Re: Vectored AIO breakage for sockets and pipes ?

[Zach Brown <zach.brown@oracle.com>]

> I'm not sure what the best way to fix this is. One option is to  
> always make
> a copy of the iovec and pass that down. Any other thoughts ?

Can we use this as another motivation to introduce an iovec container  
struct instead of passing a raw iov/seg?  The transition could turn  
hand-rolled functions like pipe_iov_copy_to_user() into functions  
that this iovec struct API provides.

I don't know if this would specifically help aio_rw_vect_retry() to  
know if it should advance the iovec on behalf of its callee who  
returned positive result codes.

Maybe it could use the API to discover a case where ret < size &&  
cur_pos(iov_struct) == initial_pos(iov_struct) via some iovec pos  
query before rw_op is called?

Or maybe the introduction of the API could normalize where the  
responsibility of advancing the iovec lies.  That might be a bit much.

Just talkin' here.

- z
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/