linux-ext4 - Re: [PATCH] ext4: fix callers of ext4_handle_release

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] [day] [month] [year] [list]

Date:	Mon, 21 Mar 2011 06:34:02 +0200
From:	Amir Goldstein <amir73il@...il.com>
To:	"Ted Ts'o" <tytso@....edu>
Cc:	Ext4 Developers List <linux-ext4@...r.kernel.org>
Subject: Re: [PATCH] ext4: fix callers of ext4_handle_release_buffer() API

On Mon, Mar 21, 2011 at 4:43 AM, Ted Ts'o <tytso@....edu> wrote:
> On Tue, Mar 01, 2011 at 03:30:24PM +0200, Amir Goldstein wrote:
>> ext4_handle_release_buffer() is an API, which is not being used properly.
>> This is not so bad considering that it calls jbd2_journal_release_buffer(),
>> which does nothing, but the snapshots implementation of this API does
>> something.
>>
>> ext4_jbd2.h defines 2 identical wrapper functions:
>> ext4_journal_release_buffer() and ext4_handle_release_buffer().
>> The former has no callers, so it was removed.
>
> Note that there are some commented-out calls to
> ext4_journal_release_buffer() in resize.c; they're probably commented
> out because at least today, jbd2_journal_release_buffer() is indeed a
> no-op.
>
> I'll update resize.c to use the preferred call.  Speaking of which,
> how does on-line resize work in the face of the snapshot feature?  Is
> that something that is or isn't supported?

Yes, it is supported. (offline resize isn't supported).
The old snapshots sizes (i_disksize) represent the file system
size at the time of the snapshot.

>
> Note also that there may be places where a buffer has
> get_write_access() called on it, only for it to be never modified.
> resize.c is an example of that.

Those places can be a problem (since COW does happen and use
up transaction credits) and calling buffer_release can help
(it tries to extend the transaction), but specifically, resize.c is not
a problem, since blocks outside the active snapshot's i_disksize
are never COWed.

>
>> There are 2 users of the API:
>> ext4_new_inode() calls ext4_handle_release_buffer() and
>> ext4_xattr_block_set() calls jbd2_journal_release_buffer() directly.
>> The latter was chagned to call the wrapper API.
>>
>> Signed-off-by: Amir Goldstein <amir73il@...rs.sf.net>
>
> I'll apply this patch the changes.
>
>                                                - 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