From patchwork Tue Sep 26 11:17:14 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hayes Wang X-Patchwork-Id: 144826 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a59:cae8:0:b0:403:3b70:6f57 with SMTP id r8csp1836953vqu; Tue, 26 Sep 2023 04:25:23 -0700 (PDT) X-Google-Smtp-Source: AGHT+IHpnQIH/v1UMOtmJ/zeUoMoU9fJHoz/eHEn25AFv6rxB2xwVgzExLRYwTdTMThWtQ8tVCTk X-Received: by 2002:a05:6870:d287:b0:1bf:df47:7b5d with SMTP id d7-20020a056870d28700b001bfdf477b5dmr9365174oae.24.1695727522945; Tue, 26 Sep 2023 04:25:22 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1695727522; cv=none; d=google.com; s=arc-20160816; b=a2yNLNog7rL56X7PxkVl/ymzF9UYAKDn6XFfNkueuLrBQoauurISBTK2phOqcAfoWT cl1syy6o+HUvcqhV4DXGCv6QFlr2315uQQVRJbvWPcMWbSnsfRvoPorYxPMjrSKmy+qK WyRhW0OpeZP6jCV+uEqOBtBz6zkY+HGaICwqp6tgYvlQTr1LCPvzg7LkudxnUtm9EUCi 4Zhl9EBK9uy/0Iq1Ttnpt0HnDVTswnF4xibAf80q+py3hK0q8JTKqjt2T/71VH5JV4qd JtsJ7UbfNv43qnEcwDG3fXAfu5uTGw+8n2ri8Cf9wnZadnGq7j3FWiRgqjvAtAUuGuso YWZg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :references:in-reply-to:message-id:date:subject:cc:to:from; bh=RRo3QZux3e9khO05WnBqhkDbWtJpQjoNBnfV8VOE5OU=; fh=ihIYiilCZXBu9x8WrnA4PYm4Oe+ipaGNn90/3600c/4=; b=e2+1k5PCbtA2siUYj5aABRY3wab5bhwzhFCl9CPfsjZtcbA46elM2j5HfAwI8q6HUv izHlbLvNPxWTsZpYvT4ByaOtc5AxJaJx6TJoLGrR8jWjQ61NRm/BNBC/04heW7MGN08m A4o+UjBJBz9PsMKm+ky7IMbSFIPkGTrI+CNFO17H5bMu/yFZ3dXDDlUQocYmxaABvU5L WrILPKweLpbzWSM5nRBVzaG0MZQxiRTDFJeJHIJhWfJLLQv9dx+jkYLNfjY4z978vKGC WYb+qsws2pQwIHhmM9CgHsC7iI8vbELDBcEPdASMcplII+ezo5whz47xw0bxSmb1K5rr qQTA== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: from howler.vger.email (howler.vger.email. [23.128.96.34]) by mx.google.com with ESMTPS id 19-20020a631753000000b005789f94cd34si12260566pgx.636.2023.09.26.04.25.22 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 26 Sep 2023 04:25:22 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) client-ip=23.128.96.34; Authentication-Results: mx.google.com; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: from out1.vger.email (depot.vger.email [IPv6:2620:137:e000::3:0]) by howler.vger.email (Postfix) with ESMTP id 86C808348234; Tue, 26 Sep 2023 04:18:42 -0700 (PDT) X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.103.10 at howler.vger.email Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231899AbjIZLSn (ORCPT + 27 others); Tue, 26 Sep 2023 07:18:43 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:33892 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231517AbjIZLSj (ORCPT ); Tue, 26 Sep 2023 07:18:39 -0400 Received: from rtits2.realtek.com.tw (rtits2.realtek.com [211.75.126.72]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7DB6211F; Tue, 26 Sep 2023 04:18:31 -0700 (PDT) X-SpamFilter-By: ArmorX SpamTrap 5.78 with qID 38QBHkQpA857841, This message is accepted by code: ctloc85258 Received: from mail.realtek.com (rtexh36505.realtek.com.tw[172.21.6.25]) by rtits2.realtek.com.tw (8.15.2/2.92/5.92) with ESMTPS id 38QBHkQpA857841 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Tue, 26 Sep 2023 19:17:46 +0800 Received: from RTEXMBS04.realtek.com.tw (172.21.6.97) by RTEXH36505.realtek.com.tw (172.21.6.25) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.32; Tue, 26 Sep 2023 19:17:44 +0800 Received: from fc38.localdomain (172.22.228.98) by RTEXMBS04.realtek.com.tw (172.21.6.97) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.7; Tue, 26 Sep 2023 19:17:43 +0800 From: Hayes Wang To: , CC: , , , , , , , Hayes Wang Subject: [PATCH net-next v3 2/2] r8152: use napi_gro_frags Date: Tue, 26 Sep 2023 19:17:14 +0800 Message-ID: <20230926111714.9448-434-nic_swsd@realtek.com> X-Mailer: git-send-email 2.41.0 In-Reply-To: <20230926111714.9448-432-nic_swsd@realtek.com> References: <20230926111714.9448-432-nic_swsd@realtek.com> MIME-Version: 1.0 X-Originating-IP: [172.22.228.98] X-ClientProxiedBy: