multifd: bugfix for migration using compression methods

When compression is enabled on the migration channel and
the pages processed are all zero pages, these pages will
not be sent and updated on the target side, resulting in
incorrect memory data on the source and target sides.

The root cause is that all compression methods call
multifd_send_prepare_common to determine whether to compress
dirty pages, but multifd_send_prepare_common does not update
the IOV of MultiFDPacket_t when all dirty pages are zero pages.

The solution is to always update the IOV of MultiFDPacket_t
regardless of whether the dirty pages are all zero pages.

Fixes: 303e6f54f9 ("migration/multifd: Implement zero page transmission on the multifd thread.")
Cc: qemu-stable@nongnu.org #9.0+
Signed-off-by: Yuan Liu <yuan1.liu@intel.com>
Reviewed-by: Jason Zeng <jason.zeng@intel.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20241218091413.140396-2-yuan1.liu@intel.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
(cherry picked from commit cdc3970f8597ebdc1a4c2090cfb4d11e297329ed)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
(Mjt: in 9.1 this code is in migration/multifd.c, not in migration/multifd-nocomp.c)

diff --git a/migration/multifd.c b/migration/multifd.c
index a6db05502aa7d44f88b93a30e9bd6701dbdab43e..0b04db9f3a42733fb65c00344a7c8fe3a41e2399 100644 (file)
--- a/migration/multifd.c
+++ b/migration/multifd.c
@@ -1690,6 +1690,7 @@ void multifd_recv_new_channel(QIOChannel *ioc, Error **errp)
 
 bool multifd_send_prepare_common(MultiFDSendParams *p)
 {
+    multifd_send_prepare_header(p);
     multifd_send_zero_page_detect(p);
 
     if (!p->pages->normal_num) {
@@ -1697,7 +1698,5 @@ bool multifd_send_prepare_common(MultiFDSendParams *p)
         return false;
     }
 
-    multifd_send_prepare_header(p);
-
     return true;
 }