kernel/linux.git/drivers/gpu/drm/scheduler, branch v5.8.10

drm/scheduler: fix drm_sched_get_cleanup_job

2020-04-15T09:09:13+00:00

We are racing to initialize sched->thread here, just always check the current thread. Signed-off-by: Christian König Reviewed-by: Andrey Grodzovsky Reviewed-by: Kent Russell Link: https://patchwork.freedesktop.org/patch/361303/

drm/scheduler: fix rare NULL ptr race

2020-03-25T21:00:11+00:00

There is one one corner case at dma_fence_signal_locked which will raise the NULL pointer problem just like below. ->dma_fence_signal ->dma_fence_signal_locked ->test_and_set_bit here trigger dma_fence_release happen due to the zero of fence refcount. ->dma_fence_put ->dma_fence_release ->drm_sched_fence_release_scheduled ->call_rcu here make the union fled “cb_list” at finished fence to NULL because struct rcu_head contains two pointer which is same as struct list_head cb_list Therefore, to hold the reference of finished fence at drm_sched_process_job to prevent the null pointer during finished fence dma_fence_signal [ 732.912867] BUG: kernel NULL pointer dereference, address: 0000000000000008 [ 732.914815] #PF: supervisor write access in kernel mode [ 732.915731] #PF: error_code(0x0002) - not-present page [ 732.916621] PGD 0 P4D 0 [ 732.917072] Oops: 0002 [#1] SMP PTI [ 732.917682] CPU: 7 PID: 0 Comm: swapper/7 Tainted: G OE 5.4.0-rc7 #1 [ 732.918980] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014 [ 732.920906] RIP: 0010:dma_fence_signal_locked+0x3e/0x100 [ 732.938569] Call Trace: [ 732.939003] [ 732.939364] dma_fence_signal+0x29/0x50 [ 732.940036] drm_sched_fence_finished+0x12/0x20 [gpu_sched] [ 732.940996] drm_sched_process_job+0x34/0xa0 [gpu_sched] [ 732.941910] dma_fence_signal_locked+0x85/0x100 [ 732.942692] dma_fence_signal+0x29/0x50 [ 732.943457] amdgpu_fence_process+0x99/0x120 [amdgpu] [ 732.944393] sdma_v4_0_process_trap_irq+0x81/0xa0 [amdgpu] v2: hold the finished fence at drm_sched_process_job instead of amdgpu_fence_process v3: resume the blank line Signed-off-by: Yintian Tao Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/sched: implement and export drm_sched_pick_best

2020-03-16T20:21:32+00:00

Remove drm_sched_entity_get_free_sched() and use the logic of picking the least loaded drm scheduler from a drm scheduler list to implement drm_sched_pick_best(). This patch also exports drm_sched_pick_best() so that it can be utilized by other drm drivers. Signed-off-by: Nirmoy Das Reviewed-by: Christian König Signed-off-by: Alex Deucher

Revert "drm/scheduler: improve job distribution with multiple queues"

2020-03-16T20:21:32+00:00

It needs to revert this patch to avoid amdgpu_test compute hang problem on picasso. This reverts commit 56822db194232c089601728d68ed078dccb97f8b. Signed-off-by: changzhu Reviewed-by: Feifei Xu Signed-off-by: Alex Deucher

drm/scheduler: fix inconsistent locking of job_list_lock

2020-03-13T15:52:36+00:00

1db8c142b6c5 (drm/scheduler: Add drm_sched_suspend/resume_timeout()) made the job_list_lock IRQ safe in as the suspend/resume calls were expected to be called from IRQ context. This usage never materialized in upstream. Instead amdgpu started locking the job_list_lock in an IRQ unsafe way in amdgpu_ib_preempt_mark_partial_job() and amdgpu_ib_preempt_job_recovery(), which leads to potential deadlock if one would actually start to call the drm_sched_suspend/resume_timeout functions from IRQ context. As no current user needs the locking to be IRQ safe, the local IRQ disable/enable is pure overhead. Fix the inconsistent locking by changing all uses of job_list_lock to use the IRQ unsafe locking primitives. Reviewed-by: Christian König Signed-off-by: Lucas Stach Signed-off-by: Alex Deucher

drm/sched: add run job trace

2020-03-13T15:52:36+00:00

Add a new trace event to show when jobs are run on the HW. Acked-by: Christian König Signed-off-by: Robert Beckett Signed-off-by: Lucas Stach Signed-off-by: Alex Deucher

drm/scheduler: implement a function to modify sched list

2020-03-09T17:51:35+00:00

Implement drm_sched_entity_modify_sched() which modifies existing sched_list with a different one. This is going to be helpful when userspace changes priority of a ctx/entity then the driver can switch to the corresponding HW scheduler list for that priority. Signed-off-by: Nirmoy Das Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: fix doc by clarifying sched_list definition

2020-01-27T21:46:44+00:00

expand sched_list definition for better understanding. Also fix a typo atleast -> at least Signed-off-by: Nirmoy Das Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/scheduler: fix documentation by replacing rq_list with sched_list

2020-01-16T18:41:00+00:00

This also replaces old artifacts with a correct one in drm_sched_entity_init() declaration Signed-off-by: Nirmoy Das Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/scheduler: improve job distribution with multiple queues

2020-01-16T18:37:54+00:00

This patch uses score based logic to select a new rq for better loadbalance between multiple rq/scheds instead of num_jobs. Below are test results after running amdgpu_test from mesa drm Before this patch: sched_name num of many times it got scheduled ========= ================================== sdma0 314 sdma1 32 comp_1.0.0 56 comp_1.0.1 0 comp_1.1.0 0 comp_1.1.1 0 comp_1.2.0 0 comp_1.2.1 0 comp_1.3.0 0 comp_1.3.1 0 After this patch: sched_name num of many times it got scheduled ========= ================================== sdma0 216 sdma1 185 comp_1.0.0 39 comp_1.0.1 9 comp_1.1.0 12 comp_1.1.1 0 comp_1.2.0 12 comp_1.2.1 0 comp_1.3.0 12 comp_1.3.1 0 Signed-off-by: Nirmoy Das Reviewed-by: Christian König Signed-off-by: Alex Deucher