RE: MPC885ADS and 2.6.13-rc5 - nogo ?

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hello!=20

Replying to myself: I'm able to see bootup messages on ttyCPM0 now=20
after i applied the "Made non-console uart work"-patch that i skipped
until now because, well, it was "non-console" ...

However, output on ttyCPM0 stops soon:

 Linux version 2.6.13-rc5 (ashaepe@rcf02) (gcc version 3.3.2) #20 Fri
Aug 5 17:48:42 CEST 2005
 Built 1 zonelists
 Kernel command line: console=3DttyCPM0,38400n8 root=3D/dev/nfs
nfsroot=3D/home/ashaepe/nfs rw ip=3D10.135.2.133:10.135.0.22:1
 0.135.0.1:255.255.252.0::eth0:off
 PID hash table entries: 64 (order: 6, 1024 bytes)
 Decrementer Frequency =3D 375000000/60
 Dentry cache hash table entries: 2048 (order: 1, 8192 bytes)
 Inode-cache hash table entries: 1024 (order: 0, 4096 bytes)
 Memory: 6272k available (1352k kernel code, 404k data, 68k init, 0k
highmem)
 Mount-cache hash table entries: 512
 Badness in dma_alloc_init at arch/ppc/kernel/dma-mapping.c:348
 Call trace:
  [c0005b10] dump_stack+0x18/0x28
  [c0003c60] check_bug_trap+0x84/0xac
  [c0003de8] ProgramCheckException+0x160/0x1a0
  [c0003260] ret_from_except_full+0x0/0x4c
  [c01a9f84] dma_alloc_init+0x44/0xa0
  [c01a66a4] do_initcalls+0x54/0xfc
  [c0002260] init+0x2c/0xf4
  [c000533c] kernel_thread+0x44/0x60
 NET: Registered protocol family 16
 JFFS2 version 2.2. (NAND) (C) 2001-2003 Red Hat, Inc.
 Serial: CPM driver $Revision: 0.01 $
 ttyCPM0 at MMIO 0xff000a80 (irq =3D 20) is a CPM UART
 ttyCPM1 at MMIO 0xff000a90 (irq =3D 19) is a CPM UART
 io scheduler noop registered

And that's all output for now...

Can anybody explain the "dma_alloc_init"-error ?

 Best regards,

 Peter

-----Original Message-----

Hello!

I gave the MPC885ADS support recently added by Andrei Konovalov=20
a try - but no luck: booting 2.6.13-rc5 with u-boot 1.1.2=20
(tftpboot & bootm) hangs after uncompressing.

Did anybody succeed with this?

Applied were the following patches from patchwork.ozlabs.org:

 * 8xx_cpm_hostalloc
 * all fixes to fec.c

The board boots an 2.4-Kernel just fine. Did i miss some patches=20
for U-Boot?

Best regards,

 Peter

Re: MPC885ADS and 2.6.13-rc5 - nogo ?

[aris]

>  Badness in dma_alloc_init at arch/ppc/kernel/dma-mapping.c:348
>  Call trace:
>   [c0005b10] dump_stack+0x18/0x28
>   [c0003c60] check_bug_trap+0x84/0xac
>   [c0003de8] ProgramCheckException+0x160/0x1a0
>   [c0003260] ret_from_except_full+0x0/0x4c
>   [c01a9f84] dma_alloc_init+0x44/0xa0
>   [c01a66a4] do_initcalls+0x54/0xfc
>   [c0002260] init+0x2c/0xf4
>   [c000533c] kernel_thread+0x44/0x60
>  NET: Registered protocol family 16
>  JFFS2 version 2.2. (NAND) (C) 2001-2003 Red Hat, Inc.
>  Serial: CPM driver $Revision: 0.01 $
>  ttyCPM0 at MMIO 0xff000a80 (irq = 20) is a CPM UART
>  ttyCPM1 at MMIO 0xff000a90 (irq = 19) is a CPM UART
just out of curiosity, try changing CONSISTENT_START to 0xe0000000

--
Aristeu

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Replying to myself: My MPC855ADS-Board has=20

 -> U-Boot 1.1.2 (Jun 27 2005 - 11:50:05)
 -> CPU:   MPC885ZPnn at 120 MHz: 8 kB I-Cache 8 kB D-Cache FEC present
 -> Board: MPC885ADS rev NR
 -> DRAM:  (8 MB SDRAM)  8 MB

... and 8 MB Flash.

AFAIK there were boards with only 2 MB RAM. Which boards do you guys
have?

