PS: Sent to invalid kml email address. This is a resend.
Hello Alan,
I have a UMSDOS system with egcs v1.1.2 and libc 5.3.12. My hardware
is a Dell 333mhz GX1 with scsi U2W (AIC 7895) and 128 megs of memory.
I don't have any swap turned on.
Here is a list of the problems with the linux v2.2.21 pre releases:
v2.2.21pre2 boots and works ok.
v2.2.21pre3 filesystem panic after root mounted and umssync is run.
v2.2.21pre4 hangs on boot after the message
"Intel machine check architecture supported"
v2.2.21rc1 Oops' on boot after the message "CPU: L2 cache = 512K
with a kernel panic. Note: I don't have any swap turned on.
Ksymoops listing
ksymoops 2.3.3 on i686 2.2.21pre2. Options used
-v ./vmlinux (specified)
-K (specified)
-L (specified)
-o /lib/modules/2.2.21rc1/ (specified)
-m /usr/src/linux/System.map (default)
No modules in ksyms, skipping objects
Unable to handle kernel NULL ptr dereference at virtual address 00000004
*pde = 00000000
Oops: 0000
Cpu: 0
Eip: 0010 : [<c0297244>]
Using defaults from ksymoops -t elf32-i386 -a i386
Eflags: 00010282
eax: c0273f00 ebx: c0273f06 ecx: c0275868 edx: c0273f00
esi: 00000004 edi: c0273f20 ebp: 00000200
ds: 0018 es: 0018 ss: 0018 esp: c0293f74
stack: c0106000 0008e000 00000010 0000000c c0293f0c 00000000 000000e8 c0293fa8
c0293fa8 00000001 00000001 00000000 00000004 03020101 00000000 00000000
0c040843 c02973ba c0273f00 00000000 000994a2 c0106000 c0294457 c0273f00
call trace: [<c0106000>] [<c0106000>] [<c0106000>] [<c010017b>]
code: ac aa 84 c0 75 fa 5b 5e 5f 5d 83 c4 3c c3 89 f6 56 53 8b 5c
>>EIP; c0297244 <init_intel+33c/34c> <=====
Trace; c0106000 <get_options+0/74>
Trace; c0106000 <get_options+0/74>
Trace; c0106000 <get_options+0/74>
Trace; c010017b <L6+0/2>
Code; c0297244 <init_intel+33c/34c>
00000000 <_EIP>:
Code; c0297244 <init_intel+33c/34c> <=====
0: ac lodsb %ds:(%esi),%al <=====
Code; c0297245 <init_intel+33d/34c>
1: aa stosb %al,%es:(%edi)
Code; c0297246 <init_intel+33e/34c>
2: 84 c0 testb %al,%al
Code; c0297248 <init_intel+340/34c>
4: 75 fa jne 0 <_EIP>
Code; c029724a <init_intel+342/34c>
6: 5b popl %ebx
Code; c029724b <init_intel+343/34c>
7: 5e popl %esi
Code; c029724c <init_intel+344/34c>
8: 5f popl %edi
Code; c029724d <init_intel+345/34c>
9: 5d popl %ebp
Code; c029724e <init_intel+346/34c>
a: 83 c4 3c addl $0x3c,%esp
Code; c0297251 <init_intel+349/34c>
d: c3 ret
Code; c0297252 <init_intel+34a/34c>
e: 89 f6 movl %esi,%esi
Code; c0297254 <table_lookup_model+0/48>
10: 56 pushl %esi
Code; c0297255 <table_lookup_model+1/48>
11: 53 pushl %ebx
Code; c0297256 <table_lookup_model+2/48>
12: 8b 5c 00 00 movl 0x0(%eax,%eax,1),%ebx
Kernel panic: Attempted to kill the idle task!
In swapper task -- not syncing
On Thu, Mar 14, 2002 at 03:19:56PM -0500, M Sweger wrote:
> v2.2.21rc1 Oops' on boot after the message "CPU: L2 cache = 512K
> with a kernel panic. Note: I don't have any swap turned on.
> >>EIP; c0297244 <init_intel+33c/34c> <=====
Fix posted to the list earlier. Add =NULL to the declaration
of the variable p in arch/i386/kernel/setup.c:init_intel()
--
| Dave Jones. http://www.codemonkey.org.uk
| SuSE Labs
> v2.2.21pre4 hangs on boot after the message
> "Intel machine check architecture supported"
Fixed..
> v2.2.21rc1 Oops' on boot after the message "CPU: L2 cache = 512K
> with a kernel panic. Note: I don't have any swap turned on.
Also fixed - can you try pre3 ?
> > v2.2.21rc1 Oops' on boot after the message "CPU: L2 cache = 512K
> > with a kernel panic. Note: I don't have any swap turned on.
>
> Also fixed - can you try pre3 ?
Duh - I mean rc2 of course. I'm patching 2.4.19pre3-ac in another window 8)
On Thu, 2002-03-14 at 23:40, Alan Cox wrote:
> > v2.2.21pre4 hangs on boot after the message
> > "Intel machine check architecture supported"
>
> Fixed..
