linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* debugfs fill_bgs command?
@ 2007-04-11 20:17 Eric Sandeen
  2007-04-11 22:52 ` Andreas Dilger
  0 siblings, 1 reply; 3+ messages in thread
From: Eric Sandeen @ 2007-04-11 20:17 UTC (permalink / raw)
  To: ext4 development

I was wondering if others think it would be useful to have a "fill_bgs"
command in debugfs; this would (minimally) mark the lower X bgs as full
for both inodes & blocks (possibly either/or), to allow testing the
higher block groups.  I'm using a hacked up version of this to do a
little ext3 testing above 8T.  Given that this would really only be a
testing option in nature, would it be accepted into e2fsprogs?  I'd
probably need some sort of "unfill" command as well, to put the bg
counters back where they should be; probably by actually reading the
bitmaps.  This way fsck would still find a consistent filesystem...

I had originally written a set_bg_field function too, to go with inode &
sb variants, though for marking the first few thousand bg's it was going
to get a bit tedious... :)

Thoughts?

Thanks,
-Eric

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: debugfs fill_bgs command?
  2007-04-11 20:17 debugfs fill_bgs command? Eric Sandeen
@ 2007-04-11 22:52 ` Andreas Dilger
  2007-04-12  1:21   ` Eric Sandeen
  0 siblings, 1 reply; 3+ messages in thread
From: Andreas Dilger @ 2007-04-11 22:52 UTC (permalink / raw)
  To: Eric Sandeen; +Cc: ext4 development

On Apr 11, 2007  15:17 -0500, Eric Sandeen wrote:
> I was wondering if others think it would be useful to have a "fill_bgs"
> command in debugfs; this would (minimally) mark the lower X bgs as full
> for both inodes & blocks (possibly either/or), to allow testing the
> higher block groups.  I'm using a hacked up version of this to do a
> little ext3 testing above 8T.  Given that this would really only be a
> testing option in nature, would it be accepted into e2fsprogs?  I'd
> probably need some sort of "unfill" command as well, to put the bg
> counters back where they should be; probably by actually reading the
> bitmaps.  This way fsck would still find a consistent filesystem...
> 
> I had originally written a set_bg_field function too, to go with inode &
> sb variants, though for marking the first few thousand bg's it was going
> to get a bit tedious... :)

Try "mke2fs -O lazy_bg XXX" and be happy.  Can be used on any kernel as
it is a COMPAT feature and marks all but first and last groups as full.

Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: debugfs fill_bgs command?
  2007-04-11 22:52 ` Andreas Dilger
@ 2007-04-12  1:21   ` Eric Sandeen
  0 siblings, 0 replies; 3+ messages in thread
From: Eric Sandeen @ 2007-04-12  1:21 UTC (permalink / raw)
  To: Andreas Dilger; +Cc: ext4 development

Andreas Dilger wrote:

> Try "mke2fs -O lazy_bg XXX" and be happy.  Can be used on any kernel as
> it is a COMPAT feature and marks all but first and last groups as full.

Ugh, how'd I miss that.  :)

Thanks...

-Eric

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2007-04-12  1:22 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-04-11 20:17 debugfs fill_bgs command? Eric Sandeen
2007-04-11 22:52 ` Andreas Dilger
2007-04-12  1:21   ` Eric Sandeen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).