qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Lieven <pl@kamp.de>
To: qemu-devel@nongnu.org
Cc: kwolf@redhat.com, pbonzini@redhat.com, Peter Lieven <pl@kamp.de>,
	stefanha@redhat.com
Subject: [Qemu-devel] [RFC PATCH] block: optimize zero writes with bdrv_write_zeroes
Date: Sat, 22 Feb 2014 14:00:22 +0100	[thread overview]
Message-ID: <1393074022-32388-1-git-send-email-pl@kamp.de> (raw)

this patch tries to optimize zero write requests
by automatically using bdrv_write_zeroes if it is
supported by the format.

i know that there is a lot of potential for discussion, but i would
like to know what the others think.

this should significantly speed up file system initialization and
should speed zero write test used to test backend storage performance.

the difference can simply be tested by e.g.

dd if=/dev/zero of=/dev/vdX bs=1M

Signed-off-by: Peter Lieven <pl@kamp.de>
---
 block.c               |    8 ++++++++
 include/qemu-common.h |    1 +
 util/iov.c            |   20 ++++++++++++++++++++
 3 files changed, 29 insertions(+)

diff --git a/block.c b/block.c
index 6f4baca..505888e 100644
--- a/block.c
+++ b/block.c
@@ -3145,6 +3145,14 @@ static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
 
     ret = notifier_with_return_list_notify(&bs->before_write_notifiers, req);
 
+    if (!ret && !(flags & BDRV_REQ_ZERO_WRITE) &&
+        drv->bdrv_co_write_zeroes && qemu_iovec_is_zero(qiov)) {
+        flags |= BDRV_REQ_ZERO_WRITE;
+        /* if the device was not opened with discard=on the below flag
+         * is immediately cleared again in bdrv_co_do_write_zeroes */
+        flags |= BDRV_REQ_MAY_UNMAP;
+    }
+
     if (ret < 0) {
         /* Do nothing, write notifier decided to fail this request */
     } else if (flags & BDRV_REQ_ZERO_WRITE) {
diff --git a/include/qemu-common.h b/include/qemu-common.h
index b0e34b2..f0ad0f9 100644
--- a/include/qemu-common.h
+++ b/include/qemu-common.h
@@ -330,6 +330,7 @@ void qemu_iovec_concat(QEMUIOVector *dst,
 void qemu_iovec_concat_iov(QEMUIOVector *dst,
                            struct iovec *src_iov, unsigned int src_cnt,
                            size_t soffset, size_t sbytes);
+bool qemu_iovec_is_zero(QEMUIOVector *qiov);
 void qemu_iovec_destroy(QEMUIOVector *qiov);
 void qemu_iovec_reset(QEMUIOVector *qiov);
 size_t qemu_iovec_to_buf(QEMUIOVector *qiov, size_t offset,
diff --git a/util/iov.c b/util/iov.c
index bb46c04..abbb374 100644
--- a/util/iov.c
+++ b/util/iov.c
@@ -342,6 +342,26 @@ void qemu_iovec_concat(QEMUIOVector *dst,
     qemu_iovec_concat_iov(dst, src->iov, src->niov, soffset, sbytes);
 }
 
+/*
+ *  check if the contents of the iovecs is all zero 
+ */
+bool qemu_iovec_is_zero(QEMUIOVector *qiov) {
+    int i;
+    for (i = 0; i < qiov->niov; i++) {
+        size_t offs = qiov->iov[i].iov_len & ~(4 * sizeof(long) - 1);
+        uint8_t *ptr = qiov->iov[i].iov_base;
+        if (offs && !buffer_is_zero(qiov->iov[i].iov_base, offs)) {
+            return false;
+        }
+        for (; offs < qiov->iov[i].iov_len; offs++) {
+             if (ptr[offs]) {
+                 return false;
+             }
+        }
+    }
+    return true;
+}
+
 void qemu_iovec_destroy(QEMUIOVector *qiov)
 {
     assert(qiov->nalloc != -1);
-- 
1.7.9.5

             reply	other threads:[~2014-02-22 13:01 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-22 13:00 Peter Lieven [this message]
2014-02-22 16:45 ` [Qemu-devel] [RFC PATCH] block: optimize zero writes with bdrv_write_zeroes Fam Zheng
2014-02-23 19:10   ` Peter Lieven
2014-02-24  1:01     ` Fam Zheng
2014-02-24 10:39       ` Paolo Bonzini
2014-02-24 11:33         ` Fam Zheng
2014-02-24 11:51           ` Paolo Bonzini
2014-02-24 12:04             ` Fam Zheng
2014-02-24 12:07             ` Kevin Wolf
2014-02-24 12:10               ` Paolo Bonzini
2014-02-24 12:22                 ` Kevin Wolf
2014-02-24 10:11 ` Kevin Wolf
2014-02-24 10:26   ` Peter Lieven
2014-02-24 10:38     ` Paolo Bonzini
2014-02-24 11:50       ` Peter Lieven
2014-02-24 13:01       ` Peter Lieven
2014-02-25 13:41         ` Kevin Wolf
2014-02-25 17:03           ` Peter Lieven

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=1393074022-32388-1-git-send-email-pl@kamp.de \
    --to=pl@kamp.de \
    --cc=kwolf@redhat.com \
    --cc=pbonzini@redhat.com \
    --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).