qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Michael Roth <mdroth@linux.vnet.ibm.com>
To: qemu-devel@nongnu.org
Cc: Fam Zheng <famz@redhat.com>,
	qemu-stable@nongnu.org, Stefan Hajnoczi <stefanha@redhat.com>
Subject: [Qemu-devel] [PATCH 31/53] mirror: Do zero write on target if sectors not allocated
Date: Thu, 30 Jul 2015 06:32:46 -0500	[thread overview]
Message-ID: <1438255988-10418-32-git-send-email-mdroth@linux.vnet.ibm.com> (raw)
In-Reply-To: <1438255988-10418-1-git-send-email-mdroth@linux.vnet.ibm.com>

From: Fam Zheng <famz@redhat.com>

If guest discards a source cluster, mirroring with bdrv_aio_readv is overkill.
Some protocols do zero upon discard, where it's best to use
bdrv_aio_write_zeroes, otherwise, bdrv_aio_discard will be enough.

Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit dcfb3beb5130694b76b57de109619fcbf9c7e5b5)
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
---
 block/mirror.c | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/block/mirror.c b/block/mirror.c
index 1814523..bd079a4 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -168,6 +168,8 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
     int64_t end, sector_num, next_chunk, next_sector, hbitmap_next_sector;
     uint64_t delay_ns = 0;
     MirrorOp *op;
+    int pnum;
+    int64_t ret;
 
     s->sector_num = hbitmap_iter_next(&s->hbi);
     if (s->sector_num < 0) {
@@ -296,8 +298,22 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
     s->in_flight++;
     s->sectors_in_flight += nb_sectors;
     trace_mirror_one_iteration(s, sector_num, nb_sectors);
-    bdrv_aio_readv(source, sector_num, &op->qiov, nb_sectors,
-                   mirror_read_complete, op);
+
+    ret = bdrv_get_block_status_above(source, NULL, sector_num,
+                                      nb_sectors, &pnum);
+    if (ret < 0 || pnum < nb_sectors ||
+            (ret & BDRV_BLOCK_DATA && !(ret & BDRV_BLOCK_ZERO))) {
+        bdrv_aio_readv(source, sector_num, &op->qiov, nb_sectors,
+                       mirror_read_complete, op);
+    } else if (ret & BDRV_BLOCK_ZERO) {
+        bdrv_aio_write_zeroes(s->target, sector_num, op->nb_sectors,
+                              s->unmap ? BDRV_REQ_MAY_UNMAP : 0,
+                              mirror_write_complete, op);
+    } else {
+        assert(!(ret & BDRV_BLOCK_DATA));
+        bdrv_aio_discard(s->target, sector_num, op->nb_sectors,
+                         mirror_write_complete, op);
+    }
     return delay_ns;
 }
 
-- 
1.9.1

  parent reply	other threads:[~2015-07-30 11:35 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-30 11:32 [Qemu-devel] Patch Round-up for stable 2.3.1, freeze on 2015-08-06 Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 01/53] bt-sdp: fix broken uuids power-of-2 calculation Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 02/53] block/iscsi: do not forget to logout from target Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 03/53] Strip brackets from vnc host Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 04/53] nbd/trivial: fix type cast for ioctl Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 05/53] vmdk: Fix next_cluster_sector for compressed write Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 06/53] vmdk: Fix overflow if l1_size is 0x20000000 Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 07/53] qcow2: Flush pending discards before allocating cluster Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 08/53] usb: fix usb-net segfault Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 09/53] virtio-net: fix the upper bound when trying to delete queues Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 10/53] target-arm: Avoid buffer overrun on UNPREDICTABLE ldrd/strd Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 11/53] fdc: force the fifo access to be in bounds of the allocated buffer Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 12/53] Revert "block: Fix unaligned zero write" Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 13/53] block: Fix NULL deference for unaligned write if qiov is NULL Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 14/53] qemu-iotests: Test unaligned sub-block zero write Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 15/53] hw/acpi/aml-build: Fix memory leak Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 16/53] qga/commands-posix: Fix bug in guest-fstrim Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 17/53] kbd: add brazil kbd keys to qemu Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 18/53] kbd: add brazil kbd keys to x11 evdev map Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 19/53] qcow2: Set MIN_L2_CACHE_SIZE to 2 Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 20/53] iotests: qcow2 COW with minimal L2 cache size Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 21/53] vmdk: Fix index_in_cluster calculation in vmdk_co_get_block_status Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 22/53] vmdk: Use vmdk_find_index_in_cluster everywhere Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 23/53] sdl2: fix crash in handle_windowevent() when restoring the screen size Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 24/53] spice-display: fix segfault in qemu_spice_create_update Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 25/53] i8254: fix out-of-bounds memory access in pit_ioport_read() Michael Roth
2015-08-03  8:40   ` [Qemu-devel] 答复: " lidonglin
2015-08-03 11:46     ` Paolo Bonzini
2015-07-30 11:32 ` [Qemu-devel] [PATCH 26/53] hw/core: rebase sysbus_get_fw_dev_path() to g_strdup_printf() Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 27/53] vhost: correctly pass error to caller in vhost_dev_enable_notifiers() Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 28/53] virtio-ccw: complete handling of guest-initiated resets Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 29/53] block: Add bdrv_get_block_status_above Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 30/53] qmp: Add optional bool "unmap" to drive-mirror Michael Roth
2015-07-30 11:32 ` Michael Roth [this message]
2015-07-30 11:32 ` [Qemu-devel] [PATCH 32/53] block: Fix dirty bitmap in bdrv_co_discard Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 33/53] qemu-iotests: Make block job methods common Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 34/53] qemu-iotests: Add test case for mirror with unmap Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 35/53] iotests: Use event_wait in wait_ready Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 36/53] iotests: add QMP event waiting queue Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 37/53] block/nfs: limit maximum readahead size to 1MB Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 38/53] s390x/ipl: Fix boot if no bootindex was specified Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 39/53] spapr_vty: lookup should only return valid VTY objects Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 40/53] target-ppc: fix hugepage support when using memory-backend-file Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 41/53] Fix irq route entries exceeding KVM_MAX_IRQ_ROUTES Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 42/53] block: Initialize local_err in bdrv_append_temp_snapshot Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 43/53] mips/kvm: Fix Big endian 32-bit register access Michael Roth
2015-07-30 11:32 ` [Qemu-devel] [PATCH 44/53] mips/kvm: Sign extend registers written to KVM Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 45/53] vfio/pci: Fix RTL8168 NIC quirks Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 46/53] virtio-net: unbreak any layout Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 47/53] vfio/pci: Fix bootindex Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 48/53] scsi: fix buffer overflow in scsi_req_parse_cdb (CVE-2015-5158) Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 49/53] block: vpc - prevent overflow if max_table_entries >= 0x40000000 Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 50/53] block: qemu-iotests - add check for multiplication overflow in vpc Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 51/53] ide: Check array bounds before writing to io_buffer (CVE-2015-5154) Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 52/53] ide/atapi: Fix START STOP UNIT command completion Michael Roth
2015-07-30 11:33 ` [Qemu-devel] [PATCH 53/53] ide: Clear DRQ after handling all expected accesses Michael Roth
2015-08-04 17:41 ` [Qemu-devel] Patch Round-up for stable 2.3.1, freeze on 2015-08-06 Michael Roth

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=1438255988-10418-32-git-send-email-mdroth@linux.vnet.ibm.com \
    --to=mdroth@linux.vnet.ibm.com \
    --cc=famz@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-stable@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).