Hi,
I came a bit further with debugging my mysterious reboots after setting up mtrr. This only happens if I build the image on an AMD64 distribution. With an older x86 distribution, the resulting image is operable.
Either my romcc changes were somewhat incomplete, or gcc -m32 on Opteron behaves different than gcc on a 32bit machine.
Stefan
* Stefan Reinauer stepan@suse.de [031006 13:48]:
I came a bit further with debugging my mysterious reboots after setting up mtrr. This only happens if I build the image on an AMD64 distribution. With an older x86 distribution, the resulting image is operable.
Either my romcc changes were somewhat incomplete, or gcc -m32 on Opteron behaves different than gcc on a 32bit machine.
argh! I just don't want to believe this. Using a more recent version of gcc helped to fix the problem. Now I get lots of these:
[tg3-5701]Ethernet addr: 00:04:76:F3:A9:CE Tigon3 [partno(3C996B-T) rev 0105 PHY(5701)] (PCI:33MHz:32-bit) Link is up at 1000 Mbps, full duplex. TX RX flow control Searching for server (DHCP)... ..UDP checksum error .UDP checksum error .UDP checksum error
After a while it works:
..Me: 192.168.0.20, Server: 192.168.0.23 Loading 192.168.0.23:legacybios ... <takes forever>
Best regards, Stefan Reinauer
* Stefan Reinauer stepan@suse.de [031006 14:13]:
the card is connected with a cross cable to a HDAMA..
etherboot 5.3.1:
[tg3-5701]Ethernet addr: 00:04:76:F3:A9:CE Tigon3 [partno(3C996B-T) rev 0105 PHY(5701)] (PCI:33MHz:32-bit) Link is up at 1000 Mbps, full duplex. TX RX flow control Searching for server (DHCP)... ..UDP checksum error .UDP checksum error .UDP checksum error ..Me: 192.168.0.20, Server: 192.168.0.23 Loading 192.168.0.23:legacybios ...
<takes forever>
etherboot 5.3.2: Etherboot 5.3.2 (GPL) http://etherboot.org Tagged ELF for [TG3][IDE] Relocating _text from: [000264d0,00058500) to [0fecdfd0,0ff00000) .ÿProbing pci nic...or (Q)uit? [tg3-5701]Ethernet addr: 00:04:76:F3:A9:CE Tigon3 [partno(3C996B-T) rev 0105 PHY(5701)] (PCI:33MHz:32-bit) Link is up at 1000 Mbps, full duplex. TX RX flow control Searching for server (DHCP)... ..transmit timed out tg3_stop_block timed out, ofs=00003c00 enable_bit=00000002 .<abort>
etherboot 5.2.2 seems to be the same as 5.2.0
Best regards, Stefan Reinauer
Greetings,
Is the cross cable fully wired for Gig ether? I've seen a lot of cables where only 2 pairs are crossed and the rest wired direct, and I've seen Gig ether NICs get terribly confused by that.
G'day, sjames
On Mon, 6 Oct 2003, Stefan Reinauer wrote:
- Stefan Reinauer stepan@suse.de [031006 14:13]:
the card is connected with a cross cable to a HDAMA..
etherboot 5.3.1:
[tg3-5701]Ethernet addr: 00:04:76:F3:A9:CE Tigon3 [partno(3C996B-T) rev 0105 PHY(5701)] (PCI:33MHz:32-bit) Link is up at 1000 Mbps, full duplex. TX RX flow control Searching for server (DHCP)... ..UDP checksum error .UDP checksum error .UDP checksum error ..Me: 192.168.0.20, Server: 192.168.0.23 Loading 192.168.0.23:legacybios ...
<takes forever>
etherboot 5.3.2: Etherboot 5.3.2 (GPL) http://etherboot.org Tagged ELF for [TG3][IDE] Relocating _text from: [000264d0,00058500) to [0fecdfd0,0ff00000) .�Probing pci nic...or (Q)uit? [tg3-5701]Ethernet addr: 00:04:76:F3:A9:CE Tigon3 [partno(3C996B-T) rev 0105 PHY(5701)] (PCI:33MHz:32-bit) Link is up at 1000 Mbps, full duplex. TX RX flow control Searching for server (DHCP)... ..transmit timed out tg3_stop_block timed out, ofs=00003c00 enable_bit=00000002 .<abort>
etherboot 5.2.2 seems to be the same as 5.2.0
Best regards, Stefan Reinauer
* steven james pyro@linuxlabs.com [031006 18:56]:
Greetings,
Is the cross cable fully wired for Gig ether? I've seen a lot of cables where only 2 pairs are crossed and the rest wired direct, and I've seen Gig ether NICs get terribly confused by that.
Hm. gotta check. I had the same cable working fine before. But I would not swear it did not use 100MBit or something..
Stefan
-
Stefan Reinauer -
steven james