Re: 2.4.18 kernel oops report

Re: 2.4.18 kernel oops report

[Benjamin Herrenschmidt]

>
>Hello! I'm not subscribed to this list, but I thought I'd send this along...I
>have a problem with kernel oopses. I'm running a Powermac G3 266 Desktop,
>with standard everything, except I've got a Voodoo3 card for my monitor and a
>PCI USB Card. I've got 392 (?) MB RAM, a SCSI Drive and two IDE drives. I'm
>running a stock 2.4.18 kernel that I got from kernel.org, and this oops has
>happened before on previous stock kernels (2.4.17 in particular). I'm sure
>that sound is part of the problem. I boot into Linux using BootX, and usually
>boot all the way into the MacOS (8.6) to get sound. Sometimes I have a hard
>crash, and when I boot back into Linux (going all the way into MacOS), I get
>kernel errors while fsck is doing its thing. If I just boot into Linux using
>the "Choose Your OS" panel at startup, fsck is able to proceed without any
>problem, but I also have no sound...and no oopses, either.
>
>I've run ksymoops through kern.log and syslog after the oops, but the syslog
>is more like what I think is necessary to be reported...if you need more, let
>me know directly, as I'm not a member of this list...I'm hesitant to send my
>System.map to the mailing list, it's kinda big!

It looks like something in macos is damaging the kernel, or a device is
still bus mastering, thus trashing memory, while linux loads. The best
way would be to fix the sound problem with BootX extension. What kind of
sound hardware do you have, awacs or burgundy ?

Can you try the following patch ? (it may applyh with a slight offset)

===== arch/ppc/kernel/pmac_feature.c 1.6 vs edited =====
--- 1.6/arch/ppc/kernel/pmac_feature.c	Fri Feb 22 01:10:43 2002
+++ edited/arch/ppc/kernel/pmac_feature.c	Sun Mar  3 17:10:01 2002
@@ -2041,10 +2045,20 @@
 		}
 	}

-	/* On all machines, switch sound off */
+	/* On all machines that support sound PM, switch sound off */
 	if (macio_chips[0].of_node)
 		pmac_do_feature_call(PMAC_FTR_SOUND_CHIP_ENABLE,
 			macio_chips[0].of_node, 0, 0);
+
+	/* While on some desktop G3s, we turn it back on */
+	if (macio_chips[0].of_node && macio_chips[0].type == macio_heathrow
+		&& (pmac_mb.model_id == PMAC_TYPE_GOSSAMER ||
+		    pmac_mb.model_id == PMAC_TYPE_SILK)) {
+		struct macio_chip* macio = &macio_chips[0];
+		MACIO_BIS(HEATHROW_FCR, HRW_SOUND_CLK_ENABLE);
+		MACIO_BIC(HEATHROW_FCR, HRW_SOUND_POWER_N);
+	}
+

 	/* On all machines, switch modem & serial ports off */
 	np = find_devices("ch-a");


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

2.4.18 kernel oops report

[Russell Hires]

[-- Attachment #1: Type: text/plain, Size: 1200 bytes --]


Hello! I'm not subscribed to this list, but I thought I'd send this along...I
have a problem with kernel oopses. I'm running a Powermac G3 266 Desktop,
with standard everything, except I've got a Voodoo3 card for my monitor and a
PCI USB Card. I've got 392 (?) MB RAM, a SCSI Drive and two IDE drives. I'm
running a stock 2.4.18 kernel that I got from kernel.org, and this oops has
happened before on previous stock kernels (2.4.17 in particular). I'm sure
that sound is part of the problem. I boot into Linux using BootX, and usually
boot all the way into the MacOS (8.6) to get sound. Sometimes I have a hard
crash, and when I boot back into Linux (going all the way into MacOS), I get
kernel errors while fsck is doing its thing. If I just boot into Linux using
the "Choose Your OS" panel at startup, fsck is able to proceed without any
problem, but I also have no sound...and no oopses, either.

