[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Y8vi8/sCvOxvLzzc@infradead.org>
Date: Sat, 21 Jan 2023 05:04:51 -0800
From: Christoph Hellwig <hch@...radead.org>
To: David Howells <dhowells@...hat.com>
Cc: Al Viro <viro@...iv.linux.org.uk>,
Christoph Hellwig <hch@...radead.org>,
Matthew Wilcox <willy@...radead.org>,
Jens Axboe <axboe@...nel.dk>, Jan Kara <jack@...e.cz>,
Jeff Layton <jlayton@...nel.org>,
Logan Gunthorpe <logang@...tatee.com>,
linux-fsdevel@...r.kernel.org, linux-block@...r.kernel.org,
linux-kernel@...r.kernel.org, Christoph Hellwig <hch@....de>
Subject: Re: [PATCH v7 4/8] block: Rename BIO_NO_PAGE_REF to BIO_PAGE_REFFED
and invert the meaning
On Fri, Jan 20, 2023 at 05:55:52PM +0000, David Howells wrote:
> Rename BIO_NO_PAGE_REF to BIO_PAGE_REFFED and invert the meaning. In a
> following patch I intend to add a BIO_PAGE_PINNED flag to indicate that the
> page needs unpinning and this way both flags have the same logic.
>
> Signed-off-by: David Howells <dhowells@...hat.com>
> Reviewed-by: Christoph Hellwig <hch@....de>
As pointed out last time, we really should not set the flag by default.
When a bio is allocated there are no pages to be released, that only
happens when we add dio-like pages to the bio. Instead of explaining
why I mean again, I've put together a version of this series that
implements this and my other suggestions here:
http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/dio-pin-pages
This also tests fine with xfs and btrfs and nvme passthrough I/O.
Powered by blists - more mailing lists