aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
AgeCommit message (Collapse)AuthorFilesLines
2017-10-19drm/amdgpu:fix duplicated setting job's vram_lostMonk Liu1-1/+1
Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-19drm/amdgpu: set -ECANCELED when dropping jobsChristian König1-0/+1
And return from the wait functions the fence error code. Signed-off-by: Christian König <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-19drm/amdgpu: keep copy of VRAM lost counter in jobChristian König1-6/+7
Instead of reading the current counter from fpriv. Signed-off-by: Christian König <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-10-09drm/amdgpu: add framework for HW specific priority settings v9Andres Rodriguez1-0/+4
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: Christian König <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Andres Rodriguez <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-07-14drm/amdgpu: allow flushing VMID0 before IB execution as wellChristian König1-0/+2
This allows us to queue IBs which needs an up to date system domain as well. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Acked-by: Felix Kuehling <[email protected]>
2017-05-24drm/amdgpu: add dep_sync for amdgpu jobChunming Zhou1-6/+11
The fence in dep_sync cannot be optimized. Signed-off-by: Chunming Zhou <[email protected]> Tested and Reviewed-by: Roger.He <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-24drm/amdgpu: skip all jobs of guilty vmChunming Zhou1-4/+11
If the vm is guilty of a GPU reset, skips all its jobs. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-24drm/amdgpu:use job* to replace voluntaryMonk Liu1-1/+1
that way we can know which job cause hang and can do per sched reset/recovery instead of all sched. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-24drm/amdgpu:don't invoke srio-gpu-reset in gpu-reset (v2)Monk Liu1-1/+5
because we don't want to do sriov-gpu-reset under certain cases, so just split those two funtion and don't invoke sr-iov one from bare-metal one. V2: remove debugfs_gpu_reset routine on SRIOV case. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-24drm/amdgpu: make pipeline sync be in same place v2Chunming Zhou1-1/+0
v2: directly return for 'if' case. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-24drm/amdgpu: add sched sync for amdgpu job v2Chunming Zhou1-4/+9
this is an improvement for previous patch, the sched_sync is to store fence that could be skipped as scheduled, when job is executed, we didn't need pipeline_sync if all fences in sched_sync are signalled, otherwise insert pipeline_sync still. v2: handle error when adding fence to sync failed. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> (v1) Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-05-10drm/amdgpu: fix dependency issueChunming Zhou1-0/+4
The problem is that executing the jobs in the right order doesn't give you the right result because consecutive jobs executed on the same engine are pipelined. In other words job B does it buffer read before job A has written it's result. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-04-28drm/amdgpu: fix no-vmid jobChunming Zhou1-1/+1
[ 132.036658] amdgpu 0000:22:00.0: VM IB without ID [ 132.036709] [drm:amdgpu_job_run [amdgpu]] *ERROR* Error scheduling IBs (-22) [ 132.036755] [drm:amd_sched_main [amdgpu]] *ERROR* Failed to run job! root cause is fence is signaled during sync transfer. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-01-27drm/amd/amdgpu: remove the uncessary parameter for ib schedulerJunwei Zhang1-2/+1
Signed-off-by: Junwei Zhang <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-10-25dma-buf: Rename struct fence to dma_fenceChris Wilson1-11/+11
I plan to usurp the short name of struct fence for a core kernel struct, and so I need to rename the specialised fence/timeline for DMA operations to make room. A consensus was reached in https://lists.freedesktop.org/archives/dri-devel/2016-July/113083.html that making clear this fence applies to DMA operations was a good thing. Since then the patch has grown a bit as usage increases, so hopefully it remains a good thing! (v2...: rebase, rerun spatch) v3: Compile on msm, spotted a manual fixup that I broke. v4: Try again for msm, sorry Daniel coccinelle script: @@ @@ - struct fence + struct dma_fence @@ @@ - struct fence_ops + struct dma_fence_ops @@ @@ - struct fence_cb + struct dma_fence_cb @@ @@ - struct fence_array + struct dma_fence_array @@ @@ - enum fence_flag_bits + enum dma_fence_flag_bits @@ @@ ( - fence_init + dma_fence_init | - fence_release + dma_fence_release | - fence_free + dma_fence_free | - fence_get + dma_fence_get | - fence_get_rcu + dma_fence_get_rcu | - fence_put + dma_fence_put | - fence_signal + dma_fence_signal | - fence_signal_locked + dma_fence_signal_locked | - fence_default_wait + dma_fence_default_wait | - fence_add_callback + dma_fence_add_callback | - fence_remove_callback + dma_fence_remove_callback | - fence_enable_sw_signaling + dma_fence_enable_sw_signaling | - fence_is_signaled_locked + dma_fence_is_signaled_locked | - fence_is_signaled + dma_fence_is_signaled | - fence_is_later + dma_fence_is_later | - fence_later + dma_fence_later | - fence_wait_timeout + dma_fence_wait_timeout | - fence_wait_any_timeout + dma_fence_wait_any_timeout | - fence_wait + dma_fence_wait | - fence_context_alloc + dma_fence_context_alloc | - fence_array_create + dma_fence_array_create | - to_fence_array + to_dma_fence_array | - fence_is_array + dma_fence_is_array | - trace_fence_emit + trace_dma_fence_emit | - FENCE_TRACE + DMA_FENCE_TRACE | - FENCE_WARN + DMA_FENCE_WARN | - FENCE_ERR + DMA_FENCE_ERR ) ( ... ) Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Gustavo Padovan <[email protected]> Acked-by: Sumit Semwal <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-09-14drm/amdgpu: mark symbols static where possibleBaoyou Xie1-1/+1
