Linaro-android

linaro-android@lists.linaro.org

1073 discussions

Non-existing kernel revision for Linaro 12.07 Android on Pandaboard (repo sync fails)

by Ilan Smith

Hi, When trying to checkout the kernel used for Linaro 12.07 on Pandaboard for Android, the required kernel revision does no longer exist in the remote git repository (it was there up till about ~2 weeks ago). Here is what I'm doing, based on the instructions in the Linaro 12.07 build for Android/Pandaboard<https://android-build.linaro.org/builds/~linaro-android/panda-ics-gcc47-til…>page: 1. Checkout branch *linaro-android-12.07-release* from *git:// android.git.linaro.org/platform/manifest.git* 2. Clone kernel from its location as stated in* tracking-panda.xml*: *git clone git://android.git.linaro.org/kernel/panda kernel* 3. Checkout kernel revision stated in *tracking-panda.xml*: *git checkout -b my_branch 1010fcd033e01ead0d43d61d30b341b0d4c68d49* I get error: *fatal: reference is not a tree: 1010fcd033e01ead0d43d61d30b341b0d4c68d49* Clearly, repo sync fails as well. What change needs to be done to the tracking-panda.xml manifest / kernel revision to checkout a correct revision?

13 years

12.09 Linaro Android Platform team plan posted

by Zach Pfeffer

Hey everyone! The 12.09 plan has been posted to: https://launchpad.net/linaro-android/+milestone/12.09 Have a look. Comments are welcome. There's a little bit of everything from benchmarking to preliminary Android 64-bit work. -- Zach Pfeffer Android Platform Team Lead, Linaro Platform Teams Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

13 years

Problem with panda images (unable to mount /systrem)

by Zygmunt Krynicki

Hey everyone. I'm having problems booting current panda JB. Any card I've tried fails to mount /system. Have you seen anything like that lately? Thanks ZK PS: I'll be back, shortly, with a new email account ;-) -- Zygmunt Krynicki Linaro Validation Team s/Validation/Android/

13 years

MediaScanner issue with JB on panda

by BELLONI Alexandre

Hi, I tried build #46 of panda-jb-gcc47-tilt-stable-blob. There is a small issue with sdcard mounting/mediascanning. Basically, the sdcard is mounted on /mnt/sdcard and MediaScanner is trying to scan /storage/sdcard0. To solve that, I created a symbolic link from /storage/sdcard0 to /mnt/sdcard in the initramfs. I really believe there is a much cleaner solution but that is a good workaround for me for now. Is that something you are willing to fix ? Regards, -- Alexandre Belloni Linux/Android Technical Leader Direct Phone : +33 4 26 49 25 08 Adeneo Embedded Adetel Group 4, ch. du Ruisseau - 69134 Ecully, France - Phone: +33 4 72 18 08 40 www.adeneo-embedded.com

13 years

About the SGX version of AOSP kernel source for panda

by YongQin Liu

Hi, All I just did a try to build the AOSP for panda locally, and found the version of the latest SGX driver binary google provides mismatches with the version in kernel/omap(android-omap-panda-3.0 branch). Below is the detail information for each: version in kernel: shell@android:/ $ cat /proc/pvr/version Version CustomerGoogle_Android_ogles1_ogles2_GPL sgxddk 18 1.8@905891 (release) omap4430_android System Version String: None shell@android:/ $ can also be checked via the file kernel/drivers/gpu/pvr/pvrversion.h The version of the SX binaries google provides: 14:10:01 liuyq:android$ strings vendor/imgtec/panda/proprietary/libGLESv2_POWERVR_SGX540_120.so |grep build OpenGL ES 2.0 build 1.8@785978 OpenGL ES GLSL ES 1.00 build 1.8@785978 14:10:26 liuyq:android$ And this mismatch will cause the the following error: PVR_K:(Error): BridgedDispatchKM: Driver initialisation not completed yet. [4807, drivers/gpu/pvr/bridged_pvr_bridge.c] which will make the monitor not work. Just for your information. -- Thanks, Yongqin Liu --------------------------------------------------------------- #mailing list linaro-android(a)lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-android linaro-validation(a)lists.linaro.org http://lists.linaro.org/pipermail/linaro-validation

