linux-kernel - Re: [PATCH net-next V10 1/6] devlink: Extend devlink rate API with traffic classes bandwidth management

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <3dddb35c-d3cb-453d-8e60-70f241abd018@nvidia.com>
Date: Sun, 25 May 2025 17:57:28 +0300
From: Carolina Jubran <cjubran@...dia.com>
To: Jakub Kicinski <kuba@...nel.org>, Tariq Toukan <tariqt@...dia.com>
Cc: "David S. Miller" <davem@...emloft.net>, Paolo Abeni <pabeni@...hat.com>,
 Eric Dumazet <edumazet@...gle.com>, Andrew Lunn <andrew+netdev@...n.ch>,
 Jiri Pirko <jiri@...dia.com>, Gal Pressman <gal@...dia.com>,
 Leon Romanovsky <leonro@...dia.com>, Donald Hunter
 <donald.hunter@...il.com>, Jiri Pirko <jiri@...nulli.us>,
 Jonathan Corbet <corbet@....net>, Saeed Mahameed <saeedm@...dia.com>,
 Leon Romanovsky <leon@...nel.org>, Shuah Khan <shuah@...nel.org>,
 netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-doc@...r.kernel.org, linux-rdma@...r.kernel.org,
 linux-kselftest@...r.kernel.org, Moshe Shemesh <moshe@...dia.com>,
 Mark Bloch <mbloch@...dia.com>, Cosmin Ratiu <cratiu@...dia.com>
Subject: Re: [PATCH net-next V10 1/6] devlink: Extend devlink rate API with
 traffic classes bandwidth management

On 21/05/2025 1:53, Jakub Kicinski wrote:
> A few quick comments here as the test is failing
> 
> On Tue, 20 May 2025 21:38:02 +0300 Tariq Toukan wrote:
>> +      -
>> +        name: rate-tc-bws
>> +        type: nest
>> +        multi-attr: true
>> +        nested-attributes: dl-rate-tc-bws
>> +      -
>> +        name: rate-tc-index
>> +        type: u8
>> +        checks:
>> +          min: 0
>> +          max: rate-tc-index-max
> 
> no need for min: 0 on an unsigned type ?
> 

Will remove them.

>> +      -
>> +        name: rate-tc-bw
>> +        type: u32
>> +        doc: |
>> +             Specifies the bandwidth allocation for the Traffic Class as a
>> +             percentage.
>> +        checks:
>> +          min: 0
>> +          max: 100
> 
> Why in percentage? I don't think any existing param in devlink rate
> or net shapers is in percentage right? Not according to what i can
> grok about the uAPI.
> 

I thought percentage might fit better here because it lets users clearly 
set the bandwidth share for each traffic class. While this isn’t the 
same as tx_weight in devlink-rate, the idea is related since both use 
relative values. If there isn’t a strong reason against it, I’d like to 
keep using percentages here.

>> +static int devlink_nl_rate_tc_bw_parse(struct nlattr *parent_nest, u32 *tc_bw,
>> +				       unsigned long *bitmap, struct netlink_ext_ack *extack)
>> +{
>> +	struct nlattr *tb[DEVLINK_ATTR_MAX + 1];
>> +	u8 tc_index;
>> +
>> +	nla_parse_nested(tb, DEVLINK_ATTR_MAX, parent_nest, devlink_dl_rate_tc_bws_nl_policy,
> 
> Let's error check this, I get that we already validated via the policy
> but what if we do memory allocations during parsing one day, or some
> other failure-prone operation.. better check the return value.
> 

Ack. will do.
> nit: over 80 chars for no good reason, the line overflows anyway.
> Please use checkpatch --max-line-width=80 for core code,
> at the very least.
> 

I noticed the current code already goes over 80 chars, but I’ll update 
it to follow the 80-char limit. Will fix, thanks.

>> +			 extack);
>> +	if (!tb[DEVLINK_ATTR_RATE_TC_INDEX]) {
>> +		NL_SET_ERR_ATTR_MISS(extack, parent_nest, DEVLINK_ATTR_RATE_TC_INDEX);
>> +		return -EINVAL;
>> +	}
>> +
>> +	tc_index = nla_get_u8(tb[DEVLINK_ATTR_RATE_TC_INDEX]);
>> +
>> +	if (!tb[DEVLINK_ATTR_RATE_TC_BW]) {
>> +		NL_SET_ERR_ATTR_MISS(extack, parent_nest, DEVLINK_ATTR_RATE_TC_BW);
>> +		return -EINVAL;
>> +	}
>> +
>> +	if (test_and_set_bit(tc_index, bitmap)) {
>> +		NL_SET_ERR_MSG_FMT(extack, "Duplicate traffic class index specified (%u)",
>> +				   tc_index);
>> +		return -EINVAL;
>> +	}
>> +
>> +	tc_bw[tc_index] = nla_get_u32(tb[DEVLINK_ATTR_RATE_TC_BW]);
>> +
>> +	return 0;
>> +}
>> +
>> +static int devlink_nl_rate_tc_bw_set(struct devlink_rate *devlink_rate,
>> +				     struct genl_info *info)
>> +{
>> +	DECLARE_BITMAP(bitmap, DEVLINK_RATE_TCS_MAX) = {};
>> +	struct devlink *devlink = devlink_rate->devlink;
>> +	const struct devlink_ops *ops = devlink->ops;
>> +	int rem, err = -EOPNOTSUPP, i, total = 0;
>> +	u32 tc_bw[DEVLINK_RATE_TCS_MAX] = {};
>> +	struct nlattr *attr;
>> +
>> +	nla_for_each_attr(attr, genlmsg_data(info->genlhdr),
>> +			  genlmsg_len(info->genlhdr), rem) {
> 
>   nla_for_each_attr_type() ?
>   or better still add a _type() version of nlmsg_for_each_attr() ?
> 
Good point, thanks. I’ll add a _type() version for nlmsg_for_each_attr. 
Do you prefer this to be part of this patch or should I send it as a 
separate patch?

>> +		if (nla_type(attr) == DEVLINK_ATTR_RATE_TC_BWS) {
>> +			err = devlink_nl_rate_tc_bw_parse(attr, tc_bw, bitmap, info->extack);
>> +			if (err)
>> +				return err;
>> +		}
>> +	}