netdev - RE: [PATCH v3 net-next] r8169: add support for rtl8168h(revid 0x2a) + rtl8211fs fiber application

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:   Wed, 24 Aug 2022 19:20:02 +0000
From:   Hau <hau@...ltek.com>
To:     Andrew Lunn <andrew@...n.ch>
CC:     "hkallweit1@...il.com" <hkallweit1@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        nic_swsd <nic_swsd@...ltek.com>,
        "kuba@...nel.org" <kuba@...nel.org>,
        "davem@...emloft.net" <davem@...emloft.net>
Subject: RE: [PATCH v3 net-next] r8169: add support for rtl8168h(revid 0x2a) + rtl8211fs fiber application

> On Tue, Aug 23, 2022 at 03:48:13PM +0000, Hau wrote:
> > > > > > @@ -914,8 +952,12 @@ static void r8168g_mdio_write(struct
> > > > > rtl8169_private *tp, int reg, int value)
> > > > > >  	if (tp->ocp_base != OCP_STD_PHY_BASE)
> > > > > >  		reg -= 0x10;
> > > > > >
> > > > > > -	if (tp->ocp_base == OCP_STD_PHY_BASE && reg ==
> MII_BMCR)
> > > > > > +	if (tp->ocp_base == OCP_STD_PHY_BASE && reg ==
> MII_BMCR) {
> > > > > > +		if (tp->sfp_if_type != RTL_SFP_IF_NONE && value &
> > > > > BMCR_PDOWN)
> > > > > > +			return;
> > > > > > +
> > > > >
> > > > > Please could you explain this change.
> > > > H/W has issue. If we power down phy, then H/W will become abnormal.
> > >
> > > Does the MAC break, or the PHY? If the PHY is the problem, the work
> > > around should be in the PHY driver, not the MAC driver.
> > >
> > It is a workaround. We did not get the root cause yet.
> 
> O.K, since you are modifying a PHY register, lets assume it is a PHY issue.
> Please put the workaround in the PHY driver.
I will try to do this.

> > > > I have tried to use alloc_mdio_bitbang(). But I will get error message "
> > > rmmod: ERROR: Module r8169 is in use " when I try to unload r8169.
> > > > After debug, I found it is cause by
> > > > "__module_get(ctrl->ops->owner); " in
> > > alloc_mdio_bitbang().
> > >
> > > void free_mdio_bitbang(struct mii_bus *bus) {
> > >         struct mdiobb_ctrl *ctrl = bus->priv;
> > >
> > >         module_put(ctrl->ops->owner);
> > >         mdiobus_free(bus);
> > > }
> > >
> > > Make sure your cleanup is symmetrical to your setup.
> > >
> > I have tried to call free_mdio_bitbang() in rtl_remove_one (). But when I
> try to unload r8169, rtl_remove_one() did not get called.
> 
> So you probably need to dig into driver/base/dd.c and maybe the drivers/pci,
> to understand why.
In command "rmmod", if module refcnt not equal to 0, It will show error message " Module r8169 is in use ". And "__module_get()" will increase module refcnt.
If module refcnt is not equal to 0, rmmod will not delete module.

Is that possible for us to call alloc_mdio_bitbang() but set owner to null? I have try this setting, command "rmmod" can be executed normally.

Thanks,
Hau

------Please consider the environment before printing this e-mail.