linux-ext4 - Re: [PATCH -V2 4/5] ext4: Use both hi and lo bits of the group desc values.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20081124021302.GB2163@mit.edu>
Date:	Sun, 23 Nov 2008 21:13:02 -0500
From:	Theodore Tso <tytso@....EDU>
To:	"Aneesh Kumar K.V" <aneesh.kumar@...ux.vnet.ibm.com>
Cc:	Eric Sandeen <sandeen@...hat.com>, Coly Li <coyli@...e.de>,
	cmm@...ibm.com, linux-ext4@...r.kernel.org
Subject: Re: [PATCH -V2 4/5] ext4: Use both hi and lo bits of the group
	desc values.

On Fri, Nov 21, 2008 at 11:11:10PM +0530, Aneesh Kumar K.V wrote:
> On Fri, Nov 21, 2008 at 11:29:49AM -0600, Eric Sandeen wrote:
> > Aneesh Kumar K.V wrote:
> > > Rename the lower bits with suffix _lo and add helper
> > > to access the values. Also rename bg_itable_unused_hi
> > > to bg_pad as in e2fsprogs. bg_itable_unused_hi is never
> > > used before
> > 
> > Maybe you can add info about under which circumstances (what sort of
> > fielsystem geometry) the hi bits are needed?  And the failure mode when
> > we don't use them, if that's known?
> 
> Coly actually said it was added for inode reservation. But he agreed to
> drop the high bits now and add them later once the performance issue
> with the patch is addressed.

The bg_itable_unused field is used for the lazy initialization of
inodes.  If the units of this field was blocks was blocks, we wouldn't
need bg_itable_unused_hi for quite a larger set of blocks.

I don't think we should nuke bg_itable_unused_hi, especially given
that we have a patch to support the kernel initializing the inodes
(yet to be reviewed).  As near as I can tell at the time when the
fields were added to e2fsprogs, we hadn't yet assigned the pad field
to bg_itable_unused_hi, but it's clearly needed if we want to support
a larger number of blocks or inodes per block group.

Instead we should create the helper functions for bg_itable_unused,
and update e2fsprogs to also support these fields.  Adding e2fsprogs
for large block groups is lower priority than large block numbers,
though.

						- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html