RPC: exit -512 [ERESTARTSYS]

["wlacey" <wlacey@goldenhindresearch.com>]

I'm bringing up a 2.4.18 version of the kernel on a tx4925 processor, using a NFS'ed root filesystem when the actual mount request RPC call fails w/-512

About line 662 in function __rpc_execute(), in file net/sunrpc/sched.c

if (task->tk_client->cl_intr && signalled())) {
        printk("RPC: got signal\n");
        task->tk_flags |= RPC_TASK_KILLED;
        rpc_exit(task, -ERESTARTSYS);
        rpc_wake_up_task(task);
  }
  }
  }

  if (task->tk_exit) {
     task->tk_exit(task);
     /* If tk_action is non-null, the user wants us to restart */
     if (task->tk_action) {
    ..
    ..

The problems is that rpc_exit() NULLS task->tk_action and I can't see how the task/RPC request gets re-generated.

Is anybody familiar w/this area of code? Why is the RPC call apparently not being re-generated?

Any help???

Thanks,
Warrick Lacey