* Intel PRO/1000 NIC
@ 2003-09-15 14:35 Lisong Xu
0 siblings, 0 replies; 3+ messages in thread
From: Lisong Xu @ 2003-09-15 14:35 UTC (permalink / raw)
To: netdev, linux-kernel
Hello,
I am sending UDP data from one PC to another PC directly through a cross
cable. The NIC of sender is Intel® PRO/1000 MT Server Adapter, and the NIC
of receiver is on-board Intel PRO/1000 (Dell PowerEdge 1600SC).
After tuning the kernel and driver parameters, the sender can send data at
1Gbps. But the receiver can only receive data at 620Mbps. (see the following
ifconfig messages, 30 seconds test)
*********** Sender *******************
eth1 Link encap:Ethernet HWaddr 00:07:E9:17:84:36
inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:9000 Metric:1
RX packets:111 errors:0 dropped:0 overruns:0 frame:0
TX packets:418072 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:30000
RX bytes:26193 (25.5 Kb) TX bytes:3759596626 (3585.4 Mb)
Interrupt:24 Base address:0xccc0 Memory:fcd00000-fcd20000
**************Receiver***********************
eth0 Link encap:Ethernet HWaddr 00:C0:9F:1E:3A:46
inet addr:192.168.1.101 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:9000 Metric:1
RX packets:261355 errors:156688 dropped:156688 overruns:156688
frame:0 <---- Look here!!!!!
TX packets:97 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:30000
RX bytes:2350019728 (2241.1 Mb) TX bytes:22821 (22.2 Kb)
Interrupt:11 Base address:0xecc0 Memory:fe100000-fe120000
***************************************************
My questions is why so many packets are "errors, dropped, overruns". I have
adjusted the "TxDescriptors, RxDescriptors, RxIntDelay,
RxAbsIntDelay, TxIntDelay, TxAbsIntDelay" for different values, but always
got similar result.
Maybe here is not the right email list to send, but I am really frustrated
with this problems, and hope to get some suggestions from the experts
here.Any suggestion is really appreciated!
Lisong
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: Intel PRO/1000 NIC
@ 2003-09-15 20:04 Feldman, Scott
0 siblings, 0 replies; 3+ messages in thread
From: Feldman, Scott @ 2003-09-15 20:04 UTC (permalink / raw)
To: Lisong Xu, netdev, linux-kernel
> I am sending UDP data from one PC to another PC directly
> through a cross cable. The NIC of sender is Intel(r) PRO/1000
> MT Server Adapter, and the NIC of receiver is on-board Intel
> PRO/1000 (Dell PowerEdge 1600SC).
A dump of lspci -vvv, ethtool -i eth<x>, and ethtool -d eth<x> will tell
us the 8254x controller version you're using; the driver version; if
you're running PCI or PCI-X, and at what speed; and what you're sharing
the bus with.
Ethtool version 1.8: http://sf.net/projects/gkernel.
Latest e1000: http://sf.net/projects/e1000.
-scott
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Intel PRO/1000 NIC
@ 2003-09-19 20:26 Lisong Xu
0 siblings, 0 replies; 3+ messages in thread
From: Lisong Xu @ 2003-09-19 20:26 UTC (permalink / raw)
To: Lisong Xu, netdev, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 2524 bytes --]
Hello,
Just in case you have a similar problem.
The possible reason is that the on-board Inter 1000 chip of Dell PowerEdge 1600SC is Intel 82540EM, which is running PCI 32/33 or PCI 32/66.
Lisong
----- Original Message -----
From: Lisong Xu
To: netdev@oss.sgi.com ; linux-kernel@vger.kernel.org
Sent: Monday, September 15, 2003 10:35 AM
Subject: Intel PRO/1000 NIC
Hello,
I am sending UDP data from one PC to another PC directly through a cross
cable. The NIC of sender is Intel® PRO/1000 MT Server Adapter, and the NIC
of receiver is on-board Intel PRO/1000 (Dell PowerEdge 1600SC).
After tuning the kernel and driver parameters, the sender can send data at
1Gbps. But the receiver can only receive data at 620Mbps. (see the following
ifconfig messages, 30 seconds test)
*********** Sender *******************
eth1 Link encap:Ethernet HWaddr 00:07:E9:17:84:36
inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:9000 Metric:1
RX packets:111 errors:0 dropped:0 overruns:0 frame:0
TX packets:418072 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:30000
RX bytes:26193 (25.5 Kb) TX bytes:3759596626 (3585.4 Mb)
Interrupt:24 Base address:0xccc0 Memory:fcd00000-fcd20000
**************Receiver***********************
eth0 Link encap:Ethernet HWaddr 00:C0:9F:1E:3A:46
inet addr:192.168.1.101 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:9000 Metric:1
RX packets:261355 errors:156688 dropped:156688 overruns:156688
frame:0 <---- Look here!!!!!
TX packets:97 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:30000
RX bytes:2350019728 (2241.1 Mb) TX bytes:22821 (22.2 Kb)
Interrupt:11 Base address:0xecc0 Memory:fe100000-fe120000
***************************************************
My questions is why so many packets are "errors, dropped, overruns". I have
adjusted the "TxDescriptors, RxDescriptors, RxIntDelay,
RxAbsIntDelay, TxIntDelay, TxAbsIntDelay" for different values, but always
got similar result.
Maybe here is not the right email list to send, but I am really frustrated
with this problems, and hope to get some suggestions from the experts
here.Any suggestion is really appreciated!
Lisong
[-- Attachment #2: Type: text/html, Size: 4808 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-09-19 20:26 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-09-15 14:35 Intel PRO/1000 NIC Lisong Xu
-- strict thread matches above, loose matches on Subject: below --
2003-09-15 20:04 Feldman, Scott
2003-09-19 20:26 Lisong Xu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).