In article <20060612.122614.78728006.masanobu@iij.ad.jp>,
	SAITOH Masanobu <masanobu@iij.ad.jp> writes:
> > and it actually detect my gigabit (i82573L iirc and according to pcictl),

My NetBSD desktop's motherboard has the same chip:

wm0 at pci3 dev 0 function 0: Intel i82573L Gigabit Ethernet, rev. 0
wm0: interrupting at ioapic0 pin 17 (irq 10)
wm0: PCI-Express bus
wm0: 256 word (8 address bits) SPI EEPROM
wm0: Ethernet address 00:15:f2:xx:xx:xx
makphy0 at wm0 phy 1: Marvell 88E1111 Gigabit PHY, rev. 2

> Other known problems (reported by tron):
> 
>  1) tso4 option cause the device timeout.

That problem is still there but IMHO not so urgent. Thanks to PCI Express
and the fast CPU driving the card TSO support isn't critical.

>  2) poor performance

That problem has been fixed. Outgoing:

ttcp-t: buflen=32768, nbuf=65536, align=16384/0, port=5001, sockbufsize=131072  tcp  -> ***
ttcp-t: socket
ttcp-t: sndbuf
ttcp-t: connect
ttcp-t: 2147483648 bytes in 18.52 real seconds = 113219.45 KB/sec +++
ttcp-t: 65536 I/O calls, msec/call = 0.29, calls/sec = 3538.11
ttcp-t: 0.0user 2.0sys 0:18real 11% 0i+0d 0maxrss 0+601481pf 102924+14csw

Incoming:

ttcp-r: buflen=8192, nbuf=2048, align=16384/0, port=5001, sockbufsize=131072  tcp
ttcp-r: socket
ttcp-r: rcvbuf
ttcp-r: accept from *.*.*.*
ttcp-r: 2147483648 bytes in 18.82 real seconds = 111408.89 KB/sec +++
ttcp-r: 308363 I/O calls, msec/call = 0.06, calls/sec = 16381.45
ttcp-r: 0.0user 1.8sys 0:18real 10% 0i+0d 0maxrss 0+2pf 57725+25csw

That's line rate.

	Kind regards

-- 
Matthias Scheler                                  http://scheler.de/~matthias/