We get a few warnings when building kernel with W=1: drivers/gpu/drm/amd/amdgpu/cz_smc.c:51:5: warning: no previous prototype for 'cz_send_msg_to_smc_async' [-Wmissing-prototypes] drivers/gpu/drm/amd/amdgpu/cz_smc.c:143:5: warning: no previous prototype for 'cz_write_smc_sram_dword' [-Wmissing-prototypes] drivers/gpu/drm/amd/amdgpu/iceland_smc.c:124:6: warning: no previous prototype for 'iceland_start_smc' [-Wmissing-prototypes] drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c:3926:6: warning: no previous prototype for 'gfx_v8_0_rlc_stop' [-Wmissing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_job.c:94:6: warning: no previous prototype for 'amdgpu_job_free_cb' [-Wmissing-prototypes] .... In fact, these functions are only used in the file in which they are declared and don't need a declaration, but can be made static. So this patch marks these functions with 'static'. Reviewed-by: Christian König <[email protected]> Acked-by: Huang Rui <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Baoyou Xie <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-09-12drm/amdgpu: change job->ctx field nameMonk Liu1-1/+1
job->ctx actually is a fence_context of the entity it belongs to, naming it as ctx is too vague, and we'll need add amdgpu_ctx into the job structure later. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-14drm/amdgpu: cleanup hw reference handling in the IB testsChristian König1-5/+3
Reference should be taken when we make the assignment, not anywhere else. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: add a bool to specify if needing vm flush V2Chunming Zhou1-1/+1
which avoids job->vm_pd_addr be changed. V2: pass job structure to amdgpu_vm_grab_id and amdgpu_vm_flush directly. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: put old hw fence of job if gpu resetChunming Zhou1-0/+2
Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: remove fence parameter from amd_sched_job_initChristian König1-3/+2
We return the fence as part of the job structur anyway, no need to do this twice. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: earlier free SA resourcesChristian König1-2/+2
Keep the time we don't have a fence associated with the resource smaller. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: shorten amdgpu_job_free_resourcesChristian König1-3/+5
The fence and the sync object are not hardware resources. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: fix user fence handling once moreChristian König1-1/+0
Same problem as with the VM page tables. The user fence address must be determined before the job is scheduled, not when the IB is executed. This fixes a security problem where user fences could be used to overwrite any part of VRAM. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: use a fence array for VMID managementChristian König1-1/+1
Just wait for any fence to become available, instead of waiting for the last entry of the LRU. Acked-by: Alex Deucher <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: add optional ring to amdgpu_sync_is_idleChristian König1-1/+1
Check if the sync object is idle depending on the ring a submission works with. Acked-by: Alex Deucher <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: remove amdgpu_sync_waitChristian König1-5/+1
Stop hiding bugs, instead print a proper error when the scheduler doesn't handle all dependencies. Acked-by: Alex Deucher <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: generalize the scheduler fenceChristian König1-2/+2
Make it two events, one for the job being scheduled and one when it is finished. Acked-by: Alex Deucher <[email protected]> Signed-off-by: Christian König <[email protected]> Acked-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: add gpu reset to timeout handlerChunming Zhou1-0/+1
so that we could actually reset the GPU when it hangs. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Junwei Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: fix and cleanup job destructionChristian König1-13/+5
Remove the job reference counting and just properly destroy it from a work item which blocks on any potential running timeout handler. Signed-off-by: Christian König <[email protected]> Reviewed-by: Monk.Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: properly abstract scheduler timeout handlingChristian König1-7/+8
The driver shouldn't mess with the scheduler internals. Signed-off-by: Christian König <[email protected]> Reviewed-by: Monk.Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: remove use_shed hack in job cleanupChristian König1-5/+8
