Commit 6e5d97d01d9da6f295f9888d4b34e29fd737861a

Authored by Nolan
Committed by Anthony Liguori
1 parent bd836776

Handle BH's queued by AIO completions in qemu_aio_flush()

Without this, the call to qemu_aio_flush during migration doesn't
actually flush all in-flight SCSI IOs.

Signed-off-by: Nolan Leake <nolan <at> sigbus.net>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Showing 1 changed file with 1 additions and 1 deletions
@@ -112,7 +112,7 @@ void qemu_aio_flush(void) @@ -112,7 +112,7 @@ void qemu_aio_flush(void)
112 LIST_FOREACH(node, &aio_handlers, node) { 112 LIST_FOREACH(node, &aio_handlers, node) {
113 ret |= node->io_flush(node->opaque); 113 ret |= node->io_flush(node->opaque);
114 } 114 }
115 - } while (ret > 0); 115 + } while (qemu_bh_poll() || ret > 0);
116 } 116 }
117 117
118 void qemu_aio_wait(void) 118 void qemu_aio_wait(void)