[PATCH v3 07/12] drm/sched: Prevent any job recoveries after device is unplugged.

Luben Tuikov luben.tuikov at amd.com
Tue Nov 24 17:11:16 UTC 2020


On 2020-11-24 2:50 a.m., Christian König wrote:
> Am 24.11.20 um 02:12 schrieb Luben Tuikov:
>> On 2020-11-23 3:06 a.m., Christian König wrote:
>>> Am 23.11.20 um 06:37 schrieb Andrey Grodzovsky:
>>>> On 11/22/20 6:57 AM, Christian König wrote:
>>>>> Am 21.11.20 um 06:21 schrieb Andrey Grodzovsky:
>>>>>> No point to try recovery if device is gone, it's meaningless.
>>>>> I think that this should go into the device specific recovery
>>>>> function and not in the scheduler.
>>>>
>>>> The timeout timer is rearmed here, so this prevents any new recovery
>>>> work to restart from here
>>>> after drm_dev_unplug was executed from amdgpu_pci_remove.It will not
>>>> cover other places like
>>>> job cleanup or starting new job but those should stop once the
>>>> scheduler thread is stopped later.
>>> Yeah, but this is rather unclean. We should probably return an error
>>> code instead if the timer should be rearmed or not.
>> Christian, this is exactly my work I told you about
>> last week on Wednesday in our weekly meeting. And
>> which I wrote to you in an email last year about this
>> time.
> 
> Yeah, that's why I'm suggesting it here as well.

It seems you're suggesting that Andrey do it, while
all too well you know I've been working on this
for some time now.

I wrote you about this last year same time
in an email. And I discussed it on the Wednesday
meeting.

You could've mentioned that here the first time.

> 
>> So what do we do now?
> 
> Split your patches into smaller parts and submit them chunk by chunk.
> 
> E.g. renames first and then functional changes grouped by area they change.

I have, but my final patch, a tiny one but which implements
the core reason for the change seems buggy, and I'm looking
for a way to debug it.

