On 24/04/12 07:52, Claudio Mignanti wrote:
Hi Christian Gagneraud,
Il 23 aprile 2012 14:52, Christian Gagneraud<ch...@techworks.ie> ha scritto:
Hi all,
I've manifested my interest in the AT91 a couple of weeks ago, and nobody
even answer any of my calls.
I would like to work hand in hand with other peoples that use AT91 based
boards.
Obviously, someone started some important modifications on the at91
directory, which look like a good idea to me, but my point is that why this
person never speak about these changes on the mailing list.
One of the big change is the linux version bump, now it seems that it has to
be the untested 3.3.2 instead of the working 2.6.38.8.
Well, important changes was applied to at91 target and subtarget, as
you can see those improvement is not only related to the at91 target
but to the whole OpenWrt project. Our intention is to update all
target to latest kernel and than make a new release soon, so the
kernel dump is a the first step in this direction.
Could this person gives details on what the plans are for the AT91 target?
Maybe we could share some thoughts and ideas.
I have access to the stamp9G20 and the portux9G20, so i can tests these two
boards. I really would like to help and contribute, but it would be nice if
the plans are known in advance.
The spare time is always a rare resource and the facto I didn't
develop this target until a couple of days ago.
So I start to read all the missing think that was remarked on mailing
list (few topics in a years, not exactly a famous platform) and also
your interest to stampg20 was noted.
The Stampg20 is now merged with the old netus subtaget inside the new
9g20 subtarget, the reorganization was mainly done to avoid the
increase of subtarget that are all using the same SoC.
https://dev.openwrt.org/browser/trunk/target/linux/at91/9g20/config-default
The default image generated by buildbot should work on StampG20, and
other G20 card enabled as well.
Hope that this can help.
Hi Claudio,
I'm updating the few patches I have here, and will send them soon to the
list.
There are 2 patches I would like your opinion.
The first one concerns the uImage generation. In the target's Makefile
the uImage is generated without compression, and then only for the
flexibility board is copied over the rootfs.
Other boards use a separate MTD partition for the kernel, and needs a
compressed kernel.
So this first patch move the mkimage call inside the "if flexibility"
and fallback for the other boards to the already built (and compressed)
image found in the kernel built tree.
Do you have any issue with that?
The second one concerns the SPI CS pins, there's a patch that "hijack"
the SPI pins definition (805-free_some_portc_pins.patch), in my case
this is not wanted. Do you know which board requires this? Is this patch
still needed since you removed some boards recently?
Regards,
Chris
Claudio
PS I just discover that some empty directories are not deleted by the
git dcommit and it is really frustating to deal with svn-git
inteoperability issues... some additional cleanup will follow.
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
--
Christian Gagneraud,
Embedded systems engineer.
Techworks Marine
1 Harbour road
Dun Laoghaire
Co. Dublin
Ireland
Tel: + 353 (0) 1 236 5990
Web: http://www.techworks.ie/
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
