Patchwork tg3: Dont dump registers if interface not ready.

login
register
mail settings
Submitter Joe Jin
Date Sept. 29, 2011, 2:10 a.m.
Message ID <4E83D3B3.7090508@oracle.com>
Download mbox | patch
Permalink /patch/116877/
State Changes Requested
Delegated to: David Miller
Headers show

Comments

Joe Jin - Sept. 29, 2011, 2:10 a.m.
When bootup the server with BCM5704 Gigabit Ethernet get below warning:

tg3 0000:03:01.0: eth0: DMA Status error.  Resetting chip.
<Registers state of device>
tg3 0000:03:01.0: eth0: 0: Host status block [00000007:00000002:(0000:0000:0000):(0000:0000)]
tg3 0000:03:01.0: eth0: 0: NAPI info [00000001:00000002:(0000:0000:01ff):0000:(00c8:0000:0000:0000)]
tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
tg3 0000:03:01.0: eth0: Link is down
tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX

If device not ready, then would not dump registers info.

Signed-off-by: Joe Jin <joe.jin@oracle.com>
Signed-off-by: Guru Anbalagane <guru.anbalagane@oracle.com>
Reported-by: Gurudas Pai <gurudas.pai@oracle.com>
Cc: Matt Carlson <mcarlson@broadcom.com>
Cc: Michael Chan <mchan@broadcom.com>
---
 drivers/net/tg3.c |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)
Xiao Jiang - Sept. 29, 2011, 6:50 a.m.
Joe Jin wrote:
> When bootup the server with BCM5704 Gigabit Ethernet get below warning:
>
> tg3 0000:03:01.0: eth0: DMA Status error.  Resetting chip.
> <Registers state of device>
> tg3 0000:03:01.0: eth0: 0: Host status block [00000007:00000002:(0000:0000:0000):(0000:0000)]
> tg3 0000:03:01.0: eth0: 0: NAPI info [00000001:00000002:(0000:0000:01ff):0000:(00c8:0000:0000:0000)]
> tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
> tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
> tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
> tg3 0000:03:01.0: eth0: Link is down
> tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
> tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
>
> If device not ready, then would not dump registers info.
>
> Signed-off-by: Joe Jin <joe.jin@oracle.com>
> Signed-off-by: Guru Anbalagane <guru.anbalagane@oracle.com>
> Reported-by: Gurudas Pai <gurudas.pai@oracle.com>
> Cc: Matt Carlson <mcarlson@broadcom.com>
> Cc: Michael Chan <mchan@broadcom.com>
> ---
>  drivers/net/tg3.c |   11 +++++++++--
>   
Maybe this one should based on net-next tree too, the tg3.c
is lived below ./drivers/net/ethernet/broadcom/ in that tree.

Thanks,
Xiao Jiang
>  1 files changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
> index 4a1374d..d79d344 100644
> --- a/drivers/net/tg3.c
> +++ b/drivers/net/tg3.c
> @@ -5475,10 +5475,15 @@ static void tg3_process_error(struct tg3 *tp)
>  {
>  	u32 val;
>  	bool real_error = false;
> +	bool dump = true;
>  
>  	if (tg3_flag(tp, ERROR_PROCESSED))
>  		return;
>  
> +	/* If interface not ready then dont dump error */
> +	if (!netif_carrier_ok(tp->dev))
> +		dump = false;
> +
>  	/* Check Flow Attention register */
>  	val = tr32(HOSTCC_FLOW_ATTN);
>  	if (val & ~HOSTCC_FLOW_ATTN_MBUF_LWM) {
> @@ -5492,14 +5497,16 @@ static void tg3_process_error(struct tg3 *tp)
>  	}
>  
>  	if (tr32(RDMAC_STATUS) || tr32(WDMAC_STATUS)) {
> -		netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
> +		if (dump)
> +			netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
>  		real_error = true;
>  	}
>  
>  	if (!real_error)
>  		return;
>  
> -	tg3_dump_state(tp);
> +	if (dump)
> +		tg3_dump_state(tp);
>  
>  	tg3_flag_set(tp, ERROR_PROCESSED);
>  	schedule_work(&tp->reset_task);
>   

