQ init and drop the KCQ disabling
via KIQ.  We disable the MEC shortly after anyway, so there is
no need to wait for all of this.  Doing so seems to leave the
MEC in a bad way.
Tested-by: NMichel Dänzer <michel.daenzer@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Newer asics with 4 SEs are not able to fit the entire bitmask in the
original field, use an array instead.

v2: keep cu_ao_mask for backward compatibility.
Signed-off-by: NFlora Cui <Flora.Cui@amd.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

caused by not program dynamic_cu_mask_addr in the KIQ MQD.

v2: create struct vi_mqd_allocation in FB which will contain
1. PM4 MQD structure.
2. Write Pointer Poll Memory.
3. Read Pointer Report Memory
4. Dynamic CU Mask.
5. Dynamic RB Mask.
Signed-off-by: NRex Zhu <Rex.Zhu@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The interrupt registers are not indexed.

Fixes: 763a47b8 (drm/amdgpu: teach amdgpu how to enable interrupts for any pipe v3)
Acked-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It was duplicated across multiple generations.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This will allow us to share more mec code.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Lots more common stuff.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

They are gfx related, not general helpers.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Make it consistent.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Same function was duplicated in all gfx IP files.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Fixes hangs on single MEC asics.

Fixes: 2ed286fb434 (drm/amdgpu: new queue policy, take first 2 queues of each pipe v2)
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The same function was duplicated in all the gfx IPs. Use
a single implementation for all.

v2: use static inline (Alex Xie)
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Suggested-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Always use the max for the family rather than the per sku limits.
This makes sure the mask is always the max size to avoid reporting
the wrong number of CUs.
Reviewed-by: NAlex Xie <AlexBin.Xie@amd.com>
Reviewed-by: NAndres Rodriguez <andresx7@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Tonga based asics may experience hangs when an HQD's EOP parameters
are modified.

Workaround this HW issue by avoiding writes to these registers for
tonga asics.

Based on the following ROCm commit:
2a0fb8 - drm/amdgpu: Synchronize KFD HQD load protocol with CP scheduler

From the ROCm git repository:
https://github.com/RadeonOpenCompute/ROCK-Kernel-Driver.git

CC: Jay Cornwall <Jay.Cornwall@amd.com>
Suggested-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The MQD structure matches the reg layout. Take advantage of this to
simplify HQD programming.

Note that the ACTIVE field still needs to be programmed last.
Suggested-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAndres Rodriguez <andresx7@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Instead of taking the first pipe and giving the rest to kfd, take the
first 2 queues of each pipe.

Effectively, amdgpu and amdkfd own the same number of queues. But
because the queues are spread over multiple pipes the hardware will be
able to better handle concurrent compute workloads.

amdgpu goes from 1 pipe to 4 pipes, i.e. from 1 compute threads to 4
amdkfd goes from 3 pipe to 4 pipes, i.e. from 3 compute threads to 4

v2: fix policy comment
Reviewed-by: NEdward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: NFelix Kuehling <Felix.Kuehling@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>