[PATCH 2/2] drm/amdgpu: patch the IBs for the second UVD instance v2

Christian König ckoenig.leichtzumerken at gmail.com
Thu Jul 26 06:24:53 UTC 2018


Am 25.07.2018 um 21:43 schrieb James Zhu:
>
> On 2018-07-25 07:05 AM, Christian König wrote:
>> Patch the IBs for the second UVD instance so that userspace don't need
>> to care about the instance they submit to.
>>
>> v2: use direct IB patching
>>
>> Signed-off-by: Christian König <christian.koenig at amd.com>
>> Reviewed-by: Alex Deucher <alexander.deucher at amd.com>
>> ---
>>   drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c | 27 +++++++++++++++++++++++++++
>>   1 file changed, 27 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c 
>> b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
>> index db5f3d78ab12..0895b80fcd37 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
>> @@ -1205,6 +1205,32 @@ static int uvd_v7_0_ring_test_ring(struct 
>> amdgpu_ring *ring)
>>       return r;
>>   }
>>   +/**
>> + * uvd_v7_0_enc_ring_test_ring - test if UVD ENC ring is working
>> + *
>> + * @ring: the engine to test on
>> + *
>> + */
>> +static int uvd_v7_0_ring_patch_cs(struct amdgpu_cs_parser *p, 
>> uint32_t ib_idx)
>> +{
>> +    struct amdgpu_ib *ib = &p->job->ibs[ib_idx];
>> +    unsigned i;
>> +
>> +    /* No patching necessary for the first instance */
>> +    if (!p->ring->me)
>> +        return 0;
>> +
>> +    for (i = 0; i < ib->length_dw; i += 2) {
>> +        uint32_t reg = amdgpu_get_ib_value(p, ib_idx, i);
>> +
>> +        reg -= p->adev->reg_offset[UVD_HWIP][0][1];
>> +        reg += p->adev->reg_offset[UVD_HWIP][1][1];
>> +
>> +        amdgpu_set_ib_value(p, ib_idx, i, reg);
>> +    }
>> +    return 0;
>> +}
>> +
>>   /**
>>    * uvd_v7_0_ring_emit_ib - execute indirect buffer
>>    *
>> @@ -1697,6 +1723,7 @@ static const struct amdgpu_ring_funcs 
>> uvd_v7_0_ring_vm_funcs = {
>>       .get_rptr = uvd_v7_0_ring_get_rptr,
>>       .get_wptr = uvd_v7_0_ring_get_wptr,
>>       .set_wptr = uvd_v7_0_ring_set_wptr,
>> +    .patch_cs = uvd_v7_0_ring_patch_cs,
> It should be .patch_cs_in_place. With this fix,

Yeah, sorry accidentally send out the original version without the 
rebase fixes.

>
> Reviewed-by: James Zhu <James.Zhu at amd.com>

Thanks,
Christian.

>
> James
>
>>       .emit_frame_size =
>>           6 + /* hdp invalidate */
>>           SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx



More information about the amd-gfx mailing list