Best regards,

 Peter

-----Original Message-----
From: linuxppc-embedded-bounces@ozlabs.org
[mailto:linuxppc-embedded-bounces@ozlabs.org] On Behalf Of
Schaefer-Hutter, Peter
Sent: Monday, August 08, 2005 11:38 AM
To: linuxppc-embedded list
Subject: RE: MPC885ADS and 2.6.13-rc5 - nogo ?

Hello!

> From: aris@cathedrallabs.org [mailto:aris@cathedrallabs.org]=20
> Sent: Friday, August 05, 2005 6:26 PM
>
> >  Badness in dma_alloc_init at arch/ppc/kernel/dma-mapping.c:348
> >  Call trace:
> >   [c0005b10] dump_stack+0x18/0x28
> >   [c0003c60] check_bug_trap+0x84/0xac
> >   [c0003de8] ProgramCheckException+0x160/0x1a0
> >   [c0003260] ret_from_except_full+0x0/0x4c
> >   [c01a9f84] dma_alloc_init+0x44/0xa0
> >   [c01a66a4] do_initcalls+0x54/0xfc
> >   [c0002260] init+0x2c/0xf4
> >   [c000533c] kernel_thread+0x44/0x60
> >  NET: Registered protocol family 16
> >  JFFS2 version 2.2. (NAND) (C) 2001-2003 Red Hat, Inc.
> >  Serial: CPM driver $Revision: 0.01 $
> >  ttyCPM0 at MMIO 0xff000a80 (irq =3D 20) is a CPM UART
> >  ttyCPM1 at MMIO 0xff000a90 (irq =3D 19) is a CPM UART
> just out of curiosity, try changing CONSISTENT_START to 0xe0000000

Well, then the error in dma_alloc_init goes away.

However, i'm still not able to run 2.6.x on the MPC885ADS.
I'm now using 2.6.13-rc6 with the latest cpm-uart-fixes=20
applied. I made the defconfig, set CONSISTENT_START to 0xe0000000
an got:

 Linux version 2.6.13-rc6 (root@rcf02) (gcc version 3.3.2) #5 Mon Aug 8
11:22:18 CEST 2005
 Built 1 zonelists
 Kernel command line: console=3DttyCPM0,38400n8 root=3D/dev/nfs
nfsroot=3D/home/test/nfs rw
ip=3D192.168.1.10:192.168.1.1:192.168.1.1:255.255.255.0::eth0:off
 PID hash table entries: 64 (order: 6, 1024 bytes)
 Decrementer Frequency =3D 450000000/60
 Dentry cache hash table entries: 2048 (order: 1, 8192 bytes)
 Inode-cache hash table entries: 1024 (order: 0, 4096 bytes)
 Memory: 6244k available (1432k kernel code, 332k data, 72k init, 0k
highmem)
 Mount-cache hash table entries: 512
 NET: Registered protocol family 16
 Serial: CPM driver $Revision: 0.01 $
 ttyCPM0 at MMIO 0xff000a80 (irq =3D 20) is a CPM UART
 ttyCPM1 at MMIO 0xff000a90 (irq =3D 19) is a CPM UART
 io scheduler noop registered
 PPP generic driver version 2.4.2
 PPP Deflate Compression module registere

Looks like the console does not start (maybe still UART-issues)...=20

As some people on this list reported success on MPC885ADS - can somebody

please hit me with the cluebat ?

Best regards,=20

  Peter

_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hello!

> From: aris@cathedrallabs.org [mailto:aris@cathedrallabs.org]=20
> Sent: Friday, August 05, 2005 6:26 PM
>
> >  Badness in dma_alloc_init at arch/ppc/kernel/dma-mapping.c:348
> >  Call trace:
> >   [c0005b10] dump_stack+0x18/0x28
> >   [c0003c60] check_bug_trap+0x84/0xac
> >   [c0003de8] ProgramCheckException+0x160/0x1a0
> >   [c0003260] ret_from_except_full+0x0/0x4c
> >   [c01a9f84] dma_alloc_init+0x44/0xa0
> >   [c01a66a4] do_initcalls+0x54/0xfc
> >   [c0002260] init+0x2c/0xf4
> >   [c000533c] kernel_thread+0x44/0x60
> >  NET: Registered protocol family 16
> >  JFFS2 version 2.2. (NAND) (C) 2001-2003 Red Hat, Inc.
> >  Serial: CPM driver $Revision: 0.01 $
> >  ttyCPM0 at MMIO 0xff000a80 (irq =3D 20) is a CPM UART
> >  ttyCPM1 at MMIO 0xff000a90 (irq =3D 19) is a CPM UART
> just out of curiosity, try changing CONSISTENT_START to 0xe0000000

Well, then the error in dma_alloc_init goes away.

