From nobody Mon Feb 9 09:57:27 2026 Received: from mail-pl1-f172.google.com (mail-pl1-f172.google.com [209.85.214.172]) (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 4D4C22147F6 for ; Thu, 9 Jan 2025 06:59:07 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736405950; cv=none; b=KgW9C1QliqDFSmd+Kwlcjjhv6f295Uhit7UGkrQdXSLaJv3R+l1k31tpoRuTWyP21su41WKera2Oxk3ABL5DtBdOsTg2PCpxZWUtbdgTtsoHpDZ6+sMXI+PQWhUcALWFzALvat3o6JVojhpUUPcwEj/lau2VsE7dPmVm6c49+uA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1736405950; c=relaxed/simple; bh=xUh8OsO82Lz+N74mTsrU5GQoEJf86sRiocBQFvqOGTc=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To; b=Ce6miVyYkGgqJfq7lACCTNHhOjP3f82soYM3yZ11Bu9xiCdg56wKYxH2f/LY0OQd8ZvsV5Gbkm+3OGsAaVgmPDVEAEIBaEfqHeuQXBsztc2325fXJa7oMfvyFot67TQcy9f5hK73phd2tv2BGH6W/TmvTChRRJEFmmRiiIpmyng= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=daynix.com; spf=pass smtp.mailfrom=daynix.com; dkim=pass (2048-bit key) header.d=daynix-com.20230601.gappssmtp.com header.i=@daynix-com.20230601.gappssmtp.com header.b=TEATUIR9; arc=none smtp.client-ip=209.85.214.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=daynix.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=daynix.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=daynix-com.20230601.gappssmtp.com header.i=@daynix-com.20230601.gappssmtp.com header.b="TEATUIR9" Received: by mail-pl1-f172.google.com with SMTP id d9443c01a7336-2166651f752so10837205ad.3 for ; Wed, 08 Jan 2025 22:59:07 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=daynix-com.20230601.gappssmtp.com; s=20230601; t=1736405947; x=1737010747; darn=vger.kernel.org; h=to:in-reply-to:references:message-id:content-transfer-encoding :mime-version:subject:date:from:from:to:cc:subject:date:message-id :reply-to; bh=oH2ziWEMe8qysnaJoX6kBUE58VGIgBpa5DNj4QNLj2A=; b=TEATUIR9FB2Ho5inyaAn39abOjCmmU5REYBTdQxdOIop94TucxSqgezGjlt6xIMjL5 LneHzhk8j91zF2jVQW5YCcBXP/y0S5XhaZ73d9VFDcR3h1qAHLbVpBapny8Q+YQY9wsC 100M23HU+fAbn9kIeVJBtoJf/kGcAm9QK4b6FyXcMJZrAjaR32C1jMzadZEyIlzvfCVU r/C33yZ/wPbFeIHAxME74RoK5TqKSxFrqTsMb5fAyf8FZeD+oj4c/+tyybGa/iH6xteI zjiZHvr7BLl+UB+YF75ziMXlENRiHTw9rhFvCtJouI9xZIyCE73c8X2SjKk/4zLsWV0m DVPQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1736405947; x=1737010747; h=to:in-reply-to:references:message-id:content-transfer-encoding :mime-version:subject:date:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=oH2ziWEMe8qysnaJoX6kBUE58VGIgBpa5DNj4QNLj2A=; b=mQT4tXbosW9+fb85YUGxsWl6CGdm13ePgDmOA8zsOheWbFhmpamnxMvhTsgf5N+gDg ZMZj3grUkWtoBOsn7MqP+2v/T6/9AU5NYLgSTgkgiMJMjU8rlQi7j2FhjnNLoUqJSZ77 jcGqZQiPiw96adANz50fZGmQkMS+i28P5ZYwViWGGJj+MTBJsG6r2jDBYroAE6giKQ5f GXxXokDz+BszghMrm/rNTR4CM8AnANmNlEqeT1YS/LxoU2jqOCcM4vvtT6ZQDgXGo0+Y CsdhpeG+umQ1IGtVtUY751R52b1MyayMYdC31VjBVk4v51ZfR+uUTXt7Fb6LBNZBI5xO 8zZQ== X-Forwarded-Encrypted: i=1; AJvYcCUfwtRieAo+O3zXUFU2VCTFVWQcb0t7fycx1dITzsBXz7vp7hba7EZPomKRIYjILQcFUR6Ukf4PWSrsO8E=@vger.kernel.org X-Gm-Message-State: AOJu0YyChXQ1QqVPaaHIyYsslIXU5nwNNjRBCJjM/pq6Zu/1q0ePO10J bt7mkkaW+/rjcVN8woagyLcDdmtnh8wMEuUFJ6fsRo+jCftmYr99W8GIs4Z2jPY= X-Gm-Gg: ASbGncvymsv5AylXasBVjjw8S5C/n6creQ8yo/98DfPdS0a4pw0Pi7vjnTpY7WHCaM9 kPJKWTK4WjgWHP1brbYxAJ1V637GipqmHyHk7NqO/uiCNLIQOSh0SybJe/cnUzXi6uKUzHwZCu6 zXsD8Q5jP5fCTiHkjIQfEfJwq4YVf5eeeD9lpUGMImEJuqBP8PgqUxUuvu29hkK++EUHL4cgUVI QKFtPohuht4D8r3NBChvd1t1zPQVRanxBBdtPXOtt9f0EwIVinVFFTSuPg= X-Google-Smtp-Source: AGHT+IGtW0ervT9Zb+HtfnwTwQjuMuiejW0TJN5qMd/qKY7fouxK2oNPYeK2WwBy9LO33JkH3d4YtQ== X-Received: by 