From mboxrd@z Thu Jan  1 00:00:00 1970
From: David Ahern <dsahern@gmail.com>
Subject: Re: [PATCH v2 iproute2 net-next] erspan: add erspan version II
 support
Date: Tue, 19 Dec 2017 15:17:17 -0700
Message-ID: <d7283b25-e7a2-76e3-3946-137bc5baf416@gmail.com>
References: <1513386366-38095-1-git-send-email-u9012063@gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
To: William Tu <u9012063@gmail.com>, netdev@vger.kernel.org
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pf0-f172.google.com ([209.85.192.172]:38038 "EHLO
        mail-pf0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751496AbdLSWRU (ORCPT
        <rfc822;netdev@vger.kernel.org>); Tue, 19 Dec 2017 17:17:20 -0500
Received: by mail-pf0-f172.google.com with SMTP id u25so11723338pfg.5
        for <netdev@vger.kernel.org>; Tue, 19 Dec 2017 14:17:20 -0800 (PST)
In-Reply-To: <1513386366-38095-1-git-send-email-u9012063@gmail.com>
Content-Language: en-US
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On 12/15/17 6:06 PM, William Tu wrote:
> @@ -343,6 +355,22 @@ get_failed:
>  				invarg("invalid erspan index\n", *argv);
>  			if (erspan_idx & ~((1<<20) - 1) || erspan_idx == 0)
>  				invarg("erspan index must be > 0 and <= 20-bit\n", *argv);
> +		} else if (strcmp(*argv, "erspan_ver") == 0) {
> +			NEXT_ARG();
> +			if (get_u8(&erspan_ver, *argv, 0))
> +				invarg("invalid erspan version\n", *argv);
> +			if (erspan_ver != 1 && erspan_ver != 2)
> +				invarg("erspan version must be 1 or 2\n", *argv);
> +		} else if (strcmp(*argv, "erspan_dir") == 0) {
> +			NEXT_ARG();
> +			if (get_u8(&erspan_dir, *argv, 0))
> +				invarg("invalid erspan direction\n", *argv);
> +			if (erspan_dir != 0 && erspan_dir != 1)
> +				invarg("erspan direction must be 0(Ingress) or 1(Egress)\n", *argv);

Why not allow ingress and egress as strings and using matches()?
e.g.,  '... erspan_dir in[gress] ...' or '... erspan_dir e[gress] ...'

Seems much more intuitive than remembering "0" = ingress and "1" = egress.

> @@ -374,8 +402,15 @@ get_failed:
>  		addattr_l(n, 1024, IFLA_GRE_TTL, &ttl, 1);
>  		addattr_l(n, 1024, IFLA_GRE_TOS, &tos, 1);
>  		addattr32(n, 1024, IFLA_GRE_FWMARK, fwmark);
> -		if (erspan_idx != 0)
> -			addattr32(n, 1024, IFLA_GRE_ERSPAN_INDEX, erspan_idx);
> +		if (erspan_ver) {
> +			addattr8(n, 1024, IFLA_GRE_ERSPAN_VER, erspan_ver);
> +			if (erspan_ver == 1 && erspan_idx != 0) {
> +				addattr32(n, 1024, IFLA_GRE_ERSPAN_INDEX, erspan_idx);
> +			} else if (erspan_ver == 2) {
> +				addattr8(n, 1024, IFLA_GRE_ERSPAN_DIR, erspan_dir);
> +				addattr16(n, 1024, IFLA_GRE_ERSPAN_HWID, erspan_hwid);
> +			}
> +		}
>  	} else {
>  		addattr_l(n, 1024, IFLA_GRE_COLLECT_METADATA, NULL, 0);
>  	}
> @@ -514,7 +549,25 @@ static void gre_print_opt(struct link_util *lu, FILE *f, struct rtattr *tb[])
>  	if (tb[IFLA_GRE_ERSPAN_INDEX]) {
>  		__u32 erspan_idx = rta_getattr_u32(tb[IFLA_GRE_ERSPAN_INDEX]);
>  
> -		fprintf(f, "erspan_index %u ", erspan_idx);
> +		print_uint(PRINT_ANY, "erspan_index", "erspan_index %u", erspan_idx);
> +	}
> +
> +	if (tb[IFLA_GRE_ERSPAN_VER]) {
> +		__u8 erspan_ver = rta_getattr_u8(tb[IFLA_GRE_ERSPAN_VER]);
> +
> +		print_uint(PRINT_ANY, "erspan_ver", "erspan_ver %u", erspan_ver);
> +	}
> +
> +	if (tb[IFLA_GRE_ERSPAN_DIR]) {
> +		__u8 erspan_dir = rta_getattr_u8(tb[IFLA_GRE_ERSPAN_DIR]);
> +
> +		print_uint(PRINT_ANY, "erspan_dir", "erspan_dir %u", erspan_dir);

similarly, here can convert the 0/1 to a human string.

Same comments for the changes to link_gre6.c