However, i'm still not able to run 2.6.x on the MPC885ADS.
I'm now using 2.6.13-rc6 with the latest cpm-uart-fixes=20
applied. I made the defconfig, set CONSISTENT_START to 0xe0000000
an got:

 Linux version 2.6.13-rc6 (root@rcf02) (gcc version 3.3.2) #5 Mon Aug 8
11:22:18 CEST 2005
 Built 1 zonelists
 Kernel command line: console=3DttyCPM0,38400n8 root=3D/dev/nfs
nfsroot=3D/home/test/nfs rw
ip=3D192.168.1.10:192.168.1.1:192.168.1.1:255.255.255.0::eth0:off
 PID hash table entries: 64 (order: 6, 1024 bytes)
 Decrementer Frequency =3D 450000000/60
 Dentry cache hash table entries: 2048 (order: 1, 8192 bytes)
 Inode-cache hash table entries: 1024 (order: 0, 4096 bytes)
 Memory: 6244k available (1432k kernel code, 332k data, 72k init, 0k
highmem)
 Mount-cache hash table entries: 512
 NET: Registered protocol family 16
 Serial: CPM driver $Revision: 0.01 $
 ttyCPM0 at MMIO 0xff000a80 (irq =3D 20) is a CPM UART
 ttyCPM1 at MMIO 0xff000a90 (irq =3D 19) is a CPM UART
 io scheduler noop registered
 PPP generic driver version 2.4.2
 PPP Deflate Compression module registere

Looks like the console does not start (maybe still UART-issues)...=20

As some people on this list reported success on MPC885ADS - can somebody

please hit me with the cluebat ?

Best regards,=20

  Peter

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hi,=20

> From: Vitaly Bordug [mailto:vbordug@ru.mvista.com]=20

> > Schaefer-Hutter, Peter wrote:=20
> >	Shouldn't that read:
> >	  #define BCSR_SIZE   ((uint)32 * 1024)

> <quotation> (This is from 8272 User Guide, but 885 I guess is the same
in this part)
>	[ ... ]
> <quotation\>

Well, i was just wondering why fads.h and my old 2.4-Kernel define it
'(32*1024)', but
this explains it.

Thank you and best regards,

 Peter

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hello,

> From: Vitaly Bordug [mailto:vbordug@ru.mvista.com]=20

> The board does not hang - there's somithing with=20
> console since you don't see anything.  Try to change=20
> BCSR_ADDR from 0xf4500000 to 0xf80...0 - the newer=20
> version of the boards may have this changed.

Hrm... arc/ppc/mpc885ads.h already reads

  /* U-Boot maps BCSR to 0xff080000 */
  #define BCSR_ADDR   ((uint)0xff080000)

And that's the same setting that my 2.4-Kernel
uses, so it should work with 0xff080000.

However, BCSR_SIZE looks suspicious in this header:

  #define BCSR_SIZE   ((uint)32)

Shouldn't that read:

  #define BCSR_SIZE   ((uint)32 * 1024)

??!?

Regards,

  Peter

Re: MPC885ADS and 2.6.13-rc5 - nogo ?

[Vitaly Bordug]

[-- Attachment #1: Type: text/plain, Size: 1348 bytes --]

Schaefer-Hutter, Peter wrote:

>Hello,
>
>  
>
>>From: Vitaly Bordug [mailto:vbordug@ru.mvista.com] 
>>    
>>
>
>  
>
>>The board does not hang - there's somithing with 
>>console since you don't see anything.  Try to change 
>>BCSR_ADDR from 0xf4500000 to 0xf80...0 - the newer 
>>version of the boards may have this changed.
>>    
>>
>
>Hrm... arc/ppc/mpc885ads.h already reads
>
>  /* U-Boot maps BCSR to 0xff080000 */
>  #define BCSR_ADDR   ((uint)0xff080000)
>
>And that's the same setting that my 2.4-Kernel
>uses, so it should work with 0xff080000.
>
>However, BCSR_SIZE looks suspicious in this header:
>
>  #define BCSR_SIZE   ((uint)32)
>
>Shouldn't that read:
>
>  #define BCSR_SIZE   ((uint)32 * 1024)
>
>??!?
>
>Regards,
>
>  Peter
>
>  
>
<quotation> (This is from 8272 User Guide, but 885 I guess is the same 
in this part)

> Most of the hardware options on the MPC8272ADS are controlled or 
> monitored by the
> BCSR, which is a 32 bit wide read / write register file. The BCSR is 
> accessed via the
> MPC8272s' memory controller (see Table 5-5) and in fact includes 8 
> registers: BCSR0 to
> BCSR7. Since the minimum block size for a CS region is 32KBytes and 
> only A(27:29)
> lines are decoded by the BCSR for register selection, BCSR0 - BCSR7 
> are duplicated inside
> that region.

