Bug#627700: New Intel Ethernet adapters (e1000e driver)
Hi,
> > > In this kernel version, the e1000e driver is missing support for
> > > i82567V-4 and i82579 and important bug fixes for i82577, i82578 and
> > > i82583.
> >
> > Here's our test results for the previously unsupported 82579V card:
>
> Thanks.
>
> [...]
Additional test results for a different system with a previously unsupported
Intel 82577 card:
00:19.0 Ethernet controller: Intel Corporation Device 1503 (rev 04)
Subsystem: Fujitsu Limited. Device 161c
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
Interrupt: pin A routed to IRQ 27
Region 0: Memory at e2600000 (32-bit, non-prefetchable) [size=128K]
Region 1: Memory at e262b000 (32-bit, non-prefetchable) [size=4K]
Region 2: I/O ports at 3080 [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: 00000000fee0f00c Data: 41b9
Capabilities: [e0] PCIe advanced features <?>
Kernel driver in use: e1000e
Kernel modules: e1000e
> 1. If the driver tries to load firmware (only required for some chips),
> does this work once the firmware file(s) are installed?
No firmware is being loaded.
> 2. Can you receive and transmit VLAN-tagged frames after creating a VLAN
> interface?
The tests outlined in #627704 were successful.
> 3. Does the interface work after suspend and resume?
I couldn't test this.
> 4. Does the interface work after removing the cable for 10 seconds and
> reinserting it?
No problems occurred.
> 5. Does multicast configuration work? (IPv6 autoconfiguration or mDNS will
cover this.)
"avahi-browse --all" worked fine.
> 6. Can the interface send and receive TCP/IP across a LAN at the same
> speed, before and after these changes? (Use e.g. netperf to test this, but
> don't forget to remove the netperf package after use.)
See below.
> 7. Are any warnings or errors logged by the kernel during the preceding
tests?
None.
>> Since the card was previously unsupported I cannot compare it to the
> > previous state. The performance of large file copies was quite ok.
> >
> > root@lynx:~# netperf
> > TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to localhost
> > (127.0.0.1) port 0 AF_INET : demo
> > Recv Send Send
> > Socket Socket Message Elapsed
> > Size Size Size Time Throughput
> > bytes bytes bytes secs. 10^6bits/sec
> >
> > 87380 16384 16384 10.00 24326.17
>
> Well you have a very fast loopback interface...
Ah...
With real world usage the performance was as expected, copying files over scp
gives a throughput of 60-70 megabytes per second for both systems, with the
CPU power of the remote host being the apparent bottleneck.
Cheers,
Moritz
--
Moritz Mühlenhoff muehlenhoff@univention.de
Open Source Software Engineer and Consultant
Univention GmbH Linux for Your Business fon: +49 421 22 232- 0
Mary-Somerville-Str.1 28359 Bremen fax: +49 421 22 232-99
http://www.univention.de
Reply to: