qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Denis V. Lunev" <den@openvz.org>
Cc: Kevin Wolf <kwolf@redhat.com>, Fam Zheng <famz@redhat.com>,
	Peter Lieven <pl@kamp.de>,
	qemu-devel@nongnu.org, Max Reitz <mreitz@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	"Denis V. Lunev" <den@openvz.org>
Subject: [Qemu-devel] [PATCH 6/7] block: use fallocate(FALLOC_FL_PUNCH_HOLE) & fallocate(0) to write zeroes
Date: Fri, 30 Jan 2015 11:42:16 +0300	[thread overview]
Message-ID: <1422607337-25335-7-git-send-email-den@openvz.org> (raw)
In-Reply-To: <1422607337-25335-1-git-send-email-den@openvz.org>

This sequence works efficiently if FALLOC_FL_ZERO_RANGE is not supported.
Unfortunately, FALLOC_FL_ZERO_RANGE is supported on really modern systems
and only for a couple of filesystems. FALLOC_FL_PUNCH_HOLE is much more
mature.

The sequence of 2 operations FALLOC_FL_PUNCH_HOLE and 0 is necessary due
to the following reasons:
- FALLOC_FL_PUNCH_HOLE creates a hole in the file, the file becomes
  sparse. In order to retain original functionality we must allocate
  disk space afterwards. This is done using fallocate(0) call
- fallocate(0) without preceeding FALLOC_FL_PUNCH_HOLE will do nothing
  if called above already allocated areas of the file, i.e. the content
  will not be zeroed

This should increase the performance a bit for not-so-modern kernels.

Signed-off-by: Denis V. Lunev <den@openvz.org>
CC: Max Reitz <mreitz@redhat.com>
CC: Kevin Wolf <kwolf@redhat.com>
CC: Stefan Hajnoczi <stefanha@redhat.com>
CC: Peter Lieven <pl@kamp.de>
CC: Fam Zheng <famz@redhat.com>
---
 block/raw-posix.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/block/raw-posix.c b/block/raw-posix.c
index 1c88ad8..7b42f37 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -967,6 +967,25 @@ static ssize_t handle_aiocb_write_zeroes(RawPosixAIOData *aiocb)
     }
 #endif
 
+#ifdef CONFIG_FALLOCATE_PUNCH_HOLE
+    if (s->has_discard && s->has_fallocate) {
+        int ret = do_fallocate(s->fd,
+                               FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
+                               aiocb->aio_offset, aiocb->aio_nbytes);
+        if (ret == 0) {
+            ret = do_fallocate(s->fd, 0, aiocb->aio_offset, aiocb->aio_nbytes);
+            if (ret == 0 || ret != -ENOTSUP) {
+                return ret;
+            }
+            s->has_fallocate = false;
+        } else if (ret != -ENOTSUP) {
+            return ret;
+        } else {
+            s->has_discard = false;
+        }
+    }
+#endif
+
 #ifdef CONFIG_FALLOCATE
     if (s->has_fallocate && aiocb->aio_offset >= bdrv_getlength(aiocb->bs)) {
         int ret = do_fallocate(s->fd, 0, aiocb->aio_offset, aiocb->aio_nbytes);
-- 
1.9.1

  parent reply	other threads:[~2015-01-30  8:42 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-30  8:42 [Qemu-devel] [PATCH v6 0/7] eliminate data write in bdrv_write_zeroes on Linux in raw-posix.c Denis V. Lunev
2015-01-30  8:42 ` [Qemu-devel] [PATCH 1/7] block/raw-posix: create translate_err helper to merge errno values Denis V. Lunev
2015-01-30  8:44   ` Peter Lieven
2015-01-30  8:42 ` [Qemu-devel] [PATCH 2/7] block/raw-posix: create do_fallocate helper Denis V. Lunev
2015-01-30  8:47   ` Peter Lieven
2015-01-30  8:49     ` Denis V. Lunev
2015-01-30  8:50       ` Peter Lieven
2015-01-30  8:42 ` [Qemu-devel] [PATCH 3/7] block/raw-posix: refactor handle_aiocb_write_zeroes a bit Denis V. Lunev
2015-01-30  8:42 ` [Qemu-devel] [PATCH 4/7] block: use fallocate(FALLOC_FL_ZERO_RANGE) in handle_aiocb_write_zeroes Denis V. Lunev
2015-01-30  8:42 ` [Qemu-devel] [PATCH 5/7] block/raw-posix: call plain fallocate " Denis V. Lunev
2015-01-30 14:58   ` Max Reitz
2015-01-30 15:41     ` Denis V. Lunev
2015-01-30 15:42       ` Max Reitz
2015-01-30 15:53         ` Denis V. Lunev
2015-01-30  8:42 ` Denis V. Lunev [this message]
2015-01-30 15:02   ` [Qemu-devel] [PATCH 6/7] block: use fallocate(FALLOC_FL_PUNCH_HOLE) & fallocate(0) to write zeroes Max Reitz
2015-01-30  8:42 ` [Qemu-devel] [PATCH 7/7] block/raw-posix: set max_write_zeroes to INT_MAX for regular files Denis V. Lunev
2015-02-02 13:23   ` Kevin Wolf
2015-02-02 13:55     ` Peter Lieven
2015-02-02 14:04       ` Kevin Wolf
2015-02-02 14:12         ` Peter Lieven
2015-02-02 14:16           ` Kevin Wolf
2015-02-02 14:20             ` Peter Lieven
2015-02-02 14:38               ` Denis V. Lunev
2015-02-02 14:49               ` Kevin Wolf
2015-02-02 15:30                 ` Denis V. Lunev
2015-02-02 13:26 ` [Qemu-devel] [PATCH v6 0/7] eliminate data write in bdrv_write_zeroes on Linux in raw-posix.c Kevin Wolf
  -- strict thread matches above, loose matches on Subject: below --
2014-12-26 12:35 [Qemu-devel] [PATCH v2 0/7] eliminate data write in bdrv_write_zeroes on Linux Denis V. Lunev
2014-12-26 12:35 ` [Qemu-devel] [PATCH 6/7] block: use fallocate(FALLOC_FL_PUNCH_HOLE) & fallocate(0) to write zeroes Denis V. Lunev

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=1422607337-25335-7-git-send-email-den@openvz.org \
    --to=den@openvz.org \
    --cc=famz@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=pl@kamp.de \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /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 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).