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

svcrdma: Reject Write/Reply chunks with segcount 0

A peer can send a Write or Reply chunk whose segcount field is zero. xdr check write chunk() only rejects segcount > rc maxpages, so zero passes the range check, and xdr inline decode(stream, 0) returns the current (non-NULL) cursor without advancing. The function returns true and pcl alloc write() then links a struct svc rdma chunk with ch segcount == 0 onto rc write pcl or rc reply pcl.

An earlier patch in this series made pcl for each segment() safe for ch segcount == 0, so this no longer drives the memory walk it used to. Rejecting the malformed frame at the decode boundary is still worthwhile as defense in depth: it keeps degenerate zero-segment chunks off the parsed chunk lists entirely, so any future consumer that walks ch segments directly cannot observe one, and it makes the zero-floor easy to backport to trees where the macro change is more intrusive. RFC 8166 has no meaning for a Write/Reply chunk that describes no remote buffer, so no legitimate client is affected.

xdr check reply chunk() funnels Reply chunks through xdr check write chunk() and inherits the same rejection.

pcl alloc write() also links each chunk onto the parsed chunk list before filling its segment array. If a future change weakens the segcount-0 rejection, an incomplete chunk is visible to consumers during the fill loop. Reorder so that list add tail() follows the segment fill loop, ensuring only fully-populated chunks appear on the list.