linux-kernel - Re: [PATCH net-next v1] hinic: add set_channels ethtool

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 PHC
Open Source and information security mailing list archives

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

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

Date:   Sat, 16 May 2020 13:35:08 -0700 (PDT)
From:   David Miller <davem@...emloft.net>
To:     luobin9@...wei.com
Cc:     linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
        luoxianjun@...wei.com, yin.yinshi@...wei.com,
        cloud.wangxiaoyun@...wei.com
Subject: Re: [PATCH net-next v1] hinic: add set_channels ethtool_ops support

From: Luo bin <luobin9@...wei.com>
Date: Fri, 15 May 2020 19:42:12 +0000

> add support to change TX/RX queue number with ethtool -L
> 
> Signed-off-by: Luo bin <luobin9@...wei.com>

I don't think you are properly following the semantics of this
ethtool command with your changes.

In fact, you are breaking the hinic_get_channels() function which
is properly advertising the ->max_* values currently.  Now it will
return zero.

Whatever is advertised in ->max_* should be the driver's maximum
capabilities.

This means that the user can request anything in the range from '1'
to these max values.

Whatever the user asks for in ->combined_count and elsewhere, you
_MUST_ provide or return an error.

That is not what hinic_set_channels() is doing.  It is using
combined_count as a "limit" rather than the value to use.