Re: 3Com 905C very slow (400Kbps)

Bogdan Costescu wrote:

>On Mon, 2 Jun 2003, Florian Hackenberger wrote:
>
>  
>
>>The mii-diag says that autonegotiated media type is 100baseTX-FD. I'm
>>using an US Robotics Hub.
>>    
>>
>
>Unless this is a _switching_ hub a.k.a. switch, the FD media setting is 
>wrong and would perfectly explain what you are seeing.
>
I've brought the 3Com card back to the shop since it was expensive and 
didn't perform that well.
I'm now testing on the Realtek 8139 Card.

>
>  
>
>>With NFS the throughput is about 300-500Mbps.
>>    
>>
>
>If you'd post something like this on the Linux NFS list you'll be regarded 
>as a hero :-) I assume that you wanted to say Kbps instead of Mbps.
>
That was my fault sorry. The M and the K are too close to each other on 
the keyboard ;-)

>
>  
>
>>The 3Com card doesn't drop or overrun packets (ifconfig claims that).
>>    
>>
>
>Could you post the content of /proc/net/dev file ? (after seeing this 
>behaviour for some time)
>  
>
Ok, now i see erroneous packets and collisions.

This is from the P75:
Inter-|   Receive                                                |  Transmit
 face |bytes    packets errs drop fifo frame compressed 
multicast|bytes    packets errs drop fifo colls carrier compressed
    lo:881263384   90552    0    0    0     0          0         0 
881263384   90552    0    0    0     0       0          0
dummy0:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
 tunl0:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
  eth0:13339582    9666    0    0    0     0          0         0   
202713    1445    0    0    0     0       0          0

This is from my NB (P4 2Ghz) Sis900 ethernet:
Inter-|   Receive                                                |  Transmit
 face |bytes    packets errs drop fifo frame compressed 
multicast|bytes    packets errs drop fifo colls carrier compressed
    lo:3360894931  306615    0    0    0     0          0         0 
3360894931  306615    0    0    0     0       0          0
dummy0:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
 tunl0:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
vmnet8:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
vmnet1:       0       0    0    0    0     0          0         0        
0       0    0    0    0     0       0          0
  eth0:10851530   20130    0    0    0     0          0         0 
50981444   51197  377    0    0     1     754          0

I'm copying a 49 MB file over NFS and get about 200-300Kbps.

When using a crossover patch cable, there are no collisions nor 
erroneous packages, but the transfer rate is still at 600Kbps, even if 
mii-diag claims 100baseTx-FD.

>  
>
>>Is the CPU to slow for the network card, or something else?
>>    
>>
>
>Yes, it might. And quite a lot of mainboards from that era had PCI 
>problems... PCI problems are usually disruptive and so they are reported 
>by the driver - do your logs contain some messages from the network card ?
>
>You could try to use ttcp or netperf or something similar to find what 
>speed the computer is able to sustain with pure network load (= without 
>disks being involved).
>  
>
This is the printout from netperf (P75):
TCP STREAM TEST to localhost
Recv   Send    Send
Socket Socket  Message  Elapsed
Size   Size    Size     Time     Throughput
bytes  bytes   bytes    secs.    10^6bits/sec

 87380  16384  16384    10.00      89.99

This is from my P4
TCP STREAM TEST to localhost
Recv   Send    Send
Socket Socket  Message  Elapsed
Size   Size    Size     Time     Throughput
bytes  bytes   bytes    secs.    10^6bits/sec

 87380  16384  16384    9.99     3964.88

BTW the values are quite the same when switching to 100baseTx-HD with 
mii-diag on both machines.

>  
>
>>I've tried the Realtek 8139C (which should be worse) as well...same
>>result.
>>    
>>
>
>This would likely indicate either the computer itself or the link (cable, 
>hub/switch) as being responsible.
>
>  
>
Here is a printout from dmesg (P75):

