linaro-android June 2012

linaro-android@lists.linaro.org

38 participants
62 discussions

Fwd: A question about Android build system.

by Kejun Zhou

---------- Forwarded message ---------- From: Kejun Zhou <kejun.zhou(a)linaro.org> Date: 11 June 2012 15:57 Subject: A question about Android build system. To: Mathieu Poirier <mathieu.poirier(a)linaro.org>, Bernhard Rosenkränzer < Bernhard.Rosenkranzer(a)linaro.org> Cc: Zach Pfeffer <zach.pfeffer(a)linaro.org> Hi, I have a question about adding pm-qa into Android. Could you give me some help? There are many scripts in pm-qa. Each script is independent test case. My question is how to make the Android build the pm-qa automatically without a modification of target/product/core.mk. Take the cpuidle which is one of pm-qa test modules. (1) There are cpuidle_01.sh, cpuidle_02.sh, cpuidle_03.sh and cpuidle_killer.c in pm-qa/cpuidle/ folder. (2) The Android.mk lists cpuidle_01.sh, cpuidle_02.sh, cpuidle_03.sh and cpuidle_killer as independent modules. ~~~~~cpuidle Android.mk~~~~~~~~~ include $(call all-subdir-makefiles) LOCAL_PATH:= $(call my-dir) module_name = cpuidle define $(module_name)_add_executable include $(CLEAR_VARS) LOCAL_MODULE_TAGS := optional LOCAL_MODULE_CLASS := tests LOCAL_MODULE := $1.sh LOCAL_SRC_FILES := $1.sh LOCAL_MODULE_PATH := $(TARGET_OUT_OPTIONAL_EXECUTABLES)/pm-qa/$(module_name) include $(BUILD_PREBUILT) endef test_num := 01 02 03 $(foreach item,$(test_num),$(eval $(call $(module_name)_add_executable, $(module_name)_$(item)))) include $(CLEAR_VARS) LOCAL_MODULE := cpuidle_killer LOCAL_SRC_FILES:= cpuidle_killer.c LOCAL_STATIC_LIBRARIES := libcutils libc LOCAL_MODULE_TAGS := tests LOCAL_MODULE_PATH := $(TARGET_OUT_OPTIONAL_EXECUTABLES)/pm-qa/$(module_name) include $(BUILD_EXECUTABLE) (3) Is it possible only add the pm-qa/cpuide folder under (Android_platform)/external/pm-qa, and the Android build system will build the pm-qa/cpuidle automatically? BRs, Kejun

12 years, 6 months

Holiday (15th)

by Zygmunt Krynicki

Hey everyone. I'm not working today, taking holiday on request. I'll be swapping this with tomorrow to catch up. Talk to you on Monday. Thanks ZK -- Zygmunt Krynicki Linaro Validation Team s/Validation/Android/

12 years, 6 months

panda build with MM enablement

by Vishal Bhoj

Hi, We now have a panda build based on AOSP kernel with complete enablement (Wif,BT,Audio,Graphics,MM) . https://android-build.linaro.org/builds/~linaro-android/panda-ics-gcc47-oma… It seems to be stable on 4430 but I am seeing issues on 4460 (pandaboard ES) when it is mounting system partitions during bootup sometimes , adding rootdelay=5 in bootargs did help some times . I would like invite people to try the build and let me know if they are facing similar issue on 4460. Regards, Vishal

12 years, 6 months

Seeded Builds for Linaro Android Toolchain

by Milo Casagrande

Hello Android Team, the Infrastructure Team has been working on activating seeded builds for the Linaro Android Toolchain: https://android-build.linaro.org/builds/~linaro-android/toolchain-trunk/ The new builds, starting from today, will use the new seeded repository, resulting in slightly faster build times. If you are experiencing any problems with this new configuration, please submit a bug report, contact the Infrastructure team, or ping me (Gwaihir) and Deepti (deepti) on IRC as we are the maintenance contacts for this month. PS: please keep me CCed in replies as I do not receive Android list traffic. Have a great day. -- Milo Casagrande Infrastructure Engineer Linaro.org <www.linaro.org> │ Open source software for ARM SoCs

12 years, 6 months

Panda TI tracking crashes with powervr stuff, kernel still alive

by Zygmunt Krynicki

Hey everyone. I've got a nasty bug (nasty for me as it hampers my android bootstrapping) that is easily reproducible on my pandaboard. I've reported it as https://bugs.launchpad.net/linaro-android/+bug/1012674/ with both the manifest file and the build log. If anyone wants to try my binaries I can provide that as well. -- Zygmunt Krynicki Linaro Validation Team

12 years, 6 months

android-build switched to EBS-based build slaves

by Paul Sokolovsky

Hello Android team, Infrastructure team currently works on switching to using custom AMIs for build slaves to increase reliability, and as first step, we're switching from instance store AMIs to EBS AMIs for the builds slaves, because EBS AMIs are more maintainable. This is transparent change, and was tested yesterday and overnight. So, this is mostly a heads-up so you were aware of the change. If you see any issues, please submit a bug or ping me or our maintenance ("on-call") team for this month - Deepti and Milo (deepti and Gwaihir on IRC). Thanks, Paul 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

12 years, 6 months

Connectivity manager tests braindump

by Zygmunt Krynicki

Hey everyone. I'm trying to figure out a way to get connectivity manager tests enabled and pushed to LAVA. I read the code briefly (please correct any misconceptions I may have) and I see a few steps that need to be taken to achieve that. 1) We need to build them into our images. That's the obvious bit. It may be done already, I'm still digging through the code and the build system. 2) We need to script them so that they can execute given the correct environment. This part feels rather simple as there are explicit ADB commands in the comments throughout that code. I will try to build a comprehensive list of tests with notes and update you as I go. Still is should be relatively easy to script the execution 3) We need to work with the LAVA team to prepare a proper environment. So far this seems to be the biggest technical challenge. From my shallow understanding of the code we'd have to setup a few wireless access points with various configurations (or alternate the configuration on a single access point but that seems harder and more error prone) and see if we can get repeatable outcome from those connection tests. The thing that worries me is the susceptibility to background noise / other transmissions / etc. It seems that most "interesting" tests are related to wireless communication. I wonder what is your opinion on this. Should we just enable all the tests that we technically can and ignore the fact that our test environment is not isolated? Thanks for the feedback ZK -- Zygmunt Krynicki Linaro Validation Team s/Validation/Android/