<quotation\>

-- 
Sincerely, 
Vitaly


[-- Attachment #2: Type: text/html, Size: 2083 bytes --]

RE: MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hello Vitaly,=20

> From: Vitaly Bordug [mailto:vbordug@ru.mvista.com]=20
>
> Mentioned support should work `out-of-the box`.
> Make sure you have console=3DttyCPM0,115200 (or whatever baudrate you=20
> have) within bootargs in the U-Boot environment. U-Boot modification
is=20
> needed only if you wish to have both FEC MAC addresses configured via=20
> U-Boot env as well. Also only the related defconfig should be used.

I have the bootargs, but still nogo - even with the mpc885ads_defconfig.

Besides, in the mpc885ads_defconfig CONFIG_SCC3_ENET is set but in the
patch
description it reads as if only SCC1 is supported...

 Best Regards,

  Peter

Re: MPC885ADS and 2.6.13-rc5 - nogo ?

[Vitaly Bordug]

[-- Attachment #1: Type: text/plain, Size: 952 bytes --]

Schaefer-Hutter, Peter wrote:

>Hello Vitaly, 
>
>  
>
>>From: Vitaly Bordug [mailto:vbordug@ru.mvista.com] 
>>
>>Mentioned support should work `out-of-the box`.
>>Make sure you have console=ttyCPM0,115200 (or whatever baudrate you 
>>have) within bootargs in the U-Boot environment. U-Boot modification
>>    
>>
>is 
>  
>
>>needed only if you wish to have both FEC MAC addresses configured via 
>>U-Boot env as well. Also only the related defconfig should be used.
>>    
>>
>
>I have the bootargs, but still nogo - even with the mpc885ads_defconfig.
>
>Besides, in the mpc885ads_defconfig CONFIG_SCC3_ENET is set but in the
>patch
>description it reads as if only SCC1 is supported...
>
> Best Regards,
>
>  Peter
>
>  
>
The board does not hang - there's somithing with console since you don't 
see anything.  Try to change BCSR_ADDR from 0xf4500000 to 0xf80...0 - 
the newer version of the boards may have this changed.


-- 
Sincerely, 
Vitaly


[-- Attachment #2: Type: text/html, Size: 1585 bytes --]

MPC885ADS and 2.6.13-rc5 - nogo ?

[Schaefer-Hutter, Peter]

Hello!

I gave the MPC885ADS support recently added by Andrei Konovalov=20
a try - but no luck: booting 2.6.13-rc5 with u-boot 1.1.2=20
(tftpboot & bootm) hangs after uncompressing.

Did anybody succeed with this?

Applied were the following patches from patchwork.ozlabs.org:

 * 8xx_cpm_hostalloc
 * all fixes to fec.c

The board boots an 2.4-Kernel just fine. Did i miss some patches=20
for U-Boot?

Best regards,

 Peter

Re: MPC885ADS and 2.6.13-rc5 - nogo ?

[Vitaly Bordug]

Schaefer-Hutter, Peter wrote:

>Hello!
>
>I gave the MPC885ADS support recently added by Andrei Konovalov 
>a try - but no luck: booting 2.6.13-rc5 with u-boot 1.1.2 
>(tftpboot & bootm) hangs after uncompressing.
>
>Did anybody succeed with this?
>
>Applied were the following patches from patchwork.ozlabs.org:
>
> * 8xx_cpm_hostalloc
> * all fixes to fec.c
>
>The board boots an 2.4-Kernel just fine. Did i miss some patches 
>for U-Boot?
>
>Best regards,
>
> Peter
>_______________________________________________
>Linuxppc-embedded mailing list
>Linuxppc-embedded@ozlabs.org
>https://ozlabs.org/mailman/listinfo/linuxppc-embedded
>
>
>  
>
Mentioned support should work `out-of-the box`.
Make sure you have console=ttyCPM0,115200 (or whatever baudrate you 
have) within bootargs in the U-Boot environment. U-Boot modification is 
needed only if you wish to have both FEC MAC addresses configured via 
U-Boot env as well. Also only the related defconfig should be used.

-- 
Sincerely, 
Vitaly