Linux version 2.4.20 (root@hacki-mobil) (gcc version 3.2.2 20030322 
(Gentoo Linux 1.4 3.2.2-r2)) #41 Wed May 28 11:12:34 CEST 2003
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
 BIOS-e820: 00000000000e8000 - 0000000000100000 (reserved)
 BIOS-e820: 0000000000100000 - 0000000003000000 (usable)
 BIOS-e820: 00000000fffe0000 - 0000000100000000 (reserved)
48MB LOWMEM available.
On node 0 totalpages: 12288
zone(0): 4096 pages.
zone(1): 8192 pages.
zone(2): 0 pages.
Kernel command line: root=/dev/hda6
Initializing CPU#0
Detected 74.704 MHz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 149.09 BogoMIPS
Memory: 45352k/49152k available (1838k kernel code, 3416k reserved, 656k 
data, 132k init, 0k highmem)
Dentry cache hash table entries: 8192 (order: 4, 65536 bytes)
Inode cache hash table entries: 4096 (order: 3, 32768 bytes)
Mount-cache hash table entries: 1024 (order: 1, 8192 bytes)
Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes)
Page-cache hash table entries: 16384 (order: 4, 65536 bytes)
Intel Pentium with F0 0F bug - workaround enabled.
CPU:     After generic, caps: 000001bf 00000000 00000000 00000000
CPU:             Common caps: 000001bf 00000000 00000000 00000000
CPU: Intel Pentium 75 - 200 stepping 05
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
PCI: PCI BIOS revision 2.10 entry at 0xfbcd0, last bus=0
PCI: Using configuration type 1
PCI: Probing PCI hardware
Limiting direct PCI/PCI transfers.
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
apm: BIOS version 1.1 Flags 0x03 (Driver version 1.16)
Starting kswapd
Journalled Block Device driver loaded
devfs: v1.12c (20020818) Richard Gooch ([email protected])
devfs: boot_options: 0x1
Installing knfsd (copyright (C) 1996 [email protected]).
NTFS driver v1.1.22 [Flags: R/O]
parport0: PC-style at 0x378 [PCSPP,EPP]
parport0: Printer, HEWLETT-PACKARD DESKJET 640C
Detected PS/2 Mouse Port.
pty: 256 Unix98 ptys configured
Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ 
SERIAL_PCI ISAPNP enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
lp0: using parport0 (polling).
Real Time Clock Driver v1.10e
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX: IDE controller on PCI bus 00 dev 38
PIIX: chipset revision 2
PIIX: not 100% native mode: will probe irqs later
PIIX: neither IDE port enabled (BIOS)
PIIX: IDE controller on PCI bus 00 dev 39
PIIX: chipset revision 2
PIIX: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xe800-0xe807, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0xe808-0xe80f, BIOS settings: hdc:pio, hdd:pio
hda: ST340016A, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
blk: queue c03bf884, I/O limit 4095Mb (mask 0xffffffff)
hda: 78165360 sectors (40021 MB) w/2048KiB Cache, CHS=4865/255/63, (U)DMA
ide-floppy driver 0.99.newide
Partition check:
 /dev/ide/host0/bus0/target0/lun0: p1 p2 < p5 p6 >
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
0 3c515 cards found.
Linux agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 16M
agpgart: no supported devices found.
[drm] Initialized tdfx 1.0.0 20010216 on minor 0
[drm] Initialized radeon 1.1.1 20010405 on minor 1
[drm:drm_init] *ERROR* Cannot initialize the agpgart module.
ide-floppy driver 0.99.newide
SCSI subsystem driver Revision: 1.00
kmod: failed to exec /sbin/modprobe -s -k scsi_hostadapter, errno = 2
Linux Kernel Card Services 3.1.22
  options:  [pci] [cardbus] [pm]
usb.c: registered new driver usbdevfs
usb.c: registered new driver hub
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 4096 bind 8192)
IPv4 over IPv4 tunneling driver
Linux IP multicast router 0.06 plus PIM-SM
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
ds: no socket drivers loaded!
FAT: bogus logical sector size 0
FAT: bogus logical sector size 0
reiserfs: checking transaction log (device 03:06) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
VFS: Mounted root (reiserfs filesystem) readonly.
Mounted devfs on /dev
Freeing unused kernel memory: 132k freed
Adding Swap: 255992k swap-space (priority -1)
8139too Fast Ethernet driver 0.9.26
eth0: RealTek RTL8139 Fast Ethernet at 0xc38d8000, 00:40:f4:34:9c:95, IRQ 11
eth0:  Identified 8139 chip type 'RTL-8139C'
hda: drive_cmd: status=0x51 { DriveReady SeekComplete Error }
hda: drive_cmd: error=0x04 { DriveStatusError }
eth0: Setting 100mbps full-duplex based on auto-negotiated partner 
ability 45e1.
svc: unknown version (3)
apm: set display: Unrecognized device ID
apm: set display: Unrecognized device ID
apm: set display: Unrecognized device ID
apm: set display: Unrecognized device ID
nfsd: last server has exited
nfsd: unexporting all filesystems
pci-scan.c:v1.11 8/31/2002  Donald Becker <[email protected]> 
http://www.scyld.com/linux/drivers.html
rtl8139.c:v1.22 11/17/2002 Donald Becker, [email protected].
 http://www.scyld.com/network/rtl8139.html
eth0: RealTek RTL8139C Fast Ethernet at 0xe400, IRQ 11, 00:40:f4:34:9c:95.
svc: unknown version (3)


-- 
Florian Hackenberger
student @
University of Technology
Graz, Austria
[email protected]


_______________________________________________
vortex mailing list
[email protected]
http://www.scyld.com/mailman/listinfo/vortex