08:40:22 <bero> #startmeeting
08:40:22 <linarobot> Meeting started Thu Nov 12 08:40:22 2015 UTC.  The chair is bero. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:40:22 <linarobot> Useful Commands: #action #agreed #help #info #idea #link #topic.
08:40:46 <bero> so, let's get started...
08:41:12 <xavierhsu> sure.
08:41:19 <bero> Qian_: do you want to go first?
08:41:37 <Qian_> What did you do?
08:41:38 <Qian_> 1. tried to runspec on Nexust5,and meet some troubles because of the differences between 5.12.3 and 5.18.1, have solved some, but now was blocked, by:
08:41:38 <Qian_> XSLoader::load('Your::Module', $Your::Module::VERSION) at /data/local/tmp/CPU2006/bin/lib/XSLoader.pm line 25.
08:41:39 <Qian_> seems cause by Fcntl module, but now, i don’t know how to fixed it.
08:41:41 <Qian_> What are you going to do?
08:41:43 <Qian_> try to fix the issue above.
08:41:45 <Qian_> try perl built by myself
08:41:47 <Qian_> What problems do you have?
08:41:49 <Qian_> I expect there is some wrong with perl environment, anyone who can help me?
08:41:51 <Qian_> seems we are close to the result if we can solve perl issues.
08:42:42 <Qian_> some issues about perl confused me.
08:42:53 <bero> nice
08:42:53 <Qian_> anyone can give some help?
08:43:33 <bero> sure will take a look
08:43:54 <Qian_> bero, i think i put the result on the server
08:44:25 <Qian_> so it's easy to share with others
08:45:34 <Qian_> bero, how can i put the SPEC on the linaro server?
08:46:34 <bero> I'll create a safe git repository
08:47:22 <Qian_> ok, please create a separate git for android.
08:48:17 <Qian_> $runspec --config=linaro.cfg --action=build --tune=base bzip2
08:48:18 <Qian_> WARNING: linker: /data/local/tmp/CPU2006/bin/specperl: unused DT entry: type 0xf arg 0x182
08:48:18 <Qian_> XSLoader::load('Your::Module', $Your::Module::VERSION) at /data/local/tmp/CPU2006/bin/lib/XSLoader.pm line 25.
08:48:19 <Qian_> Compilation failed in require at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/Seekable.pm line 104.
08:48:21 <Qian_> BEGIN failed--compilation aborted at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/Seekable.pm line 104.
08:48:23 <Qian_> Compilation failed in require at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/File.pm line 133.
08:48:25 <Qian_> BEGIN failed--compilation aborted at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/File.pm line 133.
08:48:27 <Qian_> Compilation failed in require at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/Dir.pm line 15.
08:48:29 <Qian_> BEGIN failed--compilation aborted at /data/data/com.pdaxrom.cctools/root/cctools/lib/perl5/5.18.1/armv7l-linux/IO/Dir.pm line 15.
08:48:31 <Qian_> Compilation failed in require at /data/local/tmp/CPU2006/bin/runspec line 32.
08:48:33 <Qian_> BEGIN failed--compilation aborted at /data/local/tmp/CPU2006/bin/runspec line 32.
08:48:39 <Qian_> this is the whole error msg
08:49:25 <Qian_> that's all
08:52:44 <sunao> is bero still online?
08:53:09 <yspan> ping bero; ping  bero|2   :p
08:53:31 <xavierhsu1> ping bero
08:53:33 <xavierhsu1> @@
08:56:25 <bero|2> sorry, had some connection problems
08:56:28 <bero|2> back on the other line now
08:56:52 <xavierhsu1> xavier is here
08:57:25 <bero|2> hmm, that error is almost certainly caused by it not finding one of the perl modules it wants to use
08:57:36 <bero|2> but I don't see much of a hint on which one it is trying to use
08:57:38 <Qian_> yes, IO
08:57:44 <Qian_> and Fcntl
08:58:02 <Qian_> Maybe i have get some idea now
08:58:12 <bero|2> yes, but IO::File and IO::Seekable are there and found...
