If you feel bored, test these 2.5 changes...

pa's,

If you feel good and bored, or willing to test something, the following cvs diff needs atleast a round of tests in 2.5 before I commit. Right now I'm *uber* busy with glibc, and tausq hid my kernel tree ;)

- Added the last enumerations to the cpu list.
= These are probably incorrect.

- Forward port of trap handler changes.
= Virtually identical to 2.4 code which = has been running for almost 5 months without = problems.

- Added a code comment about non-returning functions. = Purely cosmetic...

Apply, test and get back to the list? :)

c.

Index: arch/parisc/kernel/setup.c
===================================================================
RCS file: /var/cvs/linux-2.5/arch/parisc/kernel/setup.c,v retrieving revision 1.9

diff -u -p -r1.9 setup.c
@@ -212,6 +212,11 @@ static void parisc_proc_mkdir(void)
                         proc_runway_root = proc_mkdir("bus/runway", 0);
                 }
                 break;
+	case pcx:
+	case pcxs:
+	case pcxt:
+		/* Are not assured to have any of the above busses */
+		break;
 	}
 }
 

Index: arch/parisc/kernel/traps.c
===================================================================
RCS file: /var/cvs/linux-2.5/arch/parisc/kernel/traps.c,v retrieving revision 1.16

Index: arch/parisc/mm/fault.c
===================================================================
RCS file: /var/cvs/linux-2.5/arch/parisc/mm/fault.c,v retrieving revision 1.3

diff -u -p -r1.3 fault.c
@@ -257,7 +257,8 @@ no_context:
 	}
 
 	parisc_terminate("Bad Address (null pointer deref?)", regs, code, address);
-
+	/* NOT REACHED */
+	
   out_of_memory:
 	up_read(&mm->mmap_sem);
 	printk(KERN_CRIT "VM: killing process %s\n", current->comm);

Hi Carlos,

Sorry but after the boot flop on my b2000, I try on a b180 but no more success:
Freeing unused kernel memory: 416k freed
Warning: unable to open an initial console.
Unable to find swap-space signature
EXT3 FS 2.4-0.9.16, 02 Dec 2001 on md(9,0), internal journal

Stack Dump:

 178f4300:  0006ff0f 00000000 00000000 00000000 
 178f42f0:  00000000 10262010 00000000 00000000           
 178f42e0:  106698c0 178f4210 00000000 178f4230           
 178f42d0:  178f4110 00000000 178f4210 17ff7760           
 178f42c0:  fffffff4 17c6c2c0 178f3988 17c6c2a0           
 178f42b0:  178f3680 101613d4 000002b1 178f3988 
 178f42a0:  17c6c2c0 10097c00 fffffffa 00000000 
 178f4290:  1066e600 00000812 17b7b1a0 1066e680 
 178f4280:  00000931 00000003 01c048f0 00000004 
 178f4270:  1066b005 10262ee0 40177000 17aaf9a0 
 178f4260:  17ff7760 40194000 178fec20 17aaf9a0 
 178f4250:  17ff7760 178f4190 00000000 00000000 

Kernel addresses on the stack:

 [<10262010>]  [<101613d4>]  [<10262ee0>]  [<102111fc>] 
 [<1015de7c>]  [<10166e40>]  [<1010b008>]  [<1010a094>] 

Kernel Fault: Code=26 regs=178f4300 (Addr=00000204)

YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001101111111100001111 Not tainted

r00-03  00000000 1034f010 10262010 17eec260
r04-07  00000000 1034f010 00000812 1034f4b4
r08-11  1030a800 00000812 00027d40 000e0008
r12-15  00000008 000e04c8 000e0888 000e0648
r16-19  00000000 000d2430 0009e800 00000200
r20-23  1034f4b4 00000000 00000000 00000000
r24-27  00000000 00001000 178ec000 1031e010
r28-31  17eec260 b870fe25 178f4300 10044a78
sr0-3   00000000 0000010e 00000000 0000010e
sr4-7   00000000 00000000 00000000 00000000

IASQ: 00000000 00000000 IAOQ: 102620d0 102620d4
IIR: 0e681094 ISR: 00000000 IOR: 00000204
CPU: 0 CR30: 178f4000 CR31: 1035b000
ORIG_R28: 00000008

Joel

PS: I will see to backport on 2.4.20-rc1

-------------------------------------------------------------
Tiscali ADSL: Activation et 1er mois GRATUITS. Plus d'info... http://www.tiscali.be/FR/subs/adsl.asp

Joel,

Wow, you added those changes and EXT3 stopped functioning? :} Please revert the patch and try to build/boot a kernel again.

Code 26 -> "Data Memory Access Rights" trap, looks like a misbehaving EXT3.

c.

Carlos,

I test your patch on my 2.4.20-rc1(-pa24) and the kernel boot and run well :)).

I try to trigger the usage of this traps
double a=1., b=0., c;
c=a/b;
printf ("Ratio a/b: %f);

but it do not cause any fp interrupt error? (well that is an another question)

Back to 2.5.46-pa1, I revert the patch and rebuild the kernel and the kernel won't boot more. So it seems that problem comes from ext3 usage.
How may I come back to ext2 (tune2fs -O ^has_journal /dev/md0 [my boot partition] ) I nerver do it and a bit affraid to loose my system :(

Joel

-------------------------------------------------------------
Tiscali Complete, l'accès Internet moins cher que gratuit ! Plus d'info:
http://complete.tiscali.be

It should be as simple as changing your root fs type from 'ext3' or 'auto' to 'ext2' in /etc/fstab.

Andrew.

Andrew Shugg <[email protected]>                   http://www.neep.com.au/

"Just remember, Mr Fawlty, there's always someone worse off than yourself."
"Is there?  Well I'd like to meet him.  I could do with a good laugh."

Andrew,

kernel

You have perfectly right, just at boot time:
EXT2-fs warning (device md(9,0)): ext2_fill_super: mounting ext3 fs as ext2

...

But it would not help to make the system boot (I also remove ext3 support from kernel):
...
Unable to find swap-space signature

Stack Dump:
....

Thanks a lot,
Joel

-------------------------------------------------------------
Tiscali Complete, l'accès Internet moins cher que gratuit ! Plus d'info:
http://complete.tiscali.be