__i915_spin_request() sucks

[Jens Axboe]

On 11/12/2015 01:36 PM, Jens Axboe wrote:
> Hi,
>
> So a few months ago I got an XPS13 laptop, the one with the high res
> screen. GUI performance was never really that great, I attributed it to
> coming from a more powerful laptop, and the i915 driving a lot of
> pixels. But yesterday I browsed from my wife's macbook, and was blown
> away. Wow, scrolling in chrome SUCKS on the xps13. Not just scrolling,
> basically anything in chrome. Molasses. So I got sick of it, fired up a
> quick perf record, did a bunch of stuff in chrome. No super smoking
> guns, but one thing did stick out - the path leading to
> __i915_spin_request().
>
> So today, I figured I'd try just killing that spin. If it fails, we'll
> punt to normal completions, so easy change. And wow, MASSIVE difference.
> I can now scroll in chrome and not rage! It's like the laptop is 10x
> faster now.
>
> Ran git blame, and found:
>
> commit 2def4ad99befa25775dd2f714fdd4d92faec6e34
> Author: Chris Wilson <chris at chris-wilson.co.uk>
> Date:   Tue Apr 7 16:20:41 2015 +0100
>
>      drm/i915: Optimistically spin for the request completion
>
> and read the commit message. Doesn't sound that impressive. Especially
> not for something that screws up interactive performance by a LOT.
>
> What's the deal? Revert?

BTW, this:

"Limit the spinning to a single jiffie (~1us) at most"

is totally wrong. I have HZ=100 on my laptop. That's 10ms. 10ms! Even if 
I had HZ=1000, that'd still be 1ms of spinning. That's seriously screwed 
up, guys.

-- 
Jens Axboe