qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com
Subject: [Qemu-devel] [PULL 08/42] block: use fallocate(FALLOC_FL_PUNCH_HOLE) & fallocate(0) to write zeroes
Date: Fri,  6 Feb 2015 17:40:15 +0100	[thread overview]
Message-ID: <1423240849-15499-9-git-send-email-kwolf@redhat.com> (raw)
In-Reply-To: <1423240849-15499-1-git-send-email-kwolf@redhat.com>

From: "Denis V. Lunev" <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.

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>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@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);
-- 
1.8.3.1

  parent reply	other threads:[~2015-02-06 16:41 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-06 16:40 [Qemu-devel] [PULL 00/42] Block patches Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 01/42] Restore atapi_dma flag across migration Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 02/42] atapi migration: Throw recoverable error to avoid recovery Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 03/42] block/raw-posix: create translate_err helper to merge errno values Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 04/42] block/raw-posix: create do_fallocate helper Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 05/42] block/raw-posix: refactor handle_aiocb_write_zeroes a bit Kevin Wolf
2015-02-12  2:29   ` Peter Maydell
2015-02-12  4:44     ` Denis V. Lunev
2015-02-06 16:40 ` [Qemu-devel] [PULL 06/42] block: use fallocate(FALLOC_FL_ZERO_RANGE) in handle_aiocb_write_zeroes Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 07/42] block/raw-posix: call plain fallocate " Kevin Wolf
2015-02-06 16:40 ` Kevin Wolf [this message]
2015-02-06 16:40 ` [Qemu-devel] [PULL 09/42] block: change default for discard and write zeroes to INT_MAX Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 10/42] qemu-img: Add QEMU_PKGVERSION to QEMU_IMG_VERSION Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 11/42] qed: Really remove unused field QEDAIOCB.finished Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 12/42] block: add accounting for merged requests Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 13/42] hw/virtio-blk: add a constant for max number of " Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 14/42] block-backend: expose bs->bl.max_transfer_length Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 15/42] virtio-blk: introduce multiread Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 16/42] virtio-blk: add a knob to disable request merging Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 17/42] qemu-iotests: Fix supported_oses check Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 18/42] iotests: Specify format for qemu-nbd Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 19/42] block: add event when disk usage exceeds threshold Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 20/42] block/dmg: properly detect the UDIF trailer Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 21/42] block/dmg: extract mish block decoding functionality Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 22/42] block/dmg: extract processing of resource forks Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 23/42] block/dmg: process a buffer instead of reading ints Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 24/42] block/dmg: validate chunk size to avoid overflow Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 25/42] block/dmg: process XML plists Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 26/42] block/dmg: set virtual size to a non-zero value Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 27/42] block/dmg: fix sector data offset calculation Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 28/42] block/dmg: use SectorNumber from BLKX header Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 29/42] block/dmg: factor out block type check Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 30/42] block/dmg: support bzip2 block entry types Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 31/42] block/dmg: improve zeroes handling Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 32/42] qed: check for header size overflow Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 33/42] qemu-iotests: add 116 invalid QED input file tests Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 34/42] block: fix off-by-one error in qcow and qcow2 Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 35/42] iotests: Fix 083 Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 36/42] iotests: Fix 100 for nbd Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 37/42] iotests: Fix 104 for NBD Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 38/42] nbd: Improve error messages Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 39/42] block: introduce BDRV_REQUEST_MAX_SECTORS Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 40/42] nbd: fix max_discard/max_transfer_length Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 41/42] block: Give always priority to unused entries in the qcow2 L2 cache Kevin Wolf
2015-02-06 16:40 ` [Qemu-devel] [PULL 42/42] qcow2: Rewrite qcow2_alloc_bytes() Kevin Wolf

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=1423240849-15499-9-git-send-email-kwolf@redhat.com \
    --to=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.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 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).