linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Trond Myklebust <trondmy@gmail.com>
To: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs@vger.kernel.org
Subject: [PATCH 3/4] SUNRPC: Allow for some re-ordering of RPCSEC_GSS requests on the server
Date: Wed,  2 Jan 2019 17:53:12 -0500	[thread overview]
Message-ID: <20190102225313.2097-4-trond.myklebust@hammerspace.com> (raw)
In-Reply-To: <20190102225313.2097-3-trond.myklebust@hammerspace.com>

Strictly speaking, the client is probably not supposed to transmit more
requests than are allowed for in the RPCSEC_GSS window. However since
we do, try to ensure that we allow for some re-ordering by the server.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 net/sunrpc/auth_gss/auth_gss.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index ba765473d1f0..dc7124d45c8a 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -2029,9 +2029,6 @@ gss_xmit_need_reencode(struct rpc_task *task)
 	if (!ctx)
 		return true;
 
-	if (gss_seq_is_newer(req->rq_seqno, READ_ONCE(ctx->gc_seq)))
-		goto out;
-
 	seq_xmit = READ_ONCE(ctx->gc_seq_xmit);
 	while (gss_seq_is_newer(req->rq_seqno, seq_xmit)) {
 		u32 tmp = seq_xmit;
@@ -2043,7 +2040,12 @@ gss_xmit_need_reencode(struct rpc_task *task)
 		}
 	}
 
-	win = ctx->gc_win;
+	/*
+	 * Ensure the request is within 3/4 of the RPCSEC_GSS sequence
+	 * window so that we allow for some re-ordering of the requests
+	 * by the server.
+	 */
+	win = 3 * ((ctx->gc_win + 3) >> 2);
 	if (win > 0)
 		ret = !gss_seq_is_newer(req->rq_seqno, seq_xmit - win);
 out:
-- 
2.20.1


  reply	other threads:[~2019-01-02 22:54 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-02 22:53 [PATCH 0/4] bugfixes for RPCSEC_GSS client support Trond Myklebust
2019-01-02 22:53 ` [PATCH 1/4] SUNRPC: Ensure rq_bytes_sent is reset before request transmission Trond Myklebust
2019-01-02 22:53   ` [PATCH 2/4] SUNRPC: Fix the RPCSEC_GSS sequence semantics after request re-encoding Trond Myklebust
2019-01-02 22:53     ` Trond Myklebust [this message]
2019-01-02 22:53       ` [PATCH 4/4] SUNRPC: Ensure we respect the RPCSEC_GSS sequence number limit Trond Myklebust
2019-01-03 15:29   ` [PATCH 1/4] SUNRPC: Ensure rq_bytes_sent is reset before request transmission Chuck Lever
2019-01-03 16:05     ` Trond Myklebust
2019-01-03 16:17       ` Tom Talpey
2019-01-03 16:27         ` Trond Myklebust
2019-01-03 16:39         ` Chuck Lever
2019-01-03 16:41       ` Chuck Lever
2019-01-03 18:09         ` Trond Myklebust
2019-01-03 17:15 ` [PATCH 0/4] bugfixes for RPCSEC_GSS client support Chuck Lever

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190102225313.2097-4-trond.myklebust@hammerspace.com \
    --to=trondmy@gmail.com \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).