I would like to introduce myself to the group. I have recently
received an IGEPv2 board , 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
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
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) 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 .
Does this, in any way, help with the compatibility of packages
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!
I'm very happy to announce the third release (r3) of my Fedora 19 ARM
remix images for Allwinner A10, A10s, A13 and A20 based devices. This
release is based on the official Fedora 19 Final for ARM images,
with u-boot and kernel(s) from the linux-sunxi project:
New this release:
1) Fix the bad brown paper bag bug in r2 which caused it to not boot
on sun4i (A10) and sun5i (A13/A10s) devices
2) Support for the cubietruck (except for the wifi module)
3) Support for the Megafeis A08 and Mini-X with A10s
You can download it here:
It is important to read the README, the image standard comes without
u-boot pre-loaded since u-boot is board specific. The image includes
a user-friendly simple script to install the right u-boot for
your board, but if you simply xzcat the image to an sdcard, and then
boot your device with the sdcard, things will *not* work.
See the README for a list of currently supported boards.
-Many boards don't have an rtc (A10 and A20 have a builtin one),
or at least no battery backup for it, resulting in the date
+ time being wrong.
-If the date is of by more then a couple of months, "yum update"
won't work because certificate validation fails for the https
connection yum tries to make. So if yum fails to get its repodata
first check (and fix) your date
-The regular (host not otg) usb-port on A10s based boards can be a
bit quirky. It is best to plug in a hub even when using only one
device, otherwise the device may not be recognized. If this happens,
after adding a hub, often a power-cycle is needed too.
-The wifi chip on the Auxtek-T004 hdmi-stick and on the cubietruck are
And to make sure everyone reads the README, let me print it here
Fedora 19 ARM for Allwinner A10, A10s, A13 and A20 devices README
1) Insert an sdcard, note any data on the card will be destroyed!
2) Make sure the card is not mounted, run "mount" and if the card shows
up in the output umount its partitions
3) Write the img file to the card, ie as root do:
xzcat Fedora-19-a10-armhfp-r2.img.xz > /dev/mmcblk0
4) The card is not yet ready for use! Since the A10 u-boot is board
specific, the image comes without any uboot install, follow the next
steps to install the right u-boot for your board
5) Remove the card, and re-insert it. The uboot partition should get
automatically mounted, if not mount it manually,
6) As root (or through sudo) run: <uboot-part-mount>/select-board.sh, ie:
If you've dialog installed the select-board.sh script will prompt for
your board. If you don't have dialog installed, it will print the list
of supported boards. Lookup your board and re-run the script with the
shortname for your board as argument, ie:
sudo sh /run/media/hans/uboot/select-board.sh mk802
7) umount the uboot and rootfs partitions, ie:
8) Your sdcard is now ready for use
9) *Before* powering up your A10 device connect it to an hdmi or dvi monitor
10) When first booting from the sdcard inserted Fedora will automatically
reboot once, this is part of the process to resize the root partition to
fill the entire sdcard and is normal behavior.
11) After the automatic reboot Fedora will start with the initial-setup wizard:
11a) Configure networking, note:
* If you've an A10 board with wired ethernet and you want to use dhcp
you don't need to do anything.
* If you've an A20 board, your ethernet may have a random mac-address,
so if you want to configure a static ip-address and want it to stick
across reboots, go to the ethernet-tab, select the mac-address field
and delete its contents, so that the static ip address you're
configuring does not get tied to the random mac-address.
