NXP STB225 board support

NXP STB225 board support

[Ihar Hrachyshka]

Hello!
I try to start vanilla linux kernel on pnx-8335-stb225 board. What I did:
1. ARCH=mips make pnx-8335-stb225_defconfig
2. ARCH=mips make
3. Prepared U-Boot image with mkimage.

After booting it up I get the following output on my UART:

Linux version 2.6.28-rc9.netflix.PR12_RC2 (booxter@EPBYMINW0568) (gcc
version 4.2.1) #3 PREEMPT Tue Dec 23 15:52:10 EET 2008
CPU revision is: 00019068 (MIPS 4KEc)
Determined physical RAM map:
 memory: 10000000 @ 00000000 (usable)
Zone PFN ranges:
  Normal   0x00000000 -> 0x00010000
Movable zone start PFN for each node
early_node_map[1] active PFN ranges
    0: 0x00000000 -> 0x00010000
Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 65024
Kernel command line:
mtdparts=gen_nand:128k(Boot0),1M(Boot1),512k(Env),4M(Kernel),16M(Filesystem)
ip3902.mac_address=00:60:37:03:00:00 ip=10.6.2.53:10.6.27.3
Unknown boot option `ip3902.mac_address=00:60:37:03:00:00': ignoring
Primary instruction cache 16kB, VIPT, 2-way, linesize 16 bytes.
Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes
PID hash table entries: 1024 (order: 10, 4096 bytes)
CPU clock is 320 MHz
Console: colour dummy device 80x25
console [ttyS0] enabled

After this last message there is no any activity. It seems that kernel
hanged. What can I do to see what's going on (maybe stack trace)? Any
suggestions?
Has anyone started the board successfully with vanilla kernel?

Thanks in advance,
Ihar Hrachyshka

Re: NXP STB225 board support

[Ihar Hrachyshka]

On Tue, Dec 23, 2008 at 4:48 PM, Ihar Hrachyshka
<ihar.hrachyshka@gmail.com> wrote:
> Hello!
> I try to start vanilla linux kernel on pnx-8335-stb225 board. What I did:
> 1. ARCH=mips make pnx-8335-stb225_defconfig
> 2. ARCH=mips make
> 3. Prepared U-Boot image with mkimage.
>
> After booting it up I get the following output on my UART:
>
> Linux version 2.6.28-rc9.netflix.PR12_RC2 (booxter@EPBYMINW0568) (gcc
> version 4.2.1) #3 PREEMPT Tue Dec 23 15:52:10 EET 2008
> CPU revision is: 00019068 (MIPS 4KEc)
> Determined physical RAM map:
>  memory: 10000000 @ 00000000 (usable)
> Zone PFN ranges:
>  Normal   0x00000000 -> 0x00010000
> Movable zone start PFN for each node
> early_node_map[1] active PFN ranges
>    0: 0x00000000 -> 0x00010000
> Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 65024
> Kernel command line:
> mtdparts=gen_nand:128k(Boot0),1M(Boot1),512k(Env),4M(Kernel),16M(Filesystem)
> ip3902.mac_address=00:60:37:03:00:00 ip=10.6.2.53:10.6.27.3
> Unknown boot option `ip3902.mac_address=00:60:37:03:00:00': ignoring
> Primary instruction cache 16kB, VIPT, 2-way, linesize 16 bytes.
> Primary data cache 8kB, 4-way, VIPT, no aliases, linesize 16 bytes
> PID hash table entries: 1024 (order: 10, 4096 bytes)
> CPU clock is 320 MHz
> Console: colour dummy device 80x25
> console [ttyS0] enabled
>
> After this last message there is no any activity. It seems that kernel
> hanged. What can I do to see what's going on (maybe stack trace)? Any
> suggestions?
> Has anyone started the board successfully with vanilla kernel?
>
> Thanks in advance,
> Ihar Hrachyshka
>

Bisecting my Linus vanilla git, I found that the problem appeared
after the following patch was applied:

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=566f74f6b2f8b85d5b8d6caaf97e5672cecd3e3e

After reverting the patch, Linus vanilla git kernel again boots ok on
the board. Please, take a look.

Re: NXP STB225 board support

[Kevin D. Kissell]

Ihar Hrachyshka wrote:
 >
 >
 > Bisecting my Linus vanilla git, I found that the problem appeared
 > after the following patch was applied:
 >
 > 
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=566f74f6b2f8b85d5b8d6caaf97e5672cecd3e3e
 >
 > After reverting the patch, Linus vanilla git kernel again boots ok on
 > the board. Please, take a look.

Could one of the Cavium guys explain why the new code for V2 cores does
"ebase += ..."  and not "ebase = "?

          Kevin K.

Re: NXP STB225 board support

[Chad Reese]

Kevin D. Kissell wrote:
> Ihar Hrachyshka wrote:
>  >
>  >
>  > Bisecting my Linus vanilla git, I found that the problem appeared
>  > after the following patch was applied:
>  >
>  > 
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=566f74f6b2f8b85d5b8d6caaf97e5672cecd3e3e
>  >
>  > After reverting the patch, Linus vanilla git kernel again boots ok on
>  > the board. Please, take a look.
> 
> Could one of the Cavium guys explain why the new code for V2 cores does
> "ebase += ..."  and not "ebase = "?

The value "(read_c0_ebase() & 0x3ffff000)" is a physical address and
must be converted into Mips KSEG address by adding CAC_BASE. My guess is
that something about the NXP STB225 isn't fully Mips r2 and doesn't
implement the ebase register correctly. I'm afraid I don't know anything
about that processor.

Chad

Re: NXP STB225 board support

[David Daney]

Kevin D. Kissell wrote:
> Ihar Hrachyshka wrote:
>  >
>  >
>  > Bisecting my Linus vanilla git, I found that the problem appeared
>  > after the following patch was applied:
>  >
>  > 
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=566f74f6b2f8b85d5b8d6caaf97e5672cecd3e3e 
> 
>  >
>  > After reverting the patch, Linus vanilla git kernel again boots ok on
>  > the board. Please, take a look.
> 
> Could one of the Cavium guys explain why the new code for V2 cores does
> "ebase += ..."  and not "ebase = "?
> 


We need to add the offset to the proper segment (CAC_BASE).

That said, the change to set_uncached_handler looks incorrect.  It 
should be adding onto uncached_ebase not ebase.

Someone should probably fix that.

David Daney