LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Hardware (https://www.linuxquestions.org/questions/linux-hardware-18/)
-   -   Problem with symbios SCSI and 2.6 kernel series (?) (https://www.linuxquestions.org/questions/linux-hardware-18/problem-with-symbios-scsi-and-2-6-kernel-series-278228/)

GD_19 01-15-2005 06:37 PM

Problem with symbios SCSI and SMP-2.6 kernel series
 
Hello!

I 've just installed Slackware 10.0 but I need a 2.6 kernel so that I can get my DSL controller to work as well. I have tried compiling several kernel versions (mainly from the 2.6.7 and the 2.6.10 series) but I always get the following when I boot the new kernel:

sym0: <896> rev 0x1 at pci 0000:02:0c.0 irq10
sym0: using 64-bit DMA addressing
sym0: Symbios NVRAM, ID 7, Fast-40, SE, parity checking
sym0: SCSI BUS has been reset.
scsi0 : sym-2.1.18m
sym0:0:0 ABORT operation started
sym0:0:0 ABORT operation timed-out
sym0:0:0 DEVICE RESET operation started
sym0:0:0 DEVICE RESET operation timed-out
sym0:0:0 BUS RESET operation started
sym0: SCSI BUS reset detected
sym0: SCSI BUS has been reset
sym0:0:0 BUS RESET operation complete
sym0:0:0 ABORT operation started
sym0:0:0 ABORT operation timed-out
sym0:0:0 HOST RESET operation started
sym0: SCSI BUS has been reset

and I get no more output (I have to reset the system)

I have even tried compiling the kernel with slackware's stock 2.6-testing .config by only enabling SMP and selecting the option for the symbios module to be built into the kernel instead of as a module. However I keep getting the same thing...

The controller is built in the motherboard (AMI Megarum II) and connects to the intel 440GX chipset via a 64-bit PCI bus. The controller itself is a dual lsi-logic symbios 53c896 chipset.

Note that slackware's stock 2.4 non-SMP kernel works fine

Any feedback would be greately appreciated
Thanks for your time
George

GD_19 01-15-2005 08:40 PM

*UPDATE*

I just installed fedora core 3 and i get the same problem with the stock fc3 smp kernel. The system boots successfully if I select the non-SMP kernel image from GRUB.

It seems like an SMP-related issue. Does anyone have a clue of how it could be solved?

This link https://bugzilla.redhat.com/beta/show_bug.cgi?id=130783 seems to include a description of the problem although i'm not sure it's exactly the same as mine. plus it refers to a PPC architecture... I don't seem to be making much out of what is being said there anyway...

GD_19 04-17-2005 09:25 PM

UPDATE2:

This problem is here to stay, it seems...

Anyway I recently installed CentOS 4.0 (Redhat Enterprise Linux 4.0 rebuild) and finally found out that their stock SMP kernel will boot if I append the noapic option to the appropriate grub.conf line.

However I can't live on stock kernels (and having apic disabled is something i don't really like), so I tried to build a custom kernel but using the .config file supplied by CentOS as a base for my setup. This however resulted in a kernel which won't boot, even if I use the aforementioned noapic option.

Note that the case with CentOS is very different. If a kernel doesn't boot, then what I get is a system hang after the "redhat nash" message. But this is most probably due to my using the "quiet" kernel option.

It seems to be a buggy BIOS issue anyway, although I've seen similar problems with the same controller family around the net with like descriptions.

P.S. I don't know whether it makes any difference, but I always compile my smp kernels with CONFIG_NR_CPUS=2, cpu type set to Pentium II (as I have a deschutes core) and *NO* Generic x86 Support.


All times are GMT -5. The time now is 08:50 PM.