11b) Setup the time zone
11c) Set a root password
11d) Create a user
12) Log in as the just created user
13) Enjoy Fedora on your A10 device
Fedora 19 ARM for Allwinner A10 has been tested with the following devices:
* A10s-OLinuXino-MICRO (Olimex)
* A13-OLinuXino (Olimex)
* A13-OLinuXino-MICRO (Olimex)
* A20-OLinuXino-MICRO (Olimex)
* Auxtek T003 hdmi tv stick
* Auxtek T004 hdmi tv stick
* BA10 TV Box
* Cubieboard development board 1024 MB RAM
* Cubieboard2 (A20) development board
* Cubietruck development board
* Gooseberry development board
* Mele A1000G/A2000G 1024 MB RAM
* Mini-X 1024 MB RAM
* mk802 (with female mini hdmi) 512 MB RAM
* mk802 with A10s (s with a circle around it on the barcode label)
* mk802ii (with male normal hdmi) 1024 MB RAM
* r7 hdmi tv stick
* UHost U1A hdmi tv stick
* Wobo i5 TV Box
Fedora 19 ARM should also work on the following devices:
* A10 tablet sold under various names (whitelabel)
* A13 tablet sold under various names (whitelabel)
* Coby MID7042 tablet
* Coby MID8042 tablet
* Coby MID9742 tablet
* Cubieboard development board 512 MB RAM
* DNS AirTab M82 tablet
* EOMA68 A10 CPU card
* H6 netbook
* Hackberry development board
* Hyundai a7hd tablet
* iNet-97F Rev.2 (and clones) tablet
* Marsboard A10
* Megafeis A08
* Mele A1000/A2000 512 MB RAM
* Mele A3700
* Mini-X 512 MB RAM
* Mini-X with A10s soc
* mk802 (with female mini hdmi) 1024 MB RAM
* pcDuino development board
* Point of View ProTab 2 IPS 9" tablet
* Point of View ProTab 2 IPS tablet with 3g
* Sanei N90
* XZPAD700 7" tablet
Configuring the display output
Multiple video outputs at the same time are not supported. By default
hdmi output with EDID is used for all devices, except for tablets/netbooks
where the default output is the lcd.
The default hdmi output with EDID will get the native resolution of your
TV / monitor and use that. Note that in order for this to work your TV /
monitor must be connected *and turned on*, before booting your device.
The output resolution can be configured with the disp.screen0_output_mode
kernel cmdline value, which can be found in the extrargs part of uEnv.txt in
the uboot partition. The default uEnv.txt contains the following value:
This means try to use EDID and if no valid EDID info is found fallback to
The used output can be changed by adding disp.screen0_output_type=X to the
extraargs in uEnv.txt. With X being one of: 0:none; 1:lcd; 2:tv; 3:hdmi; 4:vga
Some per display type notes:
-lcd outputs: Hardcoded to the native mode, disp.screen0_output_mode is ignored
-tv: For the cvbs output disp.screen0_output_mode must be set to one of the
following: pal, pal-svideo, ntsc, ntsc-svideo, pal-m, pal-m-svideo, pal-nc,
pal-nc-svideo. Note the -svideo variants should only be used on boards with
an svideo connector, for composite out use the regular variants, ie:
-hdmi: To override the EDID detected mode, drop the "EDID:" from the
disp.screen0_output_mode value and set it to the desired mode, ie:
-vga: Does not support EDID, "EDID:" must be removed from the
disp.screen0_output_mode value otherwise it will be ignored. interlaced
progressive and refreshrate settings specified are ignored, each resolution
has hardcoded values for these. Example usage:
How to power your allwinner device
For reliable operation it is important that your allwinner device is properly
powered. Some users try to power their allwinner development boards through
the power pin on the serial port / uart connector. This is a very bad idea!
and will almost always result in unreliable operation. Instaed always power
your allwinner device over the barrel connector intended for that using,
using a quality, reliable power supply.
USB controller caveats
The OTG USB controller in host mode only supports a limited number of
devices, plugging in a hub + mouse + keyboard typically will make either
the mouse or keyboard not work. This is a hardware limitation which we
will likely not be able to work around.
On tv-sticks and top-set boxes, simply avoid the otg connector, instead
use a hub in a regular host usb connector. Note on the mini-x the otg / host
marking is not always correct. If things don't work try using the OTG
On tablets and the gooseberry unfortunately only the otg connector is
available. One solution there is using a single usb-device which is
both a keyboard and a mouse at the same time. IE the receiver for logitech
wireless desktop sets.
