On Tue, 2008-04-22 at 10:19 -0400, Chuck Lever wrote:
>
> My point is that EFAULT isn't an appropriate return code if
> rpc_run_task() fails. EFAULT is appropriate only when the I/O engine
> can't access the user's memory.
>
> If no bytes were read, an error code that reflects the problem should
> be returned.
>
> task = rpc_run_task(&task_setup_data);
> if (IS_ERR(task) {
> result = PTR_ERR(task);
> break;
> }
> rpc_put_task(task);
>
> The logic at the end of nfs_direct_read_schedule_segment() will take
> care to return the number of bytes read, or an error code if no bytes
> were read.
Then a fix should be applied to auth_gss.c: the EFAULT error message is
coming from the downcall code. If it is not appropriate to pass that on
to the RPC caller, then we should substitute an EACCES.
--
Trond Myklebust
Linux NFS client maintainer
NetApp
Trond.Myklebust@xxxxxxxxxx
www.netapp.com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
