HDAMA status

YhLu YhLu at tyan.com
Tue Aug 5 13:59:01 CEST 2003


I have checked the mptable.c and irq_table.c for hdama. They seem got some

And you maybe need to re-produce the two files.


Yinghai Lu

发件人: ron minnich [mailto:rminnich at lanl.gov] 
发送时间: 2003年8月5日 10:46
收件人: linuxbios at clustermatic.org
主题: HDAMA status

currently, I'm having PIRQ routing problems. 

Etherboot loads the kernel just fine, and here's something very odd: 
etherboot says the two MACs are at 1:3.0 and 1:4.0; the kernel says the 
macs are at 2:3.0 and 2:4.0. 

This is odd. Any one out there have a simple explanation for me? 

Also the PIRQ in the CVS looked quite wrong, but dumping the PIRQ from a 
running BIOS and using that fails as well. 

I've gone so far as to have LinuxBIOS just jam the same interrupts the 
standard bios uses by doing this in mainboard/arima/hdama/mainboard.c

static void
enable(struct chip *chip, enum chip_pass pass)

        struct mainboard_arima_hdama_config *conf =
                (struct mainboard_arima_hdama_config *)chip->chip_info;

        switch (pass) {
        default: break;
        case CONF_PASS_PRE_BOOT:
                struct device *p;
                /* PIRQ is still wrong. Just set ethernet for now */
                p = dev_find_slot(1, 0x18);
                pci_write_config8(p, 0x3c, 0xb);
                p = dev_find_slot(1, 0x20);
                pci_write_config8(p, 0x3c, 0xb);
                printk_err("JAM ethernet to IRQ 11!\n");


But it didn't help at all. I also tried doing the same crude thing to
device 2:3.0 but that did not help.

This business of the gige appearing on different busses is very odd; any 
explanation here?

Hints welcome :-)


Linuxbios mailing list
Linuxbios at clustermatic.org

More information about the coreboot mailing list