2002:a05:6a20:a109:b0:1e1:d22d:bc01 with SMTP id adf61e73a8af0-1e88d361a7dmr10424069637.30.1736405946586; Wed, 08 Jan 2025 22:59:06 -0800 (PST) Received: from localhost ([157.82.203.37]) by smtp.gmail.com with UTF8SMTPSA id d2e1a72fcca58-72aad84cfb3sm36367664b3a.89.2025.01.08.22.59.01 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Wed, 08 Jan 2025 22:59:06 -0800 (PST) From: Akihiko Odaki Date: Thu, 09 Jan 2025 15:58:43 +0900 Subject: [PATCH v2 1/3] tun: Unify vnet implementation Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20250109-tun-v2-1-388d7d5a287a@daynix.com> References: <20250109-tun-v2-0-388d7d5a287a@daynix.com> In-Reply-To: <20250109-tun-v2-0-388d7d5a287a@daynix.com> To: Jonathan Corbet , Willem de Bruijn , Jason Wang , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , "Michael S. Tsirkin" , Xuan Zhuo , Shuah Khan , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, netdev@vger.kernel.org, kvm@vger.kernel.org, virtualization@lists.linux-foundation.org, linux-kselftest@vger.kernel.org, Yuri Benditovich , Andrew Melnychenko , Stephen Hemminger , gur.stavi@huawei.com, devel@daynix.com, Akihiko Odaki X-Mailer: b4 0.14-dev-fd6e3 Both tun and tap exposes the same set of virtio-net-related features. Unify their implementations to ease future changes. Signed-off-by: Akihiko Odaki --- MAINTAINERS | 1 + drivers/net/Kconfig | 5 ++ drivers/net/Makefile | 1 + drivers/net/tap.c | 172 ++++++---------------------------------- drivers/net/tun.c | 208 ++++++++-------------------------------------= ---- drivers/net/tun_vnet.c | 186 +++++++++++++++++++++++++++++++++++++++++++ drivers/net/tun_vnet.h | 24 ++++++ 7 files changed, 273 insertions(+), 324 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 910305c11e8a..1be8a452d11f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -23903,6 +23903,7 @@ F: Documentation/networking/tuntap.rst F: arch/um/os-Linux/drivers/ F: drivers/net/tap.c F: drivers/net/tun.c +F: drivers/net/tun_vnet.h =20 TURBOCHANNEL SUBSYSTEM M: "Maciej W. Rozycki" diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 1fd5acdc73c6..255c8f9f1d7c 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -395,6 +395,7 @@ config TUN tristate "Universal TUN/TAP device driver support" depends on INET select CRC32 + select TUN_VNET help TUN/TAP provides packet reception and transmission for user space programs. It can be viewed as a simple Point-to-Point or Ethernet @@ -417,10 +418,14 @@ config TUN =20 config TAP tristate + select TUN_VNET help This option is selected by any driver implementing tap user space interface for a virtual interface to re-use core tap functionality. =20 +config TUN_VNET + tristate + config TUN_VNET_CROSS_LE bool "Support for cross-endian vnet headers on little-endian kernels" default n diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 13743d0e83b5..bc1f193eccb1 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -30,6 +30,7 @@ obj-y +=3D pcs/ obj-$(CONFIG_RIONET) +=3D rionet.o obj-$(CONFIG_NET_TEAM) +=3D team/ obj-$(CONFIG_TUN) +=3D tun.o +obj-$(CONFIG_TUN_VNET) +=3D tun_vnet.o obj-$(CONFIG_TAP) +=3D tap.o obj-$(CONFIG_VETH) +=3D veth.o obj-$(CONFIG_VIRTIO_NET) +=3D virtio_net.o diff --git a/drivers/net/tap.c b/drivers/net/tap.c index 5aa41d5f7765..60804855510b 100644 --- a/drivers/net/tap.c +++ b/drivers/net/tap.c @@ -26,74 +26,9 @@ #include #include =20 -#define TAP_IFFEATURES (IFF_VNET_HDR | IFF_MULTI_QUEUE) - -#define TAP_VNET_LE 0x80000000 -#define TAP_VNET_BE 0x40000000 - -#ifdef CONFIG_TUN_VNET_CROSS_LE -static inline bool tap_legacy_is_little_endian(struct tap_queue *q) -{ - return q->flags & TAP_VNET_BE ? false : - virtio_legacy_is_little_endian(); -} - -static long tap_get_vnet_be(struct tap_queue *q, int __user *sp) -{ - int s =3D !!(q->flags & TAP_VNET_BE); - - if (put_user(s, sp)) - return -EFAULT; - - return 0; -} - -static long tap_set_vnet_be(struct tap_queue *q, int __user *sp) -{ - int s; - - if (get_user(s, sp)) - return -EFAULT; - - if (s) - q->flags |=3D TAP_VNET_BE; - else - q->flags &=3D ~TAP_VNET_BE; - - return 0; -} -#else -static inline bool tap_legacy_is_little_endian(struct tap_queue *q) -{ - return virtio_legacy_is_little_endian(); -} - -static long tap_get_vnet_be(struct tap_queue *q, int __user *argp) -{ - return -EINVAL; -} +#include "tun_vnet.h" =20 -static long tap_set_vnet_be(struct tap_queue *q, int __user *argp) -{ - return -EINVAL; -} -#endif /* CONFIG_TUN_VNET_CROSS_LE */ - -static inline bool tap_is_little_endian(struct tap_queue *q) -{ - return q->flags & TAP_VNET_LE || - tap_legacy_is_little_endian(q); -} - -static inline u16 tap16_to_cpu(struct tap_queue *q, __virtio16 val) -{ - return __virtio16_to_cpu(tap_is_little_endian(q), val); -} - -static inline __virtio16 cpu_to_tap16(struct tap_queue *q, u16 val) -{ - return __cpu_to_virtio16(tap_is_little_endian(q), val); -} +#define TAP_IFFEATURES (IFF_VNET_HDR | IFF_MULTI_QUEUE) =20 static struct proto tap_proto =3D { .name =3D "tap", @@ -641,10 +576,10 @@ static ssize_t tap_get_user(struct tap_queue *q, void= *msg_control, struct sk_buff *skb; struct tap_dev *tap; unsigned long total_len =3D iov_iter_count(from); - unsigned long len =3D total_len; + unsigned long len; int err; struct virtio_net_hdr vnet_hdr =3D { 0 }; - int vnet_hdr_len =3D 0; + int hdr_len; int copylen =3D 0; int depth; bool zerocopy =3D false; @@ -652,38 +587,20 @@ static ssize_t tap_get_user(struct tap_queue *q, void= *msg_control, enum skb_drop_reason drop_reason; =20 if (q->flags & IFF_VNET_HDR) { - vnet_hdr_len =3D READ_ONCE(q->vnet_hdr_sz); - - err =3D -EINVAL; - if (len < vnet_hdr_len) - goto err; - len -=3D vnet_hdr_len; - - err =3D -EFAULT; - if (!copy_from_iter_full(&vnet_hdr, sizeof(vnet_hdr), from)) - goto err; - iov_iter_advance(from, vnet_hdr_len - sizeof(vnet_hdr)); - if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && - tap16_to_cpu(q, vnet_hdr.csum_start) + - tap16_to_cpu(q, vnet_hdr.csum_offset) + 2 > - tap16_to_cpu(q, vnet_hdr.hdr_len)) - vnet_hdr.hdr_len =3D cpu_to_tap16(q, - tap16_to_cpu(q, vnet_hdr.csum_start) + - tap16_to_cpu(q, vnet_hdr.csum_offset) + 2); - err =3D -EINVAL; - if (tap16_to_cpu(q, vnet_hdr.hdr_len) > len) + hdr_len =3D tun_vnet_hdr_get(READ_ONCE(q->vnet_hdr_sz), q->flags, from, = &vnet_hdr); + if (hdr_len < 0) { + err =3D hdr_len; goto err; + } + } else { + hdr_len =3D 0; } =20 - err =3D -EINVAL; - if (unlikely(len < ETH_HLEN)) - goto err; - + len =3D iov_iter_count(from); if (msg_control && sock_flag(&q->sk, SOCK_ZEROCOPY)) { struct iov_iter i; =20 - copylen =3D vnet_hdr.hdr_len ? - tap16_to_cpu(q, vnet_hdr.hdr_len) : GOODCOPY_LEN; + copylen =3D hdr_len ? hdr_len : GOODCOPY_LEN; if (copylen > good_linear) copylen =3D good_linear; else if (copylen < ETH_HLEN) @@ -697,7 +614,7 @@ static ssize_t tap_get_user(struct tap_queue *q, void *= msg_control, =20 if (!zerocopy) { copylen =3D len; - linear =3D tap16_to_cpu(q, vnet_hdr.hdr_len); + linear =3D hdr_len; if (linear > good_linear) linear =3D good_linear; else