…ted invoke
fastrpc invokes work by sending an RPC message to the DSP and blocking
in wait_for_completion_interruptible() until the DSP responds. If a
signal arrives during this wait, the syscall returns -ERESTARTSYS and
the invoke context which holds the in-flight DMA buffers and
completion state is left stranded in fl->pending.
On the next syscall attempt (either auto-restarted by the kernel via
SA_RESTART or manually retried by user-space after EINTR), a fresh
context is allocated and the RPC message is re-sent to the DSP. This
has two consequences:
- The original context leaks in fl->pending until the file is closed.
- The DSP receives a duplicate invocation. If the DSP was mid-way
through processing the first request and had issued a reverse RPC
call back to the host, the retry sends a new forward request
instead of the expected reverse-RPC response. The DSP thread
waiting for that response is never woken, causing a hang.
Fix this by saving the interrupted context to a new fl->interrupted
list on -ERESTARTSYS. When the same thread retries the invoke with a
matching sc, restore the context and jump directly to the wait,
skipping context allocation and message re-send.
Also drain fl->interrupted on process exit and complete any sleeping
contexts with -EPIPE when the rpmsg channel is removed.
Link: https://lore.kernel.org/all/43a7laqb7mnrvleunnmbxwhvzr6w3au4ofjri4r4ap7clsx6mc@jxqlr4a2lw56/
Fixes: 387f625 ("misc: fastrpc: handle interrupted contexts")
Cc: stable@kernel.org
Signed-off-by: Anandu Krishnan E <anandu.e@oss.qualcomm.com>
fastrpc invokes work by sending an RPC message to the DSP and blocking in wait_for_completion_interruptible() until the DSP responds. If a signal arrives during this wait, the syscall returns -ERESTARTSYS and the invoke context which holds the in-flight DMA buffers and completion state is left stranded in fl->pending.
On the next syscall attempt (either auto-restarted by the kernel via SA_RESTART or manually retried by user-space after EINTR), a fresh context is allocated and the RPC message is re-sent to the DSP. This has two consequences:
Fix this by saving the interrupted context to a new fl->interrupted list on -ERESTARTSYS. When the same thread retries the invoke with a matching sc, restore the context and jump directly to the wait, skipping context allocation and message re-send.
Also drain fl->interrupted on process exit and complete any sleeping contexts with -EPIPE when the rpmsg channel is removed.
Link: https://lore.kernel.org/all/43a7laqb7mnrvleunnmbxwhvzr6w3au4ofjri4r4ap7clsx6mc@jxqlr4a2lw56/
Fixes: 387f625 ("misc: fastrpc: handle interrupted contexts")
Cc: stable@kernel.org
CRs-Fixed: 4411765