netdev - Re: Regresion with dsa_port

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:   Sun, 25 Aug 2019 01:16:53 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Vivien Didelot <vivien.didelot@...il.com>
Cc:     netdev <netdev@...r.kernel.org>
Subject: Re: Regresion with dsa_port_disable

On Sat, Aug 24, 2019 at 07:12:20PM -0400, Vivien Didelot wrote:
> Hi Andrew,
> 
> On Sun, 25 Aug 2019 00:53:06 +0200, Andrew Lunn <andrew@...n.ch> wrote:
> > I just booted a ZII devel C and got a new warning splat.
> > 
> > WARNING: CPU: 0 PID: 925 at kernel/irq/manage.c:1708 __free_irq+0xc8/0x2c4
> > Trying to free already-free IRQ 0
> > Modules linked in:
> > CPU: 0 PID: 925 Comm: kworker/0:2 Not tainted 5.3.0-rc5-01151-g7ff758fcdf65 #231
> > Hardware name: Freescale Vybrid VF5xx/VF6xx (Device Tree)
> > Workqueue: events deferred_probe_work_func
> > Backtrace: 
> > [<8010d9e4>] (dump_backtrace) from [<8010dd9c>] (show_stack+0x20/0x24)
> >  r7:8016edf8 r6:00000009 r5:00000000 r4:9ec67944
> > [<8010dd7c>] (show_stack) from [<8083b03c>] (dump_stack+0x24/0x28)
> > [<8083b018>] (dump_stack) from [<8011c108>] (__warn.part.3+0xcc/0xf8)
> > [<8011c03c>] (__warn.part.3) from [<8011c1ac>] (warn_slowpath_fmt+0x78/0x94)
> >  r6:000006ac r5:80a8cbf0 r4:80d07088
> > [<8011c138>] (warn_slowpath_fmt) from [<8016edf8>] (__free_irq+0xc8/0x2c4)
> >  r3:00000000 r2:80a8cca8
> >  r7:9f486668 r6:9ee25268 r5:9f486600 r4:9ee25268
> > [<8016ed30>] (__free_irq) from [<8016f07c>] (free_irq+0x38/0x74)
> >  r10:9eeb3600 r9:9e412040 r8:00000009 r7:9ee26040 r6:9ee2404c r5:9ee242c8
> >  r4:9ee25268 r3:00000c00
> > [<8016f044>] (free_irq) from [<805a244c>] (mv88e6390x_serdes_irq_free+0x68/0x98)
> >  r5:9ee242c8 r4:9ee24040
> > [<805a23e4>] (mv88e6390x_serdes_irq_free) from [<8059bc94>] (mv88e6xxx_port_disable+0x58/0x98)
> >  r7:9ee26040 r6:00000009 r5:9ee2404c r4:9ee24040
> > [<8059bc3c>] (mv88e6xxx_port_disable) from [<80806f70>] (dsa_port_disable+0x44/0x50)
> >  r7:9ee26040 r6:9ee26d74 r5:00000009 r4:9ee26040
> > [<80806f2c>] (dsa_port_disable) from [<80805df0>] (dsa_register_switch+0x964/0xab8)
> >  r5:9efe194c r4:9ee26d38
> > [<8080548c>] (dsa_register_switch) from [<8059b734>] (mv88e6xxx_probe+0x730/0x778)
> >  r10:80943e64 r9:9fbf77d0 r8:00000000 r7:80d07088 r6:9e410040 r5:00000000
> >  r4:9e40e800
> > [<8059b004>] (mv88e6xxx_probe) from [<80582da8>] (mdio_probe+0x40/0x64)
> >  r10:00000012 r9:80d5eccc r8:00000000 r7:00000000 r6:8141f358 r5:9e40e800
> >  r4:80d5eccc
> > [<80582d68>] (mdio_probe) from [<80518858>] (really_probe+0x100/0x2d8)
> >  r5:9e40e800 r4:8141f354
> > 
> > The previous code was careful to balance mv88e6352_serdes_irq_setup()
> > with mv88e6390x_serdes_irq_free(). I _think_ your change broke this
> > balance, and we now try to free an interrupt which was never
> > allocated.
> 
> What do you mean by "balance mv88e6352_serdes_irq_setup() with
> mv88e6390x_serdes_irq_free()"?

Hi Vivien

It never called mv88e6390x_serdes_irq_free() unless
mv88e6352_serdes_irq_setup() had been called first.
mv88e6390x_serdes_irq_free() makes the assumption there actually is an
interrupt to free. I suspect your changes now call
mv88e6390x_serdes_irq_free() unconditionally.

	Andrew