When closing down all devices, we manually unset master for all attached devices, but the device's sprite info still points to the master's sprite info. This leaves us a window where the master is freed already but the device isn't yet. A signal during that window causes dereference of the already freed spriteInfo in mieqEnqueue's EnqueueScreen macro.
Simply block signals when removing all devices. It's not like we're really worrying about high-responsive input at this stage. https://bugzilla.redhat.com/show_bug.cgi?id=737031 Signed-off-by: Peter Hutterer <[email protected]> Tested-by: Paulo Zanoni <[email protected]> --- The more correct approach would be to loop through inputInfo.devices twice, removing all slaves first and then the masters. But really, if you're expecting input events to be delivered while we're shutting down all devices you're going to be disappointed anyway. dix/devices.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/dix/devices.c b/dix/devices.c index 63454fe..4a525d5 100644 --- a/dix/devices.c +++ b/dix/devices.c @@ -971,6 +971,8 @@ CloseDownDevices(void) { DeviceIntPtr dev; + OsBlockSignals(); + /* Float all SDs before closing them. Note that at this point resources * (e.g. cursors) have been freed already, so we can't just call * AttachDevice(NULL, dev, NULL). Instead, we have to forcibly set master @@ -993,6 +995,8 @@ CloseDownDevices(void) inputInfo.keyboard = NULL; inputInfo.pointer = NULL; XkbDeleteRulesDflts(); + + OsReleaseSignals(); } /** -- 1.7.7 _______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
