Benchmark of Wayland

Jerome Glisse j.glisse at gmail.com
Wed Nov 17 19:47:14 PST 2010


On Wed, Nov 17, 2010 at 10:25 PM, Russell Shaw <rjshaw at netspace.net.au> wrote:
> On 18/11/10 13:56, Justin Lee wrote:
>>
>> 2010/11/18 microcai<microcai at fedoraproject.org>:
>>>
>>>
>>> 2010/11/18 Justin Lee<justinlee5455 at gmail.com>
>>>>
>>>> This kind of latency due to OS scheduling could be eliminated by
>>>> direct-procedure call. That is, the client passes the GEM buffer to
>>>> the server just like it passes the GEM buffer as an argument to a
>>>> function. In this approach, the compositor is not a process. Rather,
>>>> the compositor is just a function live in the client program. So there
>>>> will be no context switches.
>>>
>>> So ? Move wayland entirely into the kernel ? Then there will be no delay.
>>
>> No, I don't think so. The necessary infrastructure have been there in
>> the kernel (i.e. DRM, GEM, KMS). I don't think we should bother
>> writing kernel code which is error-prone and need specific care to
>> avoid bugs and instability. Besides, I don't think moving Wayland into
>> the kernel will be any beneficial if we could do all the same thing in
>> user space.
>>
>> I pointed out the latency because I don't want to see any lag when I
>> scroll down a web page in my browser. I have noticed that in X when I
>> dragged down the scroll bar in browser, the scroll bar wasn't tied to
>> the mouse pointer and the scroll bar fell behind the mouse pointer.
>> When the scroll bar was dragged down, the web page wasn't updated
>> immediately. Instead, the page got updated until a small period of
>> time was passed after the drag happened.
>
> Laggy user interfaces are usually due to crappy inefficient
> drawing libraries doing software rendering. A lot of the blame
> can be on GPU vendors for not releasing detailed programming
> specs to enable decent drivers.
>
> If the widget toolkit isn't designed well, round-trip X11
> requests between server and client can have a large impact.
>
> Also, the kernel itself can be a bottleneck.
>

Maybe we should stop saying so, there is documentation from AMD and
Intel for their GPU, yet the community is unable to produce fast GPU
driver. Maybe it's just that we were overestimating the resource of
the community.

Cheers,
Jerome Glisse


More information about the wayland-devel mailing list