linux 2.6.12-rc1-bk5 compilation error

linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

Hi,

Once again i face an error that i got with a previous
2.6 kernel (i already posted about this in january
i think).

My config G5 bi 2Ghz and you can grab my kernel
config here : http://dj.planet-d.net/config

When compiling  2.6.12-rc1-bk5 i first get unknown
symbol _flush_disable_L1. This is in arch/ppc/kernel/l2cr.S
I changed the makefile in arch/ppc/kernel to have this :
obj-$(CONFIG_POWER4) += cpu_setup_power4.o
into :
obj-$(CONFIG_POWER4) += l2cr.o cpu_setup_power4.o

I am not sure this is good but with this i can go on.

And finaly i face back the same unknown function
than in my previous mail :
disable_6xx_mmu
used in :
arch/ppc/boot/simple/misc-prep.c

At that time i done some test defining CONFIG_6xx
which give others erros or undefining PPC_PREP
which give some others errors.

I finaly used a stupid patch to make it works (below)
Signed-off-by: Jerome Glisse <j.glisse@gmail.com>

But it seems that at some stage it gots forgotten.
Anyway what puzzle me is that it seems i am the
only one to face this ? Am i alone :) ?

diff -Naur a/arch/ppc/boot/simple/misc-prep.c
b/arch/ppc/boot/simple/misc-prep.c
--- a/arch/ppc/boot/simple/misc-prep.c      2004-12-24 22:33:51.000000000 +0100
+++ b/arch/ppc/boot/simple/misc-prep.c       2005-01-21
12:09:50.976426672 +0100
@@ -34,7 +34,11 @@
extern void serial_fixups(void);
extern struct bi_record *decompress_kernel(unsigned long load_addr,
               int num_words, unsigned long cksum);
+#ifdef CONFIG_6XX
extern void disable_6xx_mmu(void);
+#elif
+static inline void disable_6xx_mmu(void) {}
+#endif
extern unsigned long mpc10x_get_mem_size(void);

static void

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

[-- Attachment #1: Type: text/plain, Size: 134 bytes --]

I send the wrong patch (old one) and with paste & cute error.
Here is a good one.

 Signed-off-by: Jerome Glisse <j.glisse@gmail.com>

[-- Attachment #2: patch-6xx --]
[-- Type: application/octet-stream, Size: 570 bytes --]

diff -Nru a/arch/ppc/boot/simple/misc-prep.c b/arch/ppc/boot/simple/misc-prep.c
--- a/arch/ppc/boot/simple/misc-prep.c	2005-04-04 10:41:25.766786800 +0200
+++ b/arch/ppc/boot/simple/misc-prep.c	2005-04-04 10:47:23.185450904 +0200
@@ -34,7 +34,11 @@
 extern void serial_fixups(void);
 extern struct bi_record *decompress_kernel(unsigned long load_addr,
 		int num_words, unsigned long cksum);
+#ifdef CONFIG_6XX
 extern void disable_6xx_mmu(void);
+#else
+static inline void disable_6xx_mmu(void){}
+#endif
 extern unsigned long mpc10x_get_mem_size(void);
 
 static void

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

One more things i can boot with this kernel.
I got a Invalid memory access right after
kernel start...

Jerome Glisse

Re: linux 2.6.12-rc1-bk5 compilation error

[Benjamin Herrenschmidt]

On Mon, 2005-04-04 at 11:32 +0100, Jerome Glisse wrote:
> Hi,
> 
> Once again i face an error that i got with a previous
> 2.6 kernel (i already posted about this in january
> i think).
> 
> My config G5 bi 2Ghz and you can grab my kernel
> config here : http://dj.planet-d.net/config
> 
> When compiling  2.6.12-rc1-bk5 i first get unknown
> symbol _flush_disable_L1. This is in arch/ppc/kernel/l2cr.S
> I changed the makefile in arch/ppc/kernel to have this :
> obj-$(CONFIG_POWER4) += cpu_setup_power4.o
> into :
> obj-$(CONFIG_POWER4) += l2cr.o cpu_setup_power4.o
>
> .../...

You should really be using a 64 bits kernel :)

Ben.

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

> You should really be using a 64 bits kernel :)

Yes, i should, honte =E0 moi :) btw which distribution
do you use or recommand for 64bits ? gentoo ? I am
long time debian user (x86) and i think that this is not the
best choice for ppc64...

Jerome Glisse

Re: linux 2.6.12-rc1-bk5 compilation error

[Benjamin Herrenschmidt]

On Mon, 2005-04-04 at 20:11 +0100, Jerome Glisse wrote:
> > You should really be using a 64 bits kernel :)
> 
> Yes, i should, honte à moi :) btw which distribution
> do you use or recommand for 64bits ? gentoo ? I am
> long time debian user (x86) and i think that this is not the
> best choice for ppc64...

I'm using debian 32 bits and do my own 64 bits toolchain for building
kernels ;)

For toolchain, I recommend Janis scripts at penguinppc64.org (toolchain
section, there is a link to mkbiarch etc... I don't use crosstools).

Ben.

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

I almost successfully run the 2.6.12-rc2, there were some
bad old stuff lying around. But i needed the two workaround
to compile it. So maybe this two could be applied to the kernel ?

Now the only issue which remains is that the radeon framebuffer
no more set a proper mode for my screen. The screen complain
about a out of range mode. But as soon as X start i get a proper
mode set but can't switch to console.

The screen is a crt one on a radeon 9600 with dvi->vga adapter.
Which works well on 2.6.10

I will try t look at radeon framebuffer see if i can find anythings.
Any idea of which change on that code may be the origin of this ?

Jerome Glisse

Re: linux 2.6.12-rc1-bk5 compilation error

[Andreas Schwab]

Jerome Glisse <j.glisse@gmail.com> writes:

> I will try t look at radeon framebuffer see if i can find anythings.
> Any idea of which change on that code may be the origin of this ?

I'd suggest looking at the changes between rc1 and rc2 first.  On my iBook
G3 with Radeon M6 I'm also having problems with rc2, whereas rc1 is
working fine.  I haven't found the time to look closer yet.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

On Apr 5, 2005 1:39 PM, Andreas Schwab <schwab@suse.de> wrote:
> Jerome Glisse <j.glisse@gmail.com> writes:
> 
> > I will try t look at radeon framebuffer see if i can find anythings.
> > Any idea of which change on that code may be the origin of this ?
> 
> I'd suggest looking at the changes between rc1 and rc2 first.  On my iBook
> G3 with Radeon M6 I'm also having problems with rc2, whereas rc1 is
> working fine.  I haven't found the time to look closer yet.
> 
> Andreas.

Shame on me, i was passing wrong arguments to the kernel
now this works...

Jerome Glisse

Re: linux 2.6.12-rc1-bk5 compilation error

[Benjamin Herrenschmidt]

On Tue, 2005-04-05 at 14:39 +0200, Andreas Schwab wrote:
> Jerome Glisse <j.glisse@gmail.com> writes:
> 
> > I will try t look at radeon framebuffer see if i can find anythings.
> > Any idea of which change on that code may be the origin of this ?
> 
> I'd suggest looking at the changes between rc1 and rc2 first.  On my iBook
> G3 with Radeon M6 I'm also having problems with rc2, whereas rc1 is
> working fine.  I haven't found the time to look closer yet.

What problem specifically ?

Ben.

Re: linux 2.6.12-rc1-bk5 compilation error

[Andreas Schwab]

Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:

> On Tue, 2005-04-05 at 14:39 +0200, Andreas Schwab wrote:
>> Jerome Glisse <j.glisse@gmail.com> writes:
>> 
>> > I will try t look at radeon framebuffer see if i can find anythings.
>> > Any idea of which change on that code may be the origin of this ?
>> 
>> I'd suggest looking at the changes between rc1 and rc2 first.  On my iBook
>> G3 with Radeon M6 I'm also having problems with rc2, whereas rc1 is
>> working fine.  I haven't found the time to look closer yet.
>
> What problem specifically ?

<http://marc.theaimsgroup.com/?l=linux-kernel&m=111273873814456&w=2>

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: linux 2.6.12-rc1-bk5 compilation error

[Benjamin Herrenschmidt]

On Tue, 2005-04-05 at 13:52 +0200, Jerome Glisse wrote:
> I almost successfully run the 2.6.12-rc2, there were some
> bad old stuff lying around. But i needed the two workaround
> to compile it. So maybe this two could be applied to the kernel ?
> 
> Now the only issue which remains is that the radeon framebuffer
> no more set a proper mode for my screen. The screen complain
> about a out of range mode. But as soon as X start i get a proper
> mode set but can't switch to console.
> 
> The screen is a crt one on a radeon 9600 with dvi->vga adapter.
> Which works well on 2.6.10
> 
> I will try t look at radeon framebuffer see if i can find anythings.
> Any idea of which change on that code may be the origin of this ?

radeonfb picks the first mode in the list generated from DDC, which
unfortunately, on a lot of CRTs, is bullshit...

