qemu-kvm/kvm-vhost-vdpa-do-not-cleanup-the-vdpa-vhost-net-structu.patch
Miroslav Rezanina fd7488b82b * Mon Aug 28 2023 Miroslav Rezanina <mrezanin@redhat.com> - 6.2.0-39
- kvm-vhost-vdpa-do-not-cleanup-the-vdpa-vhost-net-structu.patch [bz#2215786]
- Resolves: bz#2215786
  (CVE-2023-3301 virt:rhel/qemu-kvm: QEMU: net: triggerable assertion due to race condition in hot-unplug [rhel-8])
2023-08-28 06:54:27 -04:00

82 lines
3.2 KiB
Diff

From 7b17ef78eee2b30829666f12e87ff1eee3c195b5 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Tue, 15 Aug 2023 19:00:44 -0400
Subject: [PATCH] vhost-vdpa: do not cleanup the vdpa/vhost-net structures if
peer nic is present
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 304: vhost-vdpa: do not cleanup the vdpa/vhost-net structures if peer nic is present
RH-Bugzilla: 2215786
RH-Acked-by: Ani Sinha <None>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [1/1] 16aa37efdf129f2619cedf9c030222b88eda9e26 (redhat/rhel/src/qemu-kvm/jons-qemu-kvm-2)
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2215786
CVE: CVE-2023-3301
Upstream: Merged
Conflicts: commit babf8b87127a is not present in this release, so the commit does not
apply cleanly. The two adjacent munmap() calls introduced by that commit
don't seem to be needed for the logics of this change.
commit a0d7215e339b61c7d7a7b3fcf754954d80d93eb8
Author: Ani Sinha <anisinha@redhat.com>
Date: Mon Jun 19 12:22:09 2023 +0530
vhost-vdpa: do not cleanup the vdpa/vhost-net structures if peer nic is present
When a peer nic is still attached to the vdpa backend, it is too early to free
up the vhost-net and vdpa structures. If these structures are freed here, then
QEMU crashes when the guest is being shut down. The following call chain
would result in an assertion failure since the pointer returned from
vhost_vdpa_get_vhost_net() would be NULL:
do_vm_stop() -> vm_state_notify() -> virtio_set_status() ->
virtio_net_vhost_status() -> get_vhost_net().
Therefore, we defer freeing up the structures until at guest shutdown
time when qemu_cleanup() calls net_cleanup() which then calls
qemu_del_net_client() which would eventually call vhost_vdpa_cleanup()
again to free up the structures. This time, the loop in net_cleanup()
ensures that vhost_vdpa_cleanup() will be called one last time when
all the peer nics are detached and freed.
All unit tests pass with this change.
CC: imammedo@redhat.com
CC: jusual@redhat.com
CC: mst@redhat.com
Fixes: CVE-2023-3301
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2128929
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Message-Id: <20230619065209.442185-1-anisinha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
net/vhost-vdpa.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 814f704687..ac48de9495 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -128,6 +128,14 @@ static void vhost_vdpa_cleanup(NetClientState *nc)
{
VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
+ /*
+ * If a peer NIC is attached, do not cleanup anything.
+ * Cleanup will happen as a part of qemu_cleanup() -> net_cleanup()
+ * when the guest is shutting down.
+ */
+ if (nc->peer && nc->peer->info->type == NET_CLIENT_DRIVER_NIC) {
+ return;
+ }
if (s->vhost_net) {
vhost_net_cleanup(s->vhost_net);
g_free(s->vhost_net);
--
2.39.3