Commit 154304cd authored by Alex Williamson's avatar Alex Williamson
Browse files

postcopy: Synchronize usage of the balloon inhibitor



While the qemu_balloon_inhibit() interface appears rather general purpose,
postcopy uses it in a last-caller-wins approach with no guarantee of balanced
inhibits and de-inhibits.  Wrap postcopy's usage of the inhibitor to give it
one vote overall, using the same last-caller-wins approach as previously
implemented at the balloon level.

Fixes: 01ccbec7 ("balloon: Allow multiple inhibit users")
Reported-by: default avatarChristian Borntraeger <borntraeger@de.ibm.com>
Tested-by: default avatarChristian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: default avatarCornelia Huck <cohuck@redhat.com>
Reviewed-by: default avatarJuan Quintela <quintela@redhat.com>
Signed-off-by: default avatarAlex Williamson <alex.williamson@redhat.com>
parent 8709b395
Loading
Loading
Loading
Loading
+16 −2
Original line number Diff line number Diff line
@@ -509,6 +509,20 @@ int postcopy_ram_incoming_init(MigrationIncomingState *mis)
    return 0;
}

/*
 * Manage a single vote to the QEMU balloon inhibitor for all postcopy usage,
 * last caller wins.
 */
static void postcopy_balloon_inhibit(bool state)
{
    static bool cur_state = false;

    if (state != cur_state) {
        qemu_balloon_inhibit(state);
        cur_state = state;
    }
}

/*
 * At the end of a migration where postcopy_ram_incoming_init was called.
 */
@@ -539,7 +553,7 @@ int postcopy_ram_incoming_cleanup(MigrationIncomingState *mis)
        mis->have_fault_thread = false;
    }

    qemu_balloon_inhibit(false);
    postcopy_balloon_inhibit(false);

    if (enable_mlock) {
        if (os_mlock() < 0) {
@@ -1107,7 +1121,7 @@ int postcopy_ram_enable_notify(MigrationIncomingState *mis)
     * Ballooning can mark pages as absent while we're postcopying
     * that would cause false userfaults.
     */
    qemu_balloon_inhibit(true);
    postcopy_balloon_inhibit(true);

    trace_postcopy_ram_enable_notify();