netdev - Re: [PATCH] net: dsa: microchip: KSZ9896 register regmap alignment to 32 bit boundaries

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: <CAMdwsN99s2C=qvxEO=hmpRfvjRH6E7tww0Mfp-Q044ufi8Rn-w@mail.gmail.com>
Date: Tue, 3 Dec 2024 06:29:21 +0100
From: Jesse Van Gavere <jesseevg@...il.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: netdev@...r.kernel.org, woojung.huh@...rochip.com, 
	UNGLinuxDriver@...rochip.com, andrew@...n.ch, olteanv@...il.com, 
	Jesse Van Gavere <jesse.vangavere@...oteq.com>, Tristram.Ha@...rochip.com
Subject: Re: [PATCH] net: dsa: microchip: KSZ9896 register regmap alignment to
 32 bit boundaries

Hello Jakub, all,

Op za 30 nov 2024 om 23:07 schreef Jakub Kicinski <kuba@...nel.org>:
>
> On Wed, 27 Nov 2024 23:11:29 +0100 Jesse Van Gavere wrote:
> > Commit (SHA1: 8d7ae22ae9f8c8a4407f8e993df64440bdbd0cee) fixed this issue
> > for the KSZ9477 by adjusting the regmap ranges.
>
> The correct format for referring to other commits in Linux kernel is:
>  %h (\"%s\")
> IOW:
>
>  Commit 8d7ae22ae9f8 ("net: dsa: microchip: KSZ9477 register regmap
>  alignment to 32 bit boundaries") fixed this issue...
>
> > The same issue presents itself on the KSZ9896 regs and is fixed with
> > the same regmap range modification.
>
> Could you explain the impact? What will not work / break without this
> change? Please add a Fixes tag indicating where buggy code was added
> to make sure backporters know how far to backport.
Will do, still learning the ropes of contributing, thanks for the feedback!
> --
> pw-bot: cr

What do you think I preferably do to account for Tristram's feedback
in my next patch?
Should I incorporate it as-is, keep my patch with requested changes,
or perhaps even "fix" it with below suggestion across all registers
sets?
> The port address range 0x#100-0x#13F just maps to the PHY registers 0-31,
> so it can be simply one single regmap_reg_range(0x1100, 0x113f) instead
> of many.  I suggest using regmap_reg_range(0x1100, 0x111f) and
> regmap_reg_range(0x1120, 0x113f) to remind people the high range address
> needs special handling.

> I also do not know why those _register_set are not enforced across all
> KSZ9897 related switches.

Best regards,
Jesse