netdev - Re: [PATCH] net: usb: lan78xx: add weak dependency with micrel phy module

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]

Message-ID: <20240730075529.8263-1-jtornosm@redhat.com>
Date: Tue, 30 Jul 2024 09:55:28 +0200
From: Jose Ignacio Tornos Martinez <jtornosm@...hat.com>
To: andrew@...n.ch
Cc: UNGLinuxDriver@...rochip.com,
	davem@...emloft.net,
	dsimic@...jaro.org,
	edumazet@...gle.com,
	f.fainelli@...il.com,
	gregkh@...uxfoundation.org,
	jtornosm@...hat.com,
	kuba@...nel.org,
	linux-kbuild@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	linux-usb@...r.kernel.org,
	lucas.demarchi@...el.com,
	masahiroy@...nel.org,
	mcgrof@...nel.org,
	netdev@...r.kernel.org,
	pabeni@...hat.com,
	woojung.huh@...rochip.com
Subject: Re: [PATCH] net: usb: lan78xx: add weak dependency with micrel phy module

Hello Andrew,

> So are you saying current initramfs are broken, because they don't
> include the needed PHY modules?
I am just saying that the default initramfs including the current lan78xx
driver is broken because in this case there is no information to collect
the possible phy modules. And as I commented, after the complete boot, the
only solution is to unload and load lan78xx to get the phy module from
rootfs.
 
> You can fix one example of the lan78xx
> USB dongle, but are going to leave everything else broken?
My intention was to fix the case for lan78xx because it is the one that I
have detected that does not work. Others are already working, for example
r8169, by means of a softdep with realtek phy. And my idea was to do the
same for the other detected/needed, if any (I am not aware of other similar
reported issues).
I see that you prefer to fix all the cases and always including all the phy
modules would solve the problem for lan78xx and for other possible ones.
But take into account that we should also try to avoid creating large
initramfs if not necessary, at least, if there is anyway to solve this.  
Indeed, if I am not wrong, only some phy modules are possible for
a driver and these are known.
Anyway, as it was suggested, we can explore some automatic procedure to
identify the hardware and with that, select the phy module or at least,
reduce the number of phy modules to introduce.

Thanks

Best regards
José Ignacio