13 years

Help: porting on to Android (shared memory calls)

by Naresh Kamboju

Hi, Here is a file massive_intr.c attached. which use shared memory call like (shm_open shm_close). I would like to port this file to Android. How can i do it? I have seen some links talks about share memory implementation on android, since i am new to Android not sure on the porting. http://www.androidenea.com/2010/03/share-memory-using-ashmem-and-binder-in.… http://www.netmite.com/android/mydroid/frameworks/base/libs/utils/IMemory.c… http://gitorious.org/android-eeepc/base/blobs/6340d73153190021292172e5f7281… do you feel we have to change file from .c to .cpp then only we can use shared memory in Android? please provide me the information on how to port this. Best regards Naresh Kamboju

13 years

Problem with NFS ICS booting on Pandaboard ES.

by K K

I am trying to boot ICS on Pandaboard ES over NFS. By default the linaro-android build is configured to boot images from mmc and as such, commands in 'init.rc' are tailored to mount the mmc card. This clearly needs to be modify in order to boot from NFS. I extracted and modified "uInitrd" image in the boot directory by commenting <mount ext4 /dev/block/mmcblk0p2 /system wait ro> in the "init.partitions.rc" file using following commands dd if=boot/uInitrd of=initrd.cpio.gz bs=1 skip=64 sudo gunzip -c ../initrd.cpio.gz | cpio -i sudo find . | cpio -o -H newc | gzip > ../newinitrd.cpio.gz mkimage -A arm -O linux -T ramdisk -C gzip -a 0 -e 0 -n initramfs -d ./newinitrd.cpio.gz ./uInitrd.new cp uInitrd.new boot/uInitrd Now i copied boot files on to the boot partition of the sd card. I did Server side configuration successfully. As below: 1. Install NFS packages host-PC$ sudo apt-get install nfs-kernel-server nfs-common 2. Add this line to the /etc/exports /tmp/system *(rw,sync,no_subtree_check,no_root_squash) 3. Start the NFS service host-PC$ sudo service nfs-kernel-server restart 4. Update exports for the NFS server host-PC$ sudo exportfs -a 5. Making sure the server is running host-PC$ showmount -e Export list for host-PC: /tmp/system * 6. Copied the contents of the Android system directory into /tmp/system 7. Added a text file with name "On_host_PC.txt" to the *"/tmp/system"* to verify whether the filesystem used during the boot is NFS or not. On the board I used an SD card which has ICS which is booting with out any issues. Then Rebooted the the board and entered into Uboot prompt and entered the following commands on Uboot prompt. setenv serverip 100.100.100.64 setenv ipaddr 100.100.100.50 mmc rescan setenv initrd_high "0xffffffff" setenv fdt_high "0xffffffff" fatload mmc 0:1 0x80200000 uImage fatload mmc 0:1 0x81600000 uInitrd setenv bootargs "console=ttyO2,115200n8 rootwait ro earlyprintk fixrtc nocompcache vram=48M omapfb.vram=0:24M,1:24M mem=456M@0x80000000 mem=512M@0xA0000000 init=/init androidboot.console=ttyO2 root=/dev/nfs rw nfsroot=100.100.100.64:/tmp/system,nolock,wsize=1024,rsize=1024 ip=100.100.100.50 gw=100.100.100.64 bootserver=100.100.100.64 rootdelay=2" bootm 0x80200000 0x81600000 iam getting the following messages and booting stops. [ 5.412231] iss_register_subdev_group: Unable to register subdev ov5650 [ 5.530548] IP-Config: Guessing netmask 255.0.0.0 [ 5.533172] IP-Config: Complete: [ 5.533172] device=eth0, addr=100.100.100.50, mask=255.0.0.0, gw=255.255.255.255, [ 5.547576] host=100.100.100.50, domain=, nis-domain=(none), [ 5.547576] bootserver=255.255.255.255, rootserver=100.100.100.64, rootpath= [ 5.563232] Freeing init memory: 276K [ 5.573333] init (1): /proc/1/oom_adj is deprecated, please use /proc/1/oom_score_adj instead. [ 5.911193] init: cannot open '/initlogo.rle' [ 7.348236] smsc95xx 1-1.1:1.0: eth0: link up, 100Mbps, full-duplex, lpa 0xC9E1 [ 10.006286] EXT4-fs (mmcblk0p3): mounted filesystem with ordered data mode. Opts: (null) [ 10.045043] EXT4-fs (mmcblk0p5): mounted filesystem with ordered data mode. Opts: (null) [ 10.092987] android_usb: already disabled [ 10.097900] adb_bind_config [ 10.298828] init: service 'console' requires console [ 10.304168] init: cannot find '/system/bin/servicemanager', disabling 'servicemanager' [ 10.312652] init: cannot find '/system/bin/vold', disabling 'vold' [ 10.312652] init: cannot find '/system/bin/netd', disabling 'netd' [ 10.325958] init: cannot find '/system/bin/debuggerd', disabling 'debuggerd' [ 10.333526] init: cannot find '/system/bin/rild', disabling 'ril-daemon' [ 10.340698] init: cannot find '/system/bin/surfaceflinger', disabling 'surfaceflinger' [ 10.349151] init: cannot find '/system/bin/app_process', disabling 'zygote' [ 10.356658] init: cannot find '/system/bin/drmserver', disabling 'drm' [ 10.363677] init: cannot find '/system/bin/mediaserver', disabling 'media' [ 10.371063] init: cannot find '/system/bin/dbus-daemon', disabling 'dbus' [ 10.378295] init: cannot find '/system/bin/installd', disabling 'installd' [ 10.385711] init: cannot find '/system/etc/install-recovery.sh', disabling 'flash_recovery' [ 10.394683] init: cannot find '/system/bin/keystore', disabling 'keystore' [ 10.394744] init: cannot find '/system/bin/gatord', disabling 'gatord' [ 10.394744] init: cannot find '/system/bin/aibd', disabling 'aibd' [ 10.415649] init: cannot find '/system/bin/pvrsrvinit', disabling 'pvrsrvinit' [ 10.423370] init: cannot find '/system/bin/uim', disabling 'uim' [ 10.429718] init: property (null) has no value for writing to /sys/class/android_usb/android0/iManufacturer [ 10.429748] init: property (null) has no value for writing to /sys/class/android_usb/android0/iProduct [ 10.452087] init: service 'console' requires console Can any one help me where I am doing mistake?

