From: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
To: Theodore Ts'o <tytso-3s7WtUTddSA@public.gmane.org>,
Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
Ross Zwisler
<ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org,
Matthew Wilcox <mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org>,
Dave Chinner <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org>,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
Andreas Dilger
<adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org>,
Alexander Viro
<viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>,
Jan Kara <jack-IBi9RG/b67k@public.gmane.org>,
linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH v2 2/9] ext2: tell DAX the size of allocation holes
Date: Tue, 30 Aug 2016 00:21:29 -0700 [thread overview]
Message-ID: <20160830072129.GA10997@infradead.org> (raw)
In-Reply-To: <20160829125741.cdnbb2uaditcmnw2-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
On Mon, Aug 29, 2016 at 08:57:41AM -0400, Theodore Ts'o wrote:
> It's been on my todo list. The only reason why I haven't done it yet
> is because I knew you were working on a solution, and I didn't want to
> do things one way for buffered I/O, and a different way for Direct
> I/O, and disentangling the DIO code and the different assumptions of
> how different file systems interact with the DIO code is a *mess*.
It is. I have an almost working iomap direct I/O implementation, which
simplifies a lot of this. Partially because it expects sane locking
(a shared lock should be held over read, fortunately we now have i_rwsem
for that) and allows less opt-in/out behavior, and partially just
because we have so much better infrastructure now (iomap, bio chaining,
iov_iters, ...).
The iomap version is still work in progress, but I'm about to post
a cut down version for block devices which reduces I/O latency by
20%.
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@infradead.org>
To: Theodore Ts'o <tytso@mit.edu>,
Christoph Hellwig <hch@infradead.org>,
Ross Zwisler <ross.zwisler@linux.intel.com>,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
linux-nvdimm@ml01.01.org, Matthew Wilcox <mawilcox@microsoft.com>,
Dave Chinner <david@fromorbit.com>,
linux-mm@kvack.org, Andreas Dilger <adilger.kernel@dilger.ca>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Jan Kara <jack@suse.com>,
linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org
Subject: Re: [PATCH v2 2/9] ext2: tell DAX the size of allocation holes
Date: Tue, 30 Aug 2016 00:21:29 -0700 [thread overview]
Message-ID: <20160830072129.GA10997@infradead.org> (raw)
In-Reply-To: <20160829125741.cdnbb2uaditcmnw2@thunk.org>
On Mon, Aug 29, 2016 at 08:57:41AM -0400, Theodore Ts'o wrote:
> It's been on my todo list. The only reason why I haven't done it yet
> is because I knew you were working on a solution, and I didn't want to
> do things one way for buffered I/O, and a different way for Direct
> I/O, and disentangling the DIO code and the different assumptions of
> how different file systems interact with the DIO code is a *mess*.
It is. I have an almost working iomap direct I/O implementation, which
simplifies a lot of this. Partially because it expects sane locking
(a shared lock should be held over read, fortunately we now have i_rwsem
for that) and allows less opt-in/out behavior, and partially just
because we have so much better infrastructure now (iomap, bio chaining,
iov_iters, ...).
The iomap version is still work in progress, but I'm about to post
a cut down version for block devices which reduces I/O latency by
20%.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@infradead.org>
To: "Theodore Ts'o" <tytso@mit.edu>,
Christoph Hellwig <hch@infradead.org>,
Ross Zwisler <ross.zwisler@linux.intel.com>,
linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
linux-nvdimm@ml01.01.org, Matthew Wilcox <mawilcox@microsoft.com>,
Dave Chinner <david@fromorbit.com>,
linux-mm@kvack.org, Andreas Dilger <adilger.kernel@dilger.ca>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Jan Kara <jack@suse.com>,
linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org
Subject: Re: [PATCH v2 2/9] ext2: tell DAX the size of allocation holes
Date: Tue, 30 Aug 2016 00:21:29 -0700 [thread overview]
Message-ID: <20160830072129.GA10997@infradead.org> (raw)
In-Reply-To: <20160829125741.cdnbb2uaditcmnw2@thunk.org>
On Mon, Aug 29, 2016 at 08:57:41AM -0400, Theodore Ts'o wrote:
> It's been on my todo list. The only reason why I haven't done it yet
> is because I knew you were working on a solution, and I didn't want to
> do things one way for buffered I/O, and a different way for Direct
> I/O, and disentangling the DIO code and the different assumptions of
> how different file systems interact with the DIO code is a *mess*.