if (linear < ETH_HLEN) @@ -732,9 +649,8 @@ static ssize_t tap_get_user(struct tap_queue *q, void *= msg_control, } skb->dev =3D tap->dev; =20 - if (vnet_hdr_len) { - err =3D virtio_net_hdr_to_skb(skb, &vnet_hdr, - tap_is_little_endian(q)); + if (q->flags & IFF_VNET_HDR) { + err =3D tun_vnet_hdr_to_skb(q->flags, skb, &vnet_hdr); if (err) { rcu_read_unlock(); drop_reason =3D SKB_DROP_REASON_DEV_HDR; @@ -797,23 +713,17 @@ static ssize_t tap_put_user(struct tap_queue *q, int total; =20 if (q->flags & IFF_VNET_HDR) { - int vlan_hlen =3D skb_vlan_tag_present(skb) ? VLAN_HLEN : 0; struct virtio_net_hdr vnet_hdr; =20 vnet_hdr_len =3D READ_ONCE(q->vnet_hdr_sz); - if (iov_iter_count(iter) < vnet_hdr_len) - return -EINVAL; - - if (virtio_net_hdr_from_skb(skb, &vnet_hdr, - tap_is_little_endian(q), true, - vlan_hlen)) - BUG(); =20 - if (copy_to_iter(&vnet_hdr, sizeof(vnet_hdr), iter) !=3D - sizeof(vnet_hdr)) - return -EFAULT; + ret =3D tun_vnet_hdr_from_skb(q->flags, NULL, skb, &vnet_hdr); + if (ret < 0) + goto done; =20 - iov_iter_advance(iter, vnet_hdr_len - sizeof(vnet_hdr)); + ret =3D tun_vnet_hdr_put(vnet_hdr_len, iter, &vnet_hdr); + if (ret < 0) + goto done; } total =3D vnet_hdr_len; total +=3D skb->len; @@ -1072,42 +982,6 @@ static long tap_ioctl(struct file *file, unsigned int= cmd, q->sk.sk_sndbuf =3D s; return 0; =20 - case TUNGETVNETHDRSZ: - s =3D q->vnet_hdr_sz; - if (put_user(s, sp)) - return -EFAULT; - return 0; - - case TUNSETVNETHDRSZ: - if (get_user(s, sp)) - return -EFAULT; - if (s < (int)sizeof(struct virtio_net_hdr)) - return -EINVAL; - - q->vnet_hdr_sz =3D s; - return 0; - - case TUNGETVNETLE: - s =3D !!(q->flags & TAP_VNET_LE); - if (put_user(s, sp)) - return -EFAULT; - return 0; - - case TUNSETVNETLE: - if (get_user(s, sp)) - return -EFAULT; - if (s) - q->flags |=3D TAP_VNET_LE; - else - q->flags &=3D ~TAP_VNET_LE; - return 0; - - case TUNGETVNETBE: - return tap_get_vnet_be(q, sp); - - case TUNSETVNETBE: - return tap_set_vnet_be(q, sp); - case TUNSETOFFLOAD: /* let the user check for future flags */ if (arg & ~(TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | @@ -1151,7 +1025,7 @@ static long tap_ioctl(struct file *file, unsigned int= cmd, return ret; =20 default: - return -EINVAL; + return tun_vnet_ioctl(&q->vnet_hdr_sz, &q->flags, cmd, sp); } } =20 @@ -1198,7 +1072,7 @@ static int tap_get_user_xdp(struct tap_queue *q, stru= ct xdp_buff *xdp) skb->protocol =3D eth_hdr(skb)->h_proto; =20 if (vnet_hdr_len) { - err =3D virtio_net_hdr_to_skb(skb, gso, tap_is_little_endian(q)); + err =3D tun_vnet_hdr_to_skb(q->flags, skb, gso); if (err) goto err_kfree; } diff --git a/drivers/net/tun.c b/drivers/net/tun.c index e816aaba8e5f..dbf0dee92e93 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -83,6 +83,8 @@ #include #include =20 +#include "tun_vnet.h" + static void tun_default_link_ksettings(struct net_device *dev, struct ethtool_link_ksettings *cmd); =20 @@ -94,9 +96,6 @@ static void tun_default_link_ksettings(struct net_device = *dev, * overload it to mean fasync when stored there. */ #define TUN_FASYNC IFF_ATTACH_QUEUE -/* High bits in flags field are unused. */ -#define TUN_VNET_LE 0x80000000 -#define TUN_VNET_BE 0x40000000 =20 #define TUN_FEATURES (IFF_NO_PI | IFF_ONE_QUEUE | IFF_VNET_HDR | \ IFF_MULTI_QUEUE | IFF_NAPI | IFF_NAPI_FRAGS) @@ -298,70 +297,6 @@ static bool tun_napi_frags_enabled(const struct tun_fi= le *tfile) return tfile->napi_frags_enabled; } =20 -#ifdef CONFIG_TUN_VNET_CROSS_LE -static inline bool tun_legacy_is_little_endian(struct tun_struct *tun) -{ - return tun->flags & TUN_VNET_BE ? false : - virtio_legacy_is_little_endian(); -} - -static long tun_get_vnet_be(struct tun_struct *tun, int __user *argp) -{ - int be =3D !!