13 years

An Android kernel patch appears broken - "Input: evdev - Add ioctl to block suspend while event queue is not empty."

by Jon Medhurst (Tixy)

Hi John An Android patch in the linaro-android-* branches seems to be broken in it's used of spinlocks. The patch is the one titled "Input: evdev - Add ioctl to block suspend while event queue is not empty." This is generating lockdep warning as logged in this bug https://bugs.launchpad.net/linaro-landing-team-arm/+bug/1037565 and my analysis of the problem is in this comment: https://bugs.launchpad.net/linaro-android/+bug/1037565/comments/2 Making the wakeup driver use spin_unlock_irqsave everywhere for events_lock would solve the issue, but seems like the wrong thing. A reimplementation of the evdev patch to avoid the issue might be possible, but not something I really have time for. So my suggestion would be that we should reverted the patch in the linaro-android- branches and let upstream know about the issue. I've tested that reverting doesn't appear to cause problems. This issue has only surfaced now because Jelly Bean calls the new evdev ioctls, but ICS didn't seem to. -- Tixy

13 years

Android builds timing out

by Jon Medhurst (Tixy)

The last two vexpress builds were timed out after 3 hours but I can't see anything particularly wrong in the logs, they seemed to be going fine up to the point they were aborted. Also, one of the recent Panda builds was timed out: https://android-build.linaro.org/builds/~linaro-android/panda-jb-gcc47-tilt… I notice previous successful vexpress builds have taken about 2h45m, but I can't see any recent changes which might have made pushed the builds over 3h. (I doubt newly added pm-qa tests take a significant time.) Perhaps there is some infrastructure slowness? I'm not sure what to do. -- Tixy

13 years

Adding a git to a manifest

by Zach Pfeffer

Hey Rajagopal! Sorry for not sending this mail sooner. After you go through: https://wiki.linaro.org/Platform/Android/Gerrit https://wiki.linaro.org/Platform/Android/UpdateManifestByExample You should have a pretty good idea how to include your git the the Android manifest. A couple of things: 1. We build the tip of your git, so it should always build. If things don't build or the build breaks, please be ready to fix it. 2. We can put your git under Gerrit which will allow you to review changes, etc... we also allow backdoor access for maintainers. 3. Make sure you push the manifest change to all manifests. We have many and we try and keep them aligned. 4. Please make sure everything builds locally after integrating your manifest. Feel free to add me to the review for the manifest -- Zach Pfeffer Android Platform Team Lead, Linaro Platform Teams Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

13 years

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-android