[PATCH] drm/amdgpu: optimize the padding with hw optimization

[Marek Olšák]

On Thu, Aug 1, 2024, 03:37 Christian König <christian.koenig at amd.com> wrote:

> Am 01.08.24 um 08:53 schrieb Marek Olšák:
>
> On Thu, Aug 1, 2024, 00:28 Khatri, Sunil <sukhatri at amd.com> wrote:
>
>>
>> On 8/1/2024 8:49 AM, Marek Olšák wrote:
>> >> +       /* Header is at index 0, followed by num_nops - 1 NOP packet's
>> */
>> >> +       for (i = 1; i < num_nop; i++)
>> >> +               amdgpu_ring_write(ring, ring->funcs->nop);
>> > This loop should be removed. It's unnecessary CPU overhead and we
>> > should never get more than 0x3fff NOPs (maybe use BUG_ON). Leaving the
>> > whole packet body uninitialized is the fastest option.
>> That was the original intent to just move the WPTR for the no of nops
>> and tried too. Based on Christian inputs we should not let the nops packet
>>
>> as garbage or whatever was there originally as a threat/safety measure.
>
>
> It doesn't help safety. It can only be read by the GPU with kernel-level
> permissions.
>
> Initializing the packet body is useless and adds CPU overhead, especially
> with the 256 NOPs or so that we use for no reason.
>
>
> Not filling the remaining ring buffers with NOPs is a pretty clear NAK
> from my side. Leaving garbage in the ring buffer is not even remotely
> defensive.
>

What are you defending against? You know the ring is kernel-owned memory,
right?

Marek


> What we can do is to optimize filling N DWs into the ring buffer without
> updating the WPTR each time.
>
> Regards,
> Christian.
>
>
> Marek
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20240801/c0542dc4/attachment-0001.htm>