Fedora ARM 12 on IGEPv2 (Beagle Board clone)
by Matthew Wilson
Hi all,
I would like to introduce myself to the group. I have recently
received an IGEPv2 board [1], which is based on the Beagle Board, but
with wifi, bluetooth, ethernet, and more RAM. I'm still at the "wow,
it's tiny and it runs Linux" stage. I should get a bit more time over
the next month and Christmas to play around properly with it.
I'm new to embedded development, but neither new to Linux nor ARM
(writing my first ARM assembly some 15 years ago). However, for the
past 6 years I've not even built a Linux kernel, preferring to use the
default kernel in Fedora for simplicity :)
Firstly, a thank you to those involved in Fedora ARM for getting it to
this stage. If I get the time, I'd really like to contribute some
(probably small) effort to help get Fedora ARM working well on the
IGEPv2 and Beagle Board. As I progress, I'd like to know what I can
do to help.
In the meantime, I have some questions. Apologies in advance if these
seem simple.
1) There are various different kernels from different sources. I'm
used to there being a small set of "right" kernels (that is, Fedora's
idea of "right") for x86. I fully appreciate that different ARM-based
boards are quite different in capabilities (like different instruction
set variants).
a) Is there likely to be some standardised vanilla Fedora ARM kernel
source? (Or is that simply the source RPM available for Fedora?)
Then patches /could/ be offered for the more common systems (e.g.
Beagle Board & clones, SheevaPlug).
b) Would it then make sense to offer these as pre-built RPMs for common systems?
c) Is there any guidance on which version is good to use as a base?
I've seen quite different kernel versions being used (from 2.6.27 to
2.6.31).
2) I understand a little bit about the different calling conventions,
FP differences (e.g. soft FPU versus VFP), and instruction set
differences (v5 versus v7).
a) Can the kernel can be safely built with a different instruction set
targeted? (I know there are different optimisation options passed to
GCC. Apologies if this seems a bit newbie-ish.)
b) For FP-heavy programs (e.g. ogg encoding), is it possible to build
the packages with VFP/NEON but still get them to work in a soft FPU
system? I'd imagine any call to an external library would have to
somehow be defined to use a different calling standard.
3) There seem to be some missing dependencies in the packages in the
current Fedora ARM repository. For example, emacs is requiring
libotf, which doesn't seem to be there in the repository. And
likewise with the xorg-x11-font* packages needing ttmkdir. I'm
confused as to how the RPM could have been successfully built without
it. What am I missing?
4) I see there has been some discussion over unaligned data access.
(Oh, I remember that from the ARM2 days.) It seems as if the
Cortex-A8 cores allow unaligned data access when set up to do so [2].
Does this, in any way, help with the compatibility of packages
targetting Cortex-A8?
5) I've managed to get various source packages missing from the Fedora
ARM repositories to compile successfully (natively). I guess there is
a reason why there are not in the repos right now -- is that reason
down to time and priorities, or is there some blocking bugs with many
of these packages?
I look forward to being able to contribute something back into Fedora!
Kind regards,
Matthew
[1] http://www.igep-platform.com/index.php?option=com_content&view=article&id...
[2] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344j/Beih...
7 years, 8 months
Pogoplug Pro/OXNAS
by Callum Lerwick
Hi. I've been looking for a way to get a very power efficient, but
still full featured Fedora server for a while, and the recent influx
of "plug computers" finally provides the answer. Long story short, I
ended up snagging a Pogoplug Pro on sale at Micro Center for $50,
before I learned that the Pro is NOT derived from the Kirkwood
reference design, but is a completely different architecture called
OXNAS from Oxford Semiconductor/PLX. As far as my googling can tell,
the recent "black" generation of Pogoplug hardware seems to be the
only product in the world using the OXNAS 820 SoC right now. It looks
like the WD MyBook uses/used an earlier OXNAS 810 chipset.
http://www.plxtech.com/products/consumer/nas7820
http://www.linuxfordevices.com/c/a/News/PLX-Technology-NAS-7800/
It is also "cost reduced" from earlier Pogoplugs, having only 128mb
RAM and 128mb of flash. Its a bit slower at 700mhz but is somewhat
compensated for by being dual-core. On the up side, there is in fact a
SATA port hidden inside the Pogoplug Pro. It also has some interesting
features like TCP offload and supposedly an encryption engine. (but I
haven't seen any sign of support for the encryption engine in the
Cloud Engines source so far) There's also a mini-PCIe slot containing
the Pro's wireless card.