* The broadcom sdio wifi found in the Auxtek T004 hdmi-stick and on the
Cubietruck is not supported
Supported hardware components / features:
Fedora 19 ARM for Allwinner A10 supports the following components:
* CPU + PMU + RAM
* Serial ports
* MMC cards
* Internal NAND storage
* Framebuffer on lcd / vga / hdmi / composite video
* Sound both analog out and over hdmi
* OTG USB controller
* Both standard USB host controllers
* Wired Ethernet
* IR (untested at this time)
* SPI (as module, not supported on A20)
* "tablet" keys on olinuxino boards
* 7 and 10 inch lcd displays on olinuxino boards (requires selecting the
right config in select-board.sh
Unsupported hardware components:
The following components require various proprietary blobs to be used, and
as such are not supported in the Fedora images. The kernel drivers for them
are present (usually as modules), so if you add the necessary blobs you might
get these to work:
* Mali 400 GPU
* Cedar hardware video & audio decoding and encoding engine
* G2D 2d engine
Note that the drivers for these need some memory to be reserved at boot, and
since they are not supported by default in the Fedora images, this memory
reservation has been disabled. To reserve the memory edit /boot/uEnv.txt and
remove the kernel cmdline options which disable the memory reservation.
Differences from stock Fedora
* Since the A10 is not a very powerful soc some services which are enabled by
default on Fedora are disabled in the image, see build-image.sh for a list.
* No plymouth: we log to a serial console for debugging so no pretty splash.
Also we don't use an initrd, so removing the console=ttyS0,115200 from
the extraargs in uEnv.txt will give plymouth, but so late it hardly matters.
Rebuilding the Fedora 19 ARM for Allwinner A10 disk image
Building the Fedora 19 ARM for Allwinner A10 disk image consists of 2 steps
1) Building a uboot.tar.gz and rootfs.tar.gz "overlays", this is done
bu the build-boot-root-sh script
2) Combining uboot.tar.gz and rootfs.tar.gz with an official Fedora 19 arm img,
this combining is done by the build-image.sh script
The a10 image you downloaded is based on Fedora-XFCE-armhfp-19-1-sda.raw
These scripts are hosted here:
A copy of the exact versions of these scripts used to build this Fedora A10
image can be found in the scripts directory of the uboot partition, the
kernel config used during the build can be found here too.
If you want to exactly reproduce this image it is important to use the
scripts from the scripts dir of the uboot partition, as the scripts contain
GIT tags used during the build to checkout the exact versions to build.
The pre-conditions these scripts expect to be met, and the exact usage of
them is documented in comments in the top of each script.
It looks like the BeagleBone Black is still running at 550MHz with the
latest Fedora 20. Does anyone know what is holding it back from running
at 1GHz? Is the a uboot thing, or a kernel thing, or something else? I
saw a version of uboot referred to as making the BBB run at 1GHz, but
when I tried experimenting with that I got the same 550MHz clock speed.
On 29 Dec 2013 07:07, "Ronald" <ronald.gadget(a)gmail.com> wrote:
> what about getting a wireless router from Dlink, Netgear etc and hacking
such a device? These devices are like 50$?
Those $50 devices are generally MIPS, with 32mb of ram and a single 100mb
port, if your lucky the switch chip might do vlans.
PS please keep the discussion on list
I was wondering if we planned something specific in term of presence at
FOSDEM (Brussel 1&2 Feb) for Fedora ARM ?
Maybe by then the Allwinner remix for F20 will be available ;-) and
I would be glad to come with a few Cubietruck or Cubieboard2, that we could
install, show and give a few as gifts, but I have no idea if people plan to
be present, if we can squat part of the Fedora stand (they should have 2 tables
- K area downstairs).
Opinions ? Hans, Rich do you plan to go there ? I would love to be
able to demo KVM on ARM 32bits :-)
Daniel Veillard | Open Source and Standards, Red Hat
veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/http://veillard.com/ | virtualization library http://libvirt.org/
I didn't exhaustively check them, but IIRC (this was not far from nearly two years ago, so don't hold me to it) core libraries were among the better behaved packages. I don't recall any in glibc, for example.
I'll revisit the issue over the next few weeks as I start on another distro rebuild.
