From f805bc286615d49ffd463e4a0d9914e2ca892065 Mon Sep 17 00:00:00 2001 From: Yonit Halperin Date: Wed, 14 Sep 2011 08:43:39 +0200 Subject: [PATCH 5/8] spice: set qxl->ssd.running=true before telling spice to start, RHBZ #733993 RH-Author: Yonit Halperin Message-id: <1315989819-29717-1-git-send-email-yhalperi@redhat.com> Patchwork-id: 32742 O-Subject: [PATCH] spice: set qxl->ssd.running=true before telling spice to start, RHBZ #733993 Bugzilla: 733993 RH-Acked-by: Gerd Hoffmann RH-Acked-by: Alon Levy RH-Acked-by: Uri Lublin If qxl->ssd.running=true is set after telling spice to start, the spice server thread can call qxl_send_events while qxl->ssd.running is still false. This leads to assert(d->ssd.running). Signed-off-by: Yonit Halperin Signed-off-by: Gerd Hoffmann (cherry picked from commit 7e79cf4083efa399b43f30edf23434b137fcb197) Signed-off-by: Yonit Halperin --- Since it looks like the purpose of the assert in qxl_send_event is preventing changes in the guest when the vm is stopped, I think it is not necessary for ssd.running to be exactly synchronized with the spice server status, but just be true before the spice worker starts. ui/spice-display.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) Signed-off-by: Michal Novotny Signed-off-by: Michal Novotny --- ui/spice-display.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/ui/spice-display.c b/ui/spice-display.c index 64b75a9..8c76c7f 100644 --- a/ui/spice-display.c +++ b/ui/spice-display.c @@ -256,11 +256,12 @@ void qemu_spice_vm_change_state_handler(void *opaque, int running, int reason) SimpleSpiceDisplay *ssd = opaque; if (running) { + ssd->running = true; qemu_spice_start(ssd); } else { qemu_spice_stop(ssd); + ssd->running = false; } - ssd->running = running; } void qemu_spice_display_init_common(SimpleSpiceDisplay *ssd, DisplayState *ds) -- 1.7.4.4