Re: oops ACPI in Linux-2.6.6-bk1

[Pat LaVarre]

> There were no ACPI changes between 2.6.6 and 2.6.6-bk1.

The change my newbie eye sees here is that the effect of CONFIG_ACPI=y
changes from imperceptible in 2.6.6 to kernel won't boot in 2.6.6-bk1.

> Does 2.6.6 work for you?

2.6.6 works for me, yes, as did 2.6.5, 2.6.4, etc. ...

$ grep 'CONFIG_ACPI=' .config
CONFIG_ACPI=y
$ grep 'ACPI.*=y$' .config | grep -v '^#' | wc -l
     16
$

The motherboard labels in front of me include:

"INTEL DESKTOP BOARD"
"D865GBF/D865PERC".

> > > ACPI: Subsystem revision 20040326 ...
> > > [<c01eef72>] acpi_ev_save_method_info+0x44/0x75 ...
> > > Unable to handle kernel NULL pointer dereferencec01e1194 ...
> > > Kernel panic: Aiee, killing interrupt handler! ...
> > > Unable <o ha1dleUn kernel NULL pointer dereferenceer
> > dereferenceOoops: ...
> > > Unable to handle kernel NULL pointer dereference at virtual address
> > ...
> > 
> > Theory confirmed:
> > 
> > Deleting CONFIG_ACPI=y etc. via `make xconfig` fixes this.
> 
> what theory?

Thanks for making the time to say I was unclear.

In launching a thread titled "oops ACPI" I was guessing that lossy dmesg
spew appearing immediately past "ACPI: Subsystem revision 20040326" was
caused by me living too close to `make defconfig`.  I then revisited
`make xconfig`, asked to turn off ACPI, and suddenly life got better.

I'm working from local copies of the kernel.org files:

$ ls -l *2.6.6*
-rw-rw-r--    1 pat      pat      34896138 May 10 11:14 linux-2.6.6.tar.bz2
-rw-------    1 pat      pat       1072216 May 12 08:47 patch-2.6.6-bk1.bz2
$

I can fsck the containing volume, but I haven't yet otherwise learned
how to crc-check their integrity.

Pat LaVarre


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/