Anyway, I managed to get Fedora 13 beta3 running on it by first
installing PlugApps on it, then I just swapped in a Fedora root FS:
http://archlinuxarm.org/platforms/armv6/pogoplug-provideov3
Make sure to copy /lib/modules and /usr/local over from the PlugApps
FS. The network driver is compiled as a module, and it will not auto
load. Especially if you don't have a serial console set up, before
booting you'll need to pop in a script at
/etc/sysconfig/modules/oxnas-gmac.modules :
#!/bin/sh
exec /sbin/modprobe gmac >/dev/null 2>&1
Make sure it is chmod 755
Also the driver wants firmware for the TCP offload, and seems to have
a bug where it will just lock up the system if it can't find it.
You'll find it at /lib/modules/2.6.31.6_SMP_820/gmac_copro_firmware ,
just move it to /lib/firmware/ and you should have working ethernet.
There's some weirdness with the MAC address, the driver just sets it
to a hardwired default of "00:30:E0:00:00:00" as there does not seem
to be in-kernel infrastructure to read uboot variables. The PlugApps
install script writes the MAC into /usr/local/mac_addr and the network
scripts change it before bringing the interface up. I don't understand
why they don't just read it directly from NVRAM using
/usr/local/cloudengines/bin/blparam . But currently my Pogoplug Pro is
just using the default MAC. Setting the MAC properly should be a
simple matter of scripting...
Unfortunately the PlugApps kernel is very minimal. It does not have
any NFS server support enabled, and for some reason ext4 is compiled
as a module, so your root FS has to be ext2 or ext3.
I managed to successfully compile the Cloud Engines kernel source and
write it to flash just past the original PlugApps kernel, so the
original PlugApps kernel and even the stock Pogoplug OS can still be
booted in a pinch using the uboot console. (But I may have overwritten
a backup Pogoplug kernel...)
http://www.pogoplug.com/developers-open-source.html
If you do not have a serial console set up, do NOT mess with the
flash! If you screw up the boot sequence, there is no way to un-brick
these things without one. There is NO "reset nvram to defaults"
button.
I'd really like to run btrfs on this, but the stock kernel is 2.6.31,
and my research indicates you want at least 2.6.33 for stable btrfs.
Also, support for the wireless card was merged in 2.6.34. I have
successfully upgraded the kernel through the incredibly awkward and
time consuming (but educational) method of applying the Cloud Engines
changes to 2.6.31 in a git branch, then rebasing 2.6.32 on top of it.
So my Pogoplug Pro has been running 2.6.32 for the last few months
with no problem. I'll see about pushing forward further once I have
some time to kill. (Pretty much takes a full day for my old secondary
laptop to chew through the whole rebase...)
So now I have my Pogoplug Pro working nicely as a Samba server for my
windows machines, and a NFS4 server for my Fedora machines.
I originally wanted to run a MythTV backend on it, but that may be
pushing it a bit with only 128mb RAM. Not to mention the fun of
getting MythTV compiled for Fedora arm in the first place...
11 years, 10 months
Fedora 15 ARM hardfp Virtual FAD (Fedora Activity Day) - TODAY
by Jon Masters
Folks,
We are hosting another one of our regular Fedora 15 hardfp Virtual
Fedora Activity Day today Friday July 15th, at 14:00UTC (10:00 Eastern
Daylight Time). The purpose of this session is to co-ordinate the
bootstrap of F15 hardfp (hardware floating point).
Last week, we succeeded in reaching a point where we had yum running
natively, and after that rapidly went from just under 100 native binary
RPMs built to 793 right now! We still do not have all of the
dependencies for mock, but we hope to be there soon. We need your help,
if you have the time, hardware, and interest in doing so. Together, we
can reach a point of running mock, then rebuild all of the packages we
have so far natively within a mock, and ultimately have a Koji buildroot
sufficient to officially rebuild everything within the standard Fedora
ARM Koji infrastructure.
You can find a lot more detail here about the stages involved in bringup
(we are at stage 3), along with all the pre-reqs/bits:
https://fedoraproject.org/wiki/Architectures/ARM/Fedora15_HardFP_Bootstra...
For general bootstrapping series information:
https://fedoraproject.org/wiki/Architectures/ARM/Fedora15_HardFP_Bootstrap
Be sure you follow the instructions to use an armv7hl-YOUR_FAS_USERNAME
or group name branch on the Fedora ARM git repo so that we can track who
is doing what, and more easily back out changes if you/we discover a
problem with your setup.
