Skip to contentSkip to conversation

block: Fix invalidate_cache error path for parent activation

PATCH In review Kevin Wolf 3 participants 3 messages Jan 31, 2019 – Feb 1, 2019

lore
Share & tools
lore

Kevin Wolf

author
#0 lore raw

PATCH 1/1 block: Fix invalidate_cache error path for parent activation

bdrv_co_invalidate_cache() clears the BDRV_O_INACTIVE flag before actually activating a node so that the correct permissions etc. are taken. In case of errors, the flag must be restored so that the next call to bdrv_co_invalidate_cache() retries activation.

Restoring the flag was missing in the error path for a failed parent->role->activate() call. The consequence is that this attempt to activate all images correctly fails because we still set errp, however on the next attempt BDRV_O_INACTIVE is already clear, so we return success without actually retrying the failed action.

An example where this is observable in practice is migration to a QEMU instance that has a raw format block node attached to a guest device with share-rw=off (the default) while another process holds BLK_PERM_WRITE for the same image. In this case, all activation steps before parent->role->activate() succeed because raw can tolerate other writers to the image. Only the parent callback (in particular blk_root_activate()) tries to implement the share-rw=on property and requests exclusive write permissions. This fails when the migration completes and correctly displays an error. However, a manual 'cont' will incorrectly resume the VM without calling blk_root_activate() again.

This case is described in more detail in the following bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=1531888

Fix this by correctly restoring the BDRV_O_INACTIVE flag in the error path.

---
 block.c | 1 +
 1 file changed, 1 insertion(+)
block.c +1 -0
index 0eba8ebe5c..7f5c9bb02b 100644
@@ -4549,6 +4549,7 @@ static void coroutine_fn bdrv_co_invalidate_cache(BlockDriverState *bs,
         if (parent->role->activate) {
             parent->role->activate(parent, &local_err);
             if (local_err) {
+                bs->open_flags |= BDRV_O_INACTIVE;
                 error_propagate(errp, local_err);
                 return;
             }
Signature · 1 line
2.20.1
Reviewed-by on v1 1/1
#1 lore raw
Quoted 25 lines · Kevin Wolf

bdrv_co_invalidate_cache() clears the BDRV_O_INACTIVE flag before actually activating a node so that the correct permissions etc. are taken. In case of errors, the flag must be restored so that the next call to bdrv_co_invalidate_cache() retries activation.

Restoring the flag was missing in the error path for a failed parent->role->activate() call. The consequence is that this attempt to activate all images correctly fails because we still set errp, however on the next attempt BDRV_O_INACTIVE is already clear, so we return success without actually retrying the failed action.

An example where this is observable in practice is migration to a QEMU instance that has a raw format block node attached to a guest device with share-rw=off (the default) while another process holds BLK_PERM_WRITE for the same image. In this case, all activation steps before parent->role->activate() succeed because raw can tolerate other writers to the image. Only the parent callback (in particular blk_root_activate()) tries to implement the share-rw=on property and requests exclusive write permissions. This fails when the migration completes and correctly displays an error. However, a manual 'cont' will incorrectly resume the VM without calling blk_root_activate() again.

This case is described in more detail in the following bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=1531888

Quoted 8 lines

Fix this by correctly restoring the BDRV_O_INACTIVE flag in the error path.

---
 block.c | 1 +
 1 file changed, 1 insertion(+)
Tested-by on v1 1/1
#2 lore raw
In reply to Kevin Wolf
Quoted 39 lines · Kevin Wolf

bdrv_co_invalidate_cache() clears the BDRV_O_INACTIVE flag before actually activating a node so that the correct permissions etc. are taken. In case of errors, the flag must be restored so that the next call to bdrv_co_invalidate_cache() retries activation.

Restoring the flag was missing in the error path for a failed parent->role->activate() call. The consequence is that this attempt to activate all images correctly fails because we still set errp, however on the next attempt BDRV_O_INACTIVE is already clear, so we return success without actually retrying the failed action.

An example where this is observable in practice is migration to a QEMU instance that has a raw format block node attached to a guest device with share-rw=off (the default) while another process holds BLK_PERM_WRITE for the same image. In this case, all activation steps before parent->role->activate() succeed because raw can tolerate other writers to the image. Only the parent callback (in particular blk_root_activate()) tries to implement the share-rw=on property and requests exclusive write permissions. This fails when the migration completes and correctly displays an error. However, a manual 'cont' will incorrectly resume the VM without calling blk_root_activate() again.

This case is described in more detail in the following bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=1531888

Fix this by correctly restoring the BDRV_O_INACTIVE flag in the error path.

---
 block.c | 1 +
 1 file changed, 1 insertion(+)
block.c +0 -0
index 0eba8ebe5c..7f5c9bb02b 100644

block.c static void coroutine_fn bdrv_co_invalidate_cache(BlockDriverState *bs, · line 4555

Quoted 7 lines
block.c +1 -0
@@ -4549,6 +4549,7 @@ static void coroutine_fn bdrv_co_invalidate_cache(BlockDriverState *bs,
         if (parent->role->activate) {
             parent->role->activate(parent, &local_err);
             if (local_err) {
+                bs->open_flags |= BDRV_O_INACTIVE;
                 error_propagate(errp, local_err);
                 return;
             }