08:58:22 <bero|2> It's probably another one they want to pull in
08:58:35 <Qian_> i just get some different error, now
08:59:14 <bero|2> From looking at Seekable.pm, I think it's looking for Fcntl
08:59:41 <Qian_> yes, i will try to finger it out
08:59:48 <Qian_> next one please
08:59:51 <bero|2> Does Seekable.pm line 104 say "use Fcntl qw(SEEK_SET SEEK_CUR SEEK_END);" for you too? (I'm on perl 5.20.3, so just did a quick check there... I doubt much has changed since 5.18.1)
08:59:57 <Qian_> yes
09:00:01 <bero|2> yspan: want to go next?
09:00:21 <xavierhsu1> yspan is not stay at his site.
09:00:46 <xavierhsu1> I help him to paste the report
09:00:50 <xavierhsu1> What did you do?Bugs are fixed. Some functions cannot be replaced with stub (pngwio.c). Some functions are not need to be made stub, because they are invoked only in encoder. Now amount of memory saving increases to 32K (15%).
09:01:04 <xavierhsu1> What are you going to do?Submit to gerrit review?
09:01:13 <xavierhsu1> What problems do you have?None.
09:01:26 <Qian_> bero, i will ask you on IRC
09:01:58 <yspan> I'm back :p
09:02:02 <bero|2> Qian_: you may want to try just commenting out the "use Fcntl" line and the "@EXPORT = qw(..." line a few lines below... If nothing actually uses the Fcntl function that's an ok workaround
09:02:11 <bero|2> xavierhsu1: thanks, do you want to go next with your own stuff?
09:02:37 <xavierhsu1> Bero, yspan is back
09:02:52 <xavierhsu1> he need to talk to you first?
09:03:10 <yspan> xavierhsu1, it's ok. you already paste what I want to say
09:03:21 <xavierhsu1> ok
09:03:22 <Qian_> beroI2, I will do, it maybe cause other problems
09:03:43 <xavierhsu1> What did you do? 1. I download the official image of Hikey, it looks normal.    2. I use adb command to pull the libs of mali and then push to my local side image of Hikey, the screen still will flicker.  3. I upgrade my local side source code and rebuild the android source code now.
09:03:59 <xavierhsu1> What are you going to do? 1. I upgrade and rebuild the android source code for testing libjpeg-enc.so.
09:04:09 <xavierhsu1> What problems do you have?Until now, none.
09:04:59 <sunao> hi  xavierhsu1, can you give me the link of the official image? Does it work well?
09:05:05 <bero|2> probably some change in Mali breaking things...
09:05:06 <xavierhsu1> Bero, could I use official image and only change the libjpeg.so?
09:05:10 <bero|2> xavierhsu1: sure
09:05:47 <xavierhsu1> sumao, http://builds.96boards.org/releases/hikey/linaro/aosp/latest/
09:05:56 <sunao> thanks
09:06:06 <bero|2> sunao: do you want to go next?
09:06:10 <xavierhsu1> sunao, http://builds.96boards.org/releases/hikey/linaro/aosp/latest/
09:06:10 <sunao> Ok
09:06:51 <sunao> bero, let others go first, I have to paste it from my notebook
09:06:59 <bero|2> ok
09:07:02 <xavierhsu1> sunao, Sorry I misspell your name.
09:07:06 <bero|2> liuyq: are you there?
09:07:16 <liuyq> bero|2, yes
09:07:20 <sunao> thank you xavierhsu1
09:07:26 <bero|2> liuyq: do you want to go next?
09:07:42 <liuyq> What did you do?
09:07:42 <liuyq> Merged zlib optimization patch into MARSHMALLOW-MLCR-PATCHSET.
09:07:42 <liuyq> Merged neon-fp16 fix patch for TC2 into MARSHMALLOW-MLCR-PATCHSET
09:07:42 <liuyq> Learnt instructions for building webview.apk
09:07:43 <liuyq> What are you going to do?
09:07:45 <liuyq> Finish cts test on Nexus7 with images compile with arm mode projects(started yesterday, not finished yet)
09:07:47 <liuyq> Merge the tjbench patch into MARSHMALLOW-RLCR-PATCHSET for comparison.(seems only this patch is necessary for marshmallow reference builds)
09:07:50 <liuyq> Try to build the webview.apk for marshmallow build from source
09:07:54 <liuyq> What problems do you have?
09:07:56 <liuyq> None
09:08:12 <xavierhsu1> Bero, so I can use official image and use my libjpeg-enc.so for testing the function of encoder?
