From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:37002) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TC4y1-0002RK-0D for qemu-devel@nongnu.org; Thu, 13 Sep 2012 04:40:15 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TC4xx-0007Bo-M5 for qemu-devel@nongnu.org; Thu, 13 Sep 2012 04:40:08 -0400 Received: from mx1.redhat.com ([209.132.183.28]:10955) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TC4xx-00079E-EJ for qemu-devel@nongnu.org; Thu, 13 Sep 2012 04:40:05 -0400 Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q8D8e4NC026209 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Thu, 13 Sep 2012 04:40:04 -0400 From: Gerd Hoffmann Date: Thu, 13 Sep 2012 10:39:53 +0200 Message-Id: <1347525600-28220-8-git-send-email-kraxel@redhat.com> In-Reply-To: <1347525600-28220-1-git-send-email-kraxel@redhat.com> References: <1347525600-28220-1-git-send-email-kraxel@redhat.com> Subject: [Qemu-devel] [PATCH 07/14] ehci: Walk async schedule before and after migration List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Hans de Goede , Gerd Hoffmann From: Hans de Goede Signed-off-by: Hans de Goede Signed-off-by: Gerd Hoffmann --- hw/usb/hcd-ehci.c | 28 ++++++++++++++++++++++++++++ 1 files changed, 28 insertions(+), 0 deletions(-) diff --git a/hw/usb/hcd-ehci.c b/hw/usb/hcd-ehci.c index bc86460..6a5da84 100644 --- a/hw/usb/hcd-ehci.c +++ b/hw/usb/hcd-ehci.c @@ -34,6 +34,7 @@ #include "monitor.h" #include "trace.h" #include "dma.h" +#include "sysemu.h" #define EHCI_DEBUG 0 @@ -2572,6 +2573,32 @@ static int usb_ehci_post_load(void *opaque, int version_id) return 0; } +static void usb_ehci_vm_state_change(void *opaque, int running, RunState state) +{ + EHCIState *ehci = opaque; + + /* + * We don't migrate the EHCIQueue-s, instead we rebuild them for the + * schedule in guest memory. We must do the rebuilt ASAP, so that + * USB-devices which have async handled packages have a packet in the + * ep queue to match the completion with. + */ + if (state == RUN_STATE_RUNNING) { + ehci_advance_async_state(ehci); + } + + /* + * The schedule rebuilt from guest memory could cause the migration dest + * to miss a QH unlink, and fail to cancel packets, since the unlinked QH + * will never have existed on the destination. Therefor we must flush the + * async schedule on savevm to catch any not yet noticed unlinks. + */ + if (state == RUN_STATE_SAVE_VM) { + ehci_advance_async_state(ehci); + ehci_queues_rip_unseen(ehci, 1); + } +} + static const VMStateDescription vmstate_ehci = { .name = "ehci", .version_id = 2, @@ -2721,6 +2748,7 @@ static int usb_ehci_initfn(PCIDevice *dev) usb_packet_init(&s->ipacket); qemu_register_reset(ehci_reset, s); + qemu_add_vm_change_state_handler(usb_ehci_vm_state_change, s); memory_region_init(&s->mem, "ehci", MMIO_SIZE); memory_region_init_io(&s->mem_caps, &ehci_mmio_caps_ops, s, -- 1.7.1