sunrpc: route to a populated pool in svc_pool_for_cpu()
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
≥ bfd241600a3b0db4fe43c859f1460d0a958d924a · < 9d04d64ad192439835ed9884d767353061c3ed6f| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
sunrpc: route to a populated pool in svc_pool_for_cpu()
svc_set_num_threads() spreads the requested threads evenly across the service's pools (base = nrservs / sv_nrpools). When a service runs fewer threads than it has pools -- e.g. an nfsd configured with fewer threads than the host has NUMA nodes while running in "pernode" or "percpu" mode -- the trailing pools are left with no threads at all.
svc_xprt_enqueue() selects a pool from the CPU servicing the transport, queues the transport on that pool's sp_xprts, and only wakes a thread from the same pool. Each thread services exclusively its own pool, so a transport that lands on a threadless pool is enqueued on sp_xprts and never picked up: the connection hangs indefinitely.
Have svc_pool_for_cpu() skip pools that currently have no threads, falling back to the next populated pool. This trades NUMA locality for a guarantee that the work is actually serviced. sp_nrthreads is only updated under the service mutex; the lockless read here is a best-effort routing hint, so annotate it with data_race().
From the CVE record · Linux
Linux kernel sunrpc services can leave some pools without threads when configured with fewer service threads than pools. A network connection routed to one of those pools can hang indefinitely because no thread picks it up. The input does not specify authentication or user-interaction requirements.
Operators running Linux kernel sunrpc services, including nfsd, with fewer service threads than pools should check their systems; affected versions are not specified.