CVE-2026-97952

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

sunvdc: unmap LDC cookies when the descriptor send fails

send request() maps the request's pages into the LDC channel's map table (ldc map sg()), fills in the descriptor and marks it VIO DESC READY before ringing the doorbell via vdc tx trigger(). When the trigger fails, the error path only prints a message: the descriptor stays READY and the cookies are never unmapped. The mapping is normally released in vdc end one() when the peer completes the descriptor - but a descriptor whose doorbell was never sent will never complete, and since dr->prod is not advanced on failure, the reset path (vdc requeue inflight(), which walks [cons, prod)) never visits it either. The map table entries are leaked permanently.

Since commit a11f6ca9aef9 ("sunvdc: Do not spin in an infinite loop when vio ldc send() returns EAGAIN") trigger failures occur in practice under load, so every resulting I/O error also leaks one request's worth of entries from the fixed-size (8192 entries per channel) map table. Because the allocator hands out contiguous ranges, fragmentation makes large multi-segment requests fail first as the table drains, until ldc map sg() fails permanently and the disk is dead until reboot.

It also makes any retry-based recovery unusable: requeuing the request on -EAGAIN remaps the pages on every attempt, overwriting desc->cookies and orphaning the previous mapping, so the table drains at the retry rate. This is the memory exhaustion observed when the requeue approach was first tested in October 2025.

Roll back on failure: unmap the cookies, mark the descriptor FREE again and clear the request entry. If the trigger failed with -ENOTCONN, vdc tx trigger() has already reset the port, which tears down and reallocates both the dring and the LDC channel including its map table - nothing to roll back, and the stale descriptor must not be touched.