Regards,
Luben


> 
> Regards,
> Christian.
> 
>>
>> I can submit those changes without the last part,
>> which builds on this change.
>>
>> I'm still testing the last part and was hoping
>> to submit it all in one sequence of patches,
>> after my testing.
>>
>> Regards,
>> Luben
>>
>>> Christian.
>>>
>>>> Andrey
>>>>
>>>>
>>>>> Christian.
>>>>>
>>>>>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky at amd.com>
>>>>>> ---
>>>>>>    drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c |  2 +-
>>>>>>    drivers/gpu/drm/etnaviv/etnaviv_sched.c   |  3 ++-
>>>>>>    drivers/gpu/drm/lima/lima_sched.c         |  3 ++-
>>>>>>    drivers/gpu/drm/panfrost/panfrost_job.c   |  2 +-
>>>>>>    drivers/gpu/drm/scheduler/sched_main.c    | 15 ++++++++++++++-
>>>>>>    drivers/gpu/drm/v3d/v3d_sched.c           | 15 ++++++++++-----
>>>>>>    include/drm/gpu_scheduler.h               |  6 +++++-
>>>>>>    7 files changed, 35 insertions(+), 11 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
>>>>>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
>>>>>> index d56f402..d0b0021 100644
>>>>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
>>>>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
>>>>>> @@ -487,7 +487,7 @@ int amdgpu_fence_driver_init_ring(struct
>>>>>> amdgpu_ring *ring,
>>>>>>              r = drm_sched_init(&ring->sched, &amdgpu_sched_ops,
>>>>>>                       num_hw_submission, amdgpu_job_hang_limit,
>>>>>> -                   timeout, ring->name);
>>>>>> +                   timeout, ring->name, &adev->ddev);
>>>>>>            if (r) {
>>>>>>                DRM_ERROR("Failed to create scheduler on ring %s.\n",
>>>>>>                      ring->name);
>>>>>> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c
>>>>>> b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
>>>>>> index cd46c88..7678287 100644
>>>>>> --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c
>>>>>> +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
>>>>>> @@ -185,7 +185,8 @@ int etnaviv_sched_init(struct etnaviv_gpu *gpu)
>>>>>>          ret = drm_sched_init(&gpu->sched, &etnaviv_sched_ops,
>>>>>>                     etnaviv_hw_jobs_limit, etnaviv_job_hang_limit,
>>>>>> -                 msecs_to_jiffies(500), dev_name(gpu->dev));
>>>>>> +                 msecs_to_jiffies(500), dev_name(gpu->dev),
>>>>>> +                 gpu->drm);
>>>>>>        if (ret)
>>>>>>            return ret;
>>>>>>    diff --git a/drivers/gpu/drm/lima/lima_sched.c
>>>>>> b/drivers/gpu/drm/lima/lima_sched.c
>>>>>> index dc6df9e..8a7e5d7ca 100644
>>>>>> --- a/drivers/gpu/drm/lima/lima_sched.c
>>>>>> +++ b/drivers/gpu/drm/lima/lima_sched.c
>>>>>> @@ -505,7 +505,8 @@ int lima_sched_pipe_init(struct lima_sched_pipe
>>>>>> *pipe, const char *name)
>>>>>>          return drm_sched_init(&pipe->base, &lima_sched_ops, 1,
>>>>>>                      lima_job_hang_limit, msecs_to_jiffies(timeout),
>>>>>> -                  name);
>>>>>> +                  name,
>>>>>> +                  pipe->ldev->ddev);
>>>>>>    }
>>>>>>      void lima_sched_pipe_fini(struct lima_sched_pipe *pipe)
>>>>>> diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c
>>>>>> b/drivers/gpu/drm/panfrost/panfrost_job.c
>>>>>> index 30e7b71..37b03b01 100644
>>>>>> --- a/drivers/gpu/drm/panfrost/panfrost_job.c
>>>>>> +++ b/drivers/gpu/drm/panfrost/panfrost_job.c
>>>>>> @@ -520,7 +520,7 @@ int panfrost_job_init(struct panfrost_device
>>>>>> *pfdev)
>>>>>>            ret = drm_sched_init(&js->queue[j].sched,
>>>>>>                         &panfrost_sched_ops,
>>>>>>                         1, 0, msecs_to_jiffies(500),
>>>>>> -                     "pan_js");
>>>>>> +                     "pan_js", pfdev->ddev);
>>>>>>            if (ret) {
>>>>>>                dev_err(pfdev->dev, "Failed to create scheduler: %d.",
>>>>>> ret);
>>>>>>                goto err_sched;
>>>>>> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
>>>>>> b/drivers/gpu/drm/scheduler/sched_main.c
>>>>>> index c3f0bd0..95db8c6 100644
>>>>>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>>>>>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>>>>>> @@ -53,6 +53,7 @@
>>>>>>    #include <drm/drm_print.h>
>>>>>>    #include <drm/gpu_scheduler.h>
>>>>>>    #include <drm/spsc_queue.h>
>>>>>> +#include <drm/drm_drv.h>
>>>>>>      #define CREATE_TRACE_POINTS
>>>>>>    #include "gpu_scheduler_trace.h"
>>>>>> @@ -283,8 +284,16 @@ static void drm_sched_job_timedout(struct
>>>>>> work_struct *work)
>>>>>>        struct drm_gpu_scheduler *sched;
>>>>>>        struct drm_sched_job *job;
>>>>>>    +    int idx;
>>>>>> +
>>>>>>        sched = container_of(work, struct drm_gpu_scheduler,
>>>>>> work_tdr.work);
>>>>>>    +    if (!drm_dev_enter(sched->ddev, &idx)) {
>>>>>> +        DRM_INFO("%s - device unplugged skipping recovery on
>>>>>> scheduler:%s",
>>>>>> +             __func__, sched->name);
>>>>>> +        return;
>>>>>> +    }
>>>>>> +
>>>>>>        /* Protects against concurrent deletion in
>>>>>> drm_sched_get_cleanup_job */
>>>>>>        spin_lock(&sched->job_list_lock);
>>>>>>        job = list_first_entry_or_null(&sched->ring_mirror_list,
>>>>>> @@ -316,6 +325,8 @@ static void drm_sched_job_timedout(struct
>>>>>> work_struct *work)
>>>>>>        spin_lock(&sched->job_list_lock);
>>>>>>        drm_sched_start_timeout(sched);
>>>>>>        spin_unlock(&sched->job_list_lock);
>>>>>> +
>>>>>> +    drm_dev_exit(idx);
>>>>>>    }
>>>>>>       /**
>>>>>> @@ -845,7 +856,8 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
>>>>>>               unsigned hw_submission,
>>>>>>               unsigned hang_limit,
>>>>>>               long timeout,
>>>>>> -           const char *name)
>>>>>> +           const char *name,
>>>>>> +           struct drm_device *ddev)
>>>>>>    {
>>>>>>        int i, ret;
>>>>>>        sched->ops = ops;
>>>>>> @@ -853,6 +865,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
>>>>>>        sched->name = name;
>>>>>>        sched->timeout = timeout;
>>>>>>        sched->hang_limit = hang_limit;
>>>>>> +    sched->ddev = ddev;
>>>>>>        for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_COUNT;
>>>>>> i++)
>>>>>>            drm_sched_rq_init(sched, &sched->sched_rq[i]);
>>>>>>    diff --git a/drivers/gpu/drm/v3d/v3d_sched.c
>>>>>> b/drivers/gpu/drm/v3d/v3d_sched.c
>>>>>> index 0747614..f5076e5 100644
>>>>>> --- a/drivers/gpu/drm/v3d/v3d_sched.c
>>>>>> +++ b/drivers/gpu/drm/v3d/v3d_sched.c
>>>>>> @@ -401,7 +401,8 @@ v3d_sched_init(struct v3d_dev *v3d)
>>>>>>                     &v3d_bin_sched_ops,
>>>>>>                     hw_jobs_limit, job_hang_limit,
>>>>>>                     msecs_to_jiffies(hang_limit_ms),
>>>>>> -                 "v3d_bin");
>>>>>> +                 "v3d_bin",
>>>>>> +                 &v3d->drm);
>>>>>>        if (ret) {
>>>>>>            dev_err(v3d->drm.dev, "Failed to create bin scheduler:
>>>>>> %d.", ret);
>>>>>>            return ret;
>>>>>> @@ -411,7 +412,8 @@ v3d_sched_init(struct v3d_dev *v3d)
>>>>>>                     &v3d_render_sched_ops,
>>>>>>                     hw_jobs_limit, job_hang_limit,
>>>>>>                     msecs_to_jiffies(hang_limit_ms),
>>>>>> -                 "v3d_render");
>>>>>> +                 "v3d_render",
>>>>>> +                 &v3d->drm);
>>>>>>        if (ret) {
>>>>>>            dev_err(v3d->drm.dev, "Failed to create render scheduler:
>>>>>> %d.",
>>>>>>                ret);
>>>>>> @@ -423,7 +425,8 @@ v3d_sched_init(struct v3d_dev *v3d)
>>>>>>                     &v3d_tfu_sched_ops,
>>>>>>                     hw_jobs_limit, job_hang_limit,
>>>>>>                     msecs_to_jiffies(hang_limit_ms),
>>>>>> -                 "v3d_tfu");
>>>>>> +                 "v3d_tfu",
>>>>>> +                 &v3d->drm);
>>>>>>        if (ret) {
>>>>>>            dev_err(v3d->drm.dev, "Failed to create TFU scheduler: %d.",
>>>>>>                ret);
>>>>>> @@ -436,7 +439,8 @@ v3d_sched_init(struct v3d_dev *v3d)
>>>>>>                         &v3d_csd_sched_ops,
>>>>>>                         hw_jobs_limit, job_hang_limit,
>>>>>>                         msecs_to_jiffies(hang_limit_ms),
>>>>>> -                     "v3d_csd");
>>>>>> +                     "v3d_csd",
>>>>>> +                     &v3d->drm);
