From: Alex Williamson Date: Mon, 20 Mar 2017 02:38:40 +0000 (-0600) Subject: drm/i915/kvmgt: Hold struct kvm reference X-Git-Tag: v4.11-rc5~25^2~1^2~1^2~5 X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=93a15b58cfb8a24e666ffca432f19fe65c1cd7d1;p=karo-tx-linux.git drm/i915/kvmgt: Hold struct kvm reference The kvmgt code keeps a pointer to the struct kvm associated with the device, but doesn't actually hold a reference to it. If we do unclean shutdown testing (ie. killing the user process), then we can see the kvm association to the device unset, which causes kvmgt to trigger a device release via a work queue. Naturally we cannot guarantee that the cached struct kvm pointer is still valid at this point without holding a reference. The observed failure in this case is a stuck cpu trying to acquire the spinlock from the invalid reference, but other failure modes are clearly possible. Hold a reference to avoid this. Signed-off-by: Alex Williamson Cc: stable@vger.kernel.org #v4.10 Cc: Jike Song Cc: Paolo Bonzini Cc: Zhenyu Wang Cc: Zhi Wang Reviewed-by: Jike Song Signed-off-by: Zhenyu Wang --- diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 1ea3eb270de8..d641214578a7 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -1326,6 +1326,7 @@ static int kvmgt_guest_init(struct mdev_device *mdev) vgpu->handle = (unsigned long)info; info->vgpu = vgpu; info->kvm = kvm; + kvm_get_kvm(info->kvm); kvmgt_protect_table_init(info); gvt_cache_init(vgpu); @@ -1347,6 +1348,7 @@ static bool kvmgt_guest_exit(struct kvmgt_guest_info *info) } kvm_page_track_unregister_notifier(info->kvm, &info->track_node); + kvm_put_kvm(info->kvm); kvmgt_protect_table_destroy(info); gvt_cache_destroy(info->vgpu); vfree(info);