Linux-Fsdevel Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Kanchan Joshi <joshi.k@samsung.com>
To: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Cc: Damien Le Moal <Damien.LeMoal@wdc.com>,
	linux-fsdevel@vger.kernel.org, Jens Axboe <axboe@kernel.dk>,
	linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 2/2] zonefs: use zone-append for AIO as well
Date: Tue, 21 Jul 2020 18:13:43 +0530	[thread overview]
Message-ID: <20200721122724.GA17629@test-zns> (raw)
In-Reply-To: <20200720132118.10934-3-johannes.thumshirn@wdc.com>

[-- Attachment #1: Type: text/plain, Size: 2678 bytes --]

On Mon, Jul 20, 2020 at 10:21:18PM +0900, Johannes Thumshirn wrote:
>If we get an async I/O iocb with an O_APPEND or RWF_APPEND flag set,
>submit it using REQ_OP_ZONE_APPEND to the block layer.
>
>As an REQ_OP_ZONE_APPEND bio must not be split, this does come with an
>additional constraint, namely the buffer submitted to zonefs must not be
>bigger than the max zone append size of the underlying device. For
>synchronous I/O we don't care about this constraint as we can return short
>writes, for AIO we need to return an error on too big buffers.

I wonder what part of the patch implements that constraint on large
buffer and avoids short-write.
Existing code seems to trim iov_iter in the outset. 

        max = queue_max_zone_append_sectors(bdev_get_queue(bdev));
        max = ALIGN_DOWN(max << SECTOR_SHIFT, inode->i_sb->s_blocksize);
        iov_iter_truncate(from, max);

This will prevent large-buffer seeing that error, and will lead to partial write.

>On a successful completion, the position the data is written to is
>returned via AIO's res2 field to the calling application.
>
>Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
>---
> fs/zonefs/super.c  | 143 +++++++++++++++++++++++++++++++++++++++------
> fs/zonefs/zonefs.h |   3 +
> 2 files changed, 128 insertions(+), 18 deletions(-)
>
>diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
>index 5832e9f69268..f155a658675b 100644
>--- a/fs/zonefs/super.c
>+++ b/fs/zonefs/super.c
>@@ -24,6 +24,8 @@
>
> #include "zonefs.h"
>
>+static struct bio_set zonefs_dio_bio_set;
>+
> static inline int zonefs_zone_mgmt(struct zonefs_inode_info *zi,
> 				   enum req_opf op)
> {
>@@ -700,16 +702,71 @@ static const struct iomap_dio_ops zonefs_write_dio_ops = {
> 	.end_io			= zonefs_file_write_dio_end_io,
> };
>
>+struct zonefs_dio {
>+	struct kiocb		*iocb;
>+	struct task_struct	*waiter;
>+	int			error;
>+	struct work_struct	work;
>+	size_t			size;
>+	u64			sector;
>+	struct completion	completion;
>+	struct bio		bio;
>+};

How about this (will save 32 bytes) - 
+struct zonefs_dio {
+       struct kiocb            *iocb;
+       struct task_struct      *waiter;
+       int                     error;
+	union {
+       	struct work_struct      work;   //only for async IO
+       	struct completion       completion; //only for sync IO
+	};
+       size_t                  size;
+       u64                     sector;
+       struct bio              bio;
+};
And dio->error field is not required.
I see it being used at one place -
+       ret = zonefs_file_write_dio_end_io(iocb, dio->size,
+                                          dio->error, 0);
Here error-code can be picked from dio->bio.

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]



  parent reply	other threads:[~2020-07-21 13:02 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-20 13:21 [PATCH 0/2] zonefs: use zone-append for aio with rwf append Johannes Thumshirn
2020-07-20 13:21 ` [PATCH 1/2] fs: fix kiocb ki_complete interface Johannes Thumshirn
2020-07-20 13:38   ` Christoph Hellwig
2020-07-20 13:43     ` Damien Le Moal
2020-07-20 13:47       ` Christoph Hellwig
2020-07-20 13:21 ` [PATCH 2/2] zonefs: use zone-append for AIO as well Johannes Thumshirn
2020-07-20 13:45   ` Christoph Hellwig
2020-07-20 16:48     ` Johannes Thumshirn
2020-07-21  5:54       ` Christoph Hellwig
2020-07-22 12:43         ` Johannes Thumshirn
2020-07-22 13:02           ` Damien Le Moal
2020-07-22 14:53             ` Christoph Hellwig
2020-07-22 14:51           ` Christoph Hellwig
2020-07-22 15:00             ` Johannes Thumshirn
2020-07-24 13:57             ` Kanchan Joshi
2020-07-27  3:12               ` Damien Le Moal
2020-07-21 12:43   ` Kanchan Joshi [this message]
2020-07-22 14:32     ` Johannes Thumshirn

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=20200721122724.GA17629@test-zns \
    --to=joshi.k@samsung.com \
    --cc=Damien.LeMoal@wdc.com \
    --cc=axboe@kernel.dk \
    --cc=hch@lst.de \
    --cc=johannes.thumshirn@wdc.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --subject='Re: [PATCH 2/2] zonefs: use zone-append for AIO as well' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).