xfs
[Top] [All Lists]

Re: [PATCH v4 0/8] block: prepare for multipage bvecs

To: Ming Lei <ming.lei@xxxxxxxxxxxxx>, Jens Axboe <axboe@xxxxxx>, linux-kernel@xxxxxxxxxxxxxxx
Subject: Re: [PATCH v4 0/8] block: prepare for multipage bvecs
From: Jens Axboe <axboe@xxxxxxxxx>
Date: Thu, 14 Apr 2016 08:11:03 -0600
Cc: linux-block@xxxxxxxxxxxxxxx, Christoph Hellwig <hch@xxxxxxxxxxxxx>, Al Viro <viro@xxxxxxxxxxxxxxxxxx>, "open list:DRBD DRIVER" <drbd-dev@xxxxxxxxxxxxxxxx>, Jan Kara <jack@xxxxxxx>, Keith Busch <keith.busch@xxxxxxxxx>, Kent Overstreet <kent.overstreet@xxxxxxxxx>, "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>, Mike Snitzer <snitzer@xxxxxxxxxx>, Shaohua Li <shli@xxxxxx>, Tejun Heo <tj@xxxxxxxxxx>, "open list:XFS FILESYSTEM" <xfs@xxxxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20150623.gappssmtp.com; s=20150623; h=subject:to:references:cc:from:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=2mldgws3DhKOKU9WBunxIKque0fEKqJtlNT1aI+hH7Q=; b=cMM6+u7SNIu+OaopsSq6ysKpM5ipu45ofeFjxHEIVo5KkKdj/8tGBpaaEUz4flJ34j bPwFUd94tzNPtS1UYpZHyqbd3DR71WpCbFlBCrUOkQYFCZXmd2hmc+7vWtjKvBdveU4N h0g9O1Un+urvRKe9sxQdVfNxtwLr0KSHDGQwctn5zdfUXMPEihilc8SMP2xoShU9J5KK W6OiJAEzPV4lGU6Ddge1BSXaYB/IuDzHbQuE8tpOvKgNtZrqWg7tiT35ahgR87CPwJpZ 6xsb0nnQptUrT1ynFF8ef4qSHfmAj4B6yd4ybxMFBp7Gl+n8EHr61IzS4pLPvnC4RkRu piEQ==
In-reply-to: <1460634438-26530-1-git-send-email-ming.lei@xxxxxxxxxxxxx>
References: <1460634438-26530-1-git-send-email-ming.lei@xxxxxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
On 04/14/2016 05:46 AM, Ming Lei wrote:
Hi,

Interests[1] have been shown in multipage bvecs, so this patchset
try to prepare for the support and do two things:

1) the 1st 4 patches use bvec iterator to implement iterate_bvec(),
then we can drop the non-standard way for iterating bvec, which
can be thought as a good cleanup for lib/iov_iter.c

2) remove BIO_MAX_SECTORS & BIO_MAX_SIZE, and now there is only
one user for each. Once multipage bvecs is introduced, one bio
may hold lots of sectors, and we should always use sort of BIO_MAX_VECS
which should be introduced in future and is similiar with current
BIO_MAX_PAGES.

xfstests(-a auto) have been run over ext4/xfs and no regression found
by this patchset.

We've had too many disasters in the block layer the last few series, I'm making the 4.7 round a nice and small one. I don't mind taking prep patches for the multipage bvecs, if they are simple and clean, but that's about the extent of it.

Just a heads up.

--
Jens Axboe

<Prev in Thread] Current Thread [Next in Thread>