It is. I have an almost working iomap direct I/O implementation, which
simplifies a lot of this. Partially because it expects sane locking
(a shared lock should be held over read, fortunately we now have i_rwsem
for that) and allows less opt-in/out behavior, and partially just
because we have so much better infrastructure now (iomap, bio chaining,
iov_iters, ...).
The iomap version is still work in progress, but I'm about to post
a cut down version for block devices which reduces I/O latency by
20%.
next prev parent reply other threads:[~2016-08-30 7:21 UTC|newest]
Thread overview: 135+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-23 22:04 [PATCH v2 0/9] re-enable DAX PMD support Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 2/9] ext2: tell DAX the size of allocation holes Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
[not found] ` <20160823220419.11717-3-ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2016-08-25 7:57 ` Christoph Hellwig
2016-08-25 7:57 ` Christoph Hellwig
2016-08-25 7:57 ` Christoph Hellwig
[not found] ` <20160825075728.GA11235-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-08-25 19:25 ` Ross Zwisler
2016-08-25 19:25 ` Ross Zwisler
2016-08-25 19:25 ` Ross Zwisler
2016-08-26 21:29 ` Ross Zwisler
2016-08-26 21:29 ` Ross Zwisler
2016-08-26 21:29 ` Ross Zwisler
2016-08-29 0:42 ` Dave Chinner
2016-08-29 0:42 ` Dave Chinner
[not found] ` <20160826212934.GA11265-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2016-08-29 7:41 ` Christoph Hellwig
2016-08-29 7:41 ` Christoph Hellwig
2016-08-29 7:41 ` Christoph Hellwig
[not found] ` <20160829074116.GA16491-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-08-29 12:57 ` Theodore Ts'o
2016-08-29 12:57 ` Theodore Ts'o
2016-08-29 12:57 ` Theodore Ts'o
[not found] ` <20160829125741.cdnbb2uaditcmnw2-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2016-08-30 7:21 ` Christoph Hellwig [this message]
2016-08-30 7:21 ` Christoph Hellwig
2016-08-30 7:21 ` Christoph Hellwig
2016-09-09 16:48 ` Ross Zwisler
2016-09-09 16:48 ` Ross Zwisler
2016-09-09 16:48 ` Ross Zwisler
2016-09-09 20:35 ` Matthew Wilcox
2016-09-09 20:35 ` Matthew Wilcox
[not found] ` <DM2PR21MB0089BCA980B67D8C53B25A1BCBFA0-B2pw06WL+/BAVFCO9/lqPs1VXTxX1y3OvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2016-09-09 22:34 ` Dan Williams
2016-09-09 22:34 ` Dan Williams
2016-09-09 22:34 ` Dan Williams
[not found] ` <CAPcyv4hjna08+Yw23w_V2f-RbBE6ar220+YGCuBVA-TACKWNug-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-09-10 7:31 ` Christoph Hellwig
2016-09-10 7:31 ` Christoph Hellwig
2016-09-10 7:31 ` Christoph Hellwig
2016-09-10 7:50 ` Matthew Wilcox
2016-09-10 7:50 ` Matthew Wilcox
2016-09-10 17:49 ` Theodore Ts'o
2016-09-10 17:49 ` Theodore Ts'o
[not found] ` <20160910174910.yyirb7smiob7evt5-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2016-09-11 0:42 ` Matthew Wilcox
2016-09-11 0:42 ` Matthew Wilcox
2016-09-11 0:42 ` Matthew Wilcox
2016-09-10 8:15 ` Matthew Wilcox
2016-09-10 8:15 ` Matthew Wilcox
2016-09-10 8:15 ` Matthew Wilcox
2016-09-10 14:56 ` Dan Williams
2016-09-10 14:56 ` Dan Williams
2016-09-10 7:30 ` Christoph Hellwig
2016-09-10 7:30 ` Christoph Hellwig
2016-09-10 7:33 ` Matthew Wilcox
2016-09-10 7:33 ` Matthew Wilcox
2016-09-10 7:42 ` Christoph Hellwig
2016-09-10 7:42 ` Christoph Hellwig
2016-09-10 7:52 ` Matthew Wilcox
