From patchwork Thu May 18 18:05:34 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maciej Fijalkowski X-Patchwork-Id: 13247227 X-Patchwork-Delegate: bpf@iogearbox.net Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id AF83B134A3; Thu, 18 May 2023 18:06:22 +0000 (UTC) Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5D928F9; Thu, 18 May 2023 11:06:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1684433181; x=1715969181; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Yf1SvWPjJp8YHvQYZtcKlU4iSGePPMdo2rMOgNXSfQ8=; b=jRCXw/rl6Lo+ossZl1exzmtPrsW7PDMSRXbyUsrNYA7z4+lCbN8o+DeU d+LBThF5jhkpH0WtPx0trzurM3pIVZAthL7NeNp/+oghpn5jJec7C+LPU gUMjVTohrX/A0BbN+T0f+4o/kW+1Yd2paTJ66h3OHEceFa528wN6f4BNG brHAotxsNdjMFnrb6i2GjQGNSbCGf3q9tkxFLtSuLIoa0caj3fSDGQePP CqI/Jjea5/Q41EQWD1y3gLN+M3Cdy+a4iWgkLXQA9/coQZikB/B5O9SJz U6Dk0sfqZUzyGHrDjhRA1cvtfDzhcB6NV40wpaSjkGVOerByFzcZ7OYxR g==; X-IronPort-AV: E=McAfee;i="6600,9927,10714"; a="350984900" X-IronPort-AV: E=Sophos;i="6.00,174,1681196400"; d="scan'208";a="350984900" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by fmsmga104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 18 May 2023 11:06:18 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10714"; a="948780336" X-IronPort-AV: E=Sophos;i="6.00,174,1681196400"; d="scan'208";a="948780336" Received: from boxer.igk.intel.com ([10.102.20.173]) by fmsmga006.fm.intel.com with ESMTP; 18 May 2023 11:06:16 -0700 From: Maciej Fijalkowski To: bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org Cc: netdev@vger.kernel.org, magnus.karlsson@intel.com, tirthendu.sarkar@intel.com, maciej.fijalkowski@intel.com, bjorn@kernel.org Subject: [PATCH bpf-next 10/21] xsk: support mbuf on ZC RX Date: Thu, 18 May 2023 20:05:34 +0200 Message-Id: <20230518180545.159100-11-maciej.fijalkowski@intel.com> X-Mailer: git-send-email 2.35.3 In-Reply-To: <20230518180545.159100-1-maciej.fijalkowski@intel.com> References: <20230518180545.159100-1-maciej.fijalkowski@intel.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Spam-Status: No, score=-4.4 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,RCVD_IN_DNSWL_MED, SPF_HELO_NONE,SPF_NONE,T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net X-Patchwork-Delegate: bpf@iogearbox.net Extend xdp_buff_xsk with two new list_head members - xskb_list and xskb_list_node. This means that each xdp_buff_xsk can be a node in the chain of these structs and also each xskb can carry the list by itself. This is needed so ZC drivers can add frags as xskb nodes which will make it possible to handle it both when producing AF_XDP Rx descriptors as well as freeing/recycling all the frags that a single frame carries. Speaking of latter, update xsk_buff_free() to take care of list nodes. For the former (adding as frags), introduce xsk_buff_add_frag() for ZC drivers usage that is going to be used to add a frag to the first xskb. xsk_buff_get_frag() will be utilized by XDP_TX and, on contrary, will return xdp_buff. One of the previous patches added a wrapper for ZC Rx so implement xskb list walk and production of Rx descriptors there. Signed-off-by: Maciej Fijalkowski --- include/net/xdp_sock_drv.h | 45 +++++++++++++++++++++++++++++++++++++ include/net/xsk_buff_pool.h | 2 ++ net/xdp/xsk.c | 24 +++++++++++++++++++- net/xdp/xsk_buff_pool.c | 2 ++ 4 files changed, 72 insertions(+), 1 deletion(-) diff --git a/include/net/xdp_sock_drv.h b/include/net/xdp_sock_drv.h index a3e18b958d93..1dbbe2b9c57e 100644 --- a/include/net/xdp_sock_drv.h +++ b/include/net/xdp_sock_drv.h @@ -108,10 +108,46 @@ static