On Wed, 27 Jun 2001, Stew Benedict wrote:
> The help on cmd640 mentions it's a common, buggy implementation on x86
> motherboards - probably safe to drop it. I'll throw another kernel up -
>
> ftp://ftp.mandrakesoft.com/pub/sbenedict/vmlinux-no640
Okay, this new version gets rid of all of the "IN from bad port" errors.
> c0005840 T synchronize_irq
> c0005874 T __global_cli
> c0005a78 T __global_sti
> c0005af0 T __global_save_flags
>
> This looks to be part of the IRQ implementation on PPC:
>
> arch/ppc/kernel.irq.c - don't think we'll be losing this ;^)
Guess not :) It appears once for each drive as it is starting up:
hda: Enabling Ultra DMA 4
Bogus interrupt -1 from PC = c0005ae0
hda: 20028960 sectors (10255 MB) w/1961KiB Cache, CHS=19870/16/63, (U)DMA
hdb: Enabling Ultra DMA 4
Bogus interrupt -1 from PC = c0005ae0
hdb: 50859648 sectors (26040 MB) w/418KiB Cache, CHS=50456/16/63, (U)DMA
hdc: Enabling MultiWord DMA 2
Bogus interrupt -1 from PC = c0005ae0
hdc: ATAPI 32X DVD-ROM drive, 512kB Cache, (U)DMA
After that, I haven't seen it pop up again...
> The SMP is intentional, but not for the RPM kernel. In 2.2, I couldn't
> boot an SMP kernel on a non-SMP machine, with this one it seems I can, and
> I'm considering using this kernel in the next iteration of the installer,
> so both type of machines can use it, and maybe we can get more machines
> running the graphical install, if I can get the Xpmac speed to pick up a
> bit.
Won't a UP kernel work on an SMP PPC machine? (Don't you love the
abbreviations? :) ) You won't get the benefit of the extra processors, but
I figured the install was more IO intensive than anything else...
> The AGP vs. PCI is probably due to switching to BenH's drivers. Is your
> actual card AGP or PCI?
It really is an AGP card. The dmesg now says:
aty128fb: Rage128 Pro PF (AGP) [chip rev 0x1] 16M 128-bit SDR SGRAM (1:1)
Console: switching to colour frame buffer device 128x48
fb0: ATY Rage128 frame buffer device on PCI
> Any feedback from the folks with the X lockup problem?
I'm still hoping that David will find my error with X.. I'd prefer it to
hang over not starting :)
Ian
-------------------------------------------
Ian White
email: [EMAIL PROTECTED]
