We (at OpenedHand) have been emulating ARM devices in QEMU for a while using images generated from Poky. I wondered how easy it would be to run a poky image under the Android emulator. Sadly the emulator crashes if touched by dbus and is QVGA whereas Poky’s UI (Sato) is targeted at VGA but it still roughly works. The following images show the standard emulator running a Poky built kernel and root filesystem (all entirely open source):
It gives me great pleasure to announce a new release of Poky, version 3.0 (blinky). There are lots of new features in this although the biggest change is no doubt the introduction of the new look and feel known as Sato along with the new desktop, panel and Pimlico application “Tasks”. A rough summary of the changes is:
Here are links to the release and images. See the poky website for more information including screenshots.
After much painful debugging, I finally found the cause of the segfaults I was seeing on EABI images, particularly with glibc 2.4/2.5. It turned out to be a cache flushing problem affecting StrongARM, XScale and ARMv6 processors. The bug must have been in the kernel for years and only triggered by the changed usage patterns in glibc 2.4/2.5.
I submitted a patch for the problem which has been accepted into the mainline kernel and will be in 2.6.20.
(http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=4078/1, http://www.rpsys.net/openzaurus/patches/archive/xscale_cache_workaround-r1.patch). I’d recommend anyone using the affected processors to apply this patch.
A while back I mentioned having ARM images working over NFS under the QEMU system emulator. I’m now pleased to report I have a complete toolchain working under the emulator. I’ve made a sample image and support files available so others can experiment with it. For more details see http://www.o-hand.com/~richard/qemu.html
I’m pleased to report successful booting of OpenEmbedded/Poky generated ARM images under QEMU with the root filesystem being mounted over NFS.
http://www.rpsys.net/openzaurus/qemu/ has the gory details.
My arm backtrace code has been accepted into the kernel so oprofile support for arm is complete!
The first draft of the w100fb patches have been submitted for review to linux-fbdev-devel
The c7x0 has been getting more and more sensitive to the length of time the suspend/resume button was pressed. I’ve finialy tracked this down to the suspend/resume behaviour of the keyboard driver and fixed it.