I am seeing an odd problem with a single processor board running a 3com chipset. eth0: 3Com PCI 3c905C Tornado at 0xd800, 00:e0:18:0b:10:e3, IRQ 4 product code ffff rev 00.14 date 15-31-127 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface. MII transceiver found at address 24, status 782d. Enabling bus-master transmits and whole-frame receives. eth0: scatter/gather enabled. h/w checksums enabled It is an Intel chipset (i815 according to system messages.) Not certain why eth0 is attaching itself to IRQ 4. (The bios screen is weird for this chipset. I think it believes it is a modem.) When running the stock kernel from Redhat 7.1 with IP firewalling, the card would stall out if there was no traffic on the card for 30 seconds or so. I rebuilt the kernel and stripped it down to the bare bones, with a few exceptions. I removed power management (thinking it might be a "finigan's wake on lan" issue) and the firewalling code (thinking it might be some other sort of chokage). That reduced the problem greatly, but I still get messages from the Netscreen router telling me that that machine is not responding. (I get one message every few hours.) Serial and parallel ports should be disabled on that box. I do have mtrr and the uniprocessor APIC code enabled. I am also using memory map on the network card. This is pretty frustrating. I have NEVER seen this kind of weirdness from that chipset. (Though this is the first time I have used the on-board version.) Ideas? Please Cc me on the message as I am over 3,000 messages behind on the kernel list and it will probably get buried in the mass. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
