[Stable-9.1.3 48/58] multifd: bugfix for migration using compression methods

[Michael Tokarev]

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 a6db05502a..0b04db9f3a 100644
--- 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;
 }
-- 
2.39.5