12 years, 6 months

Fwd: dhcp fails on wlan0 and ethernet with android 4.0

by Swaraj

Hi, Can anyone suggest me, why dhcp fails on ethernet and on wifi when dhcp service is started from init.rc. But its working fine when dhcpcd -B wlan0/eth0 command is given manually. Dhcp is not starting during start up and when started manually it works. D/wpa_supplicant( 1753): State: ASSOCIATING -> ASSOCIATED I/wpa_supplicant( 1753): CTRL-EVENT-STATE-CHANGE id=1 state=7 BSSID=00:00:00:00:00:00 I/wpa_supplicant( 1753): CTRL-EVENT-CONNECTED - Connection to 00:18:f8:51:45:3f completed (reauth) I//system/bin/dhcpcd( 1778): dhcpcd[1779]: Android requires an interface I//system/bin/dhcpcd( 1778): dhcpcd[1779]: stop_control: No such file or directory I/logwrapper( 1778): /system/bin/dhcpcd terminated by exit(1) D/Tethering( 1155): MasterInitialState.processMessage what=3 D/Tethering( 1155): MasterInitialState.processMessage what=3 E/DhcpStateMachine( 1155): DHCP failed on wlan0: Timed out waiting for dhcpcd to start I/ethernet( 1146): ==>android_net_ethernet_initEthernetNative I/ethernet( 1146): interface eth0:2 found V/EthernetStateTracker( 1146): Starts... I/ethernet( 1146): interface sit0:3 found E/ethernet( 1146): android_net_ethernet_initEthernetNative exited with success I/ethernet( 1146): User ask for device name on 0, list:7BC7F8, total:2 I/ethernet( 1146): User ask for device name on 1, list:7BC7F8, total:2 V/EthernetStateTracker( 1146): Successed V/EthernetStateTracker( 1146): start to monitor the ethernet devices I/EthernetManager( 1146): Init Ethernet Manager, service: com.android.server.EthernetService@4129e608 V/EthernetService( 1146): device 0 name sit0 V/EthernetService( 1146): device 1 name eth0 E/EthernetService( 1146): EthernetManager.ETHERNET_STATE_ENABLED E/MobileDataStateTracker( 1146): default: Ignoring feature request because could not acquire PhoneService E/MobileDataStateTracker( 1146): default: Could not enable APN type "default" D/NetworkManagementService( 1146): Registering observer D/NetworkManagementService( 1146): Registering observer I/WifiService( 1146): WifiService starting up with Wi-Fi enabled E/WifiStateMachine( 1146): Failed to reload STA firmware java.lang.IllegalStateException: Error communicating to native daemon I/ethernet( 1146): User ask for device name on 0, list:7BC7F8, total:2 I/ethernet( 1146): User ask for device name on 1, list:7BC7F8, total:2 I/EthernetService( 1146): Ethernet dev enabled 2 V/EthernetService( 1146): device 0 name sit0 V/EthernetService( 1146): device 1 name eth0 D/NetUtils( 1146): android_net_utils_resetConnections in env=0x7d8df8 clazz=0x2f900001 iface=sit0 mask=0x3 V/EthernetStateMachine( 1146): start to monitor the ethernet devices E/EthernetStateMachine( 1146): startMonitoring resetInterface() I/EthernetStateMachine( 1146): reset device sit0 W/CommandListener( 1068): Failed to retrieve HW addr for wlan0 (No such device) D/CommandListener( 1068): Setting iface cfg D/NetworkManagementService( 1146): rsp<213 00:00:00:00:00:00 0.0.0.0 0 [down]> D/NetworkManagementService( 1146): flags<[down]> E/WifiStateMachine( 1146): Unable to change interface settings: java.lang.IllegalStateException: Unable to communicate with native daemon to interface setcfg - com.android.server.NativeDaemonConnectorException: Cmd {interface setcfg wlan0 0.0.0.0 0 [down]} failed with code 400 : {Failed to set address (No such device)} E/WifiHW ( 1146): Cannot open "/system/etc/wifi/wpa_supplicant.conf": No such file or directory E/WifiHW ( 1146): Wi-Fi will not be enabled E/WifiStateMachine( 1146): Failed to start supplicant! [ 49.792060] init: no such service 'dhcpcd_sit0' W/EthernetStateMachine( 1146): Could not stop DHCP I/EthernetSta[ 54.808940] init: no such service 'dhcpcd_sit0:-h android-49740c444506a674 sit0' teMachine( 1146): trigger dhcp for device sit0 I/SystemServer( 1146): Throttle Service D/EthernetStateMachine( 1146): DhcpHandler: DHCP request started I/SystemServer( 1146): Mount Service E/EthernetStateMachine( 1146): DhcpHandler: DHCP request failed: Timed out waiting for dhcpcd to start I/Process ( 1146): Sending signal. PID: 1307 SIG: 3 *******init.rc********* chmod 0770 /system/etc/wifi chmod 0660 /system/etc/wifi/wpa_supplicant.conf chmod 0777 /system/etc/dhcpd/dhcpd.conf chown dhcp dhcp /system/etc/dhcpcd chown dhcp dhcp /system/etc/dhcpcd/dhcpcd.conf chmod 0775 /system/etc/dhcpcd/dhcpcd-run-hooks mkdir /data/local 0771 shell shell chown dhcp dhcp /system/etc/dhcpcd/dhcpcd-run-hooks chown dhcp dhcp /system/etc/dhcpcd/dhcpcd-hooks chown wifi wifi /system/etc/wifi/wpa_supplicant.conf #wpa_supplicant control socket for android wifi.c (android private socket) mkdir /data/misc/wifi 0770 wifi wifi mkdir /data/misc/wifi/sockets 0770 wifi wifi chmod 0770 /data/misc/wifi chmod 0770 /data/misc/wifi/wpa_supplicant.conf chown wifi wifi /data/misc/wifi chown wifi wifi /data/misc/wifi/wpa_supplicant.conf chown wifi wifi /system/release/wifi_settings.cfg mkdir /data/system/wpa_supplicant 0777 wifi wifi chmod 0777 /data/system/wpa_supplicant chown wifi wifi /data/system/wpa_supplicant # Setup paths used for socket communication with the dhcp daemon(dhcpd) mkdir /data/misc/dhcpcd 0770 dhcp dhcp chown dhcp dhcp /data/misc/dhcpcd mkdir /data/misc/dhcp 0770 dhcp dhcp chown dhcp dhcp /data/misc/dhcp service dhcpcd_eth0 /system/bin/dhcpcd -dABKL class main group dhcp disabled oneshot service dhcpcd_wlan0 /system/bin/logwrapper /system/bin/dhcpcd -dd -B wlan0 disabled oneshot Thanks Swaraj

12 years, 6 months

initial android monkeyrunner tests arrived

by Alexander Sack

Hi, I am very happy to see all those monkeyrunner tests to suddenly show up on the panda side. Thanks for everybody who worked on making those happen! https://android-build.linaro.org/builds/~linaro-android/panda-ics-gcc47-til… Now I am curious if and how we can get a description on what they do? What's the idea? I felt that having them documented on our https://wiki.linaro.org/Platform/QA/TestCases/Android page in a new "automation" section would be one option to keep all our tests documented in a single place. Other ideas? Also, in this build I see two fail for that build. Any idea what's going on? -- Alexander Sack Technical Director, Linaro Platform Teams http://www.linaro.org | Open source software for ARM SoCs http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

12 years, 6 months

tip build mmtest failures in lava

by Alexander Sack

Hi, looking at origen and snowball lava results from our tip builds I realized that mmtest's are failing: origen: https://android-build.linaro.org/builds/~linaro-android/origen-ics-gcc47-sa… -> http://validation.linaro.org/lava-server/dashboard/streams/private/team/lin… snowball: https://android-build.linaro.org/builds/~linaro-android/snowball-ics-gcc46-… -> http://validation.linaro.org/lava-server/dashboard/streams/private/team/lin… Any idea why we have those failures? -- Alexander Sack Technical Director, Linaro Platform Teams http://www.linaro.org | Open source software for ARM SoCs http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

12 years, 6 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

linaro-android June 2012