Brandeburg, Jesse wrote:
> well, you top posted first, so I'm just going to reply in outlook :-)
Hmpf... I am not that religious about that - I did not see any reason
not to, but hey - here we go properly now ;)
> I think you have (for whatever reason) ASPM still enabled. If you send us
> (and look yourself) lspci -vvv then you may see ASPM: L0s and/or L1 enabled
> make sure to look in the LinkCtl: section
Yes, ASPM is/was enabled. I thought this to be a good idea on a
notebook. At least the config help did sound good.
> The only other thing maybe to check is the bits in the eeprom are updated
> to whatever we settled on as the latest. If you send the output of ethtool
> -e ethX we can check to make sure everything is okay.
At least the well known PM flag is off - I triple checked that by now.
> do you have the ASPM option on in your .config?
I had it enabled but disabled the ASPM in the kernel now. But it did not
help the situation. I have attached the lspci output for the device - I
am unsure how to interpret the ASPM flags still mentioned there.
I have also attached the eeprom value dump, just to be sure.
Cheers
nils
> -----Original Message-----
> From: Nils Faerber [mailto:[email protected]]
> Sent: Monday, August 09, 2010 3:19 AM
> To: [email protected]
> Subject: Re: [E1000-devel] e1000e latency issues
>
> As addendum here is a short ping log:
>
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4093 ttl=64 time=0.581 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4094 ttl=64 time=18.2 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4095 ttl=64 time=186 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4096 ttl=64 time=0.565 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4097 ttl=64 time=542 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4098 ttl=64 time=0.870 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4099 ttl=64 time=792 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4100 ttl=64 time=1.52 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4101 ttl=64 time=188 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4102 ttl=64 time=1.11 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4103 ttl=64 time=648 ms
> 64 bytes from piwi.kc.de (192.168.2.28): icmp_seq=4104 ttl=64 time=112 ms
>
> As you can see every second ping (at least every third) has times >>
> 100ms, in this log up to 648ms but they can even exceed 1000ms.
>
> I really need that interface to behave so any help would really be
> appreciated...
>
> Cheers
> nils
>
>
> Am 09.08.2010 00:25, schrieb Nils Faerber:
>> Hello!
>> I am more or less "trying" to use the e1000e driver on a Thinkpad X60.
>> Up to kernel 2.6.32 it worked quite fine. The next kernel I tried was
>> 2.6.34.1 and now 2.6.35 and there I find a very nasty issue which is
>> high packet latencies, varying. If I do e.g. a ping to machine on the
>> local net ping times are sometimes OK but most of the time > 300ms and
>> going up to 1000ms - which is inacceptable.
>>
>> The very nasty and for me kind of neck breaking side effect is that I
>> cannot boot my embedded boards over tftp from this e1000e machine
>> anymore which renders the notebook useless for my work :( The embedded
>> board's tftp client timeouts on almost every packet and gives up sooner
>> or later.
>>
>> Instead of the in-kernel e1000e driver I also tried the latest release
>> from your sf.net page e1000e-1.2.10.tar.gz.
>>
>> I also already tried different module parameters like disable the power
>> management, disabling MSI, etc. but no change. The only thing I know is
>> that the driver from 2.6.32 worked.
>>
>> If I can help to diagnose this let me know.
>> If you have a possible solution, even better ;)
>>
>> Thanks!
>>
>> PS: I am not subscribed to the e1000 list so in case you reply please
>> take my address into CC directly - thanks!
>>
>> Cheers
>> nils
--
kernel concepts GbR Tel: +49-271-771091-12
Sieghuetter Hauptweg 48
D-57072 Siegen Mob: +49-176-21024535
http://www.kernelconcepts.de
02:00.0 Ethernet controller: Intel Corporation 82573L Gigabit Ethernet
Controller
Subsystem: Lenovo Device 207e
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR+ FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort-
<MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 45
Region 0: Memory at ee000000 (32-bit, non-prefetchable) [size=128K]
Region 2: I/O ports at 2000 [size=32]
Capabilities: [c8] Power Management version 2
Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA
PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 PME-Enable- DSel=0 DScale=1 PME-
Capabilities: [d0] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0
Enable+
Address: 00000000fee0300c Data: 41c1
Capabilities: [e0] Express (v1) Endpoint, MSI 00
DevCap: MaxPayload 256 bytes, PhantFunc 0, Latency L0s <512ns,
L1 <64us
ExtTag- AttnBtn- AttnInd- PwrInd- RBE- FLReset-
DevCtl: Report errors: Correctable+ Non-Fatal+ Fatal+
Unsupported+
RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop+
MaxPayload 128 bytes, MaxReadReq 512 bytes
DevSta: CorrErr- UncorrErr+ FatalErr- UnsuppReq+ AuxPwr+
TransPend-
LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Latency
L0 <128ns, L1 <64us
ClockPM+ Suprise- LLActRep- BwNot-
LnkCtl: ASPM L1 Enabled; RCB 64 bytes Disabled- Retrain-
CommClk+
ExtSynch- ClockPM+ AutWidDis- BWInt- AutBWInt-
LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+
DLActive- BWMgmt- ABWMgmt-
Capabilities: [100] Advanced Error Reporting <?>
Capabilities: [140] Device Serial Number 48-91-32-ff-ff-d3-16-00
Kernel driver in use: e1000e
Kernel modules: e1000e
Offset Values
------ ------
0x0000 00 16 d3 32 91 48 30 0b b2 ff 51 00 ff ff ff ff
0x0010 53 00 03 02 6b 02 7e 20 aa 17 9a 10 86 80 df 80
0x0020 00 00 00 20 54 7e 00 00 14 00 da 00 04 00 00 27
0x0030 c9 6c 50 31 3e 07 0b 04 8b 29 00 00 00 f0 02 0f
0x0040 08 10 00 00 04 0f ff 7f 01 4d ff ff ff ff ff ff
0x0050 14 00 1d 00 14 00 1d 00 af aa 1e 00 00 00 1d 00
0x0060 00 01 00 40 32 12 07 40 ff ff ff ff ff ff ff ff
0x0070 ff ff ff ff ff ff ff ff ff ff ff ff ff ff 85 0c
------------------------------------------------------------------------------
This SF.net email is sponsored by
Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev
_______________________________________________
E1000-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit
http://communities.intel.com/community/wired
