Hi --
--- In Alexander Clouter <> wrote:
>
> Hi,
>
> charliem_1216 <> [20080519 12:57:10 -0000]:
> >
> > > I'm deeply confused, from what I can tell the ts72xx series has
been
> > > supported natively on a 2.6.x kernel for some time. Have a look at
> > > arch/arm/mach-ep93xx/ts72xx.c. Unsure when it arrived but the
> > > copyright states 2006.
> >
> > There was initial work done by Lennert Buytenhek for the TS-7200 that
> > worked fairly well. Not all of it was pushed to mainline. There were
> > a number of show-stopper issues if you tried to use vanilla (well,
> > showstoppers if you didn't know where to look / what to do ...). From
> > memory:
> >
> > * 64M RAM required a discontig mem patch (32M boards were OK)
> >
> sparse mem support *seems* to be in for ARM platforms with a quick
grep on
> the 2.6.25 kernel tree, this good enough anyone know? If someone
in/near the
> UK/London has a spare 72x0 board for me to play with... ;)
No, sparse mem is preferred approach, but doesn't work, while
discontig mem approach is different, and does work. Sparse could be
made to work I think, but is more effort for little gain, especially
considering the long-term cirrus-arm situation. Details are in the
long-ish thread starting here:
http://www.freelists.org/archives/linux-cirrus/12-2006/msg00028.html
>
> > * Serial UART speed fix
> > * Fix nF -bit redboot problem
> >
> Got any details (patches already) or should I munch my own advice
and trawl
> the mailing list archives?
Details of the redboot nF issue are in this thread:
http://www.freelists.org/archives/linux-cirrus/01-2007/msg00073.html
>
> > * CF support (and SD of course)
> >
> ...well the SD card cannot be fixed until we get raw access to the
pins so I
> guess thats a dead end anyway.
>
> > * Machine ID
> > * NAND partition table layout
> > * 2k page NAND
> >
> Well my 7800 patch made the NAND partitioning and handling all platform
> specific with auto-magical and at *runtime* detection thrown in. I
> generalised this with the idea that the 72x0 stuff could be hacked
to use the
> 'unified' NAND driver.
Yes, autodetected would be best of course.
>
> > * YAFFS2
> >
> With the SD card I'm guessing this is the next biggy as far as
issues go.
> However it's not too bad to ask users to download a vanilla kernel
and apply
> the YAFFS2 patch manually if the rest has been done for them (I guess?).
With the 64M TS-7250 boards, and maybe others, there is NAND on the
board which uses 2k block size, and needs YAFFS2.
>
> Well I have been asked by the Debian folk and started speaking to
the orion5x
> maintainers about putting the 7800 patch I made upstream so that a
native
> Debian installer can be put togther for it. I have to split it up into
> seperate parts and until the DMA NAND issue is fixed it's probably
un-usable.
That would be great! I hope someday TS sees the benefit of helping
get their board support upstream bits, so things 'just work' with
future kernel versions.
>
> > So there evolved various sets of patches and recipes for 2.6 on the
> > M/L. Most stuff, except SD support I think, could be made to work if
> > you did some hunting, but AFAIK no-one has been pushing fixes
up-stream.
> >
> Well I probably would be happy to faff with the 72x0 if I had one
but the
> 7800 takes priority, mainly as I have one and one thing at a time is
more
> than enough for me ;)
My original notes and patch sets were described here:
http://tech.groups.yahoo.com/group/ts-7000/message/5497
Recently Mathieu C. has been having a go at a consolidated 2.6 patch
for TS-72xx boards, which look interesting, but I haven't had a chance
to try. I think he's still lacking the discontig ram patch though:
http://tech.groups.yahoo.com/group/ts-7000/message/11350
Regards, ......... Charlie
>
> Cheers
>
> Alex
>
> > > > > --- In "Steffanus Wiedjaja"
<steff_jr@>
> > > > > wrote:
> > > > > >
> > > > > > --- In "charliem_1216"
<charliem_1216@>
> > > > > > wrote:
> > > > > > > uS. I'd be happy to test some other code on 2.6 if it
would
> > > > > help,
> > > > > > > but it sounds like you want to stick to 2.4 kernels.
> > > > > > >
> > > > > > > Regards, ......... Charlie
> > > > > > >
> > > > > >
> > > > > > Hi Charlie,
> > > > > > It's seems you're successfully port 2.6 kernel into 7250,
can u
> > > > > > share how to do that? We will very appreciate.
> > > > > >
> > > > > > best regards,
> > > > > > steff
> > > > > >
> > > > >
> > > > > I wrote about the steps I used in:
> > > > > http://tech.groups.yahoo.com/group/ts-7000/message/5497
> > > > > But many other people have also written summaries about their
> > > > > approaches, which you can find in the archives.
> > > > >
> > > > > Since my post above I have moved to a more recent kernel, and
> > > > > started using Ingo's RT and HRT patchset, all without any real
> > > > > problems. Note that my testing is done mostly using EABI and
> > > > > NFSroot, and I have not yet put a modern kernel / distro
back into
> > > > > flash.
> > > > >
> > > > > Regards, ....... Charlie
> > > > >
> > > >
> > > >
> > > >
> > >
> > > --
> > > ______________________________
> > > < A watched clock never boils. >
> > > ------------------------------
> > > \ ^__^
> > > \ (oo)\_______
> > > (__)\ )\/\
> > > ||----w |
> > > || ||
> > >
> >
> >
>
> --
> ______________________________________
> / Any philosophy that can be put "in a \
> | nutshell" belongs there. |
> | |
> \ -- Sydney J. Harris /
> --------------------------------------
> \ ^__^
> \ (oo)\_______
> (__)\ )\/\
> ||----w |
> || ||
>
------------------------------------
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/
|