[PATCH v6 4/9] drm/amdgpu: create GFX-gen11 usermode queue

Shashank Sharma shashank.sharma at amd.com
Thu Sep 28 13:54:18 UTC 2023


On 28/09/2023 15:52, Alex Deucher wrote:
> On Thu, Sep 28, 2023 at 9:40 AM Shashank Sharma <shashank.sharma at amd.com> wrote:
>>
>> On 28/09/2023 15:27, Alex Deucher wrote:
>>> On Thu, Sep 28, 2023 at 9:22 AM Shashank Sharma <shashank.sharma at amd.com> wrote:
>>>> On 14/09/2023 10:24, Shashank Sharma wrote:
>>>>> On 14/09/2023 09:45, Christian König wrote:
>>>>>> Am 08.09.23 um 18:04 schrieb Shashank Sharma:
>>>>>>> A Memory queue descriptor (MQD) of a userqueue defines it in
>>>>>>> the hw's context. As MQD format can vary between different
>>>>>>> graphics IPs, we need gfx GEN specific handlers to create MQDs.
>>>>>>>
>>>>>>> This patch:
>>>>>>> - Introduces MQD handler functions for the usermode queues.
>>>>>>> - Adds new functions to create and destroy userqueue MQD for
>>>>>>>      GFX-GEN-11 IP
>>>>>>>
>>>>>>> V1: Worked on review comments from Alex:
>>>>>>>        - Make MQD functions GEN and IP specific
>>>>>>>
>>>>>>> V2: Worked on review comments from Alex:
>>>>>>>        - Reuse the existing adev->mqd[ip] for MQD creation
>>>>>>>        - Formatting and arrangement of code
>>>>>>>
>>>>>>> V3:
>>>>>>>        - Integration with doorbell manager
>>>>>>>
>>>>>>> V4: Review comments addressed:
>>>>>>>        - Do not create a new file for userq, reuse gfx_v11_0.c (Alex)
>>>>>>>        - Align name of structure members (Luben)
>>>>>>>        - Don't break up the Cc tag list and the Sob tag list in commit
>>>>>>>          message (Luben)
>>>>>>> V5:
>>>>>>>       - No need to reserve the bo for MQD (Christian).
>>>>>>>       - Some more changes to support IP specific MQD creation.
>>>>>>>
>>>>>>> V6:
>>>>>>>       - Add a comment reminding us to replace the
>>>>>>> amdgpu_bo_create_kernel()
>>>>>>>         calls while creating MQD object to amdgpu_bo_create() once
>>>>>>> eviction
>>>>>>>         fences are ready (Christian).
>>>>>>>
>>>>>>> Cc: Alex Deucher <alexander.deucher at amd.com>
>>>>>>> Cc: Christian Koenig <christian.koenig at amd.com>
>>>>>>> Signed-off-by: Shashank Sharma <shashank.sharma at amd.com>
>>>>>>> Signed-off-by: Arvind Yadav <arvind.yadav at amd.com>
>>>>>>> ---
>>>>>>>     drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c | 16 ++++
>>>>>>>     drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        | 77
>>>>>>> +++++++++++++++++++
>>>>>>>     .../gpu/drm/amd/include/amdgpu_userqueue.h    |  7 ++
>>>>>>>     3 files changed, 100 insertions(+)
>>>>>>>
>>>>>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
>>>>>>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
>>>>>>> index 44769423ba30..03fc8e89eafb 100644
>>>>>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
>>>>>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c
>>>>>>> @@ -140,12 +140,28 @@ int amdgpu_userq_ioctl(struct drm_device *dev,
>>>>>>> void *data,
>>>>>>>         return r;
>>>>>>>     }
>>>>>>>     +extern const struct amdgpu_userq_funcs userq_gfx_v11_funcs;
>>>>>>> +
>>>>>>> +static void
>>>>>>> +amdgpu_userqueue_setup_gfx(struct amdgpu_userq_mgr *uq_mgr)
>>>>>>> +{
>>>>>>> +    int maj;
>>>>>>> +    struct amdgpu_device *adev = uq_mgr->adev;
>>>>>>> +    uint32_t version = adev->ip_versions[GC_HWIP][0];
>>>>>>> +
>>>>>>> +    /* We support usermode queue only for GFX V11 as of now */
>>>>>>> +    maj = IP_VERSION_MAJ(version);
>>>>>>> +    if (maj == 11)
>>>>>>> +        uq_mgr->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_gfx_v11_funcs;
>>>>>>> +}
>>>>>> That belongs into gfx_v11.c and not here.
