[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <432FED98-2410-4760-8E7C-20675051FE96@oracle.com>
Date: Wed, 26 Dec 2018 12:44:08 -0500
From: Chuck Lever <chuck.lever@...cle.com>
To: Kangjie Lu <kjlu@....edu>
Cc: pakki001@....edu, Bruce Fields <bfields@...ldses.org>,
Jeff Layton <jlayton@...nel.org>,
Trond Myklebust <trond.myklebust@...merspace.com>,
Anna Schumaker <anna.schumaker@...app.com>,
"David S. Miller" <davem@...emloft.net>,
Linux NFS Mailing List <linux-nfs@...r.kernel.org>,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sunrpc: remove redundant code
> On Dec 25, 2018, at 10:24 PM, Kangjie Lu <kjlu@....edu> wrote:
>
> If no bytes to decode, just use "xdr->p" instead of calling
> xdr_inline_decode to get it. The fix cleans up the code.
>
> Signed-off-by: Kangjie Lu <kjlu@....edu>
> ---
> net/sunrpc/xprtrdma/rpc_rdma.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c
> index 9f53e0240035..2ef86be49bd8 100644
> --- a/net/sunrpc/xprtrdma/rpc_rdma.c
> +++ b/net/sunrpc/xprtrdma/rpc_rdma.c
> @@ -1123,7 +1123,6 @@ rpcrdma_decode_msg(struct rpcrdma_xprt *r_xprt, struct rpcrdma_rep *rep,
> {
> struct xdr_stream *xdr = &rep->rr_stream;
> u32 writelist, replychunk, rpclen;
> - char *base;
>
> /* Decode the chunk lists */
> if (decode_read_list(xdr))
> @@ -1138,10 +1137,9 @@ rpcrdma_decode_msg(struct rpcrdma_xprt *r_xprt, struct rpcrdma_rep *rep,
> return -EIO;
>
> /* Build the RPC reply's Payload stream in rqst->rq_rcv_buf */
> - base = (char *)xdr_inline_decode(xdr, 0);
> rpclen = xdr_stream_remaining(xdr);
> r_xprt->rx_stats.fixup_copy_count +=
> - rpcrdma_inline_fixup(rqst, base, rpclen, writelist & 3);
> + rpcrdma_inline_fixup(rqst, xdr->p, rpclen, writelist & 3);
I used xdr_inline_decode(xdr, 0) here because I didn't want to
embed specific knowledge about struct xdr_stream into
rpcrdma_decode_msg. IOW, adhere to the API contract.
Given the need for adding a type cast to this function call, as
reported by the kbuild robot, I don't see that this patch improves
things significantly.
> r_xprt->rx_stats.total_rdma_reply += writelist;
> return rpclen + xdr_align_size(writelist);
> --
> 2.17.2 (Apple Git-113)
>
--
Chuck Lever
Powered by blists - more mailing lists