(tun->flags & TUN_VNET_BE); - - if (put_user(be, argp)) - return -EFAULT; - - return 0; -} - -static long tun_set_vnet_be(struct tun_struct *tun, int __user *argp) -{ - int be; - - if (get_user(be, argp)) - return -EFAULT; - - if (be) - tun->flags |=3D TUN_VNET_BE; - else - tun->flags &=3D ~TUN_VNET_BE; - - return 0; -} -#else -static inline bool tun_legacy_is_little_endian(struct tun_struct *tun) -{ - return virtio_legacy_is_little_endian(); -} - -static long tun_get_vnet_be(struct tun_struct *tun, int __user *argp) -{ - return -EINVAL; -} - -static long tun_set_vnet_be(struct tun_struct *tun, int __user *argp) -{ - return -EINVAL; -} -#endif /* CONFIG_TUN_VNET_CROSS_LE */ - -static inline bool tun_is_little_endian(struct tun_struct *tun) -{ - return tun->flags & TUN_VNET_LE || - tun_legacy_is_little_endian(tun); -} - -static inline u16 tun16_to_cpu(struct tun_struct *tun, __virtio16 val) -{ - return __virtio16_to_cpu(tun_is_little_endian(tun), val); -} - -static inline __virtio16 cpu_to_tun16(struct tun_struct *tun, u16 val) -{ - return __cpu_to_virtio16(tun_is_little_endian(tun), val); -} - static inline u32 tun_hashfn(u32 rxhash) { return rxhash & TUN_MASK_FLOW_ENTRIES; @@ -1752,8 +1687,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, s= truct tun_file *tfile, struct tun_pi pi =3D { 0, cpu_to_be16(ETH_P_IP) }; struct sk_buff *skb; size_t total_len =3D iov_iter_count(from); - size_t len =3D total_len, align =3D tun->align, linear; + size_t len, align =3D tun->align, linear; struct virtio_net_hdr gso =3D { 0 }; + int hdr_len; int good_linear; int copylen; bool zerocopy =3D false; @@ -1764,37 +1700,25 @@ static ssize_t tun_get_user(struct tun_struct *tun,= struct tun_file *tfile, enum skb_drop_reason drop_reason =3D SKB_DROP_REASON_NOT_SPECIFIED; =20 if (!(tun->flags & IFF_NO_PI)) { - if (len < sizeof(pi)) + if (iov_iter_count(from) < sizeof(pi)) return -EINVAL; - len -=3D sizeof(pi); =20 if (!copy_from_iter_full(&pi, sizeof(pi), from)) return -EFAULT; } =20 if (tun->flags & IFF_VNET_HDR) { - int vnet_hdr_sz =3D READ_ONCE(tun->vnet_hdr_sz); - - if (len < vnet_hdr_sz) - return -EINVAL; - len -=3D vnet_hdr_sz; - - if (!copy_from_iter_full(&gso, sizeof(gso), from)) - return -EFAULT; - - if ((gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && - tun16_to_cpu(tun, gso.csum_start) + tun16_to_cpu(tun, gso.csum_offse= t) + 2 > tun16_to_cpu(tun, gso.hdr_len)) - gso.hdr_len =3D cpu_to_tun16(tun, tun16_to_cpu(tun, gso.csum_start) + t= un16_to_cpu(tun, gso.csum_offset) + 2); - - if (tun16_to_cpu(tun, gso.hdr_len) > len) - return -EINVAL; - iov_iter_advance(from, vnet_hdr_sz - sizeof(gso)); + hdr_len =3D tun_vnet_hdr_get(READ_ONCE(tun->vnet_hdr_sz), tun->flags, fr= om, &gso); + if (hdr_len < 0) + return hdr_len; + } else { + hdr_len =3D 0; } =20 + len =3D iov_iter_count(from); if ((tun->flags & TUN_TYPE_MASK) =3D=3D IFF_TAP) { align +=3D NET_IP_ALIGN; - if (unlikely(len < ETH_HLEN || - (gso.hdr_len && tun16_to_cpu(tun, gso.hdr_len) < ETH_HLEN))) + if (unlikely(len < ETH_HLEN || (hdr_len && hdr_len < ETH_HLEN))) return -EINVAL; } =20 @@ -1807,7 +1731,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, s= truct tun_file *tfile, * enough room for skb expand head in case it is used. * The rest of the buffer