linux-kernel - Re: [PATCH v2 5/5] block: use chunk_sectors when evaluating stacked atomic write limits

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: <98c7b752-5d09-46b0-b137-5843523f3ddf@oracle.com>
Date: Fri, 20 Jun 2025 12:35:04 +0100
From: John Garry <john.g.garry@...cle.com>
To: "Martin K. Petersen" <martin.petersen@...cle.com>
Cc: agk@...hat.com, snitzer@...nel.org, mpatocka@...hat.com, song@...nel.org,
        yukuai3@...wei.com, hch@....de, nilay@...ux.ibm.com, axboe@...nel.dk,
        dm-devel@...ts.linux.dev, linux-kernel@...r.kernel.org,
        linux-raid@...r.kernel.org, linux-block@...r.kernel.org,
        ojaswin@...ux.ibm.com
Subject: Re: [PATCH v2 5/5] block: use chunk_sectors when evaluating stacked
 atomic write limits

On 20/06/2025 03:40, Martin K. Petersen wrote:
>> Furthermore, io_min may be mutated when stacking devices, and this
>> makes it a poor candidate to hold the stripe size. Such an example (of
>> when io_min may change) would be when the io_min is less than the
>> physical block size.
> io_min is not allowed to be smaller than the physical_block_size. How
> did we end up violating that requirement?
> 
>    logical_block_size <= physical_block_size <= io_min <= io_opt

I should have been a bit less ambiguous in my words.

I meant that if we try to set the stacked device io_min (from the stripe 
size) less than the bottom device phys block size, then this leads to 
the stacked device io_min being set to the bottom device phys block 
size. That's what I mean by mutating. And that's why it's a bad idea to 
assume that the stripe size is in io_min.

Having said that, we should probably reject this even being allowed – we 
should not have physical blocks straddling stripes.