>>>>>>            if (ret) {
>>>>>>                dev_err(v3d->drm.dev, "Failed to create CSD scheduler:
>>>>>> %d.",
>>>>>>                    ret);
>>>>>> @@ -448,7 +452,8 @@ v3d_sched_init(struct v3d_dev *v3d)
>>>>>>                         &v3d_cache_clean_sched_ops,
>>>>>>                         hw_jobs_limit, job_hang_limit,
>>>>>>                         msecs_to_jiffies(hang_limit_ms),
>>>>>> -                     "v3d_cache_clean");
>>>>>> +                     "v3d_cache_clean",
>>>>>> +                     &v3d->drm);
>>>>>>            if (ret) {
>>>>>>                dev_err(v3d->drm.dev, "Failed to create CACHE_CLEAN
>>>>>> scheduler: %d.",
>>>>>>                    ret);
>>>>>> diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
>>>>>> index 9243655..a980709 100644
>>>>>> --- a/include/drm/gpu_scheduler.h
>>>>>> +++ b/include/drm/gpu_scheduler.h
>>>>>> @@ -32,6 +32,7 @@
>>>>>>      struct drm_gpu_scheduler;
>>>>>>    struct drm_sched_rq;
>>>>>> +struct drm_device;
>>>>>>      /* These are often used as an (initial) index
>>>>>>     * to an array, and as such should start at 0.
>>>>>> @@ -267,6 +268,7 @@ struct drm_sched_backend_ops {
>>>>>>     * @score: score to help loadbalancer pick a idle sched
>>>>>>     * @ready: marks if the underlying HW is ready to work
>>>>>>     * @free_guilty: A hit to time out handler to free the guilty job.
>>>>>> + * @ddev: Pointer to drm device of this scheduler.
>>>>>>     *
>>>>>>     * One scheduler is implemented for each hardware ring.
>>>>>>     */
>>>>>> @@ -288,12 +290,14 @@ struct drm_gpu_scheduler {
>>>>>>        atomic_t                        score;
>>>>>>        bool                ready;
>>>>>>        bool                free_guilty;
>>>>>> +    struct drm_device        *ddev;
>>>>>>    };
>>>>>>      int drm_sched_init(struct drm_gpu_scheduler *sched,
>>>>>>               const struct drm_sched_backend_ops *ops,
>>>>>>               uint32_t hw_submission, unsigned hang_limit, long timeout,
>>>>>> -           const char *name);
>>>>>> +           const char *name,
>>>>>> +           struct drm_device *ddev);
>>>>>>      void drm_sched_fini(struct drm_gpu_scheduler *sched);
>>>>>>    int drm_sched_job_init(struct drm_sched_job *job,
>>>> _______________________________________________
>>>> amd-gfx mailing list
>>>> amd-gfx at lists.freedesktop.org
>>>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.freedesktop.org%2Fmailman%2Flistinfo%2Famd-gfx&data=04%7C01%7Cluben.tuikov%40amd.com%7C644a4f3feb79447fd6a408d8904dab27%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637418010548375418%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=wNLdozuhVS3smIpAuWB0tjFO3XDo1OmmZEgTCxviJaI%3D&reserved=0
>>> _______________________________________________
>>> dri-devel mailing list
>>> dri-devel at lists.freedesktop.org
>>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.freedesktop.org%2Fmailman%2Flistinfo%2Fdri-devel&data=04%7C01%7Cluben.tuikov%40amd.com%7C644a4f3feb79447fd6a408d8904dab27%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637418010548385367%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=qXKgWmi%2FU042boaDF43w5uIKRLFVNgwiPYrEN%2FxV0pc%3D&reserved=0
>>>
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx at lists.freedesktop.org
>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.freedesktop.org%2Fmailman%2Flistinfo%2Famd-gfx&data=04%7C01%7Cluben.tuikov%40amd.com%7C644a4f3feb79447fd6a408d8904dab27%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637418010548385367%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=OZGMVRwFXiuhoG3%2FTP54e6vk0xpMQujqAlNxtCcX7kA%3D&reserved=0
> 
> _______________________________________________
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.freedesktop.org%2Fmailman%2Flistinfo%2Fdri-devel&data=04%7C01%7Cluben.tuikov%40amd.com%7C644a4f3feb79447fd6a408d8904dab27%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637418010548385367%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=qXKgWmi%2FU042boaDF43w5uIKRLFVNgwiPYrEN%2FxV0pc%3D&reserved=0
> 



More information about the amd-gfx mailing list