jerky mouse still in 7.0-RELEASE

[Dominic Fandrey]

Coleman Kane wrote:
> Dominic Fandrey wrote:
>> Kris Kennaway wrote:
>>> Dominic Fandrey wrote:
>>>> Kris Kennaway wrote:
>>>>> Well it doesn't rule it out.  X may be introducing latencies that
>>>>> are causing your mouse to lose sync or something.
>>>> It's not the mouse that hangs.
>>>> It's the only thing that works,
>>>> everything else hangs when I combine moused/X. This issue doesn't
>>>> exist on all my systems, though. It existed on my old Thinkpad
>>>> (Pentium-m 1.3 GHz), it exists on my new notebook (Core2 Duo with
>>>> 2.4 GHz), but it doesn't exist on my P4 with 1.6 GHz.
>>>>
>>>> Key entries, animations, they all just pile up somewhere and happen
>>>> all at once when I start using the mouse. To watch a movie I have to
>>>> keep the mouse moving all the time.
>>>>
>>>> It's not a general X and mouse problem, because without moused in
>>>> between everything works fine. I think at some point in time a bug
>>>> was either introduced in moused, or in my opinion more likely, in
>>>> the sysmouse protocol implementation in X.
>>> Could also be an interrupt issue.  Either way it's still a different
>>> issue to the ones in this thread.
>> I always thought it's about the same thing and people were just
>> imprecise in their perception. The P4 used to be affected by this,
>> too. This changed somewhere around RC1, I think. Since all of my
>> machines had encountered the problem since I switched them to
>> RELENG_7, I thought my problem was very common and it's the one
>> everyone is talking about.
> Perhaps this is related to the CX / cpu idle level problems we've been
> experiencing on a lot of hardware? What is the output of "sysctl -a |
> grep cx" on your system?

# sysctl -a|grep cx
hw.acpi.cpu.cx_lowest: C1
dev.cpu.0.cx_supported: C1/1 C2/1 C3/17
dev.cpu.0.cx_lowest: C1
dev.cpu.0.cx_usage: 100.00% 0.00% 0.00%
dev.cpu.1.cx_supported: C1/1 C2/1 C3/17
dev.cpu.1.cx_lowest: C1
dev.cpu.1.cx_usage: 100.00% 0.00% 0.00%