08:30:55 <xruxa> #startmeeting
08:30:55 <linarobot> Meeting started Tue Dec  8 08:30:55 2015 UTC.  The chair is xruxa. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:30:55 <linarobot> Useful Commands: #action #agreed #help #info #idea #link #topic.
08:31:02 <xruxa> #topic Status update round
08:31:11 <xruxa> lets see who is here today
08:31:20 <xruxa> sunao, would you go first?
08:31:20 <Qian> hi all
08:31:49 <yspan> hi
08:32:21 <xruxa> no sign of sunao
08:32:27 <xruxa> yspan, would you go first then?
08:32:55 <sunao> sorry, I am late
08:33:00 <yspan> What did you do?
08:33:00 <yspan> none.
08:33:00 <yspan> What are you going to do?
08:33:00 <yspan> Will try profiling libpng again to see if any functions that have chances of splitting
08:33:00 <yspan> What problems do you have?
08:33:01 <yspan> none
08:33:02 <sunao> yspan, pls go first
08:33:38 <xruxa> yspan, still working on the libpng?
08:33:58 <yspan> xruxa, yes.
08:34:06 <xruxa> yspan, thanks.
08:34:10 <xruxa> sunao, you ready?
08:34:14 <sunao> yes
08:34:22 <sunao> What did you do?
08:34:24 <sunao> 1。 try to port ptmalloc to bionic c, fix the build error. not finish yet.
08:34:25 <sunao> what are you going to do?
08:34:26 <sunao> 1.  port ptmalloc to Android.
08:34:28 <sunao> what problems do you have?.
08:34:30 <sunao> 1.  None.
08:34:47 <yspan> xruxa, splitting encoding part is done. Now looking for other functions that can be segmented
08:35:29 <xruxa> sunao, sounds straightforward, thanks. Looks like bit bigger task than we estimated originally, stay on it please.
08:35:46 <xruxa> Qian, please go next
08:36:36 <sunao> ls
08:36:41 <Qian> ok
08:36:46 <sunao> sorry, miss typing
08:36:50 <Qian> What did you do?
08:36:50 <Qian> Investigated LMG-914, And finish writing the script
08:36:50 <Qian> Put the results on github: https://github.com/dalvikfrank/docs/tree/qian/linaro/staticlink
08:36:52 <Qian> What are you going to do?
08:36:54 <Qian> Another work
08:36:56 <Qian> What problems do you have?
08:36:58 <Qian> one company colleague wants to know sth about LMG-kernel doing, where can i find
08:37:00 <Qian> the link page?
08:37:19 <xruxa> Qian, so you are done with the LMG-914? Did Bero review and accept that already?
08:37:40 <Qian> not yes, i put on github right now
08:37:45 <Qian> not yet
08:38:02 <xruxa> Qian, ok, lets keep it as Ongoing. You want to start working on something else?
08:38:23 <Qian> let me check for a while
08:38:58 <Qian> i will notify you later, and change the status
08:39:01 <xruxa> Qian, if you find some tasks that you like let me and Bero know and take it.
08:39:10 <Qian> i will
08:40:02 <xruxa> no sign of Luther
08:40:24 <xruxa> zhizhoutian__, would you go next?
08:40:41 <zhizhoutian__> xruxa: wait a moment pls
08:40:48 <zhizhoutian__> xruxa: i am writing it...;)
08:41:01 <xruxa> ok
08:41:21 <xruxa> I see YongQin editting the doc but not here on IRC, skipping him
08:41:34 <Qian> xruxa: where can i find the lmg-kernel are doing ?
08:41:41 <xruxa> no Eric here
08:42:18 <xruxa> Qian, I do not know of a webpage, but most of the activities are under https://projects.linaro.org/browse/LMG-395 in JIRA
08:42:34 <Qian> xruxa: thanks
08:42:38 <xruxa> wuhai, are you here?
08:43:35 <bero2> Qian: next step is identifying where switching to dynamic linking helps...
08:43:40 <wuhai> What did you do?
08:43:40 <wuhai> 1.Fixed compile error when we build user version after applied the patch of linaro,but build userdug version is OK.
08:43:40 <wuhai> 2.Analyzes the patch about new compiler of linaro.that for porting to T-shark baseline better.
08:43:40 <wuhai> what are you going to do?
08:43:40 <wuhai> Verify the linaro patch effect on user version T-Shark baseline.
08:43:40 <wuhai> what problems do you have?
08:43:40 <wuhai> None until right now
08:44:10 <Qian> bero2: you mean in details?
08:45:07 <xruxa> bero2, wuhai - do I get it right wuhai is now only validating our patches on member HW and not working on LMG-963 or other JIRA tasks anymore?
08:45:25 <bero2> Qian: yes - no point in switching to dynamic if a library is used only once etc.
08:46:21 <Qian> bero2: i have done that, if only once, it won't be created
08:46:21 * zhizhoutian__ is ok now
08:46:32 <zhizhoutian__> xruxa: this is my work log
08:46:35 <zhizhoutian__> What did you do?