2016-09-10 7:52 ` Matthew Wilcox
[not found] ` <DM2PR21MB0089C20EF469AA91A916867CCBFD0-B2pw06WL+/BAVFCO9/lqPs1VXTxX1y3OvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2016-09-11 12:47 ` Christoph Hellwig
2016-09-11 12:47 ` Christoph Hellwig
2016-09-11 12:47 ` Christoph Hellwig
2016-09-11 22:57 ` Ross Zwisler
2016-09-11 22:57 ` Ross Zwisler
2016-09-10 15:55 ` Matthew Wilcox
2016-09-10 15:55 ` Matthew Wilcox
2016-09-15 20:09 ` Ross Zwisler
2016-09-15 20:09 ` Ross Zwisler
2016-09-15 20:09 ` Ross Zwisler
[not found] ` <20160823220419.11717-1-ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2016-08-23 22:04 ` [PATCH v2 1/9] ext4: allow DAX writeback for hole punch Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-09-21 15:22 ` Ross Zwisler
2016-09-21 15:22 ` Ross Zwisler
2016-09-21 15:22 ` Ross Zwisler
2016-09-21 15:22 ` Ross Zwisler
2016-09-21 15:22 ` Ross Zwisler
2016-09-22 6:59 ` Jan Kara
2016-09-22 6:59 ` Jan Kara
2016-09-22 6:59 ` Jan Kara
2016-09-22 6:59 ` Jan Kara
2016-09-22 15:51 ` Theodore Ts'o
2016-09-22 15:51 ` Theodore Ts'o
2016-08-23 22:04 ` [PATCH v2 3/9] ext4: tell DAX the size of allocation holes Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 4/9] dax: remove buffer_size_valid() Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 5/9] dax: make 'wait_table' global variable static Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 6/9] dax: consistent variable naming for DAX entries Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 7/9] dax: coordinate locking for offsets in PMD range Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 8/9] dax: re-enable DAX PMD support Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` [PATCH v2 9/9] dax: remove "depends on BROKEN" from FS_DAX_PMD Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-23 22:04 ` Ross Zwisler
2016-08-30 23:01 ` [PATCH v2 0/9] re-enable DAX PMD support Ross Zwisler
2016-08-30 23:01 ` Ross Zwisler
2016-08-30 23:01 ` Ross Zwisler
2016-08-30 23:01 ` Ross Zwisler
2016-08-31 20:20 ` Kani, Toshimitsu
2016-08-31 20:20 ` Kani, Toshimitsu
2016-08-31 20:20 ` Kani, Toshimitsu
[not found] ` <1472674799.2092.19.camel-ZPxbGqLxI0U@public.gmane.org>
2016-08-31 21:36 ` Ross Zwisler
2016-08-31 21:36 ` Ross Zwisler
2016-08-31 21:36 ` Ross Zwisler
2016-08-31 21:36 ` Ross Zwisler
2016-08-31 21:36 ` Ross Zwisler
[not found] ` <20160831213607.GA6921-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2016-08-31 22:08 ` Kani, Toshimitsu
2016-08-31 22:08 ` Kani, Toshimitsu
2016-08-31 22:08 ` Kani, Toshimitsu
2016-08-31 22:08 ` Kani, Toshimitsu
2016-09-01 16:21 ` Ross Zwisler
2016-09-01 16:21 ` Ross Zwisler
2016-09-01 16:21 ` Ross Zwisler
2016-09-01 16:21 ` Ross Zwisler
2016-09-01 16:21 ` Ross Zwisler
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=20160830072129.GA10997@infradead.org \
--to=hch-wegcikhe2lqwvfeawa7xhq@public.gmane.org \
--cc=adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org \
--cc=jack-IBi9RG/b67k@public.gmane.org \
--cc=linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=linux-nvdimm-y27Ovi1pjclAfugRpC6u6w@public.gmane.org \
--cc=mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org \
--cc=ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org \
--cc=tytso-3s7WtUTddSA@public.gmane.org \
--cc=viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.