Andy Green <andy(a)warmcat.com> wrote:
>Gordan Bobic <gordan(a)bobich.net> wrote:
>>On 12/30/2013 11:54 AM, Andy Green wrote:
>>> Gordan Bobic <gordan(a)bobich.net> wrote:
>>>> On 12/30/2013 09:58 AM, Andy Green wrote:
>>>>> Gordan Bobic <gordan(a)bobich.net> wrote:
>>>>>> On 12/27/2013 04:02 PM, Richard W.M. Jones wrote:
>>>>>>> On Fri, Dec 27, 2013 at 09:53:54AM +0000, Gordan Bobic wrote:
>>>>>>>> How is transparent alignment fixup going to give you back the
>>>>>>>> performance you lose from accesses straddling cache lines?
>>>>>>> You can have structs straddling cache lines and causing
>>>>>>> problems without alignment issues, or structs being packed too
>>>>>>> together causing false sharing again w/o alignment being
>>>>>>> If alignment problems cause performance issues, then we should
>>>>>>> with those performance problems. If they don't, we shouldn't
>>>>>>> about them.
>>>>>>> ObHack: I once worked with an architecture [68k-based VME
>>>>>>> that not only faulted on unaligned access, but also on accesses
>>>>>>> wrong *size* (eg. using a short-sized read instruction instead of
>>>>>>> word-sized read instruction). Dealing with that nonsense
>>>>>>> lot of compiler-specific massaging of code and some inline
>>>>>> I'm very glad you mentioned compilers - this is in fact easily
>>>>>> at compiler level. Intel's ICC has an option to make all arrays
>>>>> No, if your code takes the approach to cast the struct pointer into
>>>> byte stream, the struct pointer itself can be unaligned.
>>>> It won't fix all cases, but it will fix a large chunk of them -
>>>> enough of them to make fixing the remainder a tractable problem.
>>> It's already tractable, you're choosing not to engage with solving it
>>I'll enumerate the instances of this next time I'm doing a RedSleeve
>>rebuild (might start this week when I resurrect my Koji farm of
>>devices). Last time I checked the number of instances logged was in the
>>hundreds - sufficiently high that I just gave up.
>Yeah but that's hundreds of instances of one bug in one package or one instance of a hundred bugs in different packages? If it's in a library it might show up in a few different processes but still be one bug.
>If it's in glibc it might show up many times in one session different ways but still be one issue.
>Did you try catching the sigbus or whatever you're getting in gdb?
>>>>> Your compiler can do nothing about that, you have to touch the
>>>> members using bytewise accessors to be compatible with SoCs that
>>>> fix up unaligned access properly.
>>>>>> structs always aligned to a boundary (up to 16 byte, IIRC). If GCC
>>>>>> to implement such a feature the problem could be made to go away
>>>>>> actually addressing the underlying cause of the problem. It might
>>>>>> bodge, but since complete fix of the underlying problem isn't
>>>>>> happen anyway, a good bodge would be a lot better than doing
>>>>> What's wrong with you sending patches to the upstream?
>>>> Nothing apart from the amount of man-months it would take to
>>> Nonsense... a few years ago I made my own cross distro for an ARM9
>>device without hardware fixup, entirely from source tarballs, and there
>>were almost no alignment issues in the major projects.
>>I did the same 18 months ago, and my experience was distinctly
>>different. Thankfully, with the kernel-level alignment fixup at least
>>building the distro was tractable.
>>> I guess they will tend to start to increase since more people are
>>using newer ARM SoC which all have hardware fixup - but you are the
>>backpressure against that by providing patches for the real problems
>>you found... at least if you care about it, you should be.
>>A fair point well made, but I don't think we entirely agree on the
>>of the problem.
Asked this some 6 months ago and nothing seem to exist then, but time
flies and all that.
So. Is there now any multi-port ethernet boards to turn an arm board
into a multiport router/firewall?
If at least there are instructions for saying taking a cubieboard
baseboard and attaching the ethernet ports to it, I still have my '60s
era Heathkit soldering iron handy...
I want to build at least 4 systems. 5 really so I have a
This weekend I've been moving various home systems to Fedora 20,
including the system that was running my Squeezebox server. There were
some problems running the vendor package, but I stumbled upon this
community version of the server:
Not only do they support Fedora, they use Fedora 19 armhfp on a
wandboard. I had no idea. Besides Pidora, do you know of other
community projects using Fedora on ARM? Would love to see a big list.
Brendan Conoboy / Red Hat, Inc. / blc(a)redhat.com