Re: A strange thing with yesterday's head..

On Fri, Jan 29, 2010 at 3:44 PM, Giovanni Trematerra
<giovanni.trematerra_at_gmail.com> wrote:
> On Thu, Jan 28, 2010 at 10:55 PM, Randall Stewart <rrs_at_lakerest.net> wrote:
>> I was running SCHED_ULE on an 8.0 and everything works
>> fine.
>>
>> On my 2 core head of yesterday I tried both SCHED_ULE AND
>> 4BSD.. and got the same results ;-0
>>
>> I will try my 4 core when I get home ;-)
>>
>> R
>> On Jan 28, 2010, at 11:15 AM, Gary Jennejohn wrote:
>>
>>> On Thu, 28 Jan 2010 10:30:37 -0800
>>> Randall Stewart <rrs_at_lakerest.net> wrote:
>>>
>>>> All:
>>>>
>>>> I just found a very strange thing with yesterdays head.
>>>>
>>>> The program
>>>>
>>>> http://www.freebsd.org/~rrs/my_thr.c
>>>>
>>>> I compile it:
>>>>
>>>> cc -g -o my_thr my_thr.c /usr/lib/libthr.a -lpthread
>>>>
>
> Hi Randal,
> I tried your code on an 8-core machine with a fresh head (i386)
> I have no problems with both 4BSD and ULE scheduler.
> I even upping the value of macro NUM_THREAD to 24 but I didn't notice
> nothing strange.
>
> Have you got a chance to reproduce it on your machines?
>
> --
> Gianni
> _______________________________________________
> freebsd-current_at_freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe_at_freebsd.org"
>

I ran it on my dual-core, 8-STABLE/ULE laptop. The very first time I
ran it, I experienced the temporary "seizure". It took several more
runs before I could get it to happen again.

When it did act up again, the subsequent runs resulted in VERY strange
behavior -- the machine froze, the X server displayed "anomalous"
behavior, and my mouse cursor lost it's bearing. After the my_thr
program exited, I had to switch to ttyv0 and back to "reset" the mouse
and get control of X again.

Also -- and I assume this is a side-effect of whatever in the my_thr
code (which I haven't read through and would doubtfully understand) --
the X apps I tried to launch (xpdf, Firefox) would not start-up until
after the my_thr program exited.

-Brandon