linux-kernel - Re: [Intel-wired-lan] [PATCH] igb: index regs

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] [day] [month] [year] [list]

Date:   Thu, 12 Dec 2019 16:48:49 -0800
From:   Alexander Duyck <alexander.duyck@...il.com>
To:     Vasyl Gomonovych <gomonovych@...il.com>
Cc:     Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
        David Miller <davem@...emloft.net>,
        intel-wired-lan <intel-wired-lan@...ts.osuosl.org>,
        Netdev <netdev@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [Intel-wired-lan] [PATCH] igb: index regs_buff array via index variable

On Thu, Dec 12, 2019 at 2:58 AM Vasyl Gomonovych <gomonovych@...il.com> wrote:
>
> This patch is just a preparation for additional register dump in regs_buff.
> To make new register insertion in the middle of regs_buff array easier
> change array indexing to use local counter reg_ix.
>
> ---
>
> Basically this path is just a subject to ask
> How to add a new register to dump from dataseet
> Because it is logically better to add an additional register
> in the middle of an array but that will break ABI.
> To not have the ABI problem we should just add it at the
> end of the array and increase the array size.

So I am pretty sure the patch probably breaks ABI. The reasons for the
fixed offsets is because this driver supports multiple parts that have
different register sets so we cannot have them overlapping.

We cannot change the register locations because it will break the
interface with ethtool. If you need to add additional registers you
will need to add them to the end of the array.