is mapped from userspace. */ - copylen =3D gso.hdr_len ? tun16_to_cpu(tun, gso.hdr_len) : GOODCOPY_LEN; + copylen =3D hdr_len ? hdr_len : GOODCOPY_LEN; if (copylen > good_linear) copylen =3D good_linear; linear =3D copylen; @@ -1830,10 +1754,10 @@ static ssize_t tun_get_user(struct tun_struct *tun,= struct tun_file *tfile, } else { if (!zerocopy) { copylen =3D len; - if (tun16_to_cpu(tun, gso.hdr_len) > good_linear) + if (hdr_len > good_linear) linear =3D good_linear; else - linear =3D tun16_to_cpu(tun, gso.hdr_len); + linear =3D hdr_len; } =20 if (frags) { @@ -1868,7 +1792,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, s= truct tun_file *tfile, } } =20 - if (virtio_net_hdr_to_skb(skb, &gso, tun_is_little_endian(tun))) { + if (tun_vnet_hdr_to_skb(tun->flags, skb, &gso)) { atomic_long_inc(&tun->rx_frame_errors); err =3D -EINVAL; goto free_skb; @@ -2061,29 +1985,27 @@ static ssize_t tun_put_user_xdp(struct tun_struct *= tun, struct xdp_frame *xdp_frame, struct iov_iter *iter) { + int ret; int vnet_hdr_sz =3D 0; size_t size =3D xdp_frame->len; - size_t ret; + size_t total; =20 if (tun->flags & IFF_VNET_HDR) { struct virtio_net_hdr gso =3D { 0 }; =20 vnet_hdr_sz =3D READ_ONCE(tun->vnet_hdr_sz); - if (unlikely(iov_iter_count(iter) < vnet_hdr_sz)) - return -EINVAL; - if (unlikely(copy_to_iter(&gso, sizeof(gso), iter) !=3D - sizeof(gso))) - return -EFAULT; - iov_iter_advance(iter, vnet_hdr_sz - sizeof(gso)); + ret =3D tun_vnet_hdr_put(vnet_hdr_sz, iter, &gso); + if (ret < 0) + return ret; } =20 - ret =3D copy_to_iter(xdp_frame->data, size, iter) + vnet_hdr_sz; + total =3D copy_to_iter(xdp_frame->data, size, iter) + vnet_hdr_sz; =20 preempt_disable(); - dev_sw_netstats_tx_add(tun->dev, 1, ret); + dev_sw_netstats_tx_add(tun->dev, 1, total); preempt_enable(); =20 - return ret; + return total; } =20 /* Put packet to the user space buffer */ @@ -2097,6 +2019,7 @@ static ssize_t tun_put_user(struct tun_struct *tun, int vlan_offset =3D 0; int vlan_hlen =3D 0; int vnet_hdr_sz =3D 0; + int ret; =20 if (skb_vlan_tag_present(skb)) vlan_hlen =3D VLAN_HLEN; @@ -2123,31 +2046,13 @@ static ssize_t tun_put_user(struct tun_struct *tun, if (vnet_hdr_sz) { struct virtio_net_hdr gso; =20 - if (iov_iter_count(iter) < vnet_hdr_sz) - return -EINVAL; - - if (virtio_net_hdr_from_skb(skb, &gso, - tun_is_little_endian(tun), true, - vlan_hlen)) { - struct skb_shared_info *sinfo =3D skb_shinfo(skb); - - if (net_ratelimit()) { - netdev_err(tun->dev, "unexpected GSO type: 0x%x, gso_size %d, hdr_len = %d\n", - sinfo->gso_type, tun16_to_cpu(tun, gso.gso_size), - tun16_to_cpu(tun, gso.hdr_len)); - print_hex_dump(KERN_ERR, "tun: ", - DUMP_PREFIX_NONE, - 16, 1, skb->head, - min((int)tun16_to_cpu(tun, gso.hdr_len), 64), true); - } - WARN_ON_ONCE(1); - return -EINVAL; - } - - if (copy_to_iter(&gso, sizeof(gso), iter) !=3D sizeof(gso)) - return -EFAULT; + ret =3D tun_vnet_hdr_from_skb(tun->flags, tun->dev, skb, &gso); + if (ret < 0) + goto done; =20 - iov_iter_advance(iter, vnet_hdr_sz - sizeof(gso)); + ret =3D tun_vnet_hdr_put(vnet_hdr_sz, iter, &gso); + if (ret < 0) + goto done; } =20 if (vlan_hlen) { @@ -2507,7 +2412,7 @@ static int tun_xdp_one(struct tun_struct *tun, skb_reserve(skb, xdp->data - xdp->data_hard_start); skb_put(skb, xdp->data_end - xdp->data); =20 - if (virtio_net_hdr_to_skb(skb, gso, tun_is_little_endian(tun))) { + if (tun_vnet_hdr_to_skb(tun->flags, skb, gso)) { atomic_long_inc(&tun->rx_frame_errors); kfree_skb(skb); ret =3D -EINVAL; @@ -3091,8 +2996,6 @@ static long __tun_chr_ioctl(struct file *file, unsign= ed int cmd, kgid_t group; int ifindex; int sndbuf; - int vnet_hdr_sz; - int le; int ret; bool do_notify =3D false; =20 @@ -3299,50 +3202,6 @@ static long __tun_chr_ioctl(struct file *file, unsig= ned int cmd, tun_set_sndbuf(tun); break; =20 - case TUNGETVNETHDRSZ: - vnet_hdr_sz =3D tun->vnet_hdr_sz; - if (copy_to_user(argp, &vnet_hdr_sz, sizeof(vnet_hdr_sz))) - ret =3D -EFAULT; - break; - - case TUNSETVNETHDRSZ: - if (copy_from_user(&vnet_hdr_sz, argp, sizeof(vnet_hdr_sz))) { - ret =3D -EFAULT; - break; - } - if (vnet_hdr_sz < (int)sizeof(struct virtio_net_hdr)) { - ret =3D -EINVAL; - break; - } - - tun->vnet_hdr_sz =3D vnet_hdr_sz; - break; - - case TUNGETVNETLE: - le =3D !!(tun->flags & TUN_VNET_LE); - if (put_user(le, (int __user *)argp)) - ret =3D -EFAULT; - break; - - case TUNSETVNETLE: - if (get_user(le, (int __user *)argp)) { - ret =3D -EFAULT; - break; - } - if (le) - tun->flags |=3D TUN_VNET_LE; - else - tun->flags &=3D ~TUN_VNET_LE; - break; - - case TUNGETVNETBE: - ret =3D tun_get_vnet_be(tun, argp); - break; - - case TUNSETVNETBE: - ret =3D tun_set_vnet_be(tun, argp); - break; - case TUNATTACHFILTER: /* Can be set only for TAPs */ ret =3D -EINVAL; @@ -3398,8 +3257,7 @@ static long __tun_chr_ioctl(struct file *file, unsign= ed int cmd, break; =20 default: - ret =3D -EINVAL; - break; + ret =3D tun_vnet_ioctl(&tun->vnet_hdr_sz, &tun->flags, cmd, argp); } =20 if (do_notify) diff --git a/drivers/net/tun_vnet.c b/drivers/net/tun_vnet.c new file mode 100644 index 000000000000..fe842df9e9ef --- /dev/null +++ b/drivers/net/tun_vnet.c @@ -0,0 +1,186 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +#include "tun_vnet.h" + +/* High bits in flags field are unused. */ +#define TUN_VNET_LE 0x80000000 +#define TUN_VNET_BE 0x40000000 + +static bool tun_vnet_legacy_is_little_endian(unsigned int flags) +{ + return !(IS_ENABLED(CONFIG_TUN_VNET_CROSS_LE) && (flags & TUN_VNET_BE)) && + virtio_legacy_is_little_endian(); +} + +static long tun_vnet_get_be(int flags, int __user *sp) +{ + int s =3D !!(flags & TUN_VNET_BE); + + if (!IS_ENABLED(CONFIG_TUN_VNET_CROSS_LE)) + return -EINVAL; + + if (put_user(s, sp)) + return -EFAULT; + + return 0; +} + +static long tun_vnet_set_be(int *flags, int __user *sp) +{ + int s; + + if (!IS_ENABLED(CONFIG_TUN_VNET_CROSS_LE)) + return -EINVAL; + + if (get_user(s, sp)) + return -EFAULT; + + if (s) + *flags |=3D TUN_VNET_BE; + else + *flags &=3D ~TUN_VNET_BE; + + return 0; +} + +static bool tun_vnet_is_little_endian(unsigned int flags) +{ + return flags & TUN_VNET_LE || tun_vnet_legacy_is_little_endian(flags); +} + +static u16 tun_vnet16_to_cpu(unsigned int flags, __virtio16 val) +{ + return __virtio16_to_cpu(tun_vnet_is_little_endian(flags), val); +} + +static __virtio16 cpu_to_tun_vnet16(unsigned int flags, u16 val) +{ + return __cpu_to_virtio16(tun_vnet_is_little_endian(flags), val); +} + +long tun_vnet_ioctl(int *sz, unsigned int *flags, + unsigned int cmd, int __user *sp) +{ + int s; + + switch (cmd) { + case TUNGETVNETHDRSZ: + s =3D *sz; + if (put_user(s, sp)) + return -EFAULT; + return 0; + + case TUNSETVNETHDRSZ: + if (get_user(s, sp)) + return -EFAULT; + if (s < (int)sizeof(struct virtio_net_hdr)) + return -EINVAL; + + *sz =3D s; + return 0; + + case TUNGETVNETLE: + s =3D !!