From nobody Wed Dec 24 16:03:30 2025 Received: from mail-ot1-f49.google.com (mail-ot1-f49.google.com [209.85.210.49]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id DA0FB12E72 for ; Thu, 25 Jan 2024 22:57:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.49 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1706223469; cv=none; b=fUk0QA7NN2AhxUJKUtV7uMLXTsz88GhgLLAml//W6EG4fAc5vtpaSSGpK+OlUfjXewfWZkIT2yIsfiyRcC17QjbjEfC312fh+hyhxQ6oGLFl/XJ2ob53MALE58XlCV2UMCMOOfUmLbf25CrHOU0/ZgeKbdkpkHeXCvGYebGAqQA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1706223469; c=relaxed/simple; bh=qkngexUKlfn+hMPunN78J4XlCpnaY0JzFvOY2CvOwFw=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=Xr8YJYc2GvJnz/pEDbxkGX6OFfm43PCVVLPKcfJ8MhMAb2pU2gEYc1XIrnnbafQFyMNNoicx9UWScRJ2J2Okhwo5CqSO56XgLGgdWCrAV727ycRIk1nb1ekxBuiVBYlHa8So7BAGut3t7//vaAh5mwCzRxtKxS2HbQeCFTPEFH0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=fastly.com; spf=pass smtp.mailfrom=fastly.com; dkim=pass (1024-bit key) header.d=fastly.com header.i=@fastly.com header.b=LNteaX/L; arc=none smtp.client-ip=209.85.210.49 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=fastly.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=fastly.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=fastly.com header.i=@fastly.com header.b="LNteaX/L" Received: by mail-ot1-f49.google.com with SMTP id 46e09a7af769-6dc20b4595bso71537a34.0 for ; Thu, 25 Jan 2024 14:57:47 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fastly.com; s=google; t=1706223466; x=1706828266; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=VV39OAkRpCVAHLKUI3lU3BvBDLbvOa7qJKwdNWq/qLo=; b=LNteaX/L+A7wlIGmFAxlvRJaShq9jtlCsS7lt6LO3BAt07Nb19rEMQjRcCbK+Uam8B AzOj6bZyX0A1KQiqGpZHWYCmB10uu9k3eS+FBEcf8d71RhKk3zsQkwoij+2OQbtvxmKg xwJNF9rgsMOWKsDTUZWkAu5aK+9qfLiWBUDco= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1706223466; x=1706828266; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=VV39OAkRpCVAHLKUI3lU3BvBDLbvOa7qJKwdNWq/qLo=; b=DKZudb/ViVq6RcDUlEoB40XlTxat/KNSGgswtiHK1uALt/QPZVV+/PkfL+xi+e1hnB e2XQsYTboQdaQaqD9vmmVEiSfqHxHeM8wQ7qmSEzea+5cSoVPKGEggExQOWM6cRI16Yr Bmb2x/Ao2RfPmAvSkGFRS4EiNcmxuWNncu/mD3b2mIbNaIysaMvZACNFlto6EFSni4hh bYOl7WKCSpk1fc8+YfnORb2kp5JclN9L5k4sZull3KStSRdWalO81qUimoVOAlzjEYwt rCcfCxLXI5duQt6mZu9c+2v/scQJ2fg8+L6QHAyJi7Am2aTqH8dIGb0KKerC6z7mdg+A JrJg== X-Gm-Message-State: AOJu0Yxpt3cGN6ekVT9YUZons7qpBt3WbzwVp0HZnX8EqHvACtuz2kN2 QBxqtA16c1g5dthJ5pzCyTDhxnbyIJwKXe2ItPMioq3zWkiuKlEPzVKKWyWg3GgZN/bTcu8eaUk MPa3kXhxw7Qbpep3jNNWZ/T+oCQvDmhAPD9Z98sBwdgDsf/Z3S7ONwIXd1QcPQJOgjRzk//rLHn 5zhsHx95dUSl9B9Poc8lXb4HxYVdZ4PF27z2e3vBU75fD66w== X-Google-Smtp-Source: AGHT+IEBJKdnv/0PWa1kjfk6mL3RWJeaJjGcRTqPE4k+O8aLjsFMGw7r0OXSIIiMsItoxHdR43gtew== X-Received: by 2002:a05:6359:4588:b0:176:6270:8cb6 with SMTP id no8-20020a056359458800b0017662708cb6mr549318rwb.25.1706223466122; Thu, 25 Jan 2024 14:57:46 -0800 (PST) Received: from localhost.localdomain ([2620:11a:c018:0:ea8:be91:8d1:f59b]) by smtp.gmail.com with ESMTPSA id z24-20020a631918000000b005d68962e1a7sm19948pgl.24.2024.01.25.14.57.44 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 25 Jan 2024 14:57:45 -0800 (PST) From: Joe Damato To: linux-kernel@vger.kernel.org, netdev@vger.kernel.org Cc: chuck.lever@oracle.com, jlayton@kernel.org, linux-api@vger.kernel.org, brauner@kernel.org, edumazet@google.com, davem@davemloft.net, alexander.duyck@gmail.com, sridhar.samudrala@intel.com, kuba@kernel.org, willemdebruijn.kernel@gmail.com, weiwan@google.com, Joe Damato , Alexander Viro , Jan Kara , linux-fsdevel@vger.kernel.org (open list:FILESYSTEMS (VFS and infrastructure)) Subject: [PATCH net-next v3 1/3] eventpoll: support busy poll per epoll instance Date: Thu, 25 Jan 2024 22:56:57 +0000 Message-Id: <20240125225704.12781-2-jdamato@fastly.