Jon.
12 years, 1 month
Re: [fedora-arm] Centos rootfs
by Donald Gullett
Compiling the packages is not a problem as i am compiling a load of f14 and f15 packages now.
Chris Tyler <chris(a)tylers.info> wrote:
>On Sun, 2011-07-31 at 01:32 -0400, Donald Gullett wrote:
>> Can anyone point me in the direction of information on building a centos rootfs?
>
>You'd have to compile a substantial portion of the CentOS package
>collection for ARM, then create a rootfs based on that.
>
>> How was the fedora rootfs setup derived? Was it a posted redhat
>> document or procedure or an existing fedora installed system?
>
>It's a hackish script that runs against the ARM package repo. A
>kickstart-based solution will get written in due course, I'm sure.
>
>-Chris
>
>_______________________________________________
>arm mailing list
>arm(a)lists.fedoraproject.org
>https://admin.fedoraproject.org/mailman/listinfo/arm
>
12 years, 1 month
[F15] NFS boot [howto]
by Zoran Pericic
This is how I configured NFS root on my pandaboard. There is also
complete boot partition in [1]
1) Install grubby, u-boot and dracut-network from my repo [2]
2) Install kernel-omap. This kernel have path that allow you to set mac
address for smsc95xx with smsc95xx.macaddr kernel boot argument.
3) Change KERNEL_ARG in /etc/sysconfig/uboot for dracut network boot [3]
KERNEL_ARGS="ro vram=32M root=dhcp fixrtc console=ttyO2,115200
mem=460M@0x80000000 mem=512M@0xA0000000 omapdss.def_disp=dvi
smsc95xx.macaddr=5A:3E:83:2D:BA:41 selinux=0"
4) Add option "root-path=SERVER:PATH" to dhcp server
host pandaboard1 {
hardware ethernet 5a:3e:83:2d:ba:41;
fixed-address 10.21.38.31;
option root-path "10.21.38.4:/nfs/pandaboard1-fc15";
}
5) Run "flash-kernel --update-bootscript 2.6.38.8-36.06.fc15.armv7hl.omap"
[1] http://zpericic.fedorapeople.org/boot/
[2] http://zpericic.fedorapeople.org/RPMS/
[3] http://fedoraproject.org/wiki/Dracut/Options#Network
12 years, 1 month
About arm choice of toolchain options
by Paulo César Pereira de Andrade
Hi,
Sorry for the cross post and long email :-)
Currently I am working on a very initial state build of Mandriva for arm.
Thanks to Jeff Johnson for giving me ssh access to armv7 hosts, and
Matthew Dawkins for building several Mandriva/Unity linux armv5
packages.
What I am trying to understand now is about choice of float abi.
I understand that the IHI0042D_aapcs.pdf file I donwload says
to use vfp registers for float/double arguments, but softfp seems
too good to miss, as armv5 should be around for some time yet.
So, I have two chroots, running:
softfp# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/libexec/gcc/armv7l-mandriva-linux-gnueabi/4.6.1/lto-wrapper
Target: armv7l-mandriva-linux-gnueabi
Configured with:
/home/pcpa/bootstrap/rpmbuild/BUILD/gcc-4.6-20110722/configure
--prefix=/usr --build=i586-mandriva-linux-gnu
--host=armv7l-mandriva-linux-gnueabi
--target=armv7l-mandriva-linux-gnueabi --enable-werror=no --enable-cxx
--with-cpu=cortex-a8 --with-tune=cortex-a8 --with-arch=armv7-a
--with-float=softfp --with-fpu=vfpv3-d16 --with-abi=aapcs-linux
--enable-languages=c,c++ --enable-threads=posix --disable-libssp
--disable-libmudflap
Thread model: posix
gcc version 4.6.1 20110722 (Mandriva) (GCC)
thumb# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/libexec/gcc/armv7l-mandriva-linux-gnueabi/4.6.1/lto-wrapper
Target: armv7l-mandriva-linux-gnueabi
Configured with:
/home/pcpa/bootstrap/rpmbuild/BUILD/gcc-4.6-20110722/configure
--prefix=/usr --build=i586-mandriva-linux-gnu
--host=armv7l-mandriva-linux-gnueabi
--target=armv7l-mandriva-linux-gnueabi --enable-werror=no --enable-cxx
--with-cpu=cortex-a8 --with-tune=cortex-a8 --with-arch=armv7-a
--with-mode=thumb --with-float=hard --with-fpu=vfpv3-d16
--with-abi=aapcs-linux --enable-languages=c,c++ --enable-threads=posix
--disable-libssp --disable-libmudflap
Thread model: posix
gcc version 4.6.1 20110722 (Mandriva) (GCC)
This is unmodified upstream gcc, and using a set of bootstrap
scripts from a git branch I made on a checkout of
git clone git://fedorapeople.org/~djdelorie/bootstrap.git
Since I am still very "arm noob" :-) and just yesterday did
the thumb build to learn about thumb, so far, my impression
is that the best approach should be to use thumb+softfp.
