Re: [PATCH 09/10] SUNRPC: Ensure that we read all available tcp data

[Chuck Lever <chuck.lever@oracle.com>]

[-- Attachment #1: Type: text/plain, Size: 1341 bytes --]

Trond Myklebust wrote:
> Don't stop until we run out of data, or we hit an error.

I don't understand how this patch is related to the others.

> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
> ---
> 
>  net/sunrpc/xprtsock.c |    7 +++++--
>  1 files changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
> index 30e7ac2..8bd3b0f 100644
> --- a/net/sunrpc/xprtsock.c
> +++ b/net/sunrpc/xprtsock.c
> @@ -1073,6 +1073,7 @@ static void xs_tcp_data_ready(struct sock *sk, int bytes)
>  {
>  	struct rpc_xprt *xprt;
>  	read_descriptor_t rd_desc;
> +	int read;
>  
>  	dprintk("RPC:       xs_tcp_data_ready...\n");
>  
> @@ -1084,8 +1085,10 @@ static void xs_tcp_data_ready(struct sock *sk, int bytes)
>  
>  	/* We use rd_desc to pass struct xprt to xs_tcp_data_recv */
>  	rd_desc.arg.data = xprt;
> -	rd_desc.count = 65536;
> -	tcp_read_sock(sk, &rd_desc, xs_tcp_data_recv);
> +	do {
> +		rd_desc.count = 65536;
> +		read = tcp_read_sock(sk, &rd_desc, xs_tcp_data_recv);
> +	} while (read > 0);

Why are we looping instead of simply setting rd_desc.count to a larger 
value (say, a megabyte)?

(If we keep this, it would be nice to replace the "65536" with a macro, 
and to add a comment explaining the need for the loop).

>  out:
>  	read_unlock(&sk->sk_callback_lock);
>  }

[-- Attachment #2: chuck_lever.vcf --]
[-- Type: text/x-vcard, Size: 270 bytes --]

begin:vcard
fn:Chuck Lever
n:Lever;Chuck
org:Oracle Corporation;Corporate Architecture: Linux Projects Group
adr:;;1015 Granger Avenue;Ann Arbor;MI;48104;USA
title:Principal Member of Staff
tel;work:+1 248 614 5091
x-mozilla-html:FALSE
version:2.1
end:vcard