2007/1/12, Lennert Buytenhek [EMAIL PROTECTED]:
http://armel.applieddata.net/developers/linux/eabi/armel-root-fs.tar.bz2
Excellent. Works perfectly for me on real armv4t hardware with no FPU
(using the angstrom EABI kernel)
Code speed/size comparison compared to old ABI using lame to
This is with version 2.6.18.3-ixp4xx of the kernel.
Is this something others have seen?
Yes, and its damn annoying! :) Its not just SSH sessions that randomly
hang, but any socket for me. (2.6.18-3-ixp4xx).
I got my NSLU2 on Wednesday, and this locking up was constantly
happening while
On Sat, Jan 13, 2007 at 12:39:52PM +, Martin Guy wrote:
http://armel.applieddata.net/developers/linux/eabi/armel-root-fs.tar.bz2
Excellent. Works perfectly for me on real armv4t hardware with no FPU
(using the angstrom EABI kernel)
Yay. What armv4t hardware are you using?
* James Bromberger [EMAIL PROTECTED] [2007-01-13 13:12]:
Martin, I was looking to use the rest of my disk as an iSCSI target,
but while the scsi_transport_iscsi is in your kernel, the iscsi_tcp
doesn't seem to be.
I've enabled this module in SVN now.
--
Martin Michlmayr
http://www.cyrius.com/
2007/1/13, Lennert Buytenhek [EMAIL PROTECTED]:
Yay. What armv4t hardware are you using?
That's an iPAQ h1940 with samsung S3C2410, a 233MHz arm920t.
gcc support for Crunch kind of sucks.
Well, there will be loads of developers wanting it to work well now,
which is maybe what it needed...
I've fixed the endian-ness of the environment (APEX) writing code and
I've repaired the program I wrote some time ago to show the
environment from user space. I was about to add write support to this
application when I noticed that the Loader partition is filled with
zeros. Now, it looks like
Now that I have support for alternative boot modes, do you folks know
what you want to do differently for the 'safe' boot or 'maintenance'
boot?
I can implement a kernel command line, kernel source, or ramdisk
source.
Cheers.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of
I was always envisaging the debian installer having a recovery rootfs in the
flash, and being able to switch between booting to the flashfs (especially if
the disk was not attached) or booting to the disk. Perhaps that can be done by
passing a flag on the kernel command line from Apex to the
On Sun, Jan 14, 2007 at 01:19:00PM +1030, Rod Whitby wrote:
I was always envisaging the debian installer having a recovery
rootfs in the flash, and being able to switch between booting to the
flashfs (especially if the disk was not attached) or booting to the
disk. Perhaps that can be done by
On Sun, Jan 14, 2007 at 01:14:00PM +1030, Rod Whitby wrote:
Slugimage always pads with 0xff's - it must be the debian flash-kernel script
which is padding with zeros.
Oh. I'll add support for erasing the block as well, so it *can* cope
if need be, but I'd rather start with an empty region.
10 matches
Mail list logo
