08:32:14 <bero2> #startmeeting
08:32:14 <linarobot> Meeting started Thu Nov 19 08:32:14 2015 UTC.  The chair is bero2. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:32:14 <linarobot> Useful Commands: #action #agreed #help #info #idea #link #topic.
08:32:23 <bero2> hi all...
08:32:38 <xavierhsu> Xavier is here
08:32:45 <xavierhsu> Hi Bero
08:32:46 <bero2> no big news, let's go to progress updates...
08:32:49 <Qian__> hi all
08:33:11 <bero2> yspan: do you want to go first?
08:33:34 <yspan> ok
08:33:43 <yspan> What did you do?
08:33:43 <yspan> https://dev-private-review.linaro.org/#/c/410/
08:33:43 <yspan> What are you going to do?
08:33:43 <yspan> Keep finding more functions for segment. Or start LMG-941?
08:33:43 <yspan> What problems do you have?
08:33:44 <yspan> none
08:34:43 <bero2> thanks
08:34:52 <bero2> liuyq:next?
08:35:13 <liuyq> What did you do?
08:35:13 <liuyq> Created test job templates for Hikey and enabled them
08:35:13 <liuyq> Ran benchmark tests(antutu2, antutu3, benchmarkpi, Caffeinemark, linpack, quadrandpro, vellamo3) on Nexus7 arm mode.
08:35:13 <liuyq> Investigated about the failures of Hikey jobs on LAVA(adb shell hang problem)
08:35:13 <liuyq> Investigated about the failures of AndebenchPro test job on LAVA(seems problem of AndebenchPro application)
08:35:16 <liuyq> Investigated selinux avc denials of Hikey member builds
08:35:20 <liuyq> What are you going to do?
08:35:22 <liuyq> Enable Selinux for Hikey member build
08:35:24 <liuyq> Create HOWTOs for Hikey member build
08:35:26 <liuyq> Make jobs run on LAVA for Hikey
08:35:28 <liuyq> What problems do you have?
08:35:30 <liuyq> None
08:36:38 <liuyq> bero2, hikey jobs still have problem to run on LAVA:(
08:37:04 <bero2> any reason why?
08:37:16 <bero2> and anything we can do to fix it?
08:38:05 <liuyq> bero2, seems problem related adb connections. but I don't have the permission to submit jobs for hikey on LAVA, so not able to confirmation
08:38:51 <liuyq> bero2, will check with lava/lab/qa team when they are online
08:39:05 <bero2> OK...
08:39:14 <bero2> Luther: next?
08:39:31 * luther What did you do?
08:39:32 <luther> 1. A whole day's meeting
08:39:32 * luther What are you going to do?
08:39:32 <luther> 1. Look at move and forward mechanism
08:39:32 * luther What problems do you have?
08:39:32 <luther> 1. none till now
08:40:49 <bero2> ok
08:41:16 <bero2> Qian__: next?
08:41:21 <Qian__> ok
08:41:40 <Qian__> What did you do?
08:41:40 <Qian__> Talk with Bogden, I will prepare to update my changes to linaro private server
08:41:41 <Qian__> Update scripts and write HOWTO, rebuild perl-5.22.0 and verify.
08:41:42 <Qian__> What are you going to do?
08:41:44 <Qian__> Finish updating CPU2006 to server
08:41:46 <Qian__> Build perl-5.22.0 on x86 to verify accessor error.
08:41:48 <Qian__> Write test demos for replace strings.
08:41:50 <Qian__> Write script for changing hash map to have a try
08:41:52 <Qian__> What problems do you have?
08:41:54 <Qian__> Why it causes perl hash map calling failed in android
08:43:13 <Qian__> when put spec to server, others can join in
08:43:55 <bero2> thanks, weird stuff...
08:44:13 <Qian__> bero2: any goog idea?
08:45:44 <bero2> not really... Not the biggest perl expert in the world
08:45:48 <Qian__> so, it's look like problems about SPEC framework
08:46:03 <bero2> maybe tcwg guys will know more
08:46:06 <Qian__> i think we can transplant it to android
08:46:43 <Qian__> yes, bogden said they want to transplant SPEC to android
08:47:28 <bero2> maybe we should just wait until they're done
08:47:32 <Qian__> so, hopefully they can give some help
08:47:51 <bero2> or find a way to accelerate stuff on their side
08:48:02 <Qian__> i can do some tests, and find out the root cause
08:48:42 <bero2> yes, that would be good
08:48:42 <Qian__> ok, i will keep connection with bogden
08:49:18 <Qian__> bogden said he would review my updating
08:49:37 <Qian__> so, look forward to good news
08:49:57 <Qian__> that's all from my side
08:50:47 <bero2> thanks
08:51:07 <bero2> eric___: want to go next?
08:51:21 <eric____> bero2: I have to say sorry to you
08:51:54 <eric____> I made little progress since my ubuntu  system crushed after a guy deleted the libc.so.6
08:52:20 <eric____> bero2: I have to ask IT to resinstall the system today, :(
08:53:41 <eric____> the system just work, and I will start to make  the clang work, and still hope I can finish the static analysis before this weekend
08:53:48 <bero2> shit happens... no big deal, hope you didn't lose anything important
08:54:22 <eric____> no
08:55:36 <bero2> Good..
