All of lore.kernel.org
 help / color / mirror / Atom feed
From: Max Reitz <mreitz@redhat.com>
To: qemu-block@nongnu.org
Cc: qemu-devel@nongnu.org, Max Reitz <mreitz@redhat.com>,
	Peter Maydell <peter.maydell@linaro.org>
Subject: [Qemu-devel] [PULL 12/21] block: bdrv_invalidate_cache: invalidate children first
Date: Sun, 12 Feb 2017 02:38:35 +0100	[thread overview]
Message-ID: <20170212013844.6560-3-mreitz@redhat.com> (raw)
In-Reply-To: <20170212013440.5919-1-mreitz@redhat.com>

From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

Current implementation invalidates firstly parent bds and then its
children. This leads to the following bug:

after incoming migration, in bdrv_invalidate_cache_all:
1. invalidate parent bds - reopen it with BDRV_O_INACTIVE cleared
2. child is not yet invalidated
3. parent check that its BDRV_O_INACTIVE is cleared
4. parent writes to child
5. assert in bdrv_co_pwritev, as BDRV_O_INACTIVE is set for child

This patch fixes it by just changing invalidate sequence: invalidate
children first.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20170131112308.54189-1-vsementsov@virtuozzo.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/block.c b/block.c
index b8bc2a1c68..743c349100 100644
--- a/block.c
+++ b/block.c
@@ -3248,19 +3248,18 @@ void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
     if (!(bs->open_flags & BDRV_O_INACTIVE)) {
         return;
     }
-    bs->open_flags &= ~BDRV_O_INACTIVE;
 
-    if (bs->drv->bdrv_invalidate_cache) {
-        bs->drv->bdrv_invalidate_cache(bs, &local_err);
+    QLIST_FOREACH(child, &bs->children, next) {
+        bdrv_invalidate_cache(child->bs, &local_err);
         if (local_err) {
-            bs->open_flags |= BDRV_O_INACTIVE;
             error_propagate(errp, local_err);
             return;
         }
     }
 
-    QLIST_FOREACH(child, &bs->children, next) {
-        bdrv_invalidate_cache(child->bs, &local_err);
+    bs->open_flags &= ~BDRV_O_INACTIVE;
+    if (bs->drv->bdrv_invalidate_cache) {
+        bs->drv->bdrv_invalidate_cache(bs, &local_err);
         if (local_err) {
             bs->open_flags |= BDRV_O_INACTIVE;
             error_propagate(errp, local_err);
-- 
2.11.0

  parent reply	other threads:[~2017-02-12  1:39 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-12  1:34 [Qemu-devel] [PULL 00/21] Block patches Max Reitz
2017-02-12  1:34 ` [Qemu-devel] [PULL 01/21] iotests: Fix a problem in common.filter Max Reitz
2017-02-12  1:34 ` [Qemu-devel] [PULL 02/21] qemu-img: Improve commit invalid base message Max Reitz
2017-02-12  1:36 ` [Qemu-devel] [PULL 03/21] qapi: Tweak error message of bdrv_query_image_info Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 04/21] iotests: Fix reference output for 059 Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 05/21] iotests: record separate timings per format, protocol pair Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 06/21] block/vmdk: Fix the endian problem of buf_len and lba Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 07/21] block: check full backing filename when searching protocol filenames Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 08/21] qemu-iotests: Don't create fifos / pidfiles with protocol paths Max Reitz
2017-02-12  1:36   ` [Qemu-devel] [PULL 09/21] qemu-iotest: test to lookup protocol-based image with relative backing Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 10/21] block/qapi: reduce the coupling between the bdrv_query_stats and bdrv_query_bds_stats Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 11/21] block/qapi: reduce the execution time of qmp_query_blockstats Max Reitz
2017-02-12  1:38 ` Max Reitz [this message]
2017-02-12  1:38 ` [Qemu-devel] [PULL 13/21] block/nfs: fix NULL pointer dereference in URI parsing Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 14/21] block/nfs: fix naming of runtime opts Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 15/21] qemu-io: Return non-zero exit code on failure Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 16/21] qemu-iotests: Add _unsupported_fmt helper Max Reitz
2017-02-12  1:38 ` [Qemu-devel] [PULL 17/21] qemu-io: Add failure regression tests Max Reitz
2017-02-12  1:39 ` [Qemu-devel] [PULL 18/21] qcow2: Optimize the refcount-block overlap check Max Reitz
2017-02-12  1:39 ` [Qemu-devel] [PULL 19/21] qemu-io: don't allow I/O operations larger than BDRV_REQUEST_MAX_BYTES Max Reitz
2017-02-12  1:39 ` [Qemu-devel] [PULL 20/21] qemu-img: Use qemu_strtoul() rather than raw strtoul() Max Reitz
2017-02-12  1:39 ` [Qemu-devel] [PULL 21/21] qemu-img: Avoid setting ret to unused value in img_convert() Max Reitz
2017-02-13 10:54 ` [Qemu-devel] [PULL 00/21] Block patches Peter Maydell

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=20170212013844.6560-3-mreitz@redhat.com \
    --to=mreitz@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-block@nongnu.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.