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240125225704.12781-1-jdamato@fastly.com> References: <20240125225704.12781-1-jdamato@fastly.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Allow busy polling on a per-epoll context basis. The per-epoll context usec timeout value is preferred, but the pre-existing system wide sysctl value is still supported if it specified. Note that this change uses an xor: either per epoll instance busy polling is enabled on the epoll instance or system wide epoll is enabled. Enabling both is disallowed. Signed-off-by: Joe Damato --- fs/eventpoll.c | 49 +++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 45 insertions(+), 4 deletions(-) diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 3534d36a1474..4503fec01278 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -227,6 +227,8 @@ struct eventpoll { #ifdef CONFIG_NET_RX_BUSY_POLL /* used to track busy poll napi_id */ unsigned int napi_id; + /* busy poll timeout */ + u64 busy_poll_usecs; #endif =20 #ifdef CONFIG_DEBUG_LOCK_ALLOC @@ -386,12 +388,44 @@ static inline int ep_events_available(struct eventpol= l *ep) READ_ONCE(ep->ovflist) !=3D EP_UNACTIVE_PTR; } =20 +/** + * busy_loop_ep_timeout - check if busy poll has timed out. The timeout va= lue + * from the epoll instance ep is preferred, but if it is not set fallback = to + * the system-wide global via busy_loop_timeout. + * + * @start_time: The start time used to compute the remaining time until ti= meout. + * @ep: Pointer to the eventpoll context. + * + * Return: true if the timeout has expired, false otherwise. + */ +static inline bool busy_loop_ep_timeout(unsigned long start_time, struct e= ventpoll *ep) +{ +#ifdef CONFIG_NET_RX_BUSY_POLL + unsigned long bp_usec =3D READ_ONCE(ep->busy_poll_usecs); + + if (bp_usec) { + unsigned long end_time =3D start_time + bp_usec; + unsigned long now =3D busy_loop_current_time(); + + return time_after(now, end_time); + } else { + return busy_loop_timeout(start_time); + } +#endif + return true; +} + #ifdef CONFIG_NET_RX_BUSY_POLL +static bool ep_busy_loop_on(struct eventpoll *ep) +{ + return !!ep->busy_poll_usecs ^ net_busy_loop_on(); +} + static bool ep_busy_loop_end(void *p, unsigned long start_time) { struct eventpoll *ep =3D p; =20 - return ep_events_available(ep) || busy_loop_timeout(start_time); + return ep_events_available(ep) || busy_loop_ep_timeout(start_time, ep); } =20 /* @@ -404,7 +438,7 @@ static bool ep_busy_loop(struct eventpoll *ep, int nonb= lock) { unsigned int napi_id =3D READ_ONCE(ep->napi_id); =20 - if ((napi_id >=3D MIN_NAPI_ID) && net_busy_loop_on()) { + if ((napi_id >=3D MIN_NAPI_ID) && ep_busy_loop_on(ep)) { napi_busy_loop(napi_id, nonblock ? NULL : ep_busy_loop_end, ep, false, BUSY_POLL_BUDGET); if (ep_events_available(ep)) @@ -430,7 +464,8 @@ static inline void ep_set_busy_poll_napi_id(struct epit= em *epi) struct socket *sock; struct sock *sk; =20 - if (!net_busy_loop_on()) + ep =3D epi->ep; + if (!ep_busy_loop_on(ep)) return; =20 sock =3D sock_from_file(epi->ffd.file); @@ -442,7 +477,6 @@ static inline void ep_set_busy_poll_napi_id(struct epit= em *epi) return; =20 napi_id =3D READ_ONCE(sk->sk_napi_id); - ep =3D epi->ep; =20 /* Non-NAPI IDs can be rejected * or @@ -466,6 +500,10 @@ static inline void ep_set_busy_poll_napi_id(struct epi= tem *epi) { } =20 +static inline bool ep_busy_loop_on(struct eventpoll *ep) +{ + return false; +} #endif /* CONFIG_NET_RX_BUSY_POLL */ =20 /* @@ -2058,6 +2096,9 @@ static int do_epoll_create(int flags) error =3D PTR_ERR(file); goto out_free_fd; } +#ifdef CONFIG_NET_RX_BUSY_POLL + ep->busy_poll_usecs =3D 0; +#endif ep->file =3D file; fd_install(fd, file); return fd; --=20 2.25.1