From: Keith Busch <kbusch@kernel.org>
To: Keith Busch <kbusch@meta.com>
Cc: linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org, snitzer@kernel.org,
axboe@kernel.dk, dw@davidwei.uk, brauner@kernel.org,
Hannes Reinecke <hare@suse.de>
Subject: Re: [PATCHv2 1/7] block: check for valid bio while splitting
Date: Wed, 13 Aug 2025 14:06:19 -0600 [thread overview]
Message-ID: <aJzwO9dYeBQAHnCC@kbusch-mbp> (raw)
In-Reply-To: <20250805141123.332298-2-kbusch@meta.com>
On Tue, Aug 05, 2025 at 07:11:17AM -0700, Keith Busch wrote:
> bio_for_each_bvec(bv, bio, iter) {
> + if (bv.bv_offset & lim->dma_alignment)
> + return -EINVAL;
I have a question about this part here because testing with various scsi
is showing odd results.
NVMe wants this check to actually be this:
if ((bv.bv_offset | bv.bv_len) & lim->dma_alignment)
because the dma alignment defines not only the starting address offset,
but also the length. NVMe's default alignment is 4 bytes.
But I can't make that change because many scsi devices don't set the dma
alignment and get the default 511 value. This is fine for the memory
address offset, but the lengths sent for various inquriy commands are
much smaller, like 4 and 32 byte lengths. That length wouldn't pass the
dma alignment granularity, so I think the default value is far too
conservative. Does the address start size need to be a different limit
than minimum length? I feel like they should be the same, but maybe
that's just an nvme thing.
next prev parent reply other threads:[~2025-08-13 20:06 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-05 14:11 [PATCHv2 0/7] direct-io: even more flexible io vectors Keith Busch
2025-08-05 14:11 ` [PATCHv2 1/7] block: check for valid bio while splitting Keith Busch
2025-08-10 14:37 ` Christoph Hellwig
2025-08-10 15:39 ` Keith Busch
2025-08-10 17:14 ` Christoph Hellwig
2025-08-13 20:06 ` Keith Busch [this message]
2025-08-13 20:41 ` Bart Van Assche
2025-08-13 21:01 ` Keith Busch
2025-08-13 23:17 ` Bart Van Assche
2025-08-14 1:42 ` Damien Le Moal
2025-08-13 21:23 ` Martin K. Petersen
2025-08-13 21:52 ` Keith Busch
2025-08-18 4:49 ` Christoph Hellwig
2025-08-05 14:11 ` [PATCHv2 2/7] block: align the bio after building it Keith Busch
2025-08-06 6:07 ` Hannes Reinecke
2025-08-10 14:43 ` Christoph Hellwig
2025-08-05 14:11 ` [PATCHv2 3/7] block: simplify direct io validity check Keith Busch
2025-08-05 14:11 ` [PATCHv2 4/7] iomap: " Keith Busch
2025-08-06 6:07 ` Hannes Reinecke
2025-08-05 14:11 ` [PATCHv2 5/7] block: remove bdev_iter_is_aligned Keith Busch
2025-08-05 14:11 ` [PATCHv2 6/7] blk-integrity: use simpler alignment check Keith Busch
2025-08-05 14:11 ` [PATCHv2 7/7] iov_iter: remove iov_iter_is_aligned Keith Busch
2025-08-06 6:08 ` Hannes Reinecke
2025-08-06 3:03 ` [PATCHv2 0/7] direct-io: even more flexible io vectors Martin K. Petersen
2025-08-06 14:51 ` Christoph Hellwig
2025-08-07 23:20 ` Keith Busch
2025-08-11 10:32 ` 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=aJzwO9dYeBQAHnCC@kbusch-mbp \
--to=kbusch@kernel.org \
--cc=axboe@kernel.dk \
--cc=brauner@kernel.org \
--cc=dw@davidwei.uk \
--cc=hare@suse.de \
--cc=kbusch@meta.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=snitzer@kernel.org \
/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