netdev - Re: [PATCH v3 net 1/3] net: mediatek: sgmii: ensure the SGMII PHY is powered down on configuration

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:   Tue, 24 Jan 2023 17:01:33 +0100
From:   Paolo Abeni <pabeni@...hat.com>
To:     "Russell King (Oracle)" <linux@...linux.org.uk>
Cc:     Bjørn Mork <bjorn@...k.no>,
        netdev@...r.kernel.org, Felix Fietkau <nbd@....name>,
        John Crispin <john@...ozen.org>,
        Sean Wang <sean.wang@...iatek.com>,
        Mark Lee <Mark-MC.Lee@...iatek.com>,
        Lorenzo Bianconi <lorenzo@...nel.org>,
        Daniel Golle <daniel@...rotopia.org>,
        Alexander Couzens <lynxis@...0.eu>
Subject: Re: [PATCH v3 net 1/3] net: mediatek: sgmii: ensure the SGMII PHY
 is powered down on configuration

On Tue, 2023-01-24 at 12:45 +0000, Russell King (Oracle) wrote:
> On Tue, Jan 24, 2023 at 01:19:15PM +0100, Paolo Abeni wrote:
> > On Sun, 2023-01-22 at 22:21 +0100, Bjørn Mork wrote:
> > > From: Alexander Couzens <lynxis@...0.eu>
> > > 
> > > The code expect the PHY to be in power down which is only true after reset.
> > > Allow changes of the SGMII parameters more than once.
> > > 
> > > Only power down when reconfiguring to avoid bouncing the link when there's
> > > no reason to - based on code from Russell King.
> > > 
> > > There are cases when the SGMII_PHYA_PWD register contains 0x9 which
> > > prevents SGMII from working. The SGMII still shows link but no traffic
> > > can flow. Writing 0x0 to the PHYA_PWD register fix the issue. 0x0 was
> > > taken from a good working state of the SGMII interface.
> > 
> > This looks like a legitimate fix for -net, but we need a suitable Fixes
> > tag pointing to the culprit commit.
> > 
> > Please repost including such tag. While at that you could also consider
> > including Simon's suggestion.
> > 
> > The following 2 patches looks like new features/refactor that would be
> > more suitable for net-next, and included here due to the code
> > dependency.
> 
> I'm not sure why you think that, especially for patch 2.

Because I misread the patch contents as a simple macro rename (I missed
the s/==/!=/ part).
> 
> Patch 2 corrects the sense of the duplex bit - the code originally
> set this for full duplex, but in actual fact, the bit needs to be set
> for half duplex. I can't see how one could regard that as a feature
> or a refactor.
> 
> I'm also not sure how you could regard patch 3 as a refactor. It
> could be argued that it is a new feature, but it is actually a bug
> fix for the patch converting the driver to phylink_pcs which
> omitted setting this, making the pcs_get_state() method rather
> useless.

Well, it looked more a new feature than a fix to me. The above
explanation cleared the matter. A more descriptive commit could have
avoid confusion on my side :)

> So I regard all three patches as fixes, not features or refactoring.

No objections on my side, but please add suitable fixes tag on every
patch then.

Thanks!

Paolo