I've run ksymoops through kern.log and syslog after the oops, but the syslog
is more like what I think is necessary to be reported...if you need more, let
me know directly, as I'm not a member of this list...I'm hesitant to send my
System.map to the mailing list, it's kinda big!

Thanks!

Russell


[-- Attachment #2: oops-report-syslog --]
[-- Type: text/plain, Size: 3252 bytes --]

ksymoops 2.3.4 on ppc 2.4.18.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.18/ (default)
     -m /usr/src/linux/System.map (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

Warning (compare_maps): mismatch on symbol xchg_u32  , ksyms_base says c000bc40, System.map says c0006950.  Ignoring ksyms_base entry
Warning (compare_maps): mismatch on symbol hfs_version  , hfs says dba40e60, /lib/modules/2.4.18/kernel/fs/hfs/hfs.o says dba415e0.  Ignoring /lib/modules/2.4.18/kernel/fs/hfs/hfs.o entry
Mar  2 22:30:17 localhost kernel: 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
Mar  2 22:30:18 localhost kernel:  0000 0000 0000 0000 0000 0000 0000 0000
Mar  2 22:57:14 localhost kernel: Oops: kernel access of bad area, sig: 11
Mar  2 22:57:14 localhost kernel: NIP: C0031908 XER: 00000000 LR: C0031790 SP: CD161E00 REGS: cd161d50 TRAP: 0300    Tainted: P 
Using defaults from ksymoops -t elf32-powerpc -a powerpc:common
Mar  2 22:57:14 localhost kernel: MSR: 00009032 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11
Mar  2 22:57:14 localhost kernel: TASK = cd160000[818] 'kdeinit' Last syscall: 1 
Mar  2 22:57:14 localhost kernel: last math d071c000 last altivec 00000000
Mar  2 22:57:14 localhost kernel: GPR00: 00000000 CD161E00 CD160000 C091A700 00000000 00036000 CD55C000 C0518090 
Mar  2 22:57:14 localhost kernel: GPR08: C02E0000 FFFFFFFF C091A75C C02E0000 80004044 10020C68 10010000 00000000 
Mar  2 22:57:14 localhost kernel: GPR16: 0FC79000 10079000 D02C0100 CEFAB520 0FCAF000 C02E0000 C02E0000 0FCAF000 
Mar  2 22:57:14 localhost kernel: GPR24: 10079000 D02C0100 00000001 00036000 00009000 00000000 C091A700 00000084 
Mar  2 22:57:14 localhost kernel: Call backtrace: 
Mar  2 22:57:14 localhost kernel: C0031790 C00323EC C0032A50 C0024FFC C0025464 C002836C C0013FD8 
Mar  2 22:57:14 localhost kernel: C0019668 C0019870 C000431C 0ED21E18 10004CAC 10005A5C 1000601C 
Mar  2 22:57:14 localhost kernel: 10007418 0ED0AD30 00000000 
Warning (Oops_read): Code line not seen, dumping what data is available

>>NIP; c0031908 <__free_pages_ok+1a0/3d0>   <=====
Trace; c0031790 <__free_pages_ok+28/3d0>
Trace; c00323ec <__free_pages+38/48>
Trace; c0032a50 <free_page_and_swap_cache+64/78>
Trace; c0024ffc <__free_pte+5c/70>
Trace; c0025464 <zap_page_range+1a4/258>
Trace; c002836c <exit_mmap+f0/170>
Trace; c0013fd8 <mmput+74/ac>
Trace; c0019668 <do_exit+e0/2ac>
Trace; c0019870 <sys_wait4+0/39c>
Trace; c000431c <ret_from_syscall_1+0/b4>
Trace; 0ed21e18 Before first symbol
Trace; 10004cac Before first symbol
Trace; 10005a5c Before first symbol
Trace; 1000601c Before first symbol
Trace; 10007418 Before first symbol
Trace; 0ed0ad30 Before first symbol
Trace; 00000000 Before first symbol


4 warnings issued.  Results may not be reliable.