qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v2 3/3] dump-guest-memory: Block live migration


From: Marc-André Lureau
Subject: Re: [PATCH v2 3/3] dump-guest-memory: Block live migration
Date: Fri, 27 Aug 2021 11:51:04 +0400

Hi

On Fri, Aug 27, 2021 at 11:44 AM Marc-André Lureau <marcandre.lureau@redhat.com> wrote:
Hi

On Thu, Aug 26, 2021 at 10:58 PM Peter Xu <peterx@redhat.com> wrote:
Both dump-guest-memory and live migration caches vm state at the beginning.
Either of them entering the other one will cause race on the vm state, and even
more severe on that (please refer to the crash report in the bug link).

Let's block live migration in dump-guest-memory, and that'll also block
dump-guest-memory if it detected that we're during a live migration.

Side note: migrate_del_blocker() can be called even if the blocker is not
inserted yet, so it's safe to unconditionally delete that blocker in
dump_cleanup (g_slist_remove allows no-entry-found case).

Suggested-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1996609
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 dump/dump.c | 24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

diff --git a/dump/dump.c b/dump/dump.c
index ab625909f3..9c1c1fb738 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -29,6 +29,7 @@
 #include "qemu/error-report.h"
 #include "qemu/main-loop.h"
 #include "hw/misc/vmcoreinfo.h"
+#include "migration/blocker.h"

 #ifdef TARGET_X86_64
 #include "win_dump.h"
@@ -47,6 +48,8 @@

 #define MAX_GUEST_NOTE_SIZE (1 << 20) /* 1MB should be enough */

+static Error *dump_migration_blocker;
+
 #define ELF_NOTE_SIZE(hdr_size, name_size, desc_size)   \
     ((DIV_ROUND_UP((hdr_size), 4) +                     \
       DIV_ROUND_UP((name_size), 4) +                    \
@@ -101,6 +104,7 @@ static int dump_cleanup(DumpState *s)
             qemu_mutex_unlock_iothread();
         }
     }
+    migrate_del_blocker(dump_migration_blocker);

     return 0;
 }
@@ -1927,11 +1931,6 @@ void qmp_dump_guest_memory(bool paging, const char *file,
     Error *local_err = NULL;
     bool detach_p = false;

-    if (runstate_check(RUN_STATE_INMIGRATE)) {
-        error_setg(errp, "Dump not allowed during incoming migration.");
-        return;
-    }
-
     /* if there is a dump in background, we should wait until the dump
      * finished */
     if (dump_in_progress()) {
@@ -2005,6 +2004,21 @@ void qmp_dump_guest_memory(bool paging, const char *file,
         return;
     }

+    if (!dump_migration_blocker) {
+        error_setg(&dump_migration_blocker,
+                   "Live migration disabled: dump-guest-memory in progress");
+    }
+
+    /*
+     * Allows even for -only-migratable, but forbid migration during the
+     * process of dump guest memory.
+     */
+    if (migrate_add_blocker_internal(dump_migration_blocker, errp)) {
+        /* Remember to release the fd before passing it over to dump state */
+        close(fd);
+        return;
+    }

I would move it earlier.  Why not leave it at the beginning of the function as it was?


Ah I think it's because dump_cleanup() isn't called when returning earlier. But relying on a cleanup done outside of this function is not obvious. Either dump_cleanup() should be called from here, or the blocker code should be moved to dump_init() imho.

 
Then dump_in_progress() is probably redundant too.




+
     s = &dump_state_global;
     dump_state_prepare(s);

--
2.31.1


reply via email to

[Prev in Thread] Current Thread [Next in Thread]