09:08:42 <bero|2> xavierhsu1: yes, just make sure you also put your libjpeg.so in -- without that, libjpeg-enc.so (obviously) won't be used
09:08:50 <liuyq> bero|2, started to learn instructions for chromium. CTS test on Nexus7 is still running
09:09:24 <xavierhsu1> Bero, Thanks. I will try it.
09:09:32 <bero|2> liuyq: good... I've seen your message about hikey on #linaro-release, do you know what TARGET_CPU_VARIANT is set to?
09:10:10 * bero|2 suspects an ARMv8 CPU variant is set to cortex-a15 or some other bogus combo
09:10:16 <bero|2> will look
09:10:43 <liuyq> bero|2, you mean we need to change the hikey device configuration?
09:12:16 <liuyq> bero|2, for hikey: TARGET_2ND_CPU_VARIANT := cortex-a15
09:12:20 <bero|2> liuyq: yes, it's what I thought -- "TARGET_2ND_CPU_VARIANT := cortex-a15", this is not really right (it's still an a53 even if it's in 32bit mode), just wasn't caught before
09:12:21 <bero|2> I'll fix it
09:12:58 <liuyq> bero|2,ok, thanks
09:14:47 <bero|2> liuyq: fix pushed, should build now (android-6.0 branch at least)
09:15:02 <bero|2> zhizhoutian_: do you want to go next?
09:15:14 <liuyq> bero|2, thanks, I will follow up the status
09:15:33 <zhizhoutian_> sorry, bero|2
09:15:42 <zhizhoutian_> bero|2: i need a little time to write
09:15:51 <sunao> I'm okay now, bero|2
09:15:51 <bero|2> ok
09:15:58 <bero|2> sunao: ok, go ahead...
09:16:04 <sunao> What did you do?
09:16:06 <sunao> 1. flash the hikey board, got the android run on it. But the HDMI has no output.
09:16:07 <sunao> what are you going to do?
09:16:09 <sunao> 1. try to bring the hikey up, and will test the memory allocator on the board.
09:16:10 <sunao> what problems do you have?
09:16:12 <sunao> 1.  hikey hdmi problem, the android has been booted to home screen, and can see adb on the computer, but  the LCD monitor has no signal.
09:16:13 <sunao> 2. need adb driver on Windows.
09:16:24 <sunao> I still have some problem with the hikey hdmi output
09:16:46 <bero|2> Did you pull in the Mali driver? It's not part of the git repository because we can't distribute it freely
09:17:11 <sunao> I am not build it from source ,but download the snapshot release of 96boards
09:17:31 <sunao> the adb and uart is okay, but hdmi still has no output
09:17:48 <bero|2> that's odd...
09:18:08 <bero|2> liuyq: you're probably more familiar with the hikey builds than me, do you have any idea what could be causing that?
09:18:41 <sunao> Is there a verified release and the bootloader? I use the uefi bootloader and latest snapshot release  of 2015-11-11
09:19:08 * zhizhoutian_ is ok now
09:19:17 <sunao> another question, I found no boot_fat.uefi.img from http://builds.96boards.org/releases/hikey/linaro/aosp/latest/
09:19:36 <bero|2> sunao: yes, I think the official release doesn't use UEFI yet
09:19:59 <bero|2> hikey used to have a fastboot based bootloader that is being deprecated
09:20:10 <sunao> then, which bootloader should I use? closed source or the uefi?
09:20:31 <bero|2> For the old release build, I think only the closed bootloader will work out of the box
09:20:50 <bero|2> for more current builds, uefi is the way ahead
09:20:51 <liuyq> bero|2, for the hikey hdmi output problem, guodong is looking at that I guess
09:21:24 <bero|2> zhizhoutian_: ok, ready when you are...
09:21:41 <sunao> hi liuyq, have you meet this problem before? which bootloader and release do you use?
09:21:41 <zhizhoutian_> Zhizhou Tian <zhizhou.tian@linaro.org>
09:21:41 <zhizhoutian_> What did you do?
09:21:41 <zhizhoutian_> 1.LMG-953	Investigate best settings for KSM to save memory
09:21:41 <zhizhoutian_> download tshark code base, build it, and flash it
09:21:41 <zhizhoutian_> 2. LMG-931	Support Big-Ed1 in Lava to install tshark devices.