Just so you know I am running thumb and arm builds, with
thumb using hard float and the softfp with arm instructions set:
softfp# objdump -d /usr/lib/libm.so | less
[...]
00008d30 <__ieee754_atan2>:
8d30: e3a0c000 mov ip, #0
8d34: e347cff0 movt ip, #32752 ; 0x7ff0
8d38: e92d4030 push {r4, r5, lr}
8d3c: ed2d8b10 vpush {d8-d15}
8d40: e3a05000 mov r5, #0
8d44: ec432b18 vmov d8, r2, r3
8d48: e3475ff0 movt r5, #32752 ; 0x7ff0
8d4c: e003c00c and ip, r3, ip
8d50: e15c0005 cmp ip, r5
8d54: e24dd02c sub sp, sp, #44 ; 0x2c
8d58: e1a04003 mov r4, r3
8d5c: ec410b19 vmov d9, r0, r1
8d60: e1a05002 mov r5, r2
8d64: 0a000022 beq 8df4 <__ieee754_atan2+0xc4>
[...]
thumb# objdump -d /usr/lib/libm.so | less
[...]
00007884 <__ieee754_atan2>:
7884: 2100 movs r1, #0
7886: 2000 movs r0, #0
7888: f6c7 71f0 movt r1, #32752 ; 0x7ff0
788c: ec53 2b11 vmov r2, r3, d1
7890: f6c7 70f0 movt r0, #32752 ; 0x7ff0
7894: 4019 ands r1, r3
7896: 4281 cmp r1, r0
7898: e92d 03f0 stmdb sp!, {r4, r5, r6, r7, r8, r9}
789c: ed2d 8b10 vpush {d8-d15}
78a0: 461c mov r4, r3
78a2: b08a sub sp, #40 ; 0x28
78a4: eeb0 8b41 vmov.f64 d8, d1
78a8: 4616 mov r6, r2
78aa: eeb0 9b40 vmov.f64 d9, d0
78ae: d03c beq.n 792a <__ieee754_atan2+0xa6>
[...]
I am kind of trying to figure what "The Industry" says about it,
and just checked the linaro gcc-4.6 relevant changes for me
right now, that are...
+ --with-arch=armv7-a --with-tune=cortex-a8 \
+ --with-float=$(float_abi) --with-fpu=neon \
+# check if we're building for armel or armhf
+ifeq ($(DEB_TARGET_ARCH),armhf)
+ float_abi := hard
+else ifneq (,$(filter $(DEB_TARGET_ARCH), arm armel))
+ float_abi := softfp
+endif
If I understand correctly, neon will have better support for
simd instructions right?
Either way, I used two simple benchmarks to try to sell
myself the idea of breaking compatibility with armv5 or
older binaries, but still not convinced, but, as I said, we
should use whatever "The Industry" chooses :-)
I used for benchmark http://www.tux.org/~mayer/linux/bmark.html
and http://www.linuxfordevices.com/c/a/Linux-For-Devices-Articles/Why-ARMs-EA...
and also compared with my home computer (quad)core i5 x86_64,
and attached results...
Thanks and again sorry for cross posting and long email,
Paulo
12 years, 1 month
Configuring mock
by Jon Masters
Dennis,
I suspect I just need a few hours doing something else (like sleeping), but I hit a small bump configuring mock.
I can see that you use several repos - including on i386 for group deps (cool). However, my mock refuses to recognize the Scotland armv7hl repo. I will look at it in a few hours, should be easy to fix but just annoying. If you have something I don't, like a patch to mock, let me know.
Jon.
--
Sent from my phone - message formatted and/or shortened accordingly.
12 years, 1 month
Centos rootfs
by Donald Gullett
Can anyone point me in the direction of information on building a centos rootfs?
How was the fedora rootfs setup derived? Was it a posted redhat document or procedure or an existing fedora installed system?
12 years, 1 month