>>>>> Agree,
>>>> On a second thought, we can't move it to gfx_v11.c, as this is the place
>>>> where we are setting up the gfx_userqueue functions in fpriv->uq_mgr()
>>>> for the first time, and we do not have another option but to check the
>>>> IP and setup the functions here. The only other option to do this will
>>>> be to move  uq_mgr->userq_funcs to adev->gfx.userq_funcs and setup them
>>>> with the IP init (as Alex once suggested). Please let me know your
>>>> thoughts on this.
>>> That seems cleaner to me.  They should be global anyway and could be
>>> set as part of the individual IP init sequences.  Then the presence of
>>> a pointer could be used to determine whether or not a particular IP
>>> type supports user queues.
>>>
>>> Alex
>>>
>> So if I understand this correctly, this is how we are looking to arrange
>> the userqueue IP functions:
>>
>> - Presence of adev->gfx.funcs.userqueue_funcs() will decide if this IP
>> supports userqueue or not.
>>
>> - sw_init function of the IP will setup these fptrs like:
>>
>>     in gfx_v11_0_sw_init :
>>
>>       if (GFX_MAJ == 11)
>>
>>           adev->gfx.funcs.userqueue_funcs = gfx_v11_0_userqueue_funcs
> I was thinking something more like:
>
> adev->userq_funcs[AMD_IP_BLOCK_TYPE_GFX] = gfx_v11_0_userqueue_funcs;
>
> That way there would be one place for all of the all of the fptrs and
> you could use the IP type to query the right one.
>
> And then in the IOCTLs, you could just check if the pointer is valid.  E.g.,
>
> if (!adev->userq_funcs[ip_block_type])
>     return -EINVAL;
>
> etc.
>
> You could store any metadata relevant to each userq in the per fd user
> queue manager and then just pass the state to the global userq
> functions for each IP.

Makes sense, we can do that, hope this works for Christian as well, 
@Christian ?

- Shashank

>
> Alex
>
>>
>> In amdgpu_userqueue_ioctl:
>>
>>       |
>>
>> CASE: create:
>>
>>       amdgpu_userqueue_create()
>>
>>       if (adev->gfx.funcs.userqueue_funcs) {
>>
>>           adev->gfx.funcs.userqueue_funcs.create_mqd();
>>
>>       }
>>
>>
>> CASE: destroy:
>>
>>       amdgpu_userqueue_destroy()
>>
>>       if (adev->gfx.funcs.userqueue_funcs) {
>>
>>           adev->gfx.funcs.userqueue_funcs.destroy_mqd();
>>
>>       }
>>
>> and so on ...
>>
>> Am I getting this right ?
>>
>> - Shashank
>>
>>>> - Shashank
>>>>
>>>>>>> +
>>>>>>>     int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr,
>>>>>>> struct amdgpu_device *adev)
>>>>>>>     {
>>>>>>>         mutex_init(&userq_mgr->userq_mutex);
>>>>>>>         idr_init_base(&userq_mgr->userq_idr, 1);
>>>>>>>         userq_mgr->adev = adev;
>>>>>>>     +    amdgpu_userqueue_setup_gfx(userq_mgr);
>>>>>>>         return 0;
>>>>>>>     }
>>>>>>>     diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
>>>>>>> b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
>>>>>>> index 0451533ddde4..6760abda08df 100644
>>>>>>> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
>>>>>>> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
>>>>>>> @@ -30,6 +30,7 @@
>>>>>>>     #include "amdgpu_psp.h"
>>>>>>>     #include "amdgpu_smu.h"
>>>>>>>     #include "amdgpu_atomfirmware.h"
>>>>>>> +#include "amdgpu_userqueue.h"
>>>>>>>     #include "imu_v11_0.h"
>>>>>>>     #include "soc21.h"
>>>>>>>     #include "nvd.h"
>>>>>>> @@ -6422,3 +6423,79 @@ const struct amdgpu_ip_block_version
>>>>>>> gfx_v11_0_ip_block =
>>>>>>>         .rev = 0,
>>>>>>>         .funcs = &gfx_v11_0_ip_funcs,
>>>>>>>     };
>>>>>>> +
>>>>>>> +static int gfx_v11_0_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
>>>>>>> +                      struct drm_amdgpu_userq_in *args_in,
>>>>>>> +                      struct amdgpu_usermode_queue *queue)
>>>>>>> +{
>>>>>>> +    struct amdgpu_device *adev = uq_mgr->adev;
>>>>>>> +    struct amdgpu_mqd *mqd_gfx_generic =
>>>>>>> &adev->mqds[AMDGPU_HW_IP_GFX];
>>>>>>> +    struct drm_amdgpu_userq_mqd_gfx_v11_0 mqd_user;
>>>>>>> +    struct amdgpu_mqd_prop userq_props;
>>>>>>> +    int r;
>>>>>>> +
>>>>>>> +    /* Incoming MQD parameters from userspace to be saved here */
>>>>>>> +    memset(&mqd_user, 0, sizeof(mqd_user));
>>>>>>> +
>>>>>>> +    /* Structure to initialize MQD for userqueue using generic MQD
>>>>>>> init function */
>>>>>>> +    memset(&userq_props, 0, sizeof(userq_props));
>>>>>>> +
>>>>>>> +    if (args_in->mqd_size != sizeof(struct
>>>>>>> drm_amdgpu_userq_mqd_gfx_v11_0)) {
>>>>>>> +        DRM_ERROR("MQD size mismatch\n");
>>>>>>> +        return -EINVAL;
>>>>>>> +    }
>>>>>>> +
>>>>>>> +    if (copy_from_user(&mqd_user, u64_to_user_ptr(args_in->mqd),
>>>>>>> args_in->mqd_size)) {
>>>>>>> +        DRM_ERROR("Failed to get user MQD\n");
>>>>>>> +        return -EFAULT;
>>>>>>> +    }
>>>>>>> +
>>>>>>> +    /*
>>>>>>> +     * Create BO for actual Userqueue MQD now
>>>>>>> +     * Todo: replace the calls to bo_create_kernel() with
>>>>>>> bo_create() and use
>>>>>>> +     * implicit pinning for the MQD buffers.
>>>>>> Well not implicit pinning, but rather fencing of the BO.
>>>>>>
>>>>> Noted.
>>>>>
>>>>> - Shashank
>>>>>
>>>>>
>>>>>> Regards,
>>>>>> Christian.
