Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 10 Oct 2019 17:59:55 +0200
From:   Jiri Pirko <jiri@resnulli.us>
To:     Michal Kubecek <mkubecek@suse.cz>
Cc:     David Miller <davem@davemloft.net>, netdev@vger.kernel.org,
        Jakub Kicinski <jakub.kicinski@netronome.com>,
        Andrew Lunn <andrew@lunn.ch>,
        Florian Fainelli <f.fainelli@gmail.com>,
        John Linville <linville@tuxdriver.com>,
        Stephen Hemminger <stephen@networkplumber.org>,
        Johannes Berg <johannes@sipsolutions.net>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH net-next v7 12/17] ethtool: provide link settings with
 LINKINFO_GET request
Message-ID: <20191010155955.GB2901@nanopsycho>
References: <cover.1570654310.git.mkubecek@suse.cz>
 <1568f00bf7275f1a872c177e29d5800cd73e50c8.1570654310.git.mkubecek@suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1568f00bf7275f1a872c177e29d5800cd73e50c8.1570654310.git.mkubecek@suse.cz>
User-Agent: Mutt/1.11.4 (2019-03-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Wed, Oct 09, 2019 at 10:59:37PM CEST, mkubecek@suse.cz wrote:

[...]

>+/* prepare_data() handler */

Not sure how valuable are comments like this...


>+static int linkinfo_prepare(const struct ethnl_req_info *req_base,
>+			    struct ethnl_reply_data *reply_base,
>+			    struct genl_info *info)
>+{
>+	struct linkinfo_reply_data *data =
>+		container_of(reply_base, struct linkinfo_reply_data, base);

A helper would be nice for this. For req_info too.


>+	struct net_device *dev = reply_base->dev;
>+	int ret;
>+
>+	data->lsettings = &data->ksettings.base;
>+
>+	ret = ethnl_before_ops(dev);

"before_ops"/"after_ops" sounds odd. Maybe:
ethnl_ops_begin
ethnl_ops_complete

To me in-line with ethtool_ops names?

I guess you don't want the caller (ethnl_get_doit/ethnl_get_dumpit)
to call this because it might not be needed down in prepare_data()
callback, right?


>+	if (ret < 0)
>+		return ret;
>+	ret = __ethtool_get_link_ksettings(dev, &data->ksettings);
>+	if (ret < 0 && info)
>+		GENL_SET_ERR_MSG(info, "failed to retrieve link settings");
>+	ethnl_after_ops(dev);
>+
>+	return ret;
>+}

[...]


>+const struct get_request_ops linkinfo_request_ops = {
>+	.request_cmd		= ETHTOOL_MSG_LINKINFO_GET,
>+	.reply_cmd		= ETHTOOL_MSG_LINKINFO_GET_REPLY,
>+	.hdr_attr		= ETHTOOL_A_LINKINFO_HEADER,
>+	.max_attr		= ETHTOOL_A_LINKINFO_MAX,
>+	.req_info_size		= sizeof(struct linkinfo_req_info),
>+	.reply_data_size	= sizeof(struct linkinfo_reply_data),
>+	.request_policy		= linkinfo_get_policy,
>+	.all_reqflags		= ETHTOOL_RFLAG_LINKINFO_ALL,
>+
>+	.prepare_data		= linkinfo_prepare,

Please have the ops with the same name/suffix:
	.request_policy		= linkinfo_reques_policy,
	.prepare_data		= linkinfo_prepare_data,
	.reply_size		= linkinfo_reply_size,
	.fill_reply		= linkinfo_fill_reply,

Same applies of course to the other patches.


>+	.reply_size		= linkinfo_size,
>+	.fill_reply		= linkinfo_fill,
>+};

[...]