From patchwork Thu Apr 28 12:40:01 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Willy Tarreau X-Patchwork-Id: 12830628 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 43E65C433EF for ; Thu, 28 Apr 2022 12:41:41 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1346580AbiD1Mox (ORCPT ); Thu, 28 Apr 2022 08:44:53 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46982 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1346498AbiD1Mom (ORCPT ); Thu, 28 Apr 2022 08:44:42 -0400 Received: from 1wt.eu (wtarreau.pck.nerim.net [62.212.114.60]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 55502AFAE4; Thu, 28 Apr 2022 05:41:14 -0700 (PDT) Received: (from willy@localhost) by pcw.home.local (8.15.2/8.15.2/Submit) id 23SCeFwv007486; Thu, 28 Apr 2022 14:40:15 +0200 From: Willy Tarreau To: netdev@vger.kernel.org Cc: David Miller , Jakub Kicinski , Eric Dumazet , Moshe Kol , Yossi Gilad , Amit Klein , "Jason A . Donenfeld" , linux-kernel@vger.kernel.org, Willy Tarreau Subject: [PATCH v2 net 7/7] tcp: drop the hash_32() part from the index calculation Date: Thu, 28 Apr 2022 14:40:01 +0200 Message-Id: <20220428124001.7428-8-w@1wt.eu> X-Mailer: git-send-email 2.17.5 In-Reply-To: <20220428124001.7428-1-w@1wt.eu> References: <20220428124001.7428-1-w@1wt.eu> Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org In commit 190cc82489f4 ("tcp: change source port randomizarion at connect() time"), the table_perturb[] array was introduced and an index was taken from the port_offset via hash_32(). But it turns out that hash_32() performs a multiplication while the input here comes from the output of SipHash in secure_seq, that is well distributed enough to avoid the need for yet another hash. Suggested-by: Amit Klein Reviewed-by: Eric Dumazet Signed-off-by: Willy Tarreau --- net/ipv4/inet_hashtables.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index cc5f66328b47..a5d57fa679ca 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -778,7 +778,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, net_get_random_once(table_perturb, INET_TABLE_PERTURB_SIZE * sizeof(*table_perturb)); - index = hash_32(port_offset, INET_TABLE_PERTURB_SHIFT); + index = port_offset & (INET_TABLE_PERTURB_SIZE - 1); offset = READ_ONCE(table_perturb[index]) + (port_offset >> 32); offset %= remaining;