[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: JFFS2: powerfailtesting again


Hi, David.

> You appear to have full debugging enabled. Over a 115200 serial port, that's
> also quite a severe bottleneck for JFFS2 operation.

I have done another run with no debugging - same result. I've turned debugging
on to produce additional information.

> If the only actual faults you're reporting are timeouts, then yes, I think
> you should wait longer. I don't believe JFFS2 is getting stuck in an
> endless loop - I think it's making progress with the garbage collection and
> just being very slow about it. Disabling erase suspend doesn't help you
> much here either.

How can i check this ? Can gc thread produce some info ?

> Increase the timeouts and see how long it actually takes. If the spew of
> debugging messages _stops_ and the operation hasn't finished, _that_ I'm
> interested in. It may be a deadlock which we haven't yet seen.

I have periodically checkered by tail logfile and when it was close to copy
timeouts logfile wasn't get updated (at least 4k limit was not exeeded, you
know pipe is 4k buffered)  . So I think it is _interesting_ .

Ok , i'll make another run with higher timeouts too.


To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com