Remembering the code path in a variable to cleanup differently is usually not a good idea at all. Signed-off-by: Christian König <[email protected]> Reviewed-by: Monk.Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: fix coding style in amdgpu_job_freeChristian König1-3/+4
Ther should be a new line between code and decleration. Also use amdgpu_ib_free() instead of releasing the member manually. Signed-off-by: Christian König <[email protected]> Reviewed-by: Monk.Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-07drm/amdgpu: remove begin_job/finish_jobChristian König1-2/+0
Completely pointless and confusing to use a callback to call into the same code file. Signed-off-by: Christian König <[email protected]> Reviewed-by: Monk.Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-11drm/amdgpu: fix and cleanup user fence handling v2Christian König1-1/+1
We leaked the BO in the error pass, additional to that we only have one user fence for all IBs in a job. v2: remove white space changes Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-11drm/amdgpu: move VM fields into jobChristian König1-10/+2
They are the same for all IBs. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-11drm/amdgpu: move the context from the IBs into the jobChristian König1-4/+3
We only have one context for all IBs. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-11drm/amdgpu: keep vm in job instead of ib (v2)Monk Liu1-4/+5
ib.vm is a legacy way to get vm, after scheduler implemented vm should be get from job, and all ibs from one job share the same vm, no need to keep ib.vm just move vm field to job. this patch as well add job as paramter to ib_schedule so it can get vm from job->vm. v2: agd: sqaush in: drm/amdgpu: check if ring emit_vm_flush exists in vm flush No vm flush on engines that don't support VM. bug: https://bugs.freedesktop.org/show_bug.cgi?id=95195 Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-04drm/amd/scheduler: Mark amdgpu_sched_ops constNils Wallménius1-1/+1
This marks the struct amdgpu_sched_ops const and adjusts amd_sched_init to take a const pointer for the ops param. The ops member of struct amd_gpu_scheduler is also changed to const. Reviewed-by: Christian König <[email protected]> Signed-off-by: Nils Wallménius <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-02drm/amdgpu: use ref to keep job aliveMonk Liu1-3/+12
this is to fix fatal page fault error that occured if: job is signaled/released after its timeout work is already put to the global queue (in this case the cancel_delayed_work will return false), which will lead to NX-protection error page fault during job_timeout_func. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-02drm/amdgpu: rework TDR in scheduler (v2)Monk Liu1-1/+15
Add two callbacks to scheduler to maintain jobs, and invoked for job timeout calculations. Now TDR measures time gap from job is processed by hw. v2: fix typo Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-02drm/amdgpu: delay job free to when it's finished (v2)Monk Liu1-1/+10
for those jobs submitted through scheduler, do not free it immediately after scheduled, instead free it in global workqueue by its sched fence signaling callback function. v2: call uf's bo_undef after job_run() call job's sync free after job_run() no static inline __amdgpu_job_free() anymore, just use kfree(job) to replace it. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-05-02drm/amdgpu: use sched_job_init to initialize sched_jobMonk Liu1-6/+9
Consolidate job initialization in one place rather than duplicating it in multiple places. Signed-off-by: Monk Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-03-17drm/amdgpu: use sched fence if possibleMonk Liu1-1/+4
when preemption feature lands, the SA bo should rely on sched fence, because hw fence will be invalid after its job preempted or skipped. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-03-17drm/amdgpu: move ib.fence to job.fenceMonk Liu1-2/+3
Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-03-17drm/amdgpu: give a fence param to ib_freeMonk Liu1-1/+2
thus amdgpu_ib_free() can hook sched fence to SA manager in later patches. BTW: for amdgpu_free_job(), it should only fence_put() the fence of the last ib once, so fix it as well in this patch. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-03-08drm/amdgpu: remove HW fence ownerChristian König1-1/+1
Not used any more since we now always use the sheduler. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Chunming Zhou <[email protected]>
2016-02-29drm/amdgpu: fix VM faults caused by vm_grab_id() v4Christian König1-4/+11
The owner must be per ring as long as we don't support sharing VMIDs per process. Also move the assigned VMID and page directory address into the IB structure. v3: assign the VMID to all IBs, not just the first one. v4: use correct pointer for owner Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-02-12drm/amdgpu: nuke the kernel contextChristian König1-5/+0
Not used any more. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
2016-02-12drm/amdgpu: use per VM entity for page table updates (v2)Christian König1-2/+6
Updates from different VMs can be processed independently. v2: agd: rebase on upstream Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.