[PATCH 2/3] modesetting: add dynamic connector hotplug support (MST) (v2)

Dave Airlie airlied at gmail.com
Mon Apr 13 19:59:35 PDT 2015


On 14 April 2015 at 00:51, Keith Packard <keithp at keithp.com> wrote:
> Dave Airlie <airlied at gmail.com> writes:
>
>> It uses the same option name as nvidia and the other DDXes to
>> disable tearing down outputs as it is hard to avoid racing with
>> clients.
>
> I don't think having an option here is useful -- either we figure out
> some way to make this reliable for clients, or we abandon the notion of
> ever deleting the outputs and fix the protocol spec.
>
> Can someone describe the race condition seen with clients? If it's just
> a bug in the clients, it seems like this is a fine opportunity to
> encourage such bugs to be fixed.

I don't think its really fixable, at least not in all the clients.

The problem is racing between the hot unplug and the client doing something,

So we hot unplug, we remove the output XID from the server
, in parallel the client does an operation with the output XID it has
gotten already,
and still believe is valid, and it gets BadMatch, and since
hardly anyone handles X errors it falls over.

Dave.


More information about the xorg-devel mailing list