>>>>>>
>>>>>>> +     */
>>>>>>> +    r = amdgpu_bo_create_kernel(adev, mqd_gfx_generic->mqd_size,
>>>>>>> PAGE_SIZE,
>>>>>>> +                    AMDGPU_GEM_DOMAIN_GTT,
>>>>>>> +                    &queue->mqd.obj,
>>>>>>> +                    &queue->mqd.gpu_addr,
>>>>>>> +                    &queue->mqd.cpu_ptr);
>>>>>>> +    if (r) {
>>>>>>> +        DRM_ERROR("Failed to allocate BO for userqueue (%d)", r);
>>>>>>> +        return -ENOMEM;
>>>>>>> +    }
>>>>>>> +    memset(queue->mqd.cpu_ptr, 0, mqd_gfx_generic->mqd_size);
>>>>>>> +
>>>>>>> +    /* Initialize the MQD BO with user given values */
>>>>>>> +    userq_props.wptr_gpu_addr = mqd_user.wptr_va;
>>>>>>> +    userq_props.rptr_gpu_addr = mqd_user.rptr_va;
>>>>>>> +    userq_props.queue_size = mqd_user.queue_size;
>>>>>>> +    userq_props.hqd_base_gpu_addr = mqd_user.queue_va;
>>>>>>> +    userq_props.mqd_gpu_addr = queue->mqd.gpu_addr;
>>>>>>> +    userq_props.use_doorbell = true;
>>>>>>> +
>>>>>>> +    r = mqd_gfx_generic->init_mqd(adev, (void *)queue->mqd.cpu_ptr,
>>>>>>> &userq_props);
>>>>>>> +    if (r) {
>>>>>>> +        DRM_ERROR("Failed to initialize MQD for userqueue\n");
>>>>>>> +        goto free_mqd;
>>>>>>> +    }
>>>>>>> +
>>>>>>> +    return 0;
>>>>>>> +
>>>>>>> +free_mqd:
>>>>>>> +    amdgpu_bo_free_kernel(&queue->mqd.obj, &queue->mqd.gpu_addr,
>>>>>>> &queue->mqd.cpu_ptr);
>>>>>>> +    return r;
>>>>>>> +}
>>>>>>> +
>>>>>>> +static void
>>>>>>> +gfx_v11_0_userq_mqd_destroy(struct amdgpu_userq_mgr *uq_mgr, struct
>>>>>>> amdgpu_usermode_queue *queue)
>>>>>>> +{
>>>>>>> +    struct amdgpu_userq_obj *mqd = &queue->mqd;
>>>>>>> +
>>>>>>> +    amdgpu_bo_free_kernel(&mqd->obj, &mqd->gpu_addr, &mqd->cpu_ptr);
>>>>>>> +}
>>>>>>> +
>>>>>>> +const struct amdgpu_userq_funcs userq_gfx_v11_funcs = {
>>>>>>> +    .mqd_create = gfx_v11_0_userq_mqd_create,
>>>>>>> +    .mqd_destroy = gfx_v11_0_userq_mqd_destroy,
>>>>>>> +};
>>>>>>> diff --git a/drivers/gpu/drm/amd/include/amdgpu_userqueue.h
>>>>>>> b/drivers/gpu/drm/amd/include/amdgpu_userqueue.h
>>>>>>> index 55ed6512a565..240f92796f00 100644
>>>>>>> --- a/drivers/gpu/drm/amd/include/amdgpu_userqueue.h
>>>>>>> +++ b/drivers/gpu/drm/amd/include/amdgpu_userqueue.h
>>>>>>> @@ -29,6 +29,12 @@
>>>>>>>       struct amdgpu_mqd_prop;
>>>>>>>     +struct amdgpu_userq_obj {
>>>>>>> +    void         *cpu_ptr;
>>>>>>> +    uint64_t     gpu_addr;
>>>>>>> +    struct amdgpu_bo *obj;
>>>>>>> +};
>>>>>>> +
>>>>>>>     struct amdgpu_usermode_queue {
>>>>>>>         int            queue_type;
>>>>>>>         uint64_t        doorbell_handle;
>>>>>>> @@ -37,6 +43,7 @@ struct amdgpu_usermode_queue {
>>>>>>>         struct amdgpu_mqd_prop    *userq_prop;
>>>>>>>         struct amdgpu_userq_mgr *userq_mgr;
>>>>>>>         struct amdgpu_vm    *vm;
>>>>>>> +    struct amdgpu_userq_obj mqd;
>>>>>>>     };
>>>>>>>       struct amdgpu_userq_funcs {


More information about the amd-gfx mailing list