lx164 motherboard: 
Sep  3 16:18:23 a002 /netbsd: Digital AlphaPC 164LX 533 MHz, 531MHz
Sep  3 16:18:23 a002 /netbsd: 8192 byte page size, 1 processor.
Sep  3 16:18:23 a002 /netbsd: real mem = 134217728 (1949696 reserved for 
PROM, 1
32268032 used by NetBSD)
Sep  3 16:18:23 a002 /netbsd: avail mem = 113287168
Sep  3 16:18:24 a002 /netbsd: using 1614 buffers containing 13221888 
bytes of me
mory 
Sep  3 16:18:24 a002 /netbsd: mainbus0 (root)
Sep  3 16:18:24 a002 /netbsd: cpu0 at mainbus0: ID 0 (primary), 21164A 
(pass 2)
Sep  3 16:18:24 a002 /netbsd: cia0 at mainbus0: DECchip 2117x Core Logic 
Chipset
 (Pyxis), pass 1
Sep  3 16:18:24 a002 /netbsd: cia0: extended capabilities: 1<BWEN>
Sep  3 16:18:24 a002 /netbsd: cia0: using BWX for PCI config and device 
access
Sep  3 16:18:24 a002 /netbsd: pci0 at cia0 bus 0
Sep  3 16:18:24 a002 /netbsd: pci0: i/o enabled, memory enabled
Sep  3 16:18:24 a002 /netbsd: fxp0 at pci0 dev 7 function 0: Intel 
EtherExpress
Pro 10+/100B Ethernet
Sep  3 16:18:24 a002 /netbsd: fxp0: interrupting at eb164 irq 1


performance with tulip-based ethernet cards has been right about 1 
mbit/second. now with the fxp0 it's also very slow. we are also getting 
timeout messages on fxp0. I think we're losing interrupts. 

Any thoughts here? could we have a motherboard problem (with 16 
motherboards)? This is really driving us nuts. I may just ship the 
friggin' things back and chalk it up to another DEC screwup ... we saw 
the same poor performance with both windows/nt and linux. 

ron

Ron Minnich                |"Using Windows NT, which is known to have some 
rminnich@sarnoff.com       | failure modes, on a warship is similar to hoping 
(609)-734-3120             | that luck will be in our favor"- A. Digiorgio
ftp://ftp.sarnoff.com/pub/mnfs/www/docs/cluster.html