linux-kernel - Re: [PATCH net-next v2 06/10] net: phy: print an info if a broken C45 bus is found

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]

Date:   Mon, 26 Jun 2023 08:50:51 +0200
From:   Michael Walle <mwalle@...nel.org>
To:     Simon Horman <simon.horman@...igine.com>
Cc:     Andrew Lunn <andrew@...n.ch>,
        Heiner Kallweit <hkallweit1@...il.com>,
        Russell King <linux@...linux.org.uk>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Yisen Zhuang <yisen.zhuang@...wei.com>,
        Salil Mehta <salil.mehta@...wei.com>,
        Florian Fainelli <florian.fainelli@...adcom.com>,
        Broadcom internal kernel review list 
        <bcm-kernel-feedback-list@...adcom.com>,
        Marek Behún <kabel@...nel.org>,
        Xu Liang <lxu@...linear.com>, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v2 06/10] net: phy: print an info if a broken C45
 bus is found

Am 2023-06-23 22:35, schrieb Simon Horman:
> On Fri, Jun 23, 2023 at 07:42:08PM +0200, Andrew Lunn wrote:
>> On Fri, Jun 23, 2023 at 12:29:15PM +0200, Michael Walle wrote:
>> > If there is an PHY which gets confused by C45 transactions on the MDIO
>> > bus, print an info together with the PHY identifier of the offending
>> > one.
>> >
>> > Signed-off-by: Michael Walle <mwalle@...nel.org>
>> >
>> > ---
>> > I wasn't sure if this should be phydev_dbg() or phydev_info(). I mainly
>> > see this as an info to a user why some PHYs might not be probed (or
>> > c45-over-c22 is used later).
>> 
>> The information is useful to the DT writer, not the 'user'. I would
>> assume the DT writer has a bit more kernel knowledge and can debug
>> prints on. So i would suggest phydev_dbg().

Why the DT writer? There could be no DT at all, right? But yeah, fair
enough, I thought of our hardware engineers as a user, which might be
surprised to find no C45 transactions at all for a C45 PHY.

That said, I don't have a strong opinion. I'm fine to switch that to
dev_dbg() to make the kernel output less noisy.

>> > @@ -617,10 +617,10 @@ static int mdiobus_scan_bus_c45(struct mii_bus *bus)
>> >   */
>> >  void mdiobus_scan_for_broken_c45_access(struct mii_bus *bus)
>> >  {
>> > +	struct phy_device *phydev;
>> >  	int i;
>> >
>> >  	for (i = 0; i < PHY_MAX_ADDR; i++) {
>> > -		struct phy_device *phydev;
>> >  		u32 oui;
>> 
>> It is not clear why you changed the scope of phydev. I guess another
>> version used phydev_info(), where as now you have dev_info()?
> 
> I think it is so it can be used in the dev_info() call below.

Yes, to print the PHY ID of the offending one.

> However Smatch has it's doubts that it is always initialised there.
> 
>   .../mdio_bus.c:638 mdiobus_scan_for_broken_c45_access() error: we 
> previously assumed 'phydev' could be null (see line 627)
> 
>> >  		phydev = mdiobus_get_phy(bus, i);
> 
> Line 627 immediately follows the line above, like this:
> 
> 		if (!phydev)
> 			continue;

Mhh, I see. bus->prevent_c45_access could (theoretically) set before
calling this function. I could set it to false at the beginning of
this function or I could use a new flag to indicate when to print the
warning. Any suggestions?

-michael