Hi Bob,
rwarner wrote:
Greg Ungerer wrote:
rwarner wrote:
Michael Schnell wrote:
I read that the now officially released Kernel 2.6.25 offers a lot
of goodies for embedded / soft-realtime systems.
Will this version included in the tools (Buildroot etc) ?
Is it recommended to use 2.6.25 for new projects ?
Did they fix the ARM no mmu stuff in the kernel for the ARM7TDMI?
By "they" do you mean me?
Wow, ok sorry. Didn't mean to step on toes.
No, no, its not that :-)
But "they" is someone. And we here on this list and not disjoint
from what is happening in the wider ARM community for example
(the 2 sets are not disjoint).
For quite a while now I have been the only person trying to get
the last of the non-mmu ARM changes mainlined. I am by no means
the only person working on 2.6.x ARM non-mmu in some way though.
Thought it was more then
just you. Thought you were just maintaining the uCLinux tree and the
Russel & company was maintaining the ARM Ltd tree.
Well, there is no ARM Ltd tree, as such. Russell is an independent
developer. He is the primary maintainer, almost all ARM mainline
kernel changes go through him to Linus.
I have been sending patches to Russell to get the last
remaining core pieces for non-MMU ARM included in mainline.
Otherwise no-one else is doing it.
He accepted 2 patches to fix compile problems in head.S.
So it is a little closer again. Still not complete though.
Thing is, this isn't a case of just throwing patches at Russell.
The changes need to be re-worked and acceptable to him before
they will go in.
The 2 most critical outstanding pieces is the consistent-nommu
handling, and the non-MMU fault changes. Both are not large, and
it isn't clear to me yet how Russell wants them structured to
be acceptable.
Do you need help with them as i will need this as well. Only caveat is
i don't have a fully functional system yet?
I am finishing of the next set of changes for review on the
ARM linux kernel list now (should send them out in the next
couple of days). Depending on feedback to that we can see
what direction they need to go.
I have 7 patches. 5 are next to trivial. 2 are the contentious
fault and consistent mem changes.
Now of course ARM7TDMI is too generic to really be useful
anyway. ARM7TDMI from almost every vendor is different in
terms of interrupt setup, timers and memory map. So there are
SoC specifics that need to be handled as well. I am only working
with, and maintaining, the AT91M40 currently.
Yes, it was a little to generic for me to say just ARM7TDMI. Yes the
AT91M40 will work for me. Was also curious about the AT91SAM7SE as that
will be the real core we will be using, any idea of issues with that
core and the linux kernel build?
For simple core support it should not be too hard. AT91 is well
supported, and the normal hardware variances in it are relatively
easy to setup and deal with.
Last time i tried it the build was broken and someone admitted it
was not maintained since the last attempt 2.6.14?
non-MMU ARM has never be in mainline, not in 2.6.14, not ever.
I have been keeping and updating patches (kept in the uClinux-dist
linux-2.6.x sources).
Sorry, had the wrong idea from the following link,
http://opensrc.sec.samsung.com/
It said it was mainlined.
Ok :-)
Yes that page gives the wrong impression. Hyok never followed
through and got everything merged.
In addition, the XIP portion of the kernel also appeared broken for
the ARM7TDMI.
Has it ever worked in the 2.6 series kernels?
Not sure, i thought it is fully functional for other ARM products.
I don't know, you may be right. I haven't looked at kernel
XIP at all for ARM.
Regards
Greg
------------------------------------------------------------------------
Greg Ungerer -- Chief Software Dude EMAIL: [EMAIL PROTECTED]
SnapGear -- a Secure Computing Company PHONE: +61 7 3435 2888
825 Stanley St, FAX: +61 7 3891 3630
Woolloongabba, QLD, 4102, Australia WEB: http://www.SnapGear.com
_______________________________________________
uClinux-dev mailing list
uClinux-dev@uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev@uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev
