[PATCH 16/21] block: Mark bdrv_parent_cb_change_media() GRAPH_RDLOCK

Kevin Wolf posted 21 patches 2 years, 5 months ago
Only 11 patches received!
There is a newer version of this series
[PATCH 16/21] block: Mark bdrv_parent_cb_change_media() GRAPH_RDLOCK
Posted by Kevin Wolf 2 years, 5 months ago
The function reads the parents list, so it needs to hold the graph lock.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/block.c b/block.c
index 0a28bfc3c4..7df8780d6e 100644
--- a/block.c
+++ b/block.c
@@ -3369,7 +3369,8 @@ void bdrv_unref_child(BlockDriverState *parent, BdrvChild *child)
 }
 
 
-static void bdrv_parent_cb_change_media(BlockDriverState *bs, bool load)
+static void GRAPH_RDLOCK
+bdrv_parent_cb_change_media(BlockDriverState *bs, bool load)
 {
     BdrvChild *c;
     GLOBAL_STATE_CODE();
@@ -3967,6 +3968,9 @@ bdrv_open_inherit(const char *filename, const char *reference, QDict *options,
     GLOBAL_STATE_CODE();
     assert(!qemu_in_coroutine());
 
+    /* TODO We'll eventually have to take a writer lock in this function */
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
+
     if (reference) {
         bool options_non_empty = options ? qdict_size(options) : false;
         qobject_unref(options);
-- 
2.41.0
Re: [PATCH 16/21] block: Mark bdrv_parent_cb_change_media() GRAPH_RDLOCK
Posted by Stefan Hajnoczi 2 years, 5 months ago
On Thu, Aug 17, 2023 at 02:50:15PM +0200, Kevin Wolf wrote:
> The function reads the parents list, so it needs to hold the graph lock.
> 
> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> ---
>  block.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Re: [PATCH 16/21] block: Mark bdrv_parent_cb_change_media() GRAPH_RDLOCK
Posted by Emanuele Giuseppe Esposito 2 years, 5 months ago

Am 17/08/2023 um 14:50 schrieb Kevin Wolf:
> The function reads the parents list, so it needs to hold the graph lock.
> 
> Signed-off-by: Kevin Wolf <kwolf@redhat.com>

Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>