[{"id":3674915,"web_url":"http://patchwork.ozlabs.org/comment/3674915/","msgid":"<adZGX5nA76feGRuV@redhat.com>","list_archive_url":null,"date":"2026-04-08T12:13:19","subject":"Re: [RFC v4 5/5] chardev/socket: add AF_PACKET capture path","submitter":{"id":2694,"url":"http://patchwork.ozlabs.org/api/people/2694/","name":"Daniel P. Berrangé","email":"berrange@redhat.com"},"content":"On Tue, Apr 07, 2026 at 01:05:52PM +0800, Cindy Lu wrote:\n> Add the AF_PACKET capture read path for socket chardevs. When opened\n> with af-packet-mode=capture, the read side drains raw frames with\n> recvfrom(), keeps only PACKET_OUTGOING traffic, and feeds the result\n> through the normal chardev frontend interface.\n> \n> Signed-off-by: Cindy Lu <lulu@redhat.com>\n> ---\n>  chardev/char-socket.c | 133 +++++++++++++++++++++++++++++++++++++++++-\n>  1 file changed, 131 insertions(+), 2 deletions(-)\n> \n> diff --git a/chardev/char-socket.c b/chardev/char-socket.c\n> index 45d06fda8f..76a51a853d 100644\n> --- a/chardev/char-socket.c\n> +++ b/chardev/char-socket.c\n> @@ -107,9 +107,17 @@ static void tcp_chr_accept(QIONetListener *listener,\n>  \n>  static int tcp_chr_read_poll(void *opaque);\n>  static void tcp_chr_disconnect_locked(Chardev *chr);\n> +static void tcp_chr_deliver_af_packet(Chardev *chr);\n>  \n>  #define TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE 65536\n>  \n> +static bool\n> +tcp_chr_uses_af_packet_capture(SocketChardev *s)\n> +{\n> +    return s->is_af_packet && s->af_packet_mode_set &&\n> +           s->af_packet_mode == CHARDEV_SOCKET_AF_PACKET_MODE_CAPTURE;\n> +}\n> +\n>  static bool tcp_chr_uses_af_packet_inject(SocketChardev *s)\n>  {\n>      return s->is_af_packet &&\n> @@ -300,6 +308,9 @@ static int tcp_chr_read_poll(void *opaque)\n>          return 0;\n>      }\n>      s->max_size = qemu_chr_be_can_write(chr);\n> +    if (tcp_chr_uses_af_packet_capture(s) && s->af_packet_buf_len) {\n> +        tcp_chr_deliver_af_packet(chr);\n> +    }\n>      return s->max_size;\n>  }\n>  \n> @@ -500,6 +511,98 @@ static void tcp_chr_reset_af_packet_send(SocketChardev *s)\n>      s->af_packet_send_len_bytes = 0;\n>  }\n>  \n> +/* Push buffered AF_PACKET capture data into the chardev frontend. */\n> +static void\n> +tcp_chr_deliver_af_packet(Chardev *chr)\n> +{\n> +    SocketChardev *s = SOCKET_CHARDEV(chr);\n> +\n> +    while (s->max_size > 0 && s->af_packet_buf_offset < s->af_packet_buf_len) {\n> +        size_t remaining = s->af_packet_buf_len - s->af_packet_buf_offset;\n> +        size_t chunk = MIN((size_t)s->max_size, remaining);\n> +\n> +        qemu_chr_be_write(chr, s->af_packet_buf + s->af_packet_buf_offset,\n> +                          (int)chunk);\n> +        s->af_packet_buf_offset += chunk;\n> +        s->max_size = qemu_chr_be_can_write(chr);\n> +    }\n> +\n> +    if (s->af_packet_buf_offset == s->af_packet_buf_len) {\n> +        tcp_chr_reset_af_packet_buf(s);\n> +    }\n> +}\n> +\n> +/* Copy buffered AF_PACKET capture data into a synchronous read buffer. */\n> +static int tcp_chr_copy_af_packet_buf(SocketChardev *s, uint8_t *buf,\n> +                                      int len) {\n> +    size_t remaining = s->af_packet_buf_len - s->af_packet_buf_offset;\n> +    size_t copied = MIN((size_t)len, remaining);\n> +\n> +    memcpy(buf, s->af_packet_buf + s->af_packet_buf_offset, copied);\n> +    s->af_packet_buf_offset += copied;\n> +\n> +    if (s->af_packet_buf_offset == s->af_packet_buf_len) {\n> +        tcp_chr_reset_af_packet_buf(s);\n> +    }\n> +\n> +    return (int)copied;\n> +}\n> +\n> +static ssize_t\n> +tcp_chr_capture_af_packet(Chardev *chr)\n> +{\n> +#ifdef CONFIG_LINUX\n> +    SocketChardev *s = SOCKET_CHARDEV(chr);\n> +    struct sockaddr_ll sll;\n> +    socklen_t sll_len;\n> +    ssize_t size;\n> +    uint32_t len;\n> +\n> +    if (!tcp_chr_uses_af_packet_capture(s)) {\n> +        errno = EIO;\n> +        return -1;\n> +    }\n> +\n> +    if (s->af_packet_buf_size <\n> +        sizeof(len) + TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE) {\n> +        s->af_packet_buf =\n> +            g_realloc(s->af_packet_buf,\n> +                      sizeof(len) + TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE);\n> +        s->af_packet_buf_size =\n> +            sizeof(len) + TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE;\n> +    }\n> +\n> +    for (;;) {\n> +        sll_len = sizeof(sll);\n> +        do {\n> +            size = recvfrom(s->sioc->fd, s->af_packet_buf + sizeof(len),\n> +                            TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE, 0,\n> +                            (struct sockaddr *)&sll, &sll_len);\n> +        } while (size < 0 && errno == EINTR);\n> +\n> +        if (size <= 0) {\n> +            if (size < 0 && errno != EAGAIN && errno != EWOULDBLOCK) {\n> +                trace_chr_socket_recv_err(chr, chr->label, g_strerror(errno));\n> +            }\n> +            return size;\n> +        }\n> +\n> +        if (sll.sll_pkttype != PACKET_OUTGOING) {\n> +            continue;\n> +        }\n> +\n> +        len = htonl(size);\n> +        memcpy(s->af_packet_buf, &len, sizeof(len));\n> +        s->af_packet_buf_len = sizeof(len) + size;\n> +        s->af_packet_buf_offset = 0;\n> +        return (ssize_t)s->af_packet_buf_len;\n> +    }\n> +#else\n> +    errno = EPROTONOSUPPORT;\n> +    return -1;\n> +#endif\n> +}\n> +\n>  static GSource *tcp_chr_add_watch(Chardev *chr, GIOCondition cond)\n>  {\n>      SocketChardev *s = SOCKET_CHARDEV(chr);\n> @@ -682,6 +785,22 @@ static gboolean tcp_chr_read(QIOChannel *chan, GIOCondition cond, void *opaque)\n>      if (len > s->max_size) {\n>          len = s->max_size;\n>      }\n> +    if (tcp_chr_uses_af_packet_capture(s)) {\n> +        tcp_chr_deliver_af_packet(chr);\n> +        if (s->max_size <= 0 || s->af_packet_buf_len) {\n> +            return TRUE;\n> +        }\n> +\n> +        size = tcp_chr_capture_af_packet(chr);\n> +        if (size == 0 || (size == -1 && errno != EAGAIN)) {\n> +            tcp_chr_disconnect(chr);\n> +        } else if (size > 0) {\n> +            tcp_chr_deliver_af_packet(chr);\n> +        }\n> +\n> +        return TRUE;\n> +    }\n> +\n>      size = tcp_chr_recv(chr, (void *)buf, len);\n>      if (size == 0 || (size == -1 && errno != EAGAIN)) {\n>          /* connection closed */\n> @@ -715,6 +834,10 @@ static int tcp_chr_sync_read(Chardev *chr, const uint8_t *buf, int len)\n>      int saved_errno;\n>      Error *local_err = NULL;\n>  \n> +    if (tcp_chr_uses_af_packet_capture(s) && s->af_packet_buf_len) {\n> +        return tcp_chr_copy_af_packet_buf(s, (uint8_t *)buf, len);\n> +    }\n> +\n>      if (s->state != TCP_CHARDEV_STATE_CONNECTED) {\n>          return 0;\n>      }\n> @@ -723,7 +846,14 @@ static int tcp_chr_sync_read(Chardev *chr, const uint8_t *buf, int len)\n>          error_report_err(local_err);\n>          return -1;\n>      }\n> -    size = tcp_chr_recv(chr, (void *) buf, len);\n> +    if (tcp_chr_uses_af_packet_capture(s)) {\n> +        size = tcp_chr_capture_af_packet(chr);\n> +        if (size > 0) {\n> +            size = tcp_chr_copy_af_packet_buf(s, (uint8_t *)buf, len);\n> +        }\n> +    } else {\n> +        size = tcp_chr_recv(chr, (void *)buf, len);\n> +    }\n\n\nSimilarly to the send side, I don't really think we should have this\npacket re-assembly logic in the chardev code. We should just be\ncalling the normal qio_channel_read APIs and let the netfilter code\nre-assemble packets it gets from the chardev. Mostly it seems we\nwould use TCP_CHARDEV_AF_PACKET_MAX_FRAME_SIZE instead of CHR_READ_BUF_LEN\nin the existing code paths.\n\n>      saved_errno = errno;\n>      if (s->state != TCP_CHARDEV_STATE_DISCONNECTED) {\n>          if (!qio_channel_set_blocking(s->ioc, false, &local_err)) {\n> @@ -1448,7 +1578,6 @@ static gboolean socket_reconnect_timeout(gpointer opaque)\n>      return false;\n>  }\n>  \n> -\n>  static int qmp_chardev_open_socket_server(Chardev *chr,\n>                                            bool is_telnet,\n>                                            bool is_waitconnect,\n> -- \n> 2.52.0\n> \n> \n\nWith regards,\nDaniel","headers":{"Return-Path":"<qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org>","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@legolas.ozlabs.org","Authentication-Results":["legolas.ozlabs.org;\n\tdkim=pass (1024-bit key;\n unprotected) header.d=redhat.com header.i=@redhat.com header.a=rsa-sha256\n header.s=mimecast20190719 header.b=dzDcA6ZP;\n\tdkim-atps=neutral","legolas.ozlabs.org;\n spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org\n (client-ip=209.51.188.17; helo=lists.gnu.org;\n envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n receiver=patchwork.ozlabs.org)"],"Received":["from lists.gnu.org (lists1p.gnu.org [209.51.188.17])\n\t(using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby legolas.ozlabs.org (Postfix) with ESMTPS id 4frXfh1tZRz1xv0\n\tfor <incoming@patchwork.ozlabs.org>; Thu, 09 Apr 2026 05:09:00 +1000 (AEST)","from localhost ([::1] helo=lists1p.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.90_1)\n\t(envelope-from <qemu-devel-bounces@nongnu.org>)\n\tid 1wAYDH-0001qw-Ct; Wed, 08 Apr 2026 15:05:19 -0400","from eggs.gnu.org ([2001:470:142:3::10])\n by lists1p.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256)\n (Exim 4.90_1) (envelope-from <berrange@redhat.com>)\n id 1wAYCo-0001WJ-OM\n for qemu-devel@nongnu.org; Wed, 08 Apr 2026 15:04:52 -0400","from us-smtp-delivery-124.mimecast.com ([170.10.133.124])\n by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256)\n (Exim 4.90_1) (envelope-from <berrange@redhat.com>)\n id 1wARn7-0002gU-E6\n for qemu-devel@nongnu.org; Wed, 08 Apr 2026 08:13:56 -0400","from mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com\n (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by\n relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3,\n cipher=TLS_AES_256_GCM_SHA384) id