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

Input: ims-pcu - fix heap-buffer-overflow in imspcuprocessdata()

The imspcuprocessdata() processes incoming URB data byte by byte. However, it fails to check if the readpos index exceeds IMSPCUBUFSIZE.

If a malicious USB device sends a packet larger than IMSPCUBUFSIZE, readpos will increment indefinitely. Moreover, since readpos is located immediately after readbuf, the attacker can overwrite readpos itself to arbitrarily control the index.

This manipulated readpos is subsequently used in imspcuhandleresponse() to copy data into cmdbuf, leading to a heap buffer overflow.

Specifically, an attacker can overwrite the cmddone.wait.head located at offset 136 relative to cmdbuf in the imspcuhandleresponse(). Consequently, when the driver calls complete(&pcu->cmddone), it triggers a control flow hijack by using the manipulated pointer.

Fix this by adding a bounds check for readpos before writing to readbuf. If the packet is too long, discard it, log a warning, and reset the parser state.

[dtor: factor out resetting packet state, reset checksum as well]

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

net/mlx5e: Fix HV VHCA stats zero-sized buffer allocation

mlx5ehvvhcastatscreate() is called from mlx5enicenable(), before mlx5eopen(). At that point priv->statsnch is still zero, because it is only ever incremented in mlx5echannelstatsalloc(), which is reached only from mlx5eopenchannel().

mlx5ehvvhcastatsbufsize() therefore returns 0, and kvzalloc(0, GFPKERNEL) returns ZEROSIZEPTR ((void )16) rather than NULL. The "if (!buf)" guard does not catch this, and mlx5ehvvhcastatscreate() completes "successfully" with priv->statsagent.buf set to ZEROSIZEPTR.

Once channels are opened (priv->statsnch > 0) and the hypervisor enables stats reporting, mlx5ehvvhcastatswork() recomputes buflen using the new non-zero statsnch and calls memset(buf, 0, buflen) on ZEROSIZEPTR, faulting at address 0x10.

Allocate the buffer based on priv->maxnch, which is set in mlx5eprivinit() and is the upper bound on statsnch:

- Add a separate helper mlx5ehvvhcastatsbufmaxsize() that returns sizeof(perringstats) max(maxnch, statsnch), and use it for the kvzalloc() in mlx5ehvvhcastatscreate(). - Keep mlx5ehvvhcastatsbufsize() (which returns based on statsnch) for the worker's active payload size, so the wire format (block->rings = statsnch) and the amount of data filled by mlx5ehvvhcafillstats() are unchanged.

The max(maxnch, statsnch) guard handles the rare case where mlx5eattachnetdev() recomputes maxnch downward across a detach/resume cycle while priv->statsnch persists (mlx5edetachnetdev does not call mlx5eprivcleanup, so statsnch is only reset when the netdev is destroyed). Without the guard, the worker could compute buflen from statsnch and overrun the smaller buffer allocated based on the reduced maxnch.

Allocating a non-zero buffer also makes the kvzalloc() failure path in mlx5ehvvhcastatscreate() reachable for the first time: it returns early without (re)creating the agent. Clear priv->statsagent.{agent,buf} in mlx5ehvvhcastatsdestroy() after freeing them, so that if a later create() bails out on this path, a subsequent teardown does not double-free the stale agent/buffer left from a previous enable/disable cycle.

This mirrors the existing mlx5e pattern of preallocating arrays of size maxnch (e.g. priv->channelstats) and lazily populating entries up to statsnch on demand.

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

afs: Fix UAF when sending a message

In afsmakecall(), there's a race with async call reception and destruction. If a call is dispatched that doesn't have call->writeiter set (used to specify the data content for FS.StoreData), then the first rxrpckernelsenddata() will not set MSGMORE in the msghdr.

Once rxrpcsenddata() queues the last request packet, the response could come in at any time and cause the call to be completed and put. However, afsmakecall() will look at the call again to see it ->writeiter should be handled - something it's only allowed to do if it has its own ref on the call. Whilst this is the case for synchronous calls, it isn't true for async calls such as FS.FetchData.

There's also a potential UAF in afsmakecall() in the event that an asynchronous call is being sent, but the call fails in some way (e.g. it gets aborted from the server). The problem there is that afsmakecall() tries to abort a call if the rxrpc send fails, but the asynchronous notification from rxrpc may have caused the afscall to be torn down.

generic/650 plays games with randomly taking CPUs offline, and can interject a significant delay such that the call is deallocated before afsmakecall() gets to check call->writeiter - and a UAF ensues (caught by KASAN).

BUG: KASAN: slab-use-after-free in afsmakecall+0x1c90/0x2210 [kafs] Read of size 8 at addr ffff888035e050e8 by task fsstress/1409

Fix this by making afsmakeopcall() give the op->call its own ref rather than transferring the caller's ref to it and then dropping the ref when afsmakecall() returns.

This also means that the afsmakecall() func never loses its ref on the call now.

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

ptp: vmclock: prevent read-only mappings from becoming writable

vmclockmiscdevmmap() rejects writable mappings of the shared vmclock ABI page with -EROFS, but leaves VMMAYWRITE set. Userspace can map the page read-only and then upgrade it to writable with mprotect(), after which the guest can corrupt the host-written timekeeping data (sequence counter, UTC time, TSC offset) that the vmclock ABI defines as read-only.

Clear VMMAYWRITE on the read-only path so the mapping cannot be upgraded, as i915 does for its read-only objects and as fixed in drm/vc4 (CVE-2026-68445) and drm/panthor (CVE-2024-53071).