[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250826080200.GA23095@lst.de>
Date: Tue, 26 Aug 2025 10:02:00 +0200
From: Christoph Hellwig <hch@....de>
To: Keith Busch <kbusch@...nel.org>
Cc: Keith Busch <kbusch@...a.com>, linux-block@...r.kernel.org,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
snitzer@...nel.org, axboe@...nel.dk, dw@...idwei.uk,
brauner@...nel.org, martin.petersen@...cle.com, djwong@...nel.org,
linux-xfs@...r.kernel.org, viro@...iv.linux.org.uk
Subject: Re: [PATCHv3 3/8] block: align the bio after building it
On Mon, Aug 25, 2025 at 06:37:05PM -0600, Keith Busch wrote:
> On Mon, Aug 25, 2025 at 09:47:44AM +0200, Christoph Hellwig wrote:
> > Also with this we should be able to drop the iov_iter_alignment check
> > for always COW inodes in xfs_file_dio_write. If you don't feel like
> > doing that yourself I can add it to my todo list.
>
> I'm unsure about the commit that introduced that behavior, so I think
> you should remove it if you know its okay. :)
Yeah.
> Specifically, we have this in the comments and commit message:
>
> check the alignment of each individual iovec segment, as they could
> end up with different I/Os due to the way bio_iov_iter_get_pages works
>
> bio_iov_iter_get_pages() might submit the segments as separate IO's
> anyway for other reasons. I am not sure why the alignment conditions are
> handled specifically here.
I'll take another look. Basically what this wants to prevent is
bio_iov_iter_get_pages creating bios not aligned to file system
block size.
Powered by blists - more mailing lists