Darren Hart wrote:
> Gilles Carry wrote:
> 
>>This is mainly to put back the main thread as SCHED_OTHER
>>when doing disk I/O. (stats dump...)
>>In some circumstances, disk I/O within an RT task might hang the
>>system. (deadlock)
> 
> 
> Wha?  Can you elaborate on an example case?
> 
> While it isn't advisable to perform I/O during latency sensitive 
> sections for code, or while holding certain locks, etc., the system 
> should not deadlock!  If it doesn, we have ourselves a bug.

I did this because I remembered about this problem we had earlier:
http://www.mail-archive.com/[email protected]/msg01528.html
Before the fix, the test itself was not properly designed.

For the matrix test, I don't think I/O while being RT is a problem.
However, going back to sched_other before any disk I/O is a good habit
to take. That's the reason of this patch. Consider this as a cleanup.


Gilles.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Reply via email to