Re: i915 observations

[Tobias Nygren <tnn%NetBSD.org@localhost>]

> I also wouldn?t call the keyboard input delayed. I would say
> that visual effects of keyboard input usually were delayed due
> to the fact that they tend to require just a small part of the
> screen updated. This is most visible when logging in with xdm.
>
> Once logged in, I have a clock displayed, which enforces an update
> every second, which also updates some of the remaining dashes.
> Eventu?lly, the image tends to converge to the intended one (except
> maybe for the clock itself, which tends to lag behind all the time).

I believe the effect you see is framebuffer contents written to
the CPU cache but not flushed out to memory. This is why the artefacts
lazily repair themselves when background jobs that cause frequent cache
line misses are run. (Note how the artefacts are always 1 pixel high,
64-bytes wide and seemingly randomly positioned.)