[Intel-gfx] [PATCH] drm/i915: Put all non-blocking modesets onto an ordered wq

[Chris Wilson]

Quoting Ville Syrjala (2017-11-13 13:36:22)
> From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> We have plenty of global registers and whatnot programmed without
> any further locking by the modeset code. Currently non-bocking
> modesets are allowed to execute in parallel which could corrupt
> said registers.
> 
> To avoid the problem let's run all non-blocking modesets on an
> ordered workqueue. We still put page flips etc. to system_unbound_wq
> allowing page flips on one pipe to execute in parallel with page flips
> or a modeset on a another pipe (assuming no known state is shared
> between them, at which point they would have been added to the same
> atomic commit and serialized that way).
> 
> Blocking modesets are already serialized with each other by
> connection_mutex, and thus are safe. To serialize them with
> non-blocking modesets we just flush the workqueue before executing
> blocking modesets.

I did something very similar in my patches for tracking concurrent
modesets using plane granularity. I used a fence to impose a barrier for
each level of registers (plane, crtc, global) and tracking the
dependencies/ordering via those barrier fences.

In effect, it gives the same barrier as switching between an ordered-wq,
but I feel using fences should be easier to extend down to finer
granularity. (What once was an easy patch is no more :-(
-Chris