On Tue, 20 Feb 2001, Philip Andrew Ferguson wrote:
> Hello All.
>
> I have a funny thing going on with pthreads in my Arm Linux application.
>
> First, a little background. I'm using the "nanoEngine" from Brightstar
> Engineering. Thus, I don't really know what kernel I'm running, but I can
> find it out if anyone needs it to help out.
>
> Anyway, I'm finding that I get a performance hit after a certain thread is
> woken up by a condition variable and then killed. Now, this may not have
> anything to do with the thread itself, I guess. The thread reads from a
> serial port, placing data a byte at a time into a file.
>
> After the thread is done it goes back to sleep, and my application
> continues, however it is substantially slower (about 14% slower). I have
> closed all files. Also, I have even tried killing the thread all together
> (instead of putting it back to sleep) and this did not help either.
>
> Has anyone heard of this?
>
> Thanks,
>
> Phil
> MIT - Space Systems Lab
>
>
Phil,
I am also using the NanoEngine from BSE and can reccomend
http://recycle.lbl.gov/~ldolittl/bse as worth reading. It lets you build kernel
and tools for the NanoEngine from sources, so you know what you are getting.
For reference, the version of the kernel that I got on the BSE linux dev CD was
2.4.0-test4. Can't help with the pthreads question, though.
-Ed
-----------------------------
Ed Doxat
Sony BPE
email : [EMAIL PROTECTED]
*************************************************************************
The information contained in this message or any of its
attachments may be privileged and confidential and intended
for the exclusive use of the addressee. If you are not the
addressee any disclosure, reproduction, distribution or other
dissemination or use of this communication is strictly prohibited
**************************************************************************
_______________________________________________
http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm
Please visit the above address for information on this list.