Ben.

Re: linux 2.6.12-rc1-bk5 compilation error

[Tom Rini]

On Mon, Apr 04, 2005 at 11:32:39AM +0100, Jerome Glisse wrote:

> Once again i face an error that i got with a previous
> 2.6 kernel (i already posted about this in january
> i think).
[snip]
> Anyway what puzzle me is that it seems i am the
> only one to face this ? Am i alone :) ?

I think you're the only person building a zImage for a 32bit kernel with
a 64bit toolchain for MULTIPLATFORM, yes. :)

This is a valid problem, however.  Does the following also work for you?

Signed-off-by: Tom Rini <trini@kernel.crashing.org>

Index: linux-2.6.10/arch/ppc/boot/simple/Makefile
===================================================================
--- linux-2.6.10.orig/arch/ppc/boot/simple/Makefile
+++ linux-2.6.10/arch/ppc/boot/simple/Makefile
@@ -118,10 +118,13 @@ zimageinitrd-$(pcore)			:= zImage.initrd
          end-$(pcore)			:= pcore
    cacheflag-$(pcore)			:= -include $(clear_L2_L3)
 
+# Really only valid if CONFIG_6xx=y
       zimage-$(CONFIG_PPC_PREP)		:= zImage-PPLUS
 zimageinitrd-$(CONFIG_PPC_PREP)		:= zImage.initrd-PPLUS
+ifeq ($(CONFIG_6xx),y)
      extra.o-$(CONFIG_PPC_PREP)		:= prepmap.o
         misc-$(CONFIG_PPC_PREP)		+= misc-prep.o mpc10x_memory.o
+endif
          end-$(CONFIG_PPC_PREP)		:= prep
 
          end-$(CONFIG_SANDPOINT)	:= sandpoint

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

Re: linux 2.6.12-rc1-bk5 compilation error

[Jerome Glisse]

> I think you're the only person building a zImage for a 32bit kernel with
> a 64bit toolchain for MULTIPLATFORM, yes. :)

I should move to 64bits :)
 
> This is a valid problem, however.  Does the following also work for you?

It partly works (correct the problem with disable_6xx_mmu)
but i still have to the following for resolving __flush_disable_L1

I changed the makefile in arch/ppc/kernel to have this :
obj-$(CONFIG_POWER4) += cpu_setup_power4.o
into :
obj-$(CONFIG_POWER4) += l2cr.o cpu_setup_power4.o

Jerome Glisse

[PATCH 2.6.12-rc2] ppc32: Fix building 32bit kernel for 64bit machines (Was: Re: linux 2.6.12-rc1-bk5 compilation error)

[Tom Rini]

On Fri, Apr 08, 2005 at 11:13:19AM +0100, Jerome Glisse wrote:
> > I think you're the only person building a zImage for a 32bit kernel with
> > a 64bit toolchain for MULTIPLATFORM, yes. :)
> 
> I should move to 64bits :)
>  
> > This is a valid problem, however.  Does the following also work for you?
> 
> It partly works (correct the problem with disable_6xx_mmu)
> but i still have to the following for resolving __flush_disable_L1
> 
> I changed the makefile in arch/ppc/kernel to have this :
> obj-$(CONFIG_POWER4) += cpu_setup_power4.o
> into :
> obj-$(CONFIG_POWER4) += l2cr.o cpu_setup_power4.o

OK, I think the following is a better workaround for that (based on
arch/ppc/platforms/pmac_sleep.S):

When building a ppc32 MULTIPLATFORM kernel for a 64bit pmac, we try and
build certain files or use certain functions that make no sense in that
context.  This catches the last of these.

