kern/152828: [em] poor performance on 8.1, 8.2-PRE

[Jack Vogel]

Did you eliminate that 7.0 box as part of the issue? Meaning switch to say a
Linux
or 8.2 partner to see if the problem persists or stops?

Cheers,

Jack


On Wed, Jul 20, 2011 at 10:02 AM, Steven Hartland
<killing at multiplay.co.uk>wrote:

> **
> Apologies, thought it might be worth a shot, but clearly not thanks for
> clarifying :)
>
> You absolutely correct with the msix masking the issue, we've just this
> minute had it stall again even with MSIX disabled :(
>
> I agree we have many machines using igb and em without issue even under
> lots of load.
>
> Not wishing to hijack this thread, but any advise on what else to try to
> diagnose this stalling issue would be gratefully received.
>
>     Regards
>     Steve
>
> ----- Original Message -----
> *From:* Jack Vogel <jfvogel at gmail.com>
> *To:* Steven Hartland <killing at multiplay.co.uk>
> *Cc:* Michael W. Lucas <mwlucas at blackhelicopters.org> ;
> freebsd-net at freebsd.org
> *Sent:* Wednesday, July 20, 2011 5:38 PM
> *Subject:* Re: kern/152828: [em] poor performance on 8.1, 8.2-PRE
>
> This is an em bug, igb does not share interrupt code with em, furthermore
> the adapter
> the filer of the bug is using doesn't use MSIX, only one type in the em
> driver does,
> and I might add lots of heavy users on that adapter have MSIX on without
> issues.
>
> I believe turning MSIX off is just masking whatever your real issue is, if
> you wish to
> run that way that's fine, but please don't advertise that as a 'fix' in
> some general way.
>
> Regards,
>
> Jack
>
>
> On Wed, Jul 20, 2011 at 9:07 AM, Steven Hartland <killing at multiplay.co.uk>wrote:
>
>> We're seeing tcp stalls under igb under 8.2-RELEASE and 8-STABLE (which
>> shares
>> some code with em) and the workaround for use is currently adding the
>> following
>> to /boot/loader.conf
>> hw.igb.enable_msix=0
>>
>> Might be worth trying that.
>>
>>   Regards
>>   Steve
>>
>> ----- Original Message ----- From: "Michael W. Lucas" <
>> mwlucas at blackhelicopters.org>
>> To: <freebsd-net at FreeBSD.org>
>> Sent: Wednesday, July 20, 2011 4:50 PM
>>
>> Subject: Re: kern/152828: [em] poor performance on 8.1, 8.2-PRE
>>
>>
>> The following reply was made to PR kern/152828; it has been noted by
>>> GNATS.
>>>
>>> From: "Michael W. Lucas" <mwlucas at blackhelicopters.org>
>>> To: bug-followup at FreeBSD.org
>>> Cc: rick at sloservers.com
>>> Subject: Re: kern/152828: [em] poor performance on 8.1, 8.2-PRE
>>> Date: Wed, 20 Jul 2011 11:49:08 -0400
>>>
>>> I'm seeing the exact same problem here as the original poster. Is
>>> anybody looking at this?
>>>
>>> Running FreeBSD 8.2-RELEASE-p1 amd64, from FreeNAS 8.0
>>>
>>> sysctls:
>>> dev.em.0.%desc: Intel(R) PRO/1000 Network Connection 7.1.9
>>> dev.em.0.%driver: em
>>> dev.em.0.%location: slot=0 function=0
>>> dev.em.0.%pnpinfo: vendor=0x8086 device=0x108b subvendor=0x8086
>>> subdevice=0x0000 class=0x020000
>>> dev.em.0.%parent: pci1
>>> dev.em.0.nvm: -1
>>> dev.em.0.debug: -1
>>> ...
>>>
>>> ifconfig:
>>> em0: flags=8843<UP,BROADCAST,**RUNNING,SIMPLEX,MULTICAST> metric 0 mtu
>>> 1500
>>>        options=219b<RXCSUM,TXCSUM,**VLAN_MTU,VLAN_HWTAGGING,VLAN_**
>>> HWCSUM,TSO4,WOL_MAGIC>
>>>        ether 00:15:17:31:c8:fe
>>>        inet 139.171.199.13 netmask 0xffffff00 broadcast 139.171.199.255
>>>        media: Ethernet autoselect (1000baseT <full-duplex>)
>>>        status: active
>>>
>>> dmesg | grep em0
>>> em0: <Intel(R) PRO/1000 Network Connection 7.1.9> port 0xbc00-0xbc1f mem
>>> 0xff6e0000-0xff6fffff irq 16 at device 0.0 on pci1
>>> em0: Using an MSI interrupt
>>> em0: [FILTER]
>>> em0: Ethernet address: 00:15:17:31:c8:fe
>>>
>>>
>>>
>>>
>>> --
>>> Michael W. Lucas http://www.MichaelWLucas.com/,
>>> http://blather.MichaelWLucas.**com/ <http://blather.MichaelWLucas.com/>
>>> Latest book: Network Flow Analysis http://www.**networkflowanalysis.com/<http://www.networkflowanalysis.com/>
>>> mwlucas at BlackHelicopters.org, Twitter @mwlauthor
>>> ______________________________**_________________
>>> freebsd-net at freebsd.org mailing list
>>> http://lists.freebsd.org/**mailman/listinfo/freebsd-net<http://lists.freebsd.org/mailman/listinfo/freebsd-net>
>>> To unsubscribe, send any mail to "freebsd-net-unsubscribe@**freebsd.org<freebsd-net-unsubscribe at freebsd.org>
>>> "
>>>
>>>
>> ==============================**==================
>> This e.mail is private and confidential between Multiplay (UK) Ltd. and
>> the person or entity to whom it is addressed. In the event of misdirection,
>> the recipient is prohibited from using, copying, printing or otherwise
>> disseminating it or any information contained in it.
>> In the event of misdirection, illegible or incomplete transmission please
>> telephone +44 845 868 1337 <%2B44%20845%20868%201337>
>> or return the E.mail to postmaster at multiplay.co.uk.
>>
>>
>> ______________________________**_________________
>> freebsd-net at freebsd.org mailing list
>> http://lists.freebsd.org/**mailman/listinfo/freebsd-net<http://lists.freebsd.org/mailman/listinfo/freebsd-net>
>> To unsubscribe, send any mail to "freebsd-net-unsubscribe@**freebsd.org<freebsd-net-unsubscribe at freebsd.org>
>> "
>>
>
>
> ================================================
> This e.mail is private and confidential between Multiplay (UK) Ltd. and the
> person or entity to whom it is addressed. In the event of misdirection, the
> recipient is prohibited from using, copying, printing or otherwise
> disseminating it or any information contained in it.
>
> In the event of misdirection, illegible or incomplete transmission please
> telephone +44 845 868 1337
> or return the E.mail to postmaster at multiplay.co.uk.
>