On Tue, 2022-01-18 at 21:33 +0200, Dan Aloni wrote:
> Due to change 8cfb9015280d ("NFS: Always provide aligned buffers to
> the
> RPC read layers"), a read of 0xfff is aligned up to server rsize of
> 0x1000.
>
> As a result, in a test where the server has a file of size
> 0x7fffffffffffffff, and the client tries to read from the offset
> 0x7ffffffffffff000, the read causes loff_t overflow in the server and
> it
> returns an NFS code of EINVAL to the client. The client as a result
> indefinitely retries the request.
>
> This fixes the issue by cancelling the alignment for that case.
>
NACK. This would be a server bug, not a client bug. The NFS protocol
has no notion of signed offsets, and doesn't use loff_t.
--
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@xxxxxxxxxxxxxxx
