OpenE2K/qemu-e2k
13
4
Fork 0
Code Issues 4 Pull Requests Projects 2 Releases Activity

migration: Stop migration immediately in RDMA error paths

Browse Source 
In multiple places, RDMA errors are handled in a strange way, where it only
sets qemu_file_set_error() but not stop the migration immediately.

It's not obvious what will happen later if there is already an error.  Make
all such failures stop migration immediately.

Cc: Zhijian Li (Fujitsu) <lizhijian@fujitsu.com>
Cc: Markus Armbruster <armbru@redhat.com>
Cc: Juan Quintela <quintela@redhat.com>
Cc: Fabiano Rosas <farosas@suse.de>
Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20231024163933.516546-1-peterx@redhat.com>
This commit is contained in:
Peter Xu 2023-10-24 12:39:33 -04:00 committed by Juan Quintela
parent 864128df46
commit a2326705e5
1 changed files with 10 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
migration/ram.c
View File

@ -2973,11 +2973,13 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
ret = rdma_registration_start(f, RAM_CONTROL_SETUP); ret = rdma_registration_start(f, RAM_CONTROL_SETUP);
if (ret < 0) { if (ret < 0) {
qemu_file_set_error(f, ret); qemu_file_set_error(f, ret);
return ret;
} }
ret = rdma_registration_stop(f, RAM_CONTROL_SETUP); ret = rdma_registration_stop(f, RAM_CONTROL_SETUP);
if (ret < 0) { if (ret < 0) {
qemu_file_set_error(f, ret); qemu_file_set_error(f, ret);
return ret;
} }
migration_ops = g_malloc0(sizeof(MigrationOps)); migration_ops = g_malloc0(sizeof(MigrationOps));
@ -3043,6 +3045,7 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
ret = rdma_registration_start(f, RAM_CONTROL_ROUND); ret = rdma_registration_start(f, RAM_CONTROL_ROUND);
if (ret < 0) { if (ret < 0) {
qemu_file_set_error(f, ret); qemu_file_set_error(f, ret);
goto out;
} }
t0 = qemu_clock_get_ns(QEMU_CLOCK_REALTIME); t0 = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
@ -3147,8 +3150,6 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
rs->last_stage = !migration_in_colo_state(); rs->last_stage = !migration_in_colo_state();
WITH_RCU_READ_LOCK_GUARD() { WITH_RCU_READ_LOCK_GUARD() {
int rdma_reg_ret;
if (!migration_in_postcopy()) { if (!migration_in_postcopy()) {
migration_bitmap_sync_precopy(rs, true); migration_bitmap_sync_precopy(rs, true);
} }
@ -3156,6 +3157,7 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
ret = rdma_registration_start(f, RAM_CONTROL_FINISH); ret = rdma_registration_start(f, RAM_CONTROL_FINISH);
if (ret < 0) { if (ret < 0) {
qemu_file_set_error(f, ret); qemu_file_set_error(f, ret);
return ret;
} }
/* try transferring iterative blocks of memory */ /* try transferring iterative blocks of memory */
@ -3171,23 +3173,20 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
break; break;
} }
if (pages < 0) { if (pages < 0) {
ret = pages; qemu_mutex_unlock(&rs->bitmap_mutex);
break; return pages;
} }
} }
qemu_mutex_unlock(&rs->bitmap_mutex); qemu_mutex_unlock(&rs->bitmap_mutex);
compress_flush_data(); compress_flush_data();
rdma_reg_ret = rdma_registration_stop(f, RAM_CONTROL_FINISH); ret = rdma_registration_stop(f, RAM_CONTROL_FINISH);
if (rdma_reg_ret < 0) {
qemu_file_set_error(f, rdma_reg_ret);
}
}
if (ret < 0) { if (ret < 0) {
qemu_file_set_error(f, ret);
return ret; return ret;
} }
}
ret = multifd_send_sync_main(rs->pss[RAM_CHANNEL_PRECOPY].pss_channel); ret = multifd_send_sync_main(rs->pss[RAM_CHANNEL_PRECOPY].pss_channel);
if (ret < 0) { if (ret < 0) {