[eepro100] Performance prob. (again)

[Arthur Rinkel]

Hi,

I'm having some disappointing results with a eepro100 (with i82557) and
allthough I've tried several things to fix it, nothing helped so far...

The (Rx) speed doesn't seem to exceed 2MB/s or so; the NIC is installed in
a Pentium system running 133MHz. A TCP/IP stack "test" with a prg called
ttcp resulted in a (top) speed of 5,8MB/s, so I'm guessing the CPU is 
capable of doing 4-5MB/s of network throughput. Or is this incorrect?

Tried different UTP cable...no change.
Tried Intel driver...no change.
Played with module parms (Donald's driver)...small improvement.
Forced NIC 100Mb-only...connecting switch sees 100Mb-FD.
Auto-neg. NIC...connecting switch sees 100Mb-FD, good, but still 2MB/s.
'ifconfig' reports...RX packets:214660 errors:21 dropped:0 overruns:0
                     frame:504.

Any other thing there is to test?


I have some debugging info too, but that's about 80 lines, so I'll only
post it if it's necessary. Below however, is the output of
'eepro100-diag -aaeemf':

eepro100-diag.c:v2.02 7/19/2000 Donald Becker (becker@scyld.com)
 http://www.scyld.com/diag/index.html
Index #1: Found a Intel i82557 (or i82558) EtherExpressPro100B adapter at 0xe400.
i82557 chip registers at 0xe400:
  00000050 021af010 00000000 00080002 182541e1 00000600
  No interrupt sources are pending.
   The transmit unit state is 'Suspended'.
   The receive unit state is 'Ready'.
  This status is normal for an activated but idle interface.
EEPROM contents, size 64x16:
    00: a000 a6c9 81a8 0000 0000 0101 4701 0000
  0x08: 6784 0001 4000 0001 8086 0000 0000 0000
      ...
  0x38: 0000 0000 0000 0000 0000 0000 0000 823b
 The EEPROM checksum is correct.
Intel EtherExpress Pro 10/100 EEPROM contents:
  Station address 00:A0:C9:A6:A8:81.
  Receiver lock-up bug exists. (The driver work-around *is* implemented.)
  Board assembly 678400-001, Physical connectors present: RJ45
  Primary interface chip i82555 PHY #1.
 MII PHY #1 transceiver registers:
  3000 782d 02a8 0150 01e1 41e1 0001 ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  0a03 0000 0001 0000 0000 0000 0000 0000
  0000 0000 16f6 0000 ffff ffff ffff ffff.

Any ideas?

Grtz, Arthur