From patchwork Mon Jun 5 21:20:03 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Stefan Roesch X-Patchwork-Id: 13268059 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 25F29C7EE32 for ; Mon, 5 Jun 2023 21:20:41 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229698AbjFEVUk (ORCPT ); Mon, 5 Jun 2023 17:20:40 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:40642 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231268AbjFEVUi (ORCPT ); Mon, 5 Jun 2023 17:20:38 -0400 Received: from 66-220-144-178.mail-mxout.facebook.com (66-220-144-178.mail-mxout.facebook.com [66.220.144.178]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BD866F3 for ; Mon, 5 Jun 2023 14:20:37 -0700 (PDT) Received: by devbig1114.prn1.facebook.com (Postfix, from userid 425415) id EFDBE6992199; Mon, 5 Jun 2023 14:20:21 -0700 (PDT) From: Stefan Roesch To: io-uring@vger.kernel.org, kernel-team@fb.com Cc: shr@devkernel.io, axboe@kernel.dk, ammarfaizi2@gnuweeb.org, netdev@vger.kernel.org, kuba@kernel.org, olivier@trillion01.com Subject: [PATCH v14 2/8] net: introduce napi_busy_loop_rcu() Date: Mon, 5 Jun 2023 14:20:03 -0700 Message-Id: <20230605212009.1992313-3-shr@devkernel.io> X-Mailer: git-send-email 2.39.1 In-Reply-To: <20230605212009.1992313-1-shr@devkernel.io> References: <20230605212009.1992313-1-shr@devkernel.io> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org This introduces the napi_busy_loop_rcu() function. If the caller of napi_busy_loop() function is also taking the rcu read lock, it is possible that napi_busy_loop() is releasing the read lock if it invokes schedule. However the caller is expecting that the rcu read lock is not released until the function completes. This new function avoids that problem. It expects that the caller MUST hold the rcu_read_lock while calling this function. Signed-off-by: Stefan Roesch --- include/net/busy_poll.h | 4 ++++ net/core/dev.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h index f90f0021f5f2..622623f5740e 100644 --- a/include/net/busy_poll.h +++ b/include/net/busy_poll.h @@ -47,6 +47,10 @@ void napi_busy_loop(unsigned int napi_id, bool (*loop_end)(void *, unsigned long), void *loop_end_arg, bool prefer_busy_poll, u16 budget); +void napi_busy_loop_rcu(unsigned int napi_id, + bool (*loop_end)(void *, unsigned long), + void *loop_end_arg, bool prefer_busy_poll, u16 budget); + #else /* CONFIG_NET_RX_BUSY_POLL */ static inline unsigned long net_busy_loop_on(void) { diff --git a/net/core/dev.c b/net/core/dev.c index f4677aa20f84..fcd4a6a70646 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6213,6 +6213,37 @@ static inline void __napi_busy_poll(struct napi_busy_poll_ctx *ctx, LINUX_MIB_BUSYPOLLRXPACKETS, work); local_bh_enable(); } + +/* + * Warning: can exit without calling need_resched(). + */ +void napi_busy_loop_rcu(unsigned int napi_id, + bool (*loop_end)(void *, unsigned long), + void *loop_end_arg, bool prefer_busy_poll, u16 budget) +{ + unsigned long start_time = loop_end ? busy_loop_current_time() : 0; + struct napi_busy_poll_ctx ctx = {}; + + ctx.napi = napi_by_id(napi_id); + if (!ctx.napi) + return; + + preempt_disable(); + for (;;) { + __napi_busy_poll(&ctx, prefer_busy_poll, budget); + + if (!loop_end || loop_end(loop_end_arg, start_time)) + break; + if (unlikely(need_resched())) + break; + + cpu_relax(); + } + if (ctx.napi_poll) + busy_poll_stop(ctx.napi, ctx.have_poll_lock, prefer_busy_poll, budget); + preempt_enable(); +} + void napi_busy_loop(unsigned int napi_id, bool (*loop_end)(void *, unsigned long), void *loop_end_arg, bool prefer_busy_poll, u16 budget)