netdev - Re: [PATCH net] net: phy: fix interrupt handling in non-started states

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]

Message-ID: <ceacaf47-21c9-b59a-11cf-f5d1f48aa592@gmail.com>
Date:   Tue, 12 Feb 2019 21:06:53 +0100
From:   Heiner Kallweit <hkallweit1@...il.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     Florian Fainelli <f.fainelli@...il.com>,
        David Miller <davem@...emloft.net>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Russell King - ARM Linux <linux@...linux.org.uk>
Subject: Re: [PATCH net] net: phy: fix interrupt handling in non-started
 states

On 12.02.2019 20:37, Andrew Lunn wrote:
> On Tue, Feb 12, 2019 at 07:56:15PM +0100, Heiner Kallweit wrote:
>> phylib enables interrupts before phy_start() has been called, and if
>> we receive an interrupt in a non-started state, the interrupt handler
>> returns IRQ_NONE. This causes problems with at least one Marvell chip
>> as reported by Andrew.
>> Fix this by handling interrupts the same as in phy_mac_interrupt(),
>> basically always running the phylib state machine. It knows when it
>> has to do something and when not.
>> This change allows to handle interrupts gracefully even if they
>> occur in a non-started state.
>>
>> Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
>> Reported-by: Andrew Lunn <andrew@...n.ch>
>> Signed-off-by: Heiner Kallweit <hkallweit1@...il.com>
>> ---
>>  drivers/net/phy/phy.c | 3 ---
>>  1 file changed, 3 deletions(-)
>>
>> diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
>> index 189cd2048c3a..ca5e0c0f018c 100644
>> --- a/drivers/net/phy/phy.c
>> +++ b/drivers/net/phy/phy.c
>> @@ -762,9 +762,6 @@ static irqreturn_t phy_interrupt(int irq, void *phy_dat)
>>  {
>>  	struct phy_device *phydev = phy_dat;
>>  
>> -	if (!phy_is_started(phydev))
>> -		return IRQ_NONE;		/* It can't be ours.  */
>> -
>>  	if (phydev->drv->did_interrupt && !phydev->drv->did_interrupt(phydev))
>>  		return IRQ_NONE;
> 
> Hi Heiner
> 
> Did you look at
> 
Thanks for this interesting old read.

> ommit 3c3070d713d798f7f9e7ee3614e49b47655d14d8
> Author: Maciej W. Rozycki <macro@...ux-mips.org>
> Date:   Tue Oct 3 16:18:35 2006 +0100
> 
>     [PATCH] 2.6.18: sb1250-mac: Phylib IRQ handling fixes
>     
>      This patch fixes a couple of problems discovered with interrupt handling
>     in the phylib core, namely:
>     
>     1. The driver uses timer and workqueue calls, but does not include
>        <linux/timer.h> nor <linux/workqueue.h>.
>     
>     2. The driver uses schedule_work() for handling interrupts, but does not
>        make sure any pending work scheduled thus has been completed before
>        driver's structures get freed from memory.  This is especially
>        important as interrupts may keep arriving if the line is shared with
>        another PHY.
>     
Since that time this have massively changed and I *think* this doesn't apply
any longer. cancel_delayed_work_sync() is called always before driver
structure is freed.

>        The solution is to ignore phy_interrupt() calls if the reported device
>        has already been halted and calling flush_scheduled_work() from
>        phy_stop_interrupts() (but guarded with current_is_keventd() in case
>        the function has been called through keventd from the MAC device's
>        close call to avoid a deadlock on the netlink lock).
>     
>     Signed-off-by: Maciej W. Rozycki <macro@...ux-mips.org>
>     
>     patch-mips-2.6.18-20060920-phy-irq-16
>     Signed-off-by: Jeff Garzik <jeff@...zik.org>
> 
> There has been a lot of change since then, so maybe this is no longer
> an issue?
> 
AFAICS, yes. I'll have a closer look at the scenario described by Russell,
this however should be independent of the patch here.

>    Thanks
> 	Andrew
> 
Heiner