netdev - Re: [RFC PATCH] smsc911x: protect INT_EN read-modify-write with spinlock

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sun, 25 Jan 2009 20:43:29 -0800
From:	Steve Sakoman <sakoman@...il.com>
To:	Steve Glendinning <steve.glendinning@...c.com>
Cc:	netdev@...r.kernel.org, David Miller <davem@...emloft.net>,
	Ian Saturley <ian.saturley@...c.com>
Subject: Re: [RFC PATCH] smsc911x: protect INT_EN read-modify-write with 
	spinlock

On Tue, Jan 20, 2009 at 6:01 AM, Steve Glendinning
<steve.glendinning@...c.com> wrote:
> This patch adds a new spinlock to protect read-modify-writes to the
> INT_EN register when enabling and disabling interrupt sources.
>
> I haven't actually seen any devices lock up, but I think there's a
> possibility and I'd like to eliminate it.

I'm attempting to use this driver on a new Gumstix Overo daughtercard
that uses an SMCS9221 and have been having some interrupt issues that
may or may not be related.

I am using the version in linux-omap 2.6.29-rc2.  Here's what I see at boot:

smsc911x: Driver version 2008-10-21.
eth%d: smsc911x_init: Driver Parameters:
eth%d: smsc911x_init: LAN base: 0xD086E000
eth%d: smsc911x_init: IRQ: 336
eth%d: smsc911x_init: PHY will be autodetected.
eth%d: smsc911x_init: BYTE_TEST: 0x87654321
eth%d: smsc911x_init: LAN911x identified, idrev: 0x92210000, generation: 4
eth0: smsc911x_drv_probe: Network interface: "eth0"
eth0: smsc911x_mii_init: External PHY is not supported, using internal PHY
smsc911x-mdio: probed
eth0: smsc911x_mii_probe: PHY 1: addr 1, phy_id 0x0007C0C3
eth0: attached PHY driver [SMSC LAN8700] (mii_bus:phy_addr=ffffffff:01, irq=-1)
eth0: smsc911x_phy_check_loopbackpkt: Successfully verified loopback packet
eth0: smsc911x_mii_probe: Passed Loop Back Test
eth0: smsc911x_mii_probe: phy initialised succesfully
eth0: smsc911x_drv_probe: MAC Address is set to random_ether_addr
net eth0: MAC Address: ae:b1:a7:f7:89:c2
<snip>
Configuring network interfaces...
eth0: smsc911x_open: irq polarity: active low
eth0: smsc911x_open: irq type: open drain
eth0: smsc911x_open: Testing irq handler using IRQ 336
net eth0: ISR failed signaling test (IRQ 336)

I have a scope probe hanging on the IRQ signal.  I can see that IRQ is
high at power on, then I see a flurry of interrupts,  then at around
the time the "testing irq" message comes out I see IRQ go low and
remain low.  This behavior is 100% reproducible.

I tried with and without this patch, same behavior.

Any ideas what might be going wrong?

Steve
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html