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Matt Carlson - Oct. 5, 2011, 2:14 a.m.
On Wed, Sep 28, 2011 at 11:50:01PM -0700, Xiao Jiang wrote:
> Joe Jin wrote:
> > When bootup the server with BCM5704 Gigabit Ethernet get below warning:
> >
> > tg3 0000:03:01.0: eth0: DMA Status error.  Resetting chip.
> > <Registers state of device>
> > tg3 0000:03:01.0: eth0: 0: Host status block [00000007:00000002:(0000:0000:0000):(0000:0000)]
> > tg3 0000:03:01.0: eth0: 0: NAPI info [00000001:00000002:(0000:0000:01ff):0000:(00c8:0000:0000:0000)]
> > tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
> > tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
> > tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
> > tg3 0000:03:01.0: eth0: Link is down
> > tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
> > tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
> >
> > If device not ready, then would not dump registers info.
> >
> > Signed-off-by: Joe Jin <joe.jin@oracle.com>
> > Signed-off-by: Guru Anbalagane <guru.anbalagane@oracle.com>
> > Reported-by: Gurudas Pai <gurudas.pai@oracle.com>
> > Cc: Matt Carlson <mcarlson@broadcom.com>
> > Cc: Michael Chan <mchan@broadcom.com>
> > ---
> >  drivers/net/tg3.c |   11 +++++++++--
> >   
> Maybe this one should based on net-next tree too, the tg3.c
> is lived below ./drivers/net/ethernet/broadcom/ in that tree.
> 
> Thanks,
> Xiao Jiang
> >  1 files changed, 9 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
> > index 4a1374d..d79d344 100644
> > --- a/drivers/net/tg3.c
> > +++ b/drivers/net/tg3.c
> > @@ -5475,10 +5475,15 @@ static void tg3_process_error(struct tg3 *tp)
> >  {
> >  	u32 val;
> >  	bool real_error = false;
> > +	bool dump = true;
> >  
> >  	if (tg3_flag(tp, ERROR_PROCESSED))
> >  		return;
> >  
> > +	/* If interface not ready then dont dump error */
> > +	if (!netif_carrier_ok(tp->dev))
> > +		dump = false;

Would you still experience the problem if you did the following instead
of the above link check?

		if (tg3_flag(tp, INIT_COMPLETE))
			dump = false;

> >  	/* Check Flow Attention register */
> >  	val = tr32(HOSTCC_FLOW_ATTN);
> >  	if (val & ~HOSTCC_FLOW_ATTN_MBUF_LWM) {
> > @@ -5492,14 +5497,16 @@ static void tg3_process_error(struct tg3 *tp)
> >  	}
> >  
> >  	if (tr32(RDMAC_STATUS) || tr32(WDMAC_STATUS)) {
> > -		netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
> > +		if (dump)
> > +			netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
> >  		real_error = true;
> >  	}
> >  
> >  	if (!real_error)
> >  		return;
> >  
> > -	tg3_dump_state(tp);
> > +	if (dump)
> > +		tg3_dump_state(tp);
> >  
> >  	tg3_flag_set(tp, ERROR_PROCESSED);
> >  	schedule_work(&tp->reset_task);
> >   
> 
> 

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joe Jin - Oct. 8, 2011, 3:21 a.m.
>>>  
>>> +	/* If interface not ready then dont dump error */
>>> +	if (!netif_carrier_ok(tp->dev))
>>> +		dump = false;
> 
> Would you still experience the problem if you did the following instead
> of the above link check?
> 
> 		if (tg3_flag(tp, INIT_COMPLETE))
> 			dump = false;

I'll try it then update you.

Thanks,
Joe

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joe Jin - Oct. 17, 2011, 9:52 a.m.
Hi Matt,

Sorry for late response, after apply your changes, driver did not dumped
registers info as well, but still hit tg3_stop_block timed out as below:

tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
tg3 0000:03:01.0: eth0: Link is down
tg3 0000:03:01.1: eth1: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.1: eth1: Flow control is off for TX and off for RX
tg3 0000:03:01.1: eth1: Link is down
tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
tg3 0000:03:01.0: eth0: Link is down
RPC: Registered named UNIX socket transport module.
RPC: Registered udp transport module.
RPC: Registered tcp transport module.
RPC: Registered tcp NFSv4.1 backchannel transport module.
SELinux: initialized (dev rpc_pipefs, type rpc_pipefs), uses genfs_contexts
tg3 0000:03:01.1: eth1: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.1: eth1: Flow control is off for TX and off for RX
tg3 0000:03:01.1: eth1: Link is down
tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
tg3 0000:03:01.0: eth0: Link is down
tg3 0000:03:01.1: eth1: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.1: eth1: Flow control is off for TX and off for RX
ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready


Any suggestion?

