Before scheduling a recovery due to scheduler/job hang, check if a RAS
error is detected. If so, choose RAS recovery to handle the situation. A
scheduler/job hang could be the side effect of a RAS error. In such
cases, it is required to go through the RAS error recovery process. A
RAS error recovery process in certains cases also could avoid a full
device device reset.
An error state is maintained in RAS context to detect the block
affected. Fatal Error state uses unused block id. Set the block id when
error is detected. If the interrupt handler detected a poison error,
it's not required to look for a fatal error. Skip fatal error checking
in such cases.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
NPS mode is introduced, the value of memory physical address (PA)
related to a MCA address varies per nps mode. We need to rely on
MCA address and convert it into PA accroding to nps mode.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
One UMC MCA address could map to multiply physical address (PA):
AMDGPU_RAS_EEPROM_REC_PA: one record store one PA
AMDGPU_RAS_EEPROM_REC_MCA: one record store one MCA address, PA
is not cared about
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We can set UMC node instance to invalid state if we use global channel
index, and RAS TA can choose UMC address conversion approach by checking
node_inst value.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
So the code can be simplified, and no need to expose the detail of PA
format outside address conversion.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There's no need to perform the two MMIO writes in the KIQ
Setting registers programmed period, and reducing the MMIO
writes will save the driver loading time.
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Implement sdma soft reset by sending MSG_ResetSDMA on smu 13.0.6.
v2: Add firmware version for the reset message.
v3: Add ip version check. Print inst_mask on failure.
Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add support to create user BOs with MC address for isp using the dma-buf handle
exported for the buffers allocated from system memory in isp driver.
Export amdgpu_bo_create_kernel() and amdgpu_bo_free_kernel() as well for isp to
allocate GTT internal buffers required for fw to run.
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When using MES creating a pdd will require talking to the GPU to
setup the relevant context. The code here forgot to wake up the GPU
in case it was in suspend, this causes KVM to EFAULT for passthrough
GPU for example. This issue can be masked if the GPU was woken up by
other things (e.g. opening the KMS node) first and have not yet gone to sleep.
v4: do the allocation of proc_ctx_bo in a lazy fashion
when the first queue is created in a process (Felix)
Signed-off-by: Jesse Zhang <jesse.zhang@amd.com>
Reviewed-by: Yunxiang Li <Yunxiang.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Emitting the cleaner shader must come after the check if a VM switch is
necessary or not.
Otherwise we will emit the cleaner shader every time and not just when it is
necessary because we switched between applications.
This can otherwise crash on gang submit and probably decreases performance
quite a bit.
v2: squash in fix from Srini (Alex)
Signed-off-by: Christian König <christian.koenig@amd.com>
Fixes: ee7a846ea2 ("drm/amdgpu: Emit cleaner shader at end of IB submission")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
ISP hw_init is not called with the recent changes related
to hw init levels. AMDGPU_INIT_LEVEL_DEFAULT is ignoring
the ISP IP block as AMDGPU_IP_BLK_MASK_ALL is derived using
incorrect max number of IP blocks.
Update AMDGPU_IP_BLK_MASK_ALL to use AMD_IP_BLOCK_TYPE_NUM
instead of AMDGPU_MAX_IP_NUM to fix the issue.
Fixes: 14f2fe34f5 ("drm/amdgpu: Add init levels")
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In the function pqm_uninit there is a call-assignment of "pdd =
kfd_get_process_device_data" which could be null, and this value was
later dereferenced without checking.
Fixes: fb91065851 ("drm/amdkfd: Refactor queue wptr_bo GART mapping")
Signed-off-by: Andrew Martin <Andrew.Martin@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Set the default workload type to bootup type on smu v13.0.7.
This is because of the constraint on smu v13.0.7.
Gfx activity has an even higher set point on 3D fullscreen
mode than the one on bootup mode. This causes the 3D fullscreen
mode's performance is worse than the bootup mode's performance
for the lightweighted/medium workload. For the high workload,
the performance is the same between 3D fullscreen mode and bootup
mode.
v2: set the default workload in ASIC specific file
Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.11.x
Refactor such that individual SMU IP versions can choose the startup
power profile mode. If no preference, then use the generic default power
profile selection logic.
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.11.x
base.sched may not be set for each instance and should not
be used for cases such as non-IB tests.
Fixes: 2320c9e6a7 ("drm/sched: memset() 'job' in drm_sched_job_init()")
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use second jump table in sriov for live migration or mulitple VF
support so different VF can load different version of MEC as long
as they support sjt
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Split resume into a 3rd step to handle displays when DCC is
enabled on DCN 4.0.1. Move display after the buffer funcs
have been re-enabled so that the GPU will do the move and
properly set the DCC metadata for DCN.
v2: fix fence irq resume ordering
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.11.x
smu->workload_mask is IP specific and should not be messed with in
the common code. The mask bits vary across SMU versions.
Move all handling of smu->workload_mask in to the backends and
simplify the code. Store the user's preference in smu->power_profile_mode
which will be reflected in sysfs. For internal driver profile
switches for KFD or VCN, just update the workload mask so that the
user's preference is retained. Remove all of the extra now unused
workload related elements in the smu structure.
v2: use refcounts for workload profiles
v3: rework based on feedback from Lijo
v4: fix the refcount on failure, drop backend mask
v5: rework custom handling
v6: handle failure cleanup with custom profile
v7: Update documentation
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Kenneth Feng <kenneth.feng@amd.com>
Cc: Lijo Lazar <lijo.lazar@amd.com>
Cc: stable@vger.kernel.org # 6.11.x
Use found block to call correct init/resume function on the block.
Set status.hw for resume and init.
Print re-init result again. Change to use dev_info.
Use amdgpu_device_ip_get_ip_block to get target block instead of
loop.
Fixes: 502d76308d ("drm/amdgpu: validate resume before function call")
Signed-off-by: Yiqing Yao <YiQing.Yao@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
ISP hw_init is not called with the recent changes related
to hw init levels. AMDGPU_INIT_LEVEL_DEFAULT is ignoring
the ISP IP block as AMDGPU_IP_BLK_MASK_ALL is derived using
incorrect max number of IP blocks.
Update AMDGPU_IP_BLK_MASK_ALL to use AMDGPU_MAX_IP_NUM
instead of (AMDGPU_MAX_IP_NUM - 1) to fix the issue.
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Fixes: 14f2fe34f5 ("drm/amdgpu: Add init levels")
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
On some cards when odm is used, the monitor will have 2 separate pipes
split vertically. When compression is used on the YCbCr colour space on
the second pipe to have correct colours, we need to read a pixel from the
end of first pipe to accurately display colours. Hardware was programmed
properly to account for this extra pixel but it was not calculated
properly in software causing a split screen on some monitors.
[HOW]
The fix adjusts the second pipe's viewport and timings if the pixel
encoding is YCbCr422 or YCbCr420.
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Reviewed-by: George Shen <george.shen@amd.com>
Signed-off-by: Peterson Guo <peterson.guo@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On VCN 1.0, VCN and JPEG use the same worker thread so cancel
the vcn worker rather than jpeg. On VCN 2.0 and newer
there are separate workers for each.
Fixes: 93df748737 ("drm/amdgpu/jpeg: cancel the jpeg worker")
Tested-by: George Zhang <george.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