inline bool xsk_buff_can_alloc(struct xsk_buff_pool *pool, u32 count) static inline void xsk_buff_free(struct xdp_buff *xdp) { struct xdp_buff_xsk *xskb = container_of(xdp, struct xdp_buff_xsk, xdp); + struct xdp_buff_xsk *pos, *tmp; + if (likely(!xdp_buff_has_frags(xdp))) + goto out; + + list_for_each_entry_safe(pos, tmp, &xskb->xskb_list, xskb_list_node) { + list_del(&pos->xskb_list_node); + xp_free(pos); + } + + xdp_get_shared_info_from_buff(xdp)->nr_frags = 0; +out: xp_free(xskb); } +static inline void xsk_buff_add_frag(struct xdp_buff *first, + struct xdp_buff *xdp) +{ + struct xdp_buff_xsk *xskb = container_of(first, struct xdp_buff_xsk, xdp); + struct xdp_buff_xsk *frag = container_of(xdp, struct xdp_buff_xsk, xdp); + + list_add_tail(&frag->xskb_list_node, &xskb->xskb_list); +} + +static inline struct xdp_buff *xsk_buff_get_frag(struct xdp_buff *first) +{ + struct xdp_buff_xsk *xskb = container_of(first, struct xdp_buff_xsk, xdp); + struct xdp_buff *ret = NULL; + struct xdp_buff_xsk *frag; + + frag = list_first_entry_or_null(&xskb->xskb_list, struct xdp_buff_xsk, + xskb_list_node); + if (frag) { + list_del(&frag->xskb_list_node); + ret = &frag->xdp; + } + + return ret; +} + static inline void xsk_buff_set_size(struct xdp_buff *xdp, u32 size) { xdp->data = xdp->data_hard_start + XDP_PACKET_HEADROOM; @@ -265,6 +301,15 @@ static inline void xsk_buff_free(struct xdp_buff *xdp) { } +static inline void xsk_buff_add_frag(struct xdp_buff *first, + struct xdp_buff *xdp) +{ +} + +static inline struct xdp_buff *xsk_buff_get_frag(struct xdp_buff *first) +{ +} + static inline void xsk_buff_discard(struct xdp_buff *xdp) { } diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h index 4dcca163e076..8a430163512c 100644 --- a/include/net/xsk_buff_pool.h +++ b/include/net/xsk_buff_pool.h @@ -29,6 +29,8 @@ struct xdp_buff_xsk { struct xsk_buff_pool *pool; u64 orig_addr; struct list_head free_list_node; + struct list_head xskb_list_node; + struct list_head xskb_list; }; #define XSK_CHECK_PRIV_TYPE(t) BUILD_BUG_ON(sizeof(t) > offsetofend(struct xdp_buff_xsk, cb)) diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 3df635bb2a57..fa2034c1721b 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -155,8 +155,30 @@ static int __xsk_rcv_zc(struct xdp_sock *xs, struct xdp_buff_xsk *xskb, u32 len, static int xsk_rcv_zc(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len) { struct xdp_buff_xsk *xskb = container_of(xdp, struct xdp_buff_xsk, xdp); + u32 frags = xdp_buff_has_frags(xdp); + struct xdp_buff_xsk *pos, *tmp; + u32 contd = 0; + int err; + + if (frags) + contd = XDP_PKT_CONTD; - return __xsk_rcv_zc(xs, xskb, len, 0); + err = __xsk_rcv_zc(xs, xskb, len, contd); + if (err || likely(!frags)) + goto out; + + list_for_each_entry_safe(pos, tmp, &xskb->xskb_list, xskb_list_node) { + if (list_is_singular(&xskb->xskb_list)) + contd = 0; + len = pos->xdp.data_end - pos->xdp.data; + err = __xsk_rcv_zc(xs, pos, len, contd); + if (err) + return err; + list_del(&pos->xskb_list_node); + } + +out: + return err; } static void *xsk_copy_xdp_start(struct xdp_buff *from) diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index 26f6d304451e..0a9f8ea68de3 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -99,6 +99,8 @@ struct xsk_buff_pool *xp_create_and_assign_umem(struct xdp_sock *xs, xskb->pool = pool; xskb->xdp.frame_sz = umem->chunk_size - umem->headroom; INIT_LIST_HEAD(&xskb->free_list_node); + INIT_LIST_HEAD(&xskb->xskb_list_node); + INIT_LIST_HEAD(&xskb->xskb_list); if (pool->unaligned) pool->free_heads[i] = xskb; else