08:55:54 <bero2> wuhai: next?
08:56:08 <wuhai> OK
08:56:26 <wuhai> What did you do?
08:56:26 <wuhai> 1.Download M-LCR source code, it is still ongoing.
08:56:26 <wuhai> 2.Analysis of architectural codes android PowerManager.
08:56:26 <wuhai> What are you going to do?
08:56:26 <wuhai> 1.Determine whether some of the code to be converted into C/C++ by JNI.
08:56:26 <wuhai> 2.Analysis of architectural codes other android system services.
08:56:26 <wuhai> What problems do you have?
08:56:27 <wuhai> None
08:57:00 <bero2> ok
08:58:55 <bero2> zhizhoutian__: around?
08:59:03 <zhizhoutian__> bero2: yes
08:59:12 <zhizhoutian__> What did you do?
08:59:12 <zhizhoutian__> Investigate how many memory saved by zramfs
08:59:12 <zhizhoutian__> read some zramfs article
08:59:12 <zhizhoutian__> use costmem tool
08:59:12 <zhizhoutian__> data will compress to 1/3 ?
08:59:12 <zhizhoutian__> What are you going to do?
08:59:12 <zhizhoutian__> What problems do you have?
08:59:28 <xavierhsu> What did you do?   1. I replace the libs(libjpeg.so and libjpeg-enc.so) into official image.   2. I modify the source code because my libs can not be used.
08:59:29 <xavierhsu> What are you going to do? 1. I need to modify my source code or Android.mk.
08:59:34 <xavierhsu> Sorry = =
09:00:07 <bero2> xavierhsu: thanks, you were next anyway
09:00:08 <zhizhoutian__> bero2: i can not make sure how many memory saved, but i know data come to 1/3 after compressing.
09:00:26 <zhizhoutian__> bero2: i will go on investigating it.
09:01:06 <bero2> zhizhoutian__: thanks
09:01:17 <bero2> did I forget anyone?
09:01:21 <xavierhsu> What did you do?   1. I replace the libs(libjpeg.so and libjpeg-enc.so) into official image.   2. I modify the source code because my libs can not be used.
09:01:21 <xavierhsu> What are you going to do?  1. I need to modify my source code or Android.mk.
09:01:21 <xavierhsu> What problems do you have?  1. Now the picture can be decoded but the color is wrong.   2. The picture can be encoded but our lib(libjpeg-enc.so) that is not be called.
09:01:29 <sunao> I'm waitting
09:01:39 <wangjian> What did you do?
09:01:39 <wangjian> 1.Contacted with IT to get hackbox account.
09:01:39 <wangjian> 2.I have disable live wallpapers to test and seems that more memory will be used only if the wallpaper is a live wallpaper.So if the wallpaper is a picture,disable live wallpapers is useless for reduce memory usage.But for the low-memory devices, disable live wallpaper and remove pre-install live wallpapers is necessary.
09:01:39 <wangjian> What are you going to do?
09:01:39 <wangjian> 1.Continued contact with IT to get hackbox account.
09:01:39 <wangjian> 2.Try to find other ways to reduce launcher memory usage.
09:01:39 <wangjian> What problems do you have?
09:01:40 <wangjian> none
09:02:04 <bero2> sunao: right, sorry. want to go now?
09:02:10 <sunao> Ok
09:02:11 <bero2> wangjian: thanks
09:02:18 <sunao> What did you do?
09:02:20 <sunao> 1. analyse the memory usage data, update the reulst to LMG-942.https://projects.linaro.org/browse/LMG-942.
09:02:21 <sunao> what are you going to do?
09:02:23 <sunao> 1. continue the investigate on memory allocator.
09:02:24 <sunao> what problems do you have?
09:02:26 <sunao> 1.  None other than the hikey hdmi problem.
09:02:28 <sunao> bero2, please take a look at the comments of LMG-942 to see if something more needs to be done, if you have time.
09:02:33 <sunao> that's all
09:02:38 * mingling is here
09:02:52 <bero2> sunao: sure, will do
09:03:01 <sunao> thanks you, bero2
09:03:47 <mingling> bero2: i read this today https://projects.linaro.org/i#browse/LMG-940
09:03:48 <bero2> mingling: sorry, we're getting so many people that it's becoming harder to memorize who hasn't gone yet. Want to go now?
09:03:52 <mingling> https://01.org/zh/android-ia/user-guides/android-memory-tuning-android-5.0-and-5.1?langredirect=1
09:04:13 <mingling> bero2: does not matter:)
09:04:35 <mingling> bero2: Geng.Ren want to talk with you about task assign
09:04:42 <mingling> bero2: when you time free.
09:05:29 <bero2> mingling: sure. Any time.
09:05:47 <mingling> bero2: ok, thanks, then i will tell him:)
09:07:01 <bero2> OK, anyone else?
09:08:48 <bero2> looks like we really got everyone now ;)
09:09:07 <bero2> have a great rest of the day!
09:09:14 <bero2> #endmeeting