qemu-kvm/SOURCES/kvm-block-temporarily-hold-the-new-AioContext-of-bs_top-.patch

From b952c8f1da6f8597736c0e040565830139369359 Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Tue, 14 Feb 2023 18:16:21 +0100
Subject: [PATCH] block: temporarily hold the new AioContext of bs_top in
 bdrv_append()

RH-Author: Stefano Garzarella <sgarzare@redhat.com>
RH-MergeRequest: 153: block: temporarily hold the new AioContext of bs_top in bdrv_append()
RH-Bugzilla: 2168209
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Commit: [1/1] 5b190426d996e8c9f7a781bd97aee8d25756dbd3 (sgarzarella/qemu-kvm-c-9-s)

bdrv_append() is called with bs_top AioContext held, but
bdrv_attach_child_noperm() could change the AioContext of bs_top.

bdrv_replace_node_noperm() calls bdrv_drained_begin() starting from
commit 2398747128 ("block: Don't poll in bdrv_replace_child_noperm()").
bdrv_drained_begin() can call BDRV_POLL_WHILE that assumes the new lock
is taken, so let's temporarily hold the new AioContext to prevent QEMU
from failing in BDRV_POLL_WHILE when it tries to release the wrong
AioContext.

Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2168209
Reported-by: Aihua Liang <aliang@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20230214171621.11574-1-sgarzare@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 60d90bf43c169b9d1dbcb17ed794b7b02c6862b1)
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
 block.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/block.c b/block.c
index 0d78711416..9e1dcb9e47 100644
--- a/block.c
+++ b/block.c
@@ -5275,6 +5275,8 @@ int bdrv_drop_filter(BlockDriverState *bs, Error **errp)
  * child.
  *
  * This function does not create any image files.
+ *
+ * The caller must hold the AioContext lock for @bs_top.
  */
 int bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top,
                 Error **errp)
@@ -5282,11 +5284,14 @@ int bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top,
     int ret;
     BdrvChild *child;
     Transaction *tran = tran_new();
+    AioContext *old_context, *new_context = NULL;
 
     GLOBAL_STATE_CODE();
 
     assert(!bs_new->backing);
 
+    old_context = bdrv_get_aio_context(bs_top);
+
     child = bdrv_attach_child_noperm(bs_new, bs_top, "backing",
                                      &child_of_bds, bdrv_backing_role(bs_new),
                                      tran, errp);
@@ -5295,6 +5300,19 @@ int bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top,
         goto out;
     }
 
+    /*
+     * bdrv_attach_child_noperm could change the AioContext of bs_top.
+     * bdrv_replace_node_noperm calls bdrv_drained_begin, so let's temporarily
+     * hold the new AioContext, since bdrv_drained_begin calls BDRV_POLL_WHILE
+     * that assumes the new lock is taken.
+     */
+    new_context = bdrv_get_aio_context(bs_top);
+
+    if (old_context != new_context) {
+        aio_context_release(old_context);
+        aio_context_acquire(new_context);
+    }
+
     ret = bdrv_replace_node_noperm(bs_top, bs_new, true, tran, errp);
     if (ret < 0) {
         goto out;
@@ -5306,6 +5324,11 @@ out:
 
     bdrv_refresh_limits(bs_top, NULL, NULL);
 
+    if (new_context && old_context != new_context) {
+        aio_context_release(new_context);
+        aio_context_acquire(old_context);
+    }
+
     return ret;
 }
 
-- 
2.31.1