RTEXH36506.realtek.com.tw (172.21.6.27) To RTEXMBS04.realtek.com.tw (172.21.6.97) X-KSE-ServerInfo: RTEXMBS04.realtek.com.tw, 9 X-KSE-AntiSpam-Interceptor-Info: fallback X-KSE-Antivirus-Interceptor-Info: fallback X-KSE-AntiSpam-Interceptor-Info: fallback X-KSE-ServerInfo: RTEXH36505.realtek.com.tw, 9 X-KSE-AntiSpam-Interceptor-Info: fallback X-KSE-Antivirus-Interceptor-Info: fallback X-KSE-AntiSpam-Interceptor-Info: fallback X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,SPF_HELO_NONE, SPF_PASS autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-Greylist: Sender passed SPF test, not delayed by milter-greylist-4.6.4 (howler.vger.email [0.0.0.0]); Tue, 26 Sep 2023 04:18:42 -0700 (PDT) X-getmail-retrieved-from-mailbox: INBOX X-GMAIL-THRID: 1778099183434356823 X-GMAIL-MSGID: 1778099183434356823 Use napi_gro_frags() for the skb of fragments when the work_done is less than budget. Signed-off-by: Hayes Wang --- drivers/net/usb/r8152.c | 67 ++++++++++++++++++++++++++++++----------- 1 file changed, 50 insertions(+), 17 deletions(-) diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index c4038def193f..1231bf365796 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2490,8 +2490,9 @@ static int rx_bottom(struct r8152 *tp, int budget) while (urb->actual_length > len_used) { struct net_device *netdev = tp->netdev; struct net_device_stats *stats = &netdev->stats; - unsigned int pkt_len, rx_frag_head_sz; + unsigned int pkt_len, rx_frag_head_sz, len; struct sk_buff *skb; + bool use_frags; WARN_ON_ONCE(skb_queue_len(&tp->rx_queue) >= 1000); @@ -2504,45 +2505,77 @@ static int rx_bottom(struct r8152 *tp, int budget) break; pkt_len -= ETH_FCS_LEN; + len = pkt_len; rx_data += sizeof(struct rx_desc); - if (!agg_free || tp->rx_copybreak > pkt_len) - rx_frag_head_sz = pkt_len; + if (!agg_free || tp->rx_copybreak > len) + use_frags = false; else - rx_frag_head_sz = tp->rx_copybreak; + use_frags = true; + + if (use_frags) { + /* If the budget is exhausted, the packet + * would be queued in the driver. That is, + * napi_gro_frags() wouldn't be called, so + * we couldn't use napi_get_frags(). + */ + if (work_done >= budget) { + rx_frag_head_sz = tp->rx_copybreak; + skb = napi_alloc_skb(napi, + rx_frag_head_sz); + } else { + rx_frag_head_sz = 0; + skb = napi_get_frags(napi); + } + } else { + rx_frag_head_sz = 0; + skb = napi_alloc_skb(napi, len); + } - skb = napi_alloc_skb(napi, rx_frag_head_sz); if (!skb) { stats->rx_dropped++; goto find_next_rx; } skb->ip_summed = r8152_rx_csum(tp, rx_desc); - memcpy(skb->data, rx_data, rx_frag_head_sz); - skb_put(skb, rx_frag_head_sz); - pkt_len -= rx_frag_head_sz; - rx_data += rx_frag_head_sz; - if (pkt_len) { + rtl_rx_vlan_tag(rx_desc, skb); + + if (use_frags) { + if (rx_frag_head_sz) { + memcpy(skb->data, rx_data, + rx_frag_head_sz); + skb_put(skb, rx_frag_head_sz); + len -= rx_frag_head_sz; + rx_data += rx_frag_head_sz; + skb->protocol = eth_type_trans(skb, + netdev); + } + skb_add_rx_frag(skb, 0, agg->page, agg_offset(agg, rx_data), - pkt_len, - SKB_DATA_ALIGN(pkt_len)); + len, SKB_DATA_ALIGN(len)); get_page(agg->page); + } else { + memcpy(skb->data, rx_data, len); + skb_put(skb, len); + skb->protocol = eth_type_trans(skb, netdev); } - skb->protocol = eth_type_trans(skb, netdev); - rtl_rx_vlan_tag(rx_desc, skb); if (work_done < budget) { + if (use_frags) + napi_gro_frags(napi); + else + napi_gro_receive(napi, skb); + work_done++; stats->rx_packets++; - stats->rx_bytes += skb->len; - napi_gro_receive(napi, skb); + stats->rx_bytes += pkt_len; } else { __skb_queue_tail(&tp->rx_queue, skb); } find_next_rx: - rx_data = rx_agg_align(rx_data + pkt_len + ETH_FCS_LEN); + rx_data = rx_agg_align(rx_data + len + ETH_FCS_LEN); rx_desc = (struct rx_desc *)rx_data; len_used = agg_offset(agg, rx_data); len_used += sizeof(struct rx_desc);