lkml.org 
[lkml]   [2004]   [Jun]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: drivers/block/ub.c
On Sat, 26 Jun 2004, Matthew Dharm wrote:

> > I agree that the debug logging in usb-storage is not good. A worthwhile
> > improvement would be to log only commands that fail or get an error, with
> > the logging selected by the normal USB debugging (not usb-storage verbose
> > debugging) configuration option. Matt, what do you think?
>
> This is an acknowledged weak point of usb-storage.
>
> A 'log only on error' system might be good... but it's going to be a bit
> tricky for two reasons:
>
> 1) We have to keep data around longer than currently, so we can log it if
> something goes wrong later (so we see how we got to this point).

Yep. I just wonder if logging only the current command will be enough.
Sometimes it's important to see what other, successful commands preceded
the one that didn't work. Well, we can always fall back on the old
verbose debugging.

> 2) What counts as an error? We probably only want this to kick in when the
> SCSI error-recovery does. Normal 'failed commands' probably shouldn't
> count.

An error is any time the transport routine returns TRANSPORT_ERROR or the
command is aborted by the SCSI layer. We should also log SCSI-initiated
resets, but not the auto-resets for the Bulk-only protocol.

Alan Stern

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:04    [W:1.870 / U:0.000 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site