Re: mvme5100 one more time...

3 messages, 2 authors, 2001-08-31 · open the first message on its own page

Re: mvme5100 one more time...

From: Xavier Grave <hidden>
Date: 2001-08-31 16:03:36

Le ven, 31 aoû 2001 17:49:27, Tom Rini a écrit :
On Fri, Aug 31, 2001 at 05:36:34PM +0200, Xavier Grave wrote:
quoted
quoted
That's fine..  Can you change arch/ppc/boot/simple/head.S to
something
quoted
quoted
like:
#if 0
/*
 * Determine CPU type
...
	bge	_setup_L2CR
#endif
	b	_setup_6xx
Thanks for your advice.

I do on my x86 cpu :
make CROSS_COMPILE=powerpc-unknown-linux-gnu- ARCH=ppc znetboot
on my 5100 :
PPC6-Bug>nbo 0 0 ,,, zImage.pplus
Network Booting from: I82559, Controller 0, Device 0
Device Name: /pci@fe000000/pci8086,1209@e,0:0,0
Loading: zImage.pplus

Client IP Address      = 192.168.120.38
Server IP Address      = 192.168.120.35
Gateway IP Address     = 0.0.0.0
Subnet IP Address Mask = 255.255.255.0
Boot File Name         = zImage.pplus
Argument File Name     =

Network Boot File load in progress... To abort hit <BREAK>

Bytes Received =&700757, Bytes Loaded =&700757
Bytes/Second   =&700757, Elapsed Time =1 Second(s)

Exception: Program (Illegal Instruction)
SRR0 =001F0000 SRR1 =00083040 Vector-Offset =00700
IP     =001F0000 MSR    =00003040 CR     =00000000 FPSCR  =00000000
R0     =00000000 R1     =03F00000 R2     =00000000 R3     =00000000
R4     =00000000 R5     =49504C01 R6     =00007000 R7     =001F0000
R8     =03F0CF54 R9     =03F0CD54 R10    =03F0CD60 R11    =03F0CE54
R12    =03F0CE54 R13    =00000000 R14    =00000000 R15    =00000000
R16    =00000000 R17    =00000000 R18    =00000000 R19    =00000000
R20    =00000000 R21    =00000000 R22    =00000000 R23    =00000000
R24    =00000000 R25    =00000000 R26    =00000000 R27    =00000000
R28    =00000000 R29    =00000000 R30    =00000000 R31    =00000000
SPR0   =00000000 SPR1   =00000000 SPR8   =00000000 SPR9   =00000000
001F0000 00040000  WORD        $00040000

I have done the modification you advised me and also an other one (to
compile :-(
Okay.  Try this.  Undo what I told you last time, and make the end of
the file look like:
#if 0

	.comm	.stack,4096*2,4
#endif

This may or may not compile.  I have a feeling there'll be some egg on
my face shortly...
it doesn't compile : I have the following message :
head.o: In function `start_ldr':
head.o(.text+0x176): undefined reference to `stack'
head.o(.text+0x17a): undefined reference to `stack'
make[2]: *** [zvmlinux] Erreur 1

Why the eggs ? because it doesn't compile ? :-)

--
33 (0) 1 69 15 79 59
De chacun selon ses forces, a chacun selon ses besoins

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

Re: mvme5100 one more time...

From: Tom Rini <hidden>
Date: 2001-08-31 18:09:38

On Fri, Aug 31, 2001 at 06:03:36PM +0200, Xavier Grave wrote:
Le ven, 31 ao? 2001 17:49:27, Tom Rini a ?crit :
quoted
Okay.  Try this.  Undo what I told you last time, and make the end of
the file look like:
#if 0

	.comm	.stack,4096*2,4
#endif

This may or may not compile.  I have a feeling there'll be some egg on
my face shortly...
it doesn't compile : I have the following message :
head.o: In function `start_ldr':
head.o(.text+0x176): undefined reference to `stack'
head.o(.text+0x17a): undefined reference to `stack'
make[2]: *** [zvmlinux] Erreur 1

Why the eggs ? because it doesn't compile ? :-)
Well, my guess is that .stack is being defined twice, maybe...  The reg dump
makes it look like it's about to goto decompress_kernel.  Can you try the
following patch?  And undo what Itold you to do last time..

--
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

===== arch/ppc/boot/common/util.S 1.1 vs edited =====
--- 1.1/arch/ppc/boot/common/util.S	Thu Aug  2 16:02:25 2001
+++ edited/arch/ppc/boot/common/util.S	Fri Aug 31 09:01:10 2001
@@ -116,4 +116,3 @@
 	addi	r3,r3,CACHE_LINE_SIZE	/* Next line, please */
 	bdnz	00b
 10:	blr
-	.comm	.stack,4096*2,4
===== arch/ppc/boot/prep/head.S 1.15 vs edited =====
--- 1.15/arch/ppc/boot/prep/head.S	Tue Aug 28 15:57:32 2001
+++ edited/arch/ppc/boot/prep/head.S	Fri Aug 31 09:00:56 2001
@@ -178,11 +178,9 @@
 	ori	r2,r2,initrd_end@l
 	lwz	r5,0(r2)

-
-	/* tell kernel we're prep */
-	/*
-	 * get start address of kernel code which is stored as a coff
-	 * entry.  see boot/head.S -- Cort
+	/* tell kernel we're prep, by putting 0xdeadc0de at KERNELLOAD,
+	 * and tell the kernel to start on the 4th instruction since we
+	 * overwrite the first 3 sometimes (which are 'nop').
 	 */
 	li	r9,0xc
 	mtlr	r9
@@ -191,5 +189,5 @@
 	li	r9,0
 	stw	r10,0(r9)
 	blr
-hang:
-	b	hang
+
+	.comm	.stack,4096*2,4
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: mvme5100 one more time...

From: Tom Rini <hidden>
Date: 2001-08-31 23:29:04

On Fri, Aug 31, 2001 at 11:09:38AM -0700, Tom Rini wrote:
On Fri, Aug 31, 2001 at 06:03:36PM +0200, Xavier Grave wrote:
quoted
Le ven, 31 ao? 2001 17:49:27, Tom Rini a ?crit :
quoted
Okay.  Try this.  Undo what I told you last time, and make the end of
the file look like:
#if 0

	.comm	.stack,4096*2,4
#endif

This may or may not compile.  I have a feeling there'll be some egg on
my face shortly...
it doesn't compile : I have the following message :
head.o: In function `start_ldr':
head.o(.text+0x176): undefined reference to `stack'
head.o(.text+0x17a): undefined reference to `stack'
make[2]: *** [zvmlinux] Erreur 1

Why the eggs ? because it doesn't compile ? :-)
Well, my guess is that .stack is being defined twice, maybe...  The reg dump
makes it look like it's about to goto decompress_kernel.  Can you try the
following patch?  And undo what Itold you to do last time..
And if this fails, I just thought of one more thing (I need some hw I can
test this on locally...).
Look for:
	bl	flush_instruction_cache
And add #if 0/#endif around that and try again...

--
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help