linux-kernel - Re: [PATCH v2 07/13] xfs: Introduce FORCEALIGN inode flag

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: <2fd991cc-8f29-47ce-a78a-c11c79d74e07@oracle.com>
Date: Tue, 23 Jul 2024 16:01:41 +0100
From: John Garry <john.g.garry@...cle.com>
To: Christoph Hellwig <hch@....de>
Cc: Dave Chinner <david@...morbit.com>, "Darrick J. Wong"
 <djwong@...nel.org>,
        chandan.babu@...cle.com, dchinner@...hat.com, viro@...iv.linux.org.uk,
        brauner@...nel.org, jack@...e.cz, linux-xfs@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        catherine.hoang@...cle.com, martin.petersen@...cle.com
Subject: Re: [PATCH v2 07/13] xfs: Introduce FORCEALIGN inode flag

On 23/07/2024 15:42, Christoph Hellwig wrote:
> On Tue, Jul 23, 2024 at 11:11:28AM +0100, John Garry wrote:
>> I am looking at something like this to implement read-only for those inodes:
> 
> Yikes.  Treating individual inodes in a file systems as read-only
> is about the most confusing and harmful behavior we could do.

That was the suggestion which I was given earlier in this thread.

> 
> Just treat it as any other rocompat feature please an mount the entire
> file system read-only if not supported.
> 
> Or even better let this wait a little, and work with Darrick to work
> on the rextsize > 1 reflіnk patches and just make the thing work.

I'll let Darrick comment on this.

> 
>>> So what about forcealign and RT?
>>
>> Any opinion on this?
> 
> What about forcealign and RT?

In this series version I was mounting the whole FS as RO if 
XFS_FEAT_FORCEALIGN and XFS_FEAT_REFLINK was found in the SB. And so 
very different to how I was going to individual treat inodes which 
happen to be forcealign and reflink, above.

So I was asking guidance when whether that approach (for RT and 
forcealign) is sound.