* add build-dependency on libcap-dev [linux-any] to enable virtfs support
  which has been dropped in 1.1. (Closes: #677654)
* intel_hda-do-not-call-msi_reset-when-only-device-state-needs-resetting.patch
  patch to fix Fixing reset of MSI function in intel-hda virtual device.
  The fix (applied to stable-1.1.1) was partially wrong, as it actually
  added the msi_reset() call to two code paths instead of one as planned.
  Fix this by splitting the function in question into two parts.
  (Closes: #688964)
* blockdev-preserve-readonly-and-snapshot-states-across-media-changes.patch:
  allow opening of read-only cdrom images/devices (Closes: #686776)
* ahci-properly-reset-PxCMD-on-HBA-reset.patch: fix windows install on ahci
  (Closes: #696052)
* e1000-discard-packets-that-are-too-long-if-not-SBP-and-not-LPE.patch:
  discard too long rx packets which may overflow guest buffer
  (Closes: #696051)
* eepro100-fix-network-hang-when-rx-buffers-run-out.patch:
  fix e100 stall (Closes: #696061)
* fix possible network stalls/slowness in e1000 device emulation:
  net-notify-iothread-after-flushing-queue.patch
  e1000-flush-queue-whenever-can_receive-can-go-from-false-to-true.patch
  (Closes: #696063)
* fixes-related-to-processing-of-qemu-s-numa-option.patch:
  fixes numa handling (Closes: #691343)
* qcow2-fix-avail_sectors-in-cluster-allocation-code.patch:
  fixes data corruption in stacked qcow2 (Closes: #695905)
* qcow2-fix-refcount-table-size-calculation.patch: another possible
  corruption or crash in qcow2 (Closes: #691569)
* tap-reset-vnet-header-size-on-open.patch: always ensure tap device is
  in known state initially (Closes: #696057)
* vmdk-fix-data-corruption-bug-in-WRITE-and-READ-handling.patch:
  possible data corruption bug in vmdk image format (Closes: #696050)