[PATCH] netdev/fec: fix ifconfig eth0 down hang issue (v5)

[Bryan Wu]

On 04/29/2010 05:15 AM, Colin Ian King wrote:
> On Wed, 2010-04-28 at 09:29 +0800, Bryan Wu wrote:
>> BugLink: http://bugs.launchpad.net/bugs/559065
>>
>> v5:
>> Only call fec_enet_mii_probe() in open function, otherwise the first open
>> action will cause NULL pointer error.
>>
>> v4:
>>   - don't touch suspend/resume functions in this patch
>>   - set fep->phy_dev = NULL in the fec_enet_mii_probe() entry
>>
>> v3:
>> if fec_enet_mii_probe() fails, we should exit the open and resume functions
>> properly and give up further operations without a proper PHY connected.
>>
>> v2:
>> Check fec_enet_mii_probe() return value directly, instead of fep->phy_dev
>>
>> v1:
>> In ethernet connection open/close function, we need to use phy_connect
>> and phy_disconnect operation before we start/stop phy. Otherwise it will
>> cause system hang. Also suspend/resume needs the same fix.
>>
>> Signed-off-by: Bryan Wu<bryan.wu at canonical.com>
>> ---
>>   drivers/net/fec.c |   28 ++++++++++++++++------------
>>   1 files changed, 16 insertions(+), 12 deletions(-)
>>
>> diff --git a/drivers/net/fec.c b/drivers/net/fec.c
>> index 2280373..692d3c4 100644
>> --- a/drivers/net/fec.c
>> +++ b/drivers/net/fec.c
>> @@ -691,6 +691,8 @@ static int fec_enet_mii_probe(struct net_device *dev)
>>   	struct phy_device *phy_dev = NULL;
>>   	int phy_addr;
>>
>> +	fep->phy_dev = NULL;
>> +
>>   	/* find the first phy */
>>   	for (phy_addr = 0; phy_addr<  PHY_MAX_ADDR; phy_addr++) {
>>   		if (fep->mii_bus->phy_map[phy_addr]) {
>> @@ -721,6 +723,11 @@ static int fec_enet_mii_probe(struct net_device *dev)
>>   	fep->link = 0;
>>   	fep->full_duplex = 0;
>>
>> +	printk(KERN_INFO "%s: Freescale FEC PHY driver [%s] "
>> +		"(mii_bus:phy_addr=%s, irq=%d)\n", dev->name,
>> +		fep->phy_dev->drv->name, dev_name(&fep->phy_dev->dev),
>> +		fep->phy_dev->irq);
>> +
>>   	return 0;
>>   }
>>
>> @@ -768,13 +775,8 @@ static int fec_enet_mii_init(struct platform_device *pdev)
>>   	if (mdiobus_register(fep->mii_bus))
>>   		goto err_out_free_mdio_irq;
>>
>> -	if (fec_enet_mii_probe(dev) != 0)
>> -		goto err_out_unregister_bus;
>> -
>>   	return 0;
>>
>> -err_out_unregister_bus:
>> -	mdiobus_unregister(fep->mii_bus);
>>   err_out_free_mdio_irq:
>>   	kfree(fep->mii_bus->irq);
>>   err_out_free_mdiobus:
>> @@ -928,7 +930,12 @@ fec_enet_open(struct net_device *dev)
>>   	if (ret)
>>   		return ret;
>>
>> -	/* schedule a link state check */
>> +	/* Probe and connect to PHY when open the interface */
>> +	ret = fec_enet_mii_probe(dev);
>> +	if (ret) {
>> +		fec_enet_free_buffers(dev);
>> +		return ret;
>> +	}
>>   	phy_start(fep->phy_dev);
>>   	netif_start_queue(dev);
>>   	fep->opened = 1;
>> @@ -942,10 +949,12 @@ fec_enet_close(struct net_device *dev)
>>
>>   	/* Don't know what to do yet. */
>>   	fep->opened = 0;
>> -	phy_stop(fep->phy_dev);
>>   	netif_stop_queue(dev);
>>   	fec_stop(dev);
>>
>> +	if (fep->phy_dev)
>> +		phy_disconnect(fep->phy_dev);
>> +
>>           fec_enet_free_buffers(dev);
>>   	clk_disable(fep->clk);
>>
>> @@ -1337,11 +1346,6 @@ fec_probe(struct platform_device *pdev)
>>
>>   	clk_disable(fep->clk);
>>
>> -	printk(KERN_INFO "%s: Freescale FEC PHY driver [%s] "
>> -		"(mii_bus:phy_addr=%s, irq=%d)\n", ndev->name,
>> -		fep->phy_dev->drv->name, dev_name(&fep->phy_dev->dev),
>> -		fep->phy_dev->irq);
>> -
>>   	return 0;
>>
>>   failed_register:
>> --
>> 1.7.0.4
>>
>>
> Minor question - it seems that fec_enet_close() is no longer doing a
> phy_stop().  Is that intentional and why can we get away with not doing
> a phy_stop()?
>

The original design uses phy_stop() here, we faced an ifdown hang issue. 
phy_disconnect will disconnect from the phy and stop it via it's internal 
statues machine. And when we open fec, phy_connect will be called in 
fec_enet_mii_probe().

Actually, I was a little bit confused by this phy_stop() and phy_disconnect() at 
beginning. But from our test, phy_stop() will cause ifdown hang issue. After 
looking at other driver such as drivers/net/davinci_emac.c, I think 
phy_disconnect is the right way here. Or maybe we can call phy_stop then 
followed by phy_disconnect, but phy_disconnect here is good enough. Let me ping 
upstream for help with the patch later.

Thanks,
-- 
Bryan Wu <bryan.wu at canonical.com>
Kernel Developer    +86.138-1617-6545 Mobile
Ubuntu Kernel Team | Hardware Enablement Team
Canonical Ltd.      www.canonical.com
Ubuntu - Linux for human beings | www.ubuntu.com