Signed-off-by: Tom Rini <trini@kernel.crashing.org>

--- linux-2.6.orig/arch/ppc/platforms/pmac_cache.S
+++ linux-2.6/arch/ppc/platforms/pmac_cache.S
@@ -28,6 +28,9 @@
  */
 
 _GLOBAL(flush_disable_caches)
+#ifndef CONFIG_6xx
+	blr
+#else
 BEGIN_FTR_SECTION
 	b	flush_disable_745x
 END_FTR_SECTION_IFSET(CPU_FTR_SPEC7450)
@@ -323,3 +326,4 @@ END_FTR_SECTION_IFSET(CPU_FTR_L3CR)
 	mtmsr	r11		/* restore DR and EE */
 	isync
 	blr
+#endif	/* CONFIG_6xx */
--- linux-2.6.orig/arch/ppc/boot/simple/Makefile
+++ linux-2.6/arch/ppc/boot/simple/Makefile
@@ -123,10 +123,13 @@ zimageinitrd-$(pcore)			:= zImage.initrd
          end-$(pcore)			:= pcore
    cacheflag-$(pcore)			:= -include $(clear_L2_L3)
 
+# Really only valid if CONFIG_6xx=y
       zimage-$(CONFIG_PPC_PREP)		:= zImage-PPLUS
 zimageinitrd-$(CONFIG_PPC_PREP)		:= zImage.initrd-PPLUS
+ifeq ($(CONFIG_6xx),y)
      extra.o-$(CONFIG_PPC_PREP)		:= prepmap.o
         misc-$(CONFIG_PPC_PREP)		+= misc-prep.o mpc10x_memory.o
+endif
          end-$(CONFIG_PPC_PREP)		:= prep
 
          end-$(CONFIG_SANDPOINT)	:= sandpoint

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

Re: [PATCH 2.6.12-rc2] ppc32: Fix building 32bit kernel for 64bit machines (Was: Re: linux 2.6.12-rc1-bk5 compilation error)

[Jerome Glisse]

> OK, I think the following is a better workaround for that (based on
> arch/ppc/platforms/pmac_sleep.S):
> 
> When building a ppc32 MULTIPLATFORM kernel for a 64bit pmac, we try and
> build certain files or use certain functions that make no sense in that
> context.  This catches the last of these.
> 
> Signed-off-by: Tom Rini <trini@kernel.crashing.org>

Works fine over here. Thx.

Jerome Glisse

Re: [PATCH 2.6.12-rc2] ppc32: Fix building 32bit kernel for 64bit machines (Was: Re: linux 2.6.12-rc1-bk5 compilation error)

[Benjamin Herrenschmidt]

Go for it i

> When building a ppc32 MULTIPLATFORM kernel for a 64bit pmac, we try and
> build certain files or use certain functions that make no sense in that
> context.  This catches the last of these.
> 
> Signed-off-by: Tom Rini <trini@kernel.crashing.org>
> 
> --- linux-2.6.orig/arch/ppc/platforms/pmac_cache.S
> +++ linux-2.6/arch/ppc/platforms/pmac_cache.S
> @@ -28,6 +28,9 @@
>   */
>  
>  _GLOBAL(flush_disable_caches)
> +#ifndef CONFIG_6xx
> +	blr
> +#else
>  BEGIN_FTR_SECTION
>  	b	flush_disable_745x
>  END_FTR_SECTION_IFSET(CPU_FTR_SPEC7450)
> @@ -323,3 +326,4 @@ END_FTR_SECTION_IFSET(CPU_FTR_L3CR)
>  	mtmsr	r11		/* restore DR and EE */
>  	isync
>  	blr
> +#endif	/* CONFIG_6xx */
> --- linux-2.6.orig/arch/ppc/boot/simple/Makefile
> +++ linux-2.6/arch/ppc/boot/simple/Makefile
> @@ -123,10 +123,13 @@ zimageinitrd-$(pcore)			:= zImage.initrd
>           end-$(pcore)			:= pcore
>     cacheflag-$(pcore)			:= -include $(clear_L2_L3)
>  
> +# Really only valid if CONFIG_6xx=y
>        zimage-$(CONFIG_PPC_PREP)		:= zImage-PPLUS
>  zimageinitrd-$(CONFIG_PPC_PREP)		:= zImage.initrd-PPLUS
> +ifeq ($(CONFIG_6xx),y)
>       extra.o-$(CONFIG_PPC_PREP)		:= prepmap.o
>          misc-$(CONFIG_PPC_PREP)		+= misc-prep.o mpc10x_memory.o
> +endif
>           end-$(CONFIG_PPC_PREP)		:= prep
>  
>           end-$(CONFIG_SANDPOINT)	:= sandpoint
> 
-- 
Benjamin Herrenschmidt <benh@kernel.crashing.org>