From: Keith Busch <kbusch@kernel.org>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Keith Busch <kbusch@meta.com>,
linux-block@vger.kernel.org, axboe@kernel.dk, hch@lst.de,
ebiggers@kernel.org
Subject: Re: [PATCHv6] blk-integrity: support arbitrary buffer alignment
Date: Mon, 24 Nov 2025 10:57:34 -0700 [thread overview]
Message-ID: <aSScjjoOZswC35nR@kbusch-mbp> (raw)
In-Reply-To: <yq1o6or2y11.fsf@ca-mkp.ca.oracle.com>
On Mon, Nov 24, 2025 at 11:31:54AM -0500, Martin K. Petersen wrote:
> > A bio segment may have partial interval block data with the rest
> > continuing into the next segments because direct-io data payloads only
> > need to aligned in memory to the device's DMA limits.
>
> No objections from me if NVMe needs this.
Truthfully, the users I'm enabling rely on the NVMe io_uring passthrough
interface, so much of this path isn't in the path. I just want to remove
the dma_alignment limit when a csum is used because the incoming data is
naturally unaligned in memory to the lba size. Most of this patch is
prep work to safely modify one line of code :) But I'm still really
happy with the result!
This should fix a real bug too: requests with data integrity and merged
bio's would fail without this. It's unlikely you could merge such bio's
due to integrity segment and virtual boundary limits so it's not
surprising issues haven't come up, but there are patches staged for 6.19
that make it easier to do that for capable NVMe devices.
next prev parent reply other threads:[~2025-11-24 17:57 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-24 16:17 [PATCHv6] blk-integrity: support arbitrary buffer alignment Keith Busch
2025-11-24 16:31 ` Martin K. Petersen
2025-11-24 17:57 ` Keith Busch [this message]
2025-11-24 20:49 ` Martin K. Petersen
2025-11-24 16:58 ` Christoph Hellwig
2025-11-24 21:34 ` Caleb Sander Mateos
2025-11-24 22:56 ` Keith Busch
2025-11-25 3:41 ` Caleb Sander Mateos
2025-11-25 11:31 ` Christoph Hellwig
2025-11-25 14:24 ` Keith Busch
2025-11-25 16:46 ` Christoph Hellwig
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=aSScjjoOZswC35nR@kbusch-mbp \
--to=kbusch@kernel.org \
--cc=axboe@kernel.dk \
--cc=ebiggers@kernel.org \
--cc=hch@lst.de \
--cc=kbusch@meta.com \
--cc=linux-block@vger.kernel.org \
--cc=martin.petersen@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox