ts-7000
[Top] [All Lists]

Re: [ts-7000] Re: TS-7250 + Linux 2.6.31 reprise

To:
Subject: Re: [ts-7000] Re: TS-7250 + Linux 2.6.31 reprise
From:
Date: Wed, 23 Sep 2009 23:48:22 -0600
"Charles" <> writes:

> Hi --
> --- In   wrote:
>> Strongly encouraged by seeing Martin Guy's recent post on working crunch
>> support in gcc 4.3 ( http://martinwguy.co.uk/martin/crunch/ )
>>
>> And some very clean, simple patches for my ts7250...
>>
>> http://martinwguy.co.uk/martin/ts7250/KERNEL
>
> Martin has done an incredible amount of work to get crunch support working
> around all the corner cases! But as far as up-to-date patches for mainline,
> take a look at:
> http://mcrapet.free.fr/

Darn. I'd spent most of the day digging into the sparsemem problem,
having fun, and then you tell me the problem is already mostly solved. :(

That url has to bubble up closer to the top of google somehow.

>> The fourth patch, involving the discontiguous memory support, was too
>> scary to apply at present and not required to just try out 2.6.31.
>
> For this, the right way is to enable sparsemem support after applying (at
> least) the base patch of Mathieu's patchset above. That is the direction all 
> of
> the ARM platforms are going. I don't think it's in vanilla yet for ep93xx (it
> wasn't in 2.6.29 at least), but check before applying the patch anyway.

There is sparsemem support for various arms in 2.6.31 but not the
ep93xx. It does look like a clean solution overall, for systems that
have a slightly less spreadeagled memory map than the ts7250.

Where I was stuck on introducing sparsemem was in trying to map
VMALLOC_END below physical ram, but I suspect I was attacking a symptom
rather than a cause.

More or less blindly applying 0003-ep93xx_base.patch introduced a world
of hurt I'm not going to fix today:

/home/d/src/git/new/linux-2.6/include/linux/mm.h:493:2: error: #error 
SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > BITS_PER_LONG - NR_PAGEFLAGS
make[2]: *** [arch/arm/kernel/asm-offsets.s] Error 1

Which is due to

#define SECTION_SIZE_BITS 24

In memory.h.

You'd think it should be 23 (8MB). That doesn't work either.

Setting it to 25 compiles only to blow up

In file included from 
/home/d/src/git/new/linux-2.6/arch/arm/kernel/process.c:55:
/home/d/src/git/new/linux-2.6/arch/arm/mach-ep93xx/include/mach/system.h: In 
function âarch_resetâ:
/home/d/src/git/new/linux-2.6/arch/arm/mach-ep93xx/include/mach/system.h:16: 
error: âdevicecfgâ undeclared (first use in this function)

I will clean up my tree, go back to the start, and be more thorough and
careful in the morning.

>
> Select sparsemem, after 0003-ep93xx_base.patch
>
>>
>> 2) Correctly identifying the flash. My assumption is that the flash
>> partition is actually on 0 not not the last block. Is there a known
>> working 2.6.25 .config for this board out there to start from (googling
>> deeper and recompiling now)
>
> Apply 0023-ts7250_nand.patch. Mathieu also has a bunch of different configs to
> choose from.

My concern is this (very old) dev board has a messed up redboot set of
partitions in the first place:

RedBoot> fis list
Name              FLASH addr  Mem addr    Length      Entry point
(reserved)        0x60000000  0x60000000  0x07D04000  0x00000000
RedBoot           0x67D04000  0x67D04000  0x00040000  0x00000000
vmlinux           0x67D44000  0x00218000  0x00160000  0x00218000
RedBoot config    0x67FF8000  0x67FF8000  0x00001000  0x00000000
FIS directory     0x67FFC000  0x67FFC000  0x00004000  0x00000000


>> 3) It would be nice if I knew how to get it to recognise the usb stick in
>> it without having to having to do a switch_root or whatever the
>> equivalent is nowadays. That way I could (hopefully) just boot right
>> into debian...
>
> Don't think the TS boards will boot from USB directly.

Well, it was my hope that compiling in the usb and scsi and usb-storage
modules along with the filesystems would get me there, but so far no luck.
>
>> Creating 3 MTD partitions on "ts7250-nand":
>> 0x000000000000-0x000000004000 : "TS-BOOTROM"
>> 0x000000004000-0x000007d04000 : "Linux"
>> 0x000007d04000-0x000008000000 : "RedBoot"
>
> Yep, you need the patch above for 2k blocks.

The code I'd looked at didn't look at the device for partitioning
information. I'll look over the 2k patch in the morning.

>> usb 1-3: new full speed USB device using ep93xx-ohci and address 2
>> usb 1-3: no configurations
>> usb 1-3: can't read configurations, error -22
>> usb 1-3: new full speed USB device using ep93xx-ohci and address 3
>> usb 1-3: no configurations
>> usb 1-3: can't read configurations, error -22
>> usb 1-3: new full speed USB device using ep93xx-ohci and address 4
>> usb 1-3: no configurations
>> usb 1-3: can't read configurations, error -22
>> usb 1-3: new full speed USB device using ep93xx-ohci and address 5
>> usb 1-3: no configurations
>> usb 1-3: can't read configurations, error -22
>> hub 1-0:1.0: unable to enumerate USB device on port 3
>
> Don't know about these.

Trying to convince a usb stick to be recognised on boot.

>
> regards .......... Charlie
>

Thanks very much for the steer!

--
Dave Taht
http://the-edge.blogspot.com


------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
    
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU