linux-kernel - Re: [PATCH 2/2] fsmap: use blocksize units instead of cluster units

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]

Message-ID: <aN65nsxrpMoK33h5@li-dc0c254c-257c-11b2-a85c-98b6c1322444.ibm.com>
Date: Thu, 2 Oct 2025 23:12:54 +0530
From: Ojaswin Mujoo <ojaswin@...ux.ibm.com>
To: "Theodore Ts'o" <tytso@....edu>
Cc: linux-ext4@...r.kernel.org, Ritesh Harjani <ritesh.list@...il.com>,
        Zhang Yi <yi.zhang@...wei.com>, linux-kernel@...r.kernel.org,
        "Darrick J . Wong" <djwong@...nel.org>
Subject: Re: [PATCH 2/2] fsmap: use blocksize units instead of cluster units

On Fri, Sep 26, 2025 at 08:35:36AM -0400, Theodore Ts'o wrote:
> On Fri, Sep 05, 2025 at 01:44:47PM +0530, Ojaswin Mujoo wrote:
> > Currently, ext4's implementation of fsmap converts the ranges to cluster
> > units if bigalloc is enabled and then converts to block units whenever
> > needed. However, this back and forth conversion has known to cause
> > several edge case issues since even with bigalloc the metadata is still
> > in block size unit....
> 
> This commit causes ext4/028 to fail with a 1k blocksize.  The failure
> happens after just under 45 minutes; before this commit, ext4/028
> would complete after a second.
> 
> Do you have a preference regarding whether I just drop this commit, or
> drop the whole series?  The previous patch looks fine to me and fixes
> a real problem, so my plan is to keep the 1/2 commit and drop this
> one.

Hey Ted,

Thanks for pointing this out, I'll look into the failure. Sorry for the
late reply I've been on vacation this week. 

I'll get to it as soon as possible.

Regards,
ojaswin

> 
> Cheers,
> 
> 						- Ted