[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZMn7Px9P1Jep01Qm@nanopsycho>
Date: Wed, 2 Aug 2023 08:44:15 +0200
From: Jiri Pirko <jiri@...nulli.us>
To: Jakub Kicinski <kuba@...nel.org>
Cc: netdev@...r.kernel.org, pabeni@...hat.com, davem@...emloft.net,
edumazet@...gle.com, moshe@...dia.com, saeedm@...dia.com,
idosch@...dia.com, petrm@...dia.com
Subject: Re: [patch net-next 4/8] devlink: add split ops generated according
to spec
Tue, Aug 01, 2023 at 08:56:55PM CEST, kuba@...nel.org wrote:
>On Tue, 1 Aug 2023 16:19:03 +0200 Jiri Pirko wrote:
>> Improve the existing devlink spec in order to serve as a source fot
>
>s/fot/for/
>
>> generation of valid devlink split ops for the existing commands.
>> Add the generated sources.
>
>> +/* DEVLINK_CMD_GET - do */
>> +const struct nla_policy devlink_get_nl_policy[DEVLINK_ATTR_DEV_NAME + 1] = {
>> + [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING, },
>> + [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING, },
>> +};
>
>What's the impact of narrowing down the policies? Could you describe it
>in the commit message?
Should be no impact afaik. The code does not care about the rest of the
attributes and dont-validate-strict will allow any garbage to be passed
by the user. I will put a note in the commit message as you asked.
Powered by blists - more mailing lists