Re: Kernel 2.4.21

[Marek S. <ma...@st...>]

On Thu, 31 Jul 2003, [ISO-8859-1] Andreas W=FCst wrote:

> > Today I found that also the built-in parranel port makes kernel oopses.
> FWIW, back in the days I tried to use the parport, I've got an endless
> sequence of irq_unbalanced(..) messages, until I killed the task which
> was trying to access the parport (i.e. lpr). This was with 2.4.17 on a
> A4000.

I have something like this when printer is not turned on.

> But it did never oops when using the parport!

So here is a log with oops from parport:

parport0: Amiga built-in port using irq
lp0: using parport0 (interrupt-driven).
Oops: kernel access of bad area, sig: 11
NIP: C000E708 XER: 20000000 LR: C000E708 SP: C0A39B50 REGS: c0a39aa0 TRAP:
0300    Not tainted
MSR: 00009072 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11
DAR: 00000018, DSISR: 20000000
TASK =3D c0a38000[409] 'StylusColor@360' Last syscall: 11
last math c3cd4000 last altivec 00000000
GPR00: 00000000 C0A39B50 C0A38000 C01ACE40 C0400000 0FFF0000 00000000
48B73285
GPR08: 00000000 C01B0000 00000000 0FFFFFFF 24048028 1001BAA0 00000005
103EB000
GPR16: 0FFEB000 C09AA100 C01ACE40 0FFF0000 C01B0000 00000005 0FFF0000
103EB000
GPR24: C09AA100 00000005 00000192 0FFF0000 00005000 C09AA100 0FFEB000
00000000
Call backtrace:
C0031898 C0023844 C0026A20 C0040688 C0040940 C0055C4C C0040EF4
C0041260 C0006DD4 C0003D5C 00000000 0FF56CDC 10000BAC 10001410
10002538 0FECAC30 00000000
kernel BUG at mmap.c:1163!
Oops: Exception in kernel mode, sig: 4
NIP: C0026A6C XER: 00000000 LR: C0026A6C SP: C0A39960 REGS: c0a398b0 TRAP:
0700    Not tainted
MSR: 00089072 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11
TASK =3D c0a38000[409] 'StylusColor@360' Last syscall: 11
last math c3cd4000 last altivec 00000000
GPR00: C0026A6C C0A39960 C0A38000 0000001B 00001072 00000001 C0A3981C
C01B0000
GPR08: 000018F5 00000000 00000000 C0A39880 8410C004 1001BAA0 00000005
103EB000
GPR16: 0FFEB000 C09AA100 C01ACE40 0FFF0000 00009072 47A39A90 00000000
C0003F90
GPR24: 00000000 C01ACE40 00030001 20000000 0000000B C0A39AA0 C01ACE40
00000000
Call backtrace:
C0026A6C C00114F8 C0016C1C C000411C C000D3DC C000D290 C0003F90
C0031898 C0023844 C0026A20 C0040688 C0040940 C0055C4C C0040EF4
C0041260 C0006DD4 C0003D5C 00000000 0FF56CDC 10000BAC 10001410
10002538 0FECAC30 00000000

Nice, isn't it? ;)

> > So
> > the above discussion is pointless, because the main problem lies in the
> > interrupt code, not in drivers... Do you remember when the problems
> > with
> > interrupts started?
>
> So maybe with 2.4.17 it was at least still half working.

Well, I used 2.4.18 for a very long time (actually I'm still using it,
because there is no better kernel - you know in what state is 2.4.21
kernels, all earlier working 2.4.x kernels don't have ext3 support, same
for 2.2.10...).

Regards
--=20
Marek Szyprowski ........... mailto:ma...@st... ...