(*flags & TUN_VNET_LE); + if (put_user(s, sp)) + return -EFAULT; + return 0; + + case TUNSETVNETLE: + if (get_user(s, sp)) + return -EFAULT; + if (s) + *flags |=3D TUN_VNET_LE; + else + *flags &=3D ~TUN_VNET_LE; + return 0; + + case TUNGETVNETBE: + return tun_vnet_get_be(*flags, sp); + + case TUNSETVNETBE: + return tun_vnet_set_be(flags, sp); + + default: + return -EINVAL; + } +} +EXPORT_SYMBOL_GPL(tun_vnet_ioctl); + +int tun_vnet_hdr_get(int sz, unsigned int flags, struct iov_iter *from, + struct virtio_net_hdr *hdr) +{ + if (iov_iter_count(from) < sz) + return -EINVAL; + + if (!copy_from_iter_full(hdr, sizeof(*hdr), from)) + return -EFAULT; + + iov_iter_advance(from, sz - sizeof(*hdr)); + if ((hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) && + tun_vnet16_to_cpu(flags, hdr->csum_start) + + tun_vnet16_to_cpu(flags, hdr->csum_offset) + 2 > + tun_vnet16_to_cpu(flags, hdr->hdr_len)) + hdr->hdr_len =3D cpu_to_tun_vnet16(flags, + tun_vnet16_to_cpu(flags, hdr->csum_start) + + tun_vnet16_to_cpu(flags, hdr->csum_offset) + 2); + if (tun_vnet16_to_cpu(flags, hdr->hdr_len) > iov_iter_count(from)) + return -EINVAL; + + return tun_vnet16_to_cpu(flags, hdr->hdr_len); +} +EXPORT_SYMBOL_GPL(tun_vnet_hdr_get); + +int tun_vnet_hdr_put(int sz, struct iov_iter *iter, + const struct virtio_net_hdr *hdr) +{ + if (iov_iter_count(iter) < sz) + return -EINVAL; + + if (copy_to_iter(hdr, sizeof(*hdr), iter) !=3D sizeof(*hdr)) + return -EFAULT; + + iov_iter_advance(iter, sz - sizeof(*hdr)); + + return 0; +} +EXPORT_SYMBOL_GPL(tun_vnet_hdr_put); + +int tun_vnet_hdr_to_skb(unsigned int flags, struct sk_buff *skb, + const struct virtio_net_hdr *hdr) +{ + return virtio_net_hdr_to_skb(skb, hdr, tun_vnet_is_little_endian(flags)); +} +EXPORT_SYMBOL_GPL(tun_vnet_hdr_to_skb); + +int tun_vnet_hdr_from_skb(unsigned int flags, const struct net_device *dev, + const struct sk_buff *skb, + struct virtio_net_hdr *hdr) +{ + int vlan_hlen =3D skb_vlan_tag_present(skb) ? VLAN_HLEN : 0; + + if (virtio_net_hdr_from_skb(skb, hdr, + tun_vnet_is_little_endian(flags), true, + vlan_hlen)) { + struct skb_shared_info *sinfo =3D skb_shinfo(skb); + + if (net_ratelimit()) { + netdev_err(dev, "unexpected GSO type: 0x%x, gso_size %d, hdr_len %d\n", + sinfo->gso_type, tun_vnet16_to_cpu(flags, hdr->gso_size), + tun_vnet16_to_cpu(flags, hdr->hdr_len)); + print_hex_dump(KERN_ERR, "tun: ", + DUMP_PREFIX_NONE, + 16, 1, skb->head, + min(tun_vnet16_to_cpu(flags, hdr->hdr_len), 64), true); + } + WARN_ON_ONCE(1); + return -EINVAL; + } + + return 0; +} +EXPORT_SYMBOL_GPL(tun_vnet_hdr_from_skb); + +MODULE_DESCRIPTION("Common library for drivers implementing TUN/TAP's virt= io-related features"); +MODULE_AUTHOR("Max Krasnyansky "); +MODULE_AUTHOR("Arnd Bergmann "); +MODULE_AUTHOR("Sainath Grandhi "); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/tun_vnet.h b/drivers/net/tun_vnet.h new file mode 100644 index 000000000000..2dfdbe92bb24 --- /dev/null +++ b/drivers/net/tun_vnet.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef TUN_VNET_H +#define TUN_VNET_H + +#include +#include + +long tun_vnet_ioctl(int *sz, unsigned int *flags, + unsigned int cmd, int __user *sp); + +int tun_vnet_hdr_get(int sz, unsigned int flags, struct iov_iter *from, + struct virtio_net_hdr *hdr); + +int tun_vnet_hdr_put(int sz, struct iov_iter *iter, + const struct virtio_net_hdr *hdr); + +int tun_vnet_hdr_to_skb(unsigned int flags, struct sk_buff *skb, + const struct virtio_net_hdr *hdr); + +int tun_vnet_hdr_from_skb(unsigned int flags, const struct net_device *dev, + const struct sk_buff *skb, + struct virtio_net_hdr *hdr); + +#endif /* TUN_VNET_H */ --=20 2.47.1