[PATCH v2 2/5] drm/virtio: Add a helper to map and note the dma addrs and lengths

[Dmitry Osipenko]

On 11/20/24 06:44, Kasireddy, Vivek wrote:
> Hi Dmitry,
> 
>> Subject: Re: [PATCH v2 2/5] drm/virtio: Add a helper to map and note the
>> dma addrs and lengths
>>
>> ...
>>> After rebasing v2 of this patch series on top of the above patch, I see that
>>> this use-case works as expected with Qemu master. Let me send out v3,
>>> which would be a rebase of v2 on top of the above patch.
>> ...
>>>>> Am I doing anything wrong? Suggestions are welcome.
>>>> Could you please share your Qemu launch parameters? I'll try to recreate
>> the
>>>> issue you are seeing.
>>
>> Thanks a lot for sharing your Qemu command. I haven't used the
>> `host-phys-bits` flags, will try again soon and then also check host
>> errors if it still won't work.
> Any chance you would be able to finish reviewing this patch series before
> your upcoming vacation? 

Let's try, we have a week.

Good news, I got it to work with latest misc-next+v3 on guest and by
using passthroughed AMD RDNA3 card instead of DG2. Haven't yet checked
whether DG2 works with this guest kernel combination, maybe it will.

A bit bad news is that display flickers on each framebuffer flip (shows
a black screen) using dynamic dmabuf attachment. Pinning imported
dma-buf permanently fixes the flicker, i.e. dynamic dmabuf attachment
doesn't work as expected. Disabling CONFIG_DMABUF_MOVE_NOTIFY helps.

There are these virtio-gpu errors on each flicker:

[drm:virtio_gpu_dequeue_ctrl_func] *ERROR* response 0x1200 (command 0x106)

Overall, patches are good to me. I'll leave couple comments on the v3.
The flickering issue isn't a blocker given that MOVE_NOTIFY is
considered to be experimental today, we can fix it later.

-- 
Best regards,
Dmitry