CVE-2022-50738

In the Linux kernel, the following vulnerability has been resolved:

vhost-vdpa: fix an iotlb memory leak

Before commit 3d5698793897 ("vhost-vdpa: introduce asid based IOTLB") we called vhostvdpaiotlbunmap(v, iotlb, 0ULL, 0ULL - 1) during release to free all the resources allocated when processing user IOTLB messages through vhostvdpaprocessiotlb_update(). That commit changed the handling of IOTLB a bit, and we accidentally removed some code called during the release.

We partially fixed this with commit 037d4305569a ("vhost-vdpa: call vhostvdpacleanup during the release") but a potential memory leak is still there as showed by kmemleak if the application does not send VHOSTIOTLBINVALIDATE or crashes:

unreferenced object 0xffff888007fbaa30 (size 16): comm "blkio-bench", pid 914, jiffies 4294993521 (age 885.500s) hex dump (first 16 bytes): 40 73 41 07 80 88 ff ff 00 00 00 00 00 00 00 00 @sA............. backtrace: [<0000000087736d2a>] kmemcachealloctrace+0x142/0x1c0 [<0000000060740f50>] vhostvdpaprocessiotlbmsg+0x68c/0x901 [vhostvdpa] [<0000000083e8e205>] vhostchrwriteiter+0xc0/0x4a0 [vhost] [<000000008f2f414a>] vhostvdpachrwriteiter+0x18/0x20 [vhostvdpa] [<00000000de1cd4a0>] vfswrite+0x216/0x4b0 [<00000000a2850200>] ksyswrite+0x71/0xf0 [<00000000de8e720b>] _x64syswrite+0x19/0x20 [<0000000018b12cbb>] dosyscall64+0x3f/0x90 [<00000000986ec465>] entrySYSCALL64after_hwframe+0x63/0xcd

Let's fix this calling vhostvdpaiotlbunmap() on the whole range in vhostvdparemoveas(). We move that call before vhostdevcleanup() since we need a valid v->vdev.mm in vhostvdpapaunmap(). vhostiotlbreset() call can be removed, since vhostvdpaiotlbunmap() on the whole range removes all the entries.

The kmemleak log reported was observed with a vDPA device that has use_va set to true (e.g. VDUSE). This patch has been tested with both types of devices.