Thanks,
Joe

On 10/05/11 10:14, Matt Carlson wrote:
> On Wed, Sep 28, 2011 at 11:50:01PM -0700, Xiao Jiang wrote:
>> Joe Jin wrote:
>>> When bootup the server with BCM5704 Gigabit Ethernet get below warning:
>>>
>>> tg3 0000:03:01.0: eth0: DMA Status error.  Resetting chip.
>>> <Registers state of device>
>>> tg3 0000:03:01.0: eth0: 0: Host status block [00000007:00000002:(0000:0000:0000):(0000:0000)]
>>> tg3 0000:03:01.0: eth0: 0: NAPI info [00000001:00000002:(0000:0000:01ff):0000:(00c8:0000:0000:0000)]
>>> tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
>>> tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
>>> tg3 0000:03:01.0: tg3_stop_block timed out, ofs=4800 enable_bit=2
>>> tg3 0000:03:01.0: eth0: Link is down
>>> tg3 0000:03:01.0: eth0: Link is up at 1000 Mbps, full duplex
>>> tg3 0000:03:01.0: eth0: Flow control is on for TX and on for RX
>>>
>>> If device not ready, then would not dump registers info.
>>>
>>> Signed-off-by: Joe Jin <joe.jin@oracle.com>
>>> Signed-off-by: Guru Anbalagane <guru.anbalagane@oracle.com>
>>> Reported-by: Gurudas Pai <gurudas.pai@oracle.com>
>>> Cc: Matt Carlson <mcarlson@broadcom.com>
>>> Cc: Michael Chan <mchan@broadcom.com>
>>> ---
>>>  drivers/net/tg3.c |   11 +++++++++--
>>>   
>> Maybe this one should based on net-next tree too, the tg3.c
>> is lived below ./drivers/net/ethernet/broadcom/ in that tree.
>>
>> Thanks,
>> Xiao Jiang
>>>  1 files changed, 9 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
>>> index 4a1374d..d79d344 100644
>>> --- a/drivers/net/tg3.c
>>> +++ b/drivers/net/tg3.c
>>> @@ -5475,10 +5475,15 @@ static void tg3_process_error(struct tg3 *tp)
>>>  {
>>>  	u32 val;
>>>  	bool real_error = false;
>>> +	bool dump = true;
>>>  
>>>  	if (tg3_flag(tp, ERROR_PROCESSED))
>>>  		return;
>>>  
>>> +	/* If interface not ready then dont dump error */
>>> +	if (!netif_carrier_ok(tp->dev))
>>> +		dump = false;
> 
> Would you still experience the problem if you did the following instead
> of the above link check?
> 
> 		if (tg3_flag(tp, INIT_COMPLETE))
> 			dump = false;
> 
>>>  	/* Check Flow Attention register */
>>>  	val = tr32(HOSTCC_FLOW_ATTN);
>>>  	if (val & ~HOSTCC_FLOW_ATTN_MBUF_LWM) {
>>> @@ -5492,14 +5497,16 @@ static void tg3_process_error(struct tg3 *tp)
>>>  	}
>>>  
>>>  	if (tr32(RDMAC_STATUS) || tr32(WDMAC_STATUS)) {
>>> -		netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
>>> +		if (dump)
>>> +			netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
>>>  		real_error = true;
>>>  	}
>>>  
>>>  	if (!real_error)
>>>  		return;
>>>  
>>> -	tg3_dump_state(tp);
>>> +	if (dump)
>>> +		tg3_dump_state(tp);
>>>  
>>>  	tg3_flag_set(tp, ERROR_PROCESSED);
>>>  	schedule_work(&tp->reset_task);
>>>   
>>
>>
>
Luis Henriques - March 22, 2012, 1:17 p.m.
Joe Jin <joe.jin@oracle.com> writes:

Hi,

> Hi Matt,
>
> Sorry for late response, after apply your changes, driver did not dumped
> registers info as well, but still hit tg3_stop_block timed out as below:
>
...
>
> Any suggestion?

I was wondering if there any progress with this patch.  Is this
stalled, or has it been dropped?

Cheers,

Patch

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 4a1374d..d79d344 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -5475,10 +5475,15 @@  static void tg3_process_error(struct tg3 *tp)
 {
 	u32 val;
 	bool real_error = false;
+	bool dump = true;
 
 	if (tg3_flag(tp, ERROR_PROCESSED))
 		return;
 
+	/* If interface not ready then dont dump error */
+	if (!netif_carrier_ok(tp->dev))
+		dump = false;
+
 	/* Check Flow Attention register */
 	val = tr32(HOSTCC_FLOW_ATTN);
 	if (val & ~HOSTCC_FLOW_ATTN_MBUF_LWM) {
@@ -5492,14 +5497,16 @@  static void tg3_process_error(struct tg3 *tp)
 	}
 
 	if (tr32(RDMAC_STATUS) || tr32(WDMAC_STATUS)) {
-		netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
+		if (dump)
+			netdev_err(tp->dev, "DMA Status error.  Resetting chip.\n");
 		real_error = true;
 	}
 
 	if (!real_error)
 		return;
 
-	tg3_dump_state(tp);
+	if (dump)
+		tg3_dump_state(tp);
 
 	tg3_flag_set(tp, ERROR_PROCESSED);
 	schedule_work(&tp->reset_task);