linux-stable.git/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c, branch linux-4.19.y Linux kernel stable tree drm/amdkfd: fix a memory leak issue 2020-10-01T11:14:50+00:00 Dennis Li Dennis.Li@amd.com 2020-09-02T09:11:09+00:00 ce81be26d33f32b5e6edf02abade4259165223c9 [ Upstream commit 087d764159996ae378b08c0fdd557537adfd6899 ] In the resume stage of GPU recovery, start_cpsch will call pm_init which set pm->allocated as false, cause the next pm_release_ib has no chance to release ib memory. Add pm_release_ib in stop_cpsch which will be called in the suspend stage of GPU recovery. Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Dennis Li <Dennis.Li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 087d764159996ae378b08c0fdd557537adfd6899 ]

In the resume stage of GPU recovery, start_cpsch will call pm_init
which set pm->allocated as false, cause the next pm_release_ib has
no chance to release ib memory.

Add pm_release_ib in stop_cpsch which will be called in the suspend
stage of GPU recovery.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdkfd: Fix sdma queue map issue 2019-07-31T05:26:58+00:00 Oak Zeng Oak.Zeng@amd.com 2019-02-08T21:44:35+00:00 210dfe6309112dd028bf06561b828a749b6e1169 [ Upstream commit 065e4bdfa1f3ab2884c110394d8b7e7ebe3b988c ] Previous codes assumes there are two sdma engines. This is not true e.g., Raven only has 1 SDMA engine. Fix the issue by using sdma engine number info in device_info. Signed-off-by: Oak Zeng <Oak.Zeng@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 065e4bdfa1f3ab2884c110394d8b7e7ebe3b988c ]

Previous codes assumes there are two sdma engines.
This is not true e.g., Raven only has 1 SDMA engine.
Fix the issue by using sdma engine number info in
device_info.

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drm/amdkfd: Fix incorrect use of process->mm 2018-10-04T15:37:25+00:00 Felix Kuehling Felix.Kuehling@amd.com 2018-10-02T22:41:12+00:00 11b29c9e25788d0afb2ddb67bcd89424bd25f2f7 This mm_struct pointer should never be dereferenced. If running in a user thread, just use current->mm. If running in a kernel worker use get_task_mm to get a safe reference to the mm_struct. Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
This mm_struct pointer should never be dereferenced. If running in
a user thread, just use current->mm. If running in a kernel worker
use get_task_mm to get a safe reference to the mm_struct.

Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Call kfd2kgd.set_compute_idle 2018-07-16T23:10:37+00:00 Felix Kuehling Felix.Kuehling@amd.com 2018-07-16T23:10:37+00:00 b5aa3f4aef724e9c0f626dcf69948b22efcc5176 User mode queue submissions don't go through KFD. Therefore we don't know exactly when compute is idle or not idle. We use the existence of user mode queues on a device as an approximation. register_process is called when the first queue of a process is created. Conversely unregister_process is called when the last queue is destroyed. The first process that is registered takes compute out of idle. The last process that is unregisters sets compute back to idle. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Eric Huang <JinHuiEric.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
User mode queue submissions don't go through KFD. Therefore we don't
know exactly when compute is idle or not idle. We use the existence
of user mode queues on a device as an approximation.

register_process is called when the first queue of a process is
created. Conversely unregister_process is called when the last queue
is destroyed. The first process that is registered takes compute
out of idle. The last process that is unregisters sets compute back
to idle.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Eric Huang <JinHuiEric.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Make SDMA engine number an ASIC-dependent variable 2018-07-13T20:17:44+00:00 Yong Zhao yong.zhao@amd.com 2018-07-13T20:17:44+00:00 98bb92222eef6ac022f352aa4224f4c94a119199 On Raven there is only one SDMA engine instead of previously assumed two, so we need to adapt our code to this new scenario. Signed-off-by: Yong Zhao <yong.zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
On Raven there is only one SDMA engine instead of previously assumed two,
so we need to adapt our code to this new scenario.

Signed-off-by: Yong Zhao <yong.zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Replace mqd with mqd_mgr as the variable name for mqd_manager 2018-07-12T02:33:07+00:00 Yong Zhao yong.zhao@amd.com 2018-07-12T02:33:07+00:00 8d5f355290880331e265d4c3f3b66c805969f18e This will make reading code much easier. Signed-off-by: Yong Zhao <yong.zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
This will make reading code much easier.

Signed-off-by: Yong Zhao <yong.zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Introduce KFD module parameter halt_if_hws_hang 2018-07-12T02:33:05+00:00 Yong Zhao yong.zhao@amd.com 2018-07-12T02:33:05+00:00 0e9a860c72ec387140a0feb4b8d9a6d0004e9316 This avoids triggering a GPU reset or otherwise changing the HW state. Instead KFD will hang, which allows HW debugging tools to analyze the problem. Signed-off-by: Yong Zhao <yong.zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
This avoids triggering a GPU reset or otherwise changing the HW
state. Instead KFD will hang, which allows HW debugging tools to
analyze the problem.

Signed-off-by: Yong Zhao <yong.zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Add debugfs interface to trigger HWS hang 2018-07-12T02:33:04+00:00 Shaoyun Liu Shaoyun.Liu@amd.com 2018-07-12T02:33:04+00:00 a29ec470b19e58044005973301f233e0b20ed8c4 Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Implement hang detection in KFD and call amdgpu 2018-07-12T02:32:58+00:00 Shaoyun Liu Shaoyun.Liu@amd.com 2018-07-12T02:32:58+00:00 73ea648d921ed5c58895c8592321456fe12b697f The reset will be performed in a new hw_exception work thread to handle HWS hang without blocking the thread that detected the hang. Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
The reset will be performed in a new hw_exception work thread to
handle HWS hang without blocking the thread that detected the hang.

Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
drm/amdkfd: Handle VM faults in KFD 2018-07-12T02:32:50+00:00 shaoyunl Shaoyun.Liu@amd.com 2018-07-12T02:32:50+00:00 2640c3facbd6e21e63c95f19588cc24913a263cd 1. Pre-GFX9 the amdgpu ISR saves the vm-fault status and address per per-vmid. amdkfd needs to get the information from amdgpu through the new get_vm_fault_info interface. On GFX9 and later, all the required information is in the IH ring 2. amdkfd unmaps all queues from the faulting process and create new run-list without the guilty process 3. amdkfd notifies the runtime of the vm fault trap via EVENT_TYPE_MEMORY Signed-off-by: shaoyun liu <shaoyun.liu@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com> 
1. Pre-GFX9 the amdgpu ISR saves the vm-fault status and address per
   per-vmid. amdkfd needs to get the information from amdgpu through the
   new get_vm_fault_info interface. On GFX9 and later, all the required
   information is in the IH ring
2. amdkfd unmaps all queues from the faulting process and create new
   run-list without the guilty process
3. amdkfd notifies the runtime of the vm fault trap via EVENT_TYPE_MEMORY

Signed-off-by: shaoyun liu <shaoyun.liu@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>