[vfio-users] After update, input-linux errors with Operation not permitted.

[sL1pKn07 SpinFlo]

if remove repeat=on all works without mouse freezes :S

2016-09-12 23:33 GMT+02:00 Jonathan Scruggs <j.scruggs at gmail.com>:
> I don't have many problems with it even after restarting the VM Guest
> (windows). My config settings are in this email thread. I use autorotate
> even.  I play games. I'll try world of tanks tomorrow. My mouse and keyboard
> are passed through as PS/2 and I don't know how to make them USB. I'm
> wondering if I can pass them as usb that I could make them MSI interrupts to
> test performance difference.
>
>
> On 12 Sep 2016 22:11, "sL1pKn07 SpinFlo" <sl1pkn07 at gmail.com> wrote:
>>
>> one question about the mouse and the input-linux.
>>
>>  you have problems with it? for me can't play decent any game (for
>> example Wold Of Tanks, this game use the mouse for rote the turret and
>> shot).. every battle the mouse is bloked without any error. only take
>> drive again if restart the guest. drop&grab between host/guest don't
>> work
>>
>> you have this behavior?
>>
>> Gerd, is a knowed issue?
>>
>> greetings
>>
>> 2016-09-12 22:16 GMT+02:00 Jonathan Scruggs <j.scruggs at gmail.com>:
>> > I pulled that commit to make a patch and applied it to the stable 2.1
>> > release and it works perfectly!
>> >
>> > Thanks. :)
>> >
>> > On 12 September 2016 at 20:03, sL1pKn07 SpinFlo <sl1pkn07 at gmail.com>
>> > wrote:
>> >>
>> >> pushed!
>> >>
>> >>
>> >>
>> >> https://libvirt.org/git/?p=libvirt.git;a=commit;h=0276015325a746f2cd26ef0fddfc682cf641423f
>> >>
>> >> 2016-09-12 20:08 GMT+02:00 Jonathan Scruggs <j.scruggs at gmail.com>:
>> >> > Excellent!
>> >> >
>> >> >
>> >> > On 12 Sep 2016 18:35, "sL1pKn07 SpinFlo" <sl1pkn07 at gmail.com> wrote:
>> >> >>
>> >> >> [19:16:39] <laine> sL1pKn07: I found the problem. Before commit
>> >> >> a9331394, cfg->cgroupControllers was set to 0 before it went through
>> >> >> a
>> >> >> loop to set a bit for each controller listed in qemu.conf. That init
>> >> >> was removed in commit a9331394
>> >> >> [19:17:26] <laine> Normally this would be okay, because most things
>> >> >> are initialized to 0 when the memory is allocated. But in this case,
>> >> >> for some reason it's explicitly set to -1 when the memory is
>> >> >> allocated.
>> >> >> [19:24:58] <sL1pKn07> then is fixeable?
>> >> >> [19:25:31] <laine> Yes. I'm typing the commit log message for the
>> >> >> patch right now.
>> >> >> [19:25:48] <laine> I'll pastebin it for you as soon as I hit save.
>> >> >>
>> >> >> YAY!
>> >> >>
>> >> >> soon more notices
>> >> >>
>> >> >> 2016-09-11 23:40 GMT+02:00 Jonathan Scruggs <j.scruggs at gmail.com>:
>> >> >> > Not on the computer at the moment, but if you go to the archives
>> >> >> > website,
>> >> >> > you can link the email thread to him. Might be useful.
>> >> >> >
>> >> >> >
>> >> >> > On 11 Sep 2016 22:27, "sL1pKn07 SpinFlo" <sl1pkn07 at gmail.com>
>> >> >> > wrote:
>> >> >> >>
>> >> >> >> sorry the typo. is irc.oftc.net
>> >
>> >