[PATCH v3 2/6] user32: Implement GetRegisteredRawInputDevices.

[Zhiyi Zhang]

On 8/7/20 10:54 PM, Francois Gouget wrote:
> On Fri, 7 Aug 2020, Zhiyi Zhang wrote:
> [...]
>>> I grabbed the package list of debiant before and after the june upgrade 
>>> for comparison. But the closest I get to relevant package upgrades 
>>> are a bunch of OpenGL (libegl, libgl1, libgles2) and Mesa driver 
>>> upgrades. Maybe I missed something.
>>>
>>> 2020-05-04 -> https://termbin.com/br19
>>> 2020-06-15 -> https://termbin.com/8cb9
>> It looks like a libX11 bug. Comparing the package list from 2020-05-04 to 2020-06-15,
>> note that libx11-6 was upgraded from 1.6.9-2 to 1.6.9-2+b1. There is XReply bug in both the lastest and 1.6.9
>> version of libX11 and I think it's being triggered somehow on TestBots. See https://gitlab.freedesktop.org/xorg/lib/libx11/-/issues/93
>> There is a patch in https://gitlab.freedesktop.org/xorg/lib/libx11/-/merge_requests/29 but it's not merged yet.
>>
>> Maybe we should downgrade the package for the moment and wait for the 
>> fix to land. Or install a custom build with the fix.
> The most workable options appear to be to either downgrade to 1.6.7-1 
> (from Debian 10) or upgrade to 1.6.10-3 (current in Debian Testing). 
> From what you said the latter would not have the fix anyway but I don't 
> know if the former has the bug already.
Both 1.6.7-1 and 1.6.10-3 have the bug already. But 1.6.7-1 on Debian 10 TestBots didn't have the monitor
timeouts before so I think it was using a different code path. I think we should downgrade to 1.6.7-1 first
and use a custom build with the fix if we have to, assuming libx11 is the real culprit of course.

>
> WineTest has certainly been wrecking havoc with my desktop's X session 
> lately [1] and it has 1.6.7-1.
>
>
> [1] I'd typically find it at some very low resolution in the morning. 
>     Sometimes xrandr -s 0 would not even work, sometimes xrandr --auto 
>     would work instead, and sometimes nothing short of a reboot would 
>     revive the monitor.
>