us-mta-532-2qL6uBlsM0C1Im3PqM84XQ-1; Wed,\n 08 Apr 2026 08:13:48 -0400","from mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com\n (mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.12])\n (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)\n key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest\n SHA256)\n (No client certificate requested)\n by mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS\n id 83F491955DD2; Wed,  8 Apr 2026 12:13:24 +0000 (UTC)","from redhat.com (headnet01.pony-001.prod.iad2.dc.redhat.com\n [10.2.32.101])\n by mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with\n ESMTPS\n id 8195D1955F2B; Wed,  8 Apr 2026 12:13:22 +0000 (UTC)"],"DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n s=mimecast20190719; t=1775650431;\n h=from:from:reply-to:reply-to:subject:subject:date:date:\n message-id:message-id:to:to:cc:cc:mime-version:mime-version:\n content-type:content-type:in-reply-to:in-reply-to:  references:references;\n bh=dMCBNtxVArYttWyy7cP/2A+U1IRE/oIt+BGty3J210g=;\n b=dzDcA6ZP2LqXwkClJt6Fx49KwwfDbMKbx9bWAif7qVqeaRx0P133Yvu2eNJgfZyK3A7uXl\n Bxb86f1Q4xDRud8nOOK06guaxN82yWyjZmckMERo2mFSEYIo2ej1r+cwrJSxSIUTAD9waD\n nz/Ss6z5Da+Z5IuhqaPcv2msVZ8yOPI=","X-MC-Unique":"2qL6uBlsM0C1Im3PqM84XQ-1","X-Mimecast-MFC-AGG-ID":"2qL6uBlsM0C1Im3PqM84XQ_1775650427","Date":"Wed, 8 Apr 2026 13:13:19 +0100","From":"Daniel =?utf-8?b?UC4gQmVycmFuZ8Op?= <berrange@redhat.com>","To":"Cindy Lu <lulu@redhat.com>","Cc":"mst@redhat.com, jasowang@redhat.com, zhangckid@gmail.com,\n lizhijian@fujitsu.com, jmarcin@redhat.com, qemu-devel@nongnu.org","Subject":"Re: [RFC v4 5/5] chardev/socket: add AF_PACKET capture path","Message-ID":"<adZGX5nA76feGRuV@redhat.com>","References":"<20260407050818.2249570-1-lulu@redhat.com>\n <20260407050818.2249570-6-lulu@redhat.com>","MIME-Version":"1.0","Content-Type":"text/plain; charset=utf-8","Content-Disposition":"inline","In-Reply-To":"<20260407050818.2249570-6-lulu@redhat.com>","User-Agent":"Mutt/2.2.14 (2025-02-20)","X-Scanned-By":"MIMEDefang 3.0 on 10.30.177.12","Received-SPF":"pass client-ip=170.10.133.124;\n envelope-from=berrange@redhat.com;\n helo=us-smtp-delivery-124.mimecast.com","X-Spam_score_int":"7","X-Spam_score":"0.7","X-Spam_bar":"/","X-Spam_report":"(0.7 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.54,\n DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1,\n RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H2=0.001, RCVD_IN_SBL_CSS=3.335,\n RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001,\n SPF_HELO_PASS=-0.001, SPF_PASS=-0.001 autolearn=no autolearn_force=no","X-Spam_action":"no action","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.29","Precedence":"list","List-Id":"qemu development <qemu-devel.nongnu.org>","List-Unsubscribe":"<https://lists.nongnu.org/mailman/options/qemu-devel>,\n <mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>","List-Archive":"<https://lists.nongnu.org/archive/html/qemu-devel>","List-Post":"<mailto:qemu-devel@nongnu.org>","List-Help":"<mailto:qemu-devel-request@nongnu.org?subject=help>","List-Subscribe":"<https://lists.nongnu.org/mailman/listinfo/qemu-devel>,\n <mailto:qemu-devel-request@nongnu.org?subject=subscribe>","Reply-To":"Daniel =?utf-8?b?UC4gQmVycmFuZ8Op?= <berrange@redhat.com>","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org"}}]