Re: I/O Scheduler

On 2008-08-19 at 00:21:46 [+0200], Salvatore Benedetto <[email protected]> 
wrote:
> 
> while trying to hunt down some bugs in the file cache I found some
> TODOs (not necessarely
> in the file cache code) about rewriting some stuff once the I/O
> scheduler is in place.
> 
> Now I'm wondering what's the status of the scheduler and if there is some 
> work
> going on behind the scene on these parts that were meant to be "fixed", 
> before
> I continue hunting these bugs, as it may be related and save me some time.
> 
> In particular, I'm interested in fixing the bugs I reported: #2618 ,
> #2631 and #2634.

You seem to confuse file and block cache. Anyway, I haven't really looked 
closer into it, but the explanation you've added to #2631 seems right. I 
cannot say much wrt. #2634; it doesn't look like the new I/O request 
framework is to blame, though. No idea at all regarding #2618; assuming that 
an on-disk structure corruption was detected, the recent I/O changes 
certainly are candidates. If you're able to reproduce it, you can check back 
with an earlier revision (e.g. the one tagged "somewhat stable").

> Beside, while we are at it, whats the policy used by the scheduler? :-)

Currently no-op. This will change, but with what strategy (or strategies) 
we'll end up, has to be seen.

CU, Ingo

-------------------------------------------------------------------------
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=/