>
what causes that 'cause I've been battling with that for a while
on 2.4.19-pre3 and 2.4.19-pre3-aa1. I put a couple of printk's to see
where the problem was (arch/i386/bluesmoke.c), but after rebooting it
did not show up. I just vompiled 2.4.19-pre3-aa and 2.4.19-pre3 +
preempt, I would like o have it solved before I reboot though
thanx
-Bongani
> on 2.4.19-pre3 and 2.4.19-pre3-aa1. I put a couple of printk's to see
> where the problem was (arch/i386/bluesmoke.c), but after rebooting it
> did not show up. I just vompiled 2.4.19-pre3-aa and 2.4.19-pre3 +
> preempt, I would like o have it solved before I reboot though
The intel cpus get deeply upset if you write to bank zero of the error
reporting registers used by MCE exceptions. The AMD ones want you to do so
and the change that went in was tested on AMD boxes but not Intel.
Revert the bluesmoke.c change and all should be well
On Mar 14, 2002 23:53 +0200, Bongani wrote:
> ...I just vompiled 2.4.19...
Is this what you do when the code is so bad it makes you sick ;-)
Cheers, Andreas
--
Andreas Dilger \ "If a man ate a pound of pasta and a pound of antipasto,
\ would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/ -- Dogbert
On Thu, 14 Mar 2002, Alan Cox wrote:
> > > v2.2.21rc1 Oops' on boot after the message "CPU: L2 cache = 512K
> > > with a kernel panic. Note: I don't have any swap turned on.
> >
> > Also fixed - can you try pre3 ?
>
> Duh - I mean rc2 of course. I'm patching 2.4.19pre3-ac in another window 8)
>
patch-2.2.21-rc2 boots and works ok on a UMSDOS system.
Ready for patch-2.2.21-pre5 !
Hopefully, linux 2.2.x will get the 160gig IDE patch that 2.4.x has.
> Hopefully, linux 2.2.x will get the 160gig IDE patch that 2.4.x has.
Andre's 2.2 patch has been picked up and maintained (I forget by who). I
don't consider it ever a mainstream 2.2 candidate
On Fri, Mar 15, 2002 at 02:50:37PM +0000, you [Alan Cox] wrote:
> > Hopefully, linux 2.2.x will get the 160gig IDE patch that 2.4.x has.
>
> Andre's 2.2 patch has been picked up and maintained (I forget by who). I
> don't consider it ever a mainstream 2.2 candidate
I think you mean Krzysztof Oledzki ([email protected])
From: Krzysztof Oledzki ([email protected])
Subject: ANNOUNCE - 2.4.x ide backport for 2.2.21pre2 kernel
http://groups.google.com/groups?q=ANNOUNCE+-+2.4.x+ide+backport+for&hl=en&ie=ISO-8859-1&oe=ISO-8859-1&selm=linux.kernel.Pine.LNX.4.33.0201291654480.17318-100000%40dark.pcgames.pl&rnum=1
BTW: I've heard rumours that some older ide chipsets support 48-bit
addressing with a bios upgrade. HPT370 and Via 686B have been mentioned.
Highpoint windows driver readme:
v2.1 11/15/2001
* Add 48bit LBA (Big Drive) support
But I'm not sure whether this means HPT370 or 372 only.
Is this possible and will it be possible to support that functionality in
Linux?
-- v --
[email protected]
Ok! I downloaded it, and patched the kernel. Although I don't have a
160gig drive yet (was waiting for backport), the linux 2.2.21rc2 with
the pre2 ide patch (latest and greatest) booted ok and hasn't crashed.
I'll need to try it on a new 160 gig drive sometime.
Hmmm. Hope consideration is given to back port this patch to 2.2.x.
On Fri, 15 Mar 2002, Ville Herva wrote:
> On Fri, Mar 15, 2002 at 02:50:37PM +0000, you [Alan Cox] wrote:
> > > Hopefully, linux 2.2.x will get the 160gig IDE patch that 2.4.x has.
> >
> > Andre's 2.2 patch has been picked up and maintained (I forget by who). I
> > don't consider it ever a mainstream 2.2 candidate
>
> I think you mean Krzysztof Oledzki ([email protected])
>
> From: Krzysztof Oledzki ([email protected])
> Subject: ANNOUNCE - 2.4.x ide backport for 2.2.21pre2 kernel
>
> http://groups.google.com/groups?q=ANNOUNCE+-+2.4.x+ide+backport+for&hl=en&ie=ISO-8859-1&oe=ISO-8859-1&selm=linux.kernel.Pine.LNX.4.33.0201291654480.17318-100000%40dark.pcgames.pl&rnum=1
>
>
> BTW: I've heard rumours that some older ide chipsets support 48-bit
> addressing with a bios upgrade. HPT370 and Via 686B have been mentioned.
>
> Highpoint windows driver readme:
>
> v2.1 11/15/2001
> * Add 48bit LBA (Big Drive) support
>
> But I'm not sure whether this means HPT370 or 372 only.
>
> Is this possible and will it be possible to support that functionality in
> Linux?
>
>
> -- v --
>
> [email protected]
>