<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<p style="font-family:Arial;font-size:11pt;color:#0078D7;margin:5pt;" align="Left">
[AMD Official Use Only - Internal Distribution Only]<br>
</p>
<br>
<div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Good question. I think push it to drm-misc-next for upstream. We can carry it internally in amd-staging-drm-next for internal testing and I can coordinate with drm-next. I think the amdgpu changes are pretty straightforward, so shouldn't be too hard keep
the merge straight. I dunno. <br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Alex</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div id="appendonsend"></div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>From:</b> Christian König <ckoenig.leichtzumerken@gmail.com><br>
<b>Sent:</b> Friday, February 5, 2021 4:58 AM<br>
<b>To:</b> Liu, Leo <Leo.Liu@amd.com>; amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>; dri-devel@lists.freedesktop.org <dri-devel@lists.freedesktop.org>; Deucher, Alexander <Alexander.Deucher@amd.com><br>
<b>Subject:</b> Re: [PATCH 3/3] drm/amdgpu: share scheduler score on VCN3 instances</font>
<div> </div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
<div class="PlainText">Alex how do we want to merge this?<br>
<br>
I've just pushed the first patch to drm-misc-next since that needed a <br>
rebase because it touches other drivers as well.<br>
<br>
But the rest is really AMD specific and I'm not sure if the dependent <br>
stuff is already in there as well.<br>
<br>
So if I push it to drm-misc-next you will probably need to merge and if <br>
I push it to amd-staging-drm-next somebody else might need to merge when <br>
drm-misc-next is merged.<br>
<br>
Ideas?<br>
<br>
Christian.<br>
<br>
Am 04.02.21 um 19:34 schrieb Leo Liu:<br>
> The series are:<br>
><br>
> Reviewed-and-Tested-by: Leo Liu <leo.liu@amd.com><br>
><br>
><br>
> On 2021-02-04 9:44 a.m., Christian König wrote:<br>
>> The VCN3 instances can do both decode as well as encode.<br>
>><br>
>> Share the scheduler load balancing score and remove fixing encode to<br>
>> only the second instance.<br>
>><br>
>> Signed-off-by: Christian König <christian.koenig@amd.com><br>
>> ---<br>
>> drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 1 +<br>
>> drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 11 +++++++----<br>
>> 2 files changed, 8 insertions(+), 4 deletions(-)<br>
>><br>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h <br>
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h<br>
>> index 13aa417f6be7..d10bc4f0a05f 100644<br>
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h<br>
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h<br>
>> @@ -211,6 +211,7 @@ struct amdgpu_vcn_inst {<br>
>> void *saved_bo;<br>
>> struct amdgpu_ring ring_dec;<br>
>> struct amdgpu_ring ring_enc[AMDGPU_VCN_MAX_ENC_RINGS];<br>
>> + atomic_t sched_score;<br>
>> struct amdgpu_irq_src irq;<br>
>> struct amdgpu_vcn_reg external;<br>
>> struct amdgpu_bo *dpg_sram_bo;<br>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c <br>
>> b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c<br>
>> index 239a4eb52c61..b33f513fd2ac 100644<br>
>> --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c<br>
>> +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c<br>
>> @@ -171,6 +171,7 @@ static int vcn_v3_0_sw_init(void *handle)<br>
>> for (i = 0; i < adev->vcn.num_vcn_inst; i++) {<br>
>> volatile struct amdgpu_fw_shared *fw_shared;<br>
>> +<br>
>> if (adev->vcn.harvest_config & (1 << i))<br>
>> continue;<br>
>> @@ -198,6 +199,8 @@ static int vcn_v3_0_sw_init(void *handle)<br>
>> if (r)<br>
>> return r;<br>
>> + atomic_set(&adev->vcn.inst[i].sched_score, 0);<br>
>> +<br>
>> ring = &adev->vcn.inst[i].ring_dec;<br>
>> ring->use_doorbell = true;<br>
>> if (amdgpu_sriov_vf(adev)) {<br>
>> @@ -209,7 +212,8 @@ static int vcn_v3_0_sw_init(void *handle)<br>
>> ring->no_scheduler = true;<br>
>> sprintf(ring->name, "vcn_dec_%d", i);<br>
>> r = amdgpu_ring_init(adev, ring, 512, <br>
>> &adev->vcn.inst[i].irq, 0,<br>
>> - AMDGPU_RING_PRIO_DEFAULT, NULL);<br>
>> + AMDGPU_RING_PRIO_DEFAULT,<br>
>> + &adev->vcn.inst[i].sched_score);<br>
>> if (r)<br>
>> return r;<br>
>> @@ -227,11 +231,10 @@ static int vcn_v3_0_sw_init(void *handle)<br>
>> } else {<br>
>> ring->doorbell_index = <br>
>> (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + j + 8 * i;<br>
>> }<br>
>> - if (adev->asic_type == CHIP_SIENNA_CICHLID && i != 1)<br>
>> - ring->no_scheduler = true;<br>
>> sprintf(ring->name, "vcn_enc_%d.%d", i, j);<br>
>> r = amdgpu_ring_init(adev, ring, 512, <br>
>> &adev->vcn.inst[i].irq, 0,<br>
>> - AMDGPU_RING_PRIO_DEFAULT, NULL);<br>
>> + AMDGPU_RING_PRIO_DEFAULT,<br>
>> + &adev->vcn.inst[i].sched_score);<br>
>> if (r)<br>
>> return r;<br>
>> }<br>
<br>
</div>
</span></font></div>
</div>
</body>
</html>