08:46:35 <zhizhoutian__> 1. investigate zramfs
08:46:35 <zhizhoutian__> go on tracing code with virtual machine.
08:46:35 <zhizhoutian__> 2. work with yfw、liuyq and tinno/sprd guys on apply latest lianro patches
08:46:35 <zhizhoutian__> save ~30M memory on userdebug version and ~20M memory on user version
08:47:40 <Qian> bero2: and the prefix of xxstatic is the count used
08:49:16 <Qian> bero2: also i have listed the test-case and module counts reference
08:49:42 <xruxa> zhizhoutian__, thanks.
08:50:27 <xruxa> wangjian, are you here with status report?
08:50:35 <wangjian> xruxa:
08:50:39 <wangjian> xruxa: yes
08:50:53 <wangjian> What did you do?
08:50:53 <wangjian> 1.Has compared the memory usage on userdebug version on T-Shark device.
08:50:53 <wangjian> 2.Fix the error in user build on T-Shark code.
08:50:53 <wangjian> 3.Analysis of the details of the patches in marshmallow-gcc5-patchset
08:50:53 <wangjian> What are you going to do?
08:50:54 <wangjian> 1.Compare the memory usage on user version on T-Shark device.
08:50:54 <wangjian> 2.Remove dlmalloc patches in marshmallow-gcc5-patchset then build and compare memory usage.
08:50:55 <wangjian> What problems do you have?
08:50:55 <wangjian> none
08:51:35 <xruxa> wangjian, thanks. I guess you and wuhai are now mostly validating our improvements on t-shark and not working on any JIRA tasks, right?
08:51:49 <wangjian> xruxa: yes
08:51:55 <Qian> bero2: Please see latest comment, the linkage of github
08:52:05 <amit_khare> bero2, xruxa: Sorry to interrupt in between. Milosz suggested that we should have keyword for LCR bugs, It will be easier to search LCR bugs and will help in our test report automation( we have template ready)
08:52:19 <xruxa> so the work on Launcher and Services is not really in Progress but abandoned
08:52:25 <xruxa> amit_khare, there is LCR keyword alreadt
08:53:17 <wangjian> xruxa: we are work with liuyq ,zhizhoutian__ and YinFengwei to verify the linaro patch effect on T-shark
08:54:24 <xruxa> wangjian, wuhai - ack, I will update JIRA for the same effect
08:54:42 <xruxa> liuyq, do you wnat to go next?
08:54:55 <liuyq> What did you do?
08:54:56 <liuyq> Investigated problems about the linaro ram patch for TShark user build
08:54:56 <liuyq> Created benchmark test scripts for AndebenchPro2015
08:54:56 <liuyq> Verified linaro gcc5.3 and gcc5.2 with member hikey and juno builds.(gcc5.3 can build for both builds)
08:54:56 <liuyq> What are you going to do?
08:54:57 <liuyq> upload patches for TShark user build to gerrit for review
08:55:01 <liuyq> Following up status/jobs with update to linaro gcc5.3 and new android version tag(6.0.1_r3)
08:55:03 <liuyq> disable the triggers for fvp and tc2 builds, add triggers for r-lcr hikey builds
08:55:05 <liuyq> What problems do you have?
08:55:07 <liuyq> None
08:55:29 <amit_khare> xruxa: thanks.
08:56:02 <liuyq> xruxa, there might be something I missed about LCR, if you find, please let me know
08:57:13 <xruxa> liuyq, should we remove the older andebench (it does not work anyway in LAVA) and try to get the 2015 one working?
08:58:10 <xruxa> I do not see Yu Xing here on chat but he is adding report on the doc.
08:58:51 <liuyq> xruxa, I will check if nkambo can help to verify the AndebenchPro2015 scripts with juno, if that's no problem, I will update the config file
08:59:34 <xruxa> thanks liuyq
08:59:45 <xruxa> xavierhsu, please go next
09:00:00 <xavierhsu> ok
09:00:14 <xavierhsu> What did you do?I trace the code.
09:00:27 <liuyq> xruxa, for Hikey, will wait reply from Milosz about the tests on Hikey 2GB version, and if he has any idea about turn the vc.py on Hikey 1GB ram version
09:00:35 <xavierhsu> What are you going to do? 1. I try to understand the jpeg encode's problem.     2. The problem is the upper function using "while", so I guess that I can not call "dlopen" and "dlclose" every time when the upper function call the function "jpeg_write_scanlines".
09:00:47 <xavierhsu> What problems do you have?Debug now
09:01:18 <xruxa> xavierhsu, ok, thanks.
09:02:02 <xruxa> did I forget about someone? eric____ ? Luther? Yu Xing, Ming Ling ?
09:04:00 <xruxa> ok, thank you everyone for joining, see you tomorrow
09:04:42 <xruxa> #info wangjian and wuhai are now not working on JIRA tasks and are validating out patches on T-Sharks.
09:04:48 <xruxa> #endmeeting