[KLUG Members] 2.4.22 On The Verge

[Adam Williams]

> >>>>ACPI! ACPI! ACPI!  Laptop Linux users rejoice!
> >>>Are you going to try it?  If so, PLEASE let me know how it works!
> >>I intend to try it, ACPI would fix several problems I have.
> >Cool.  Please let me know.  There is a debate on the DL-develop mailing
> >list, if it's stable enough to include ACPI support in our kernel.
> Will do.  I'll probably try in on Sunday, tearing the roof off my porch tomorrow
> and have bad movie night tonight.  I'd really just like power off / reboot to
> work, which doesn't on my laptop without ACPI enabled.

I tried it, it sucks.  I can't get a working kernel out of 2.4.22 for
anything.  Either basic modules don't build (no lvm-mod) or dies with
bizarre errors like -

/usr/src/linux-2.4.22/include/linux/sched.h:159:1: warning: this is the location of the previous definition
In file included from ksyms.c:17:
/usr/src/linux-2.4.22/include/linux/kernel_stat.h: In function
`kstat_irqs':
/usr/src/linux-2.4.22/include/linux/kernel_stat.h:62: `smp_num_cpus'
undeclared (first use in this function)
/usr/src/linux-2.4.22/include/linux/kernel_stat.h:62: (Each undeclared
identifier is reported only once
/usr/src/linux-2.4.22/include/linux/kernel_stat.h:62: for each function
it appears in.)
In file included from
/usr/src/linux-2.4.22/include/linux/interrupt.h:46,
                 from ksyms.c:21:
/usr/src/linux-2.4.22/include/asm/hardirq.h:37:1: warning:
"synchronize_irq" redefined
In file included from
/usr/src/linux-2.4.22/include/linux/modversions.h:132,
                 from /usr/src/linux-2.4.22/include/linux/module.h:21,
                 from ksyms.c:14:
/usr/src/linux-2.4.22/include/linux/modules/i386_ksyms.ver:92:1:
warning: this is the location of the previous definition
make[2]: *** [ksyms.o] Error 1
make[2]: Leaving directory `/usr/src/linux-2.4.22/kernel'
make[1]: *** [first_rule] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.22/kernel'
make: *** [_dir_kernel] Error 2

I've gotten one to boot as far as saying my root volume is invalid. 
Possibly mkinitrd problem?  But a generated a initrd with the standard
process - mkinitrd /boot/initrd-2.4.22.img 2.4.22 after an "install
modules" (during one pass when the compile managed to complete, or at
least it looked like it).

One can get it to compile, but that involves fishing around to make sure
certain modules that break it are disabled - guessing these based upon
the source file where the error occured.  That does not constitute a
good time.

Pretty disgusted that this is a "production" release kernel.  I can only
suspect that the kernel gods have come to depend on the distro's to
clean up the mess.

Thanks, I'll just try my luck with a 2.6.x release candidate.  It can't
be worse.