09:21:42 <zhizhoutian_> support Lava’s Elsa to install T8 device in lava
09:21:42 <zhizhoutian_> What are you going to do?
09:21:43 <zhizhoutian_> 1.help IT to build a website that linaro engineer can access tshark code base
09:21:43 <zhizhoutian_> What problems do you have?
09:22:12 <bero|2> luther: do you want to go next?
09:22:37 <luther> ok
09:22:46 <luther> 1. A whole day's meeting
09:22:46 * luther What are you going to do?
09:22:46 <luther> 1. Look at move and forward mechanism
09:22:46 * luther What problems do you have?
09:22:46 <luther> 1. none till now
09:22:52 * luther What did you do?
09:23:18 <bero|2> ok...
09:23:25 <bero|2> eric____: Do you have any news already?
09:26:14 <liuyq> sunao, I used uefi before, the document about usefi is here: https://github.com/96boards/documentation/wiki/HiKeyUEFI
09:27:02 <sunao> thanks liuyq, the official releae or the snapshot?
09:27:13 <bero|2> apparently no news from eric____...
09:27:21 <bero|2> lijixin: do you have any news?
09:27:33 <lijixin> ok
09:28:10 <lijixin> What did you do?
09:28:10 <lijixin> look at the code about external/compiler-rt/lib/builtins/arm,
09:28:10 <lijixin> download Linara MLCR source code for Researching the changes of memory related operations in Bionic .
09:28:10 <lijixin> But I don't download MLCR source code because of network problem.
09:28:27 <liuyq> sunao, I tried images built on my local side. but that's a long time ago. will start the build for hikey at weekend
09:28:50 <lijixin> What are you going to do?
09:28:50 <lijixin> resolve the download problem.
09:28:50 <lijixin> continus to my task,see whether can optimized the memory related operations in  compiler-rt.
09:28:59 <eric____> bero|2: not yet, I look into libwebp, and know that invesgate need a tool called coverage, I'm try to learn it
09:29:21 <sunao> Thanks liuyq
09:29:23 <lijixin> bero: What problems do you have?
09:29:23 <lijixin> none at present.
09:30:37 <bero|2> eric____: ok, good
09:31:07 <bero|2> lijixin: good, let me know if there's problems
09:31:14 <bero|2> wuhai: any news?
09:31:59 <wuhai> bero I think i can do this :
09:31:59 <wuhai> Find more time-consuming system Service Code,and change it to achieve through JNI,I think it would be more efficient.
09:32:15 <eric____> bero|2: I have no permission to access "google drive", so there's no way to update my progress on it
09:32:19 <wuhai> Do you think feasible?
09:32:37 <bero|2> eric____: Thanks for letting me know, will fix...
09:32:57 <bero|2> wuhai: yes, that sounds interesting... go ahead
09:32:58 <eric____> bero|2: thanks
09:33:29 <wuhai> OK,i will do it
09:33:51 <bero|2> ok, everyone should have access to the doc now
09:34:02 <bero|2> wangjian: any news?
09:34:13 <liuyq> sunao, you can check with  docularxu on your hdmi problem, he is the expert on that.
09:34:38 <eric____> bero|2: yes, it is
09:34:45 <wangjian> What did you do?
09:34:46 <wangjian> I have got the tshark device and code from Spreadtrum , build code and flash into the device.
09:34:46 <wangjian> What are you going to do?
09:34:46 <wangjian> Start to analyze launcher app code.
09:34:46 <wangjian> what problems do you have?
09:34:46 <wangjian> none
09:35:27 <bero|2> good...
09:35:33 <bero|2> yuxing_: any news?
09:37:19 <eric____> bero|2: I know you have much experience on segment lib, so do you have any doc for that?
09:38:03 <eric____> bero|2: I think it could be much more efficient to start with your guide
09:38:54 <bero|2> eric____: We don't really have a doc for it, but we have the patches for segmenting libpng and libjpeg... they're probably a good place to start
09:39:04 <bero|2> yspan: do you have a link to your current patch?
09:39:30 <eric____> bero|2: I think so!  how can I get it?
09:39:47 <bero|2> eric____: https://projects.linaro.org/browse/LMG-415 has a patch attached
09:40:07 <yspan> no, where can I put it temporarily?
09:40:23 <bero|2> not sure if it's the latest version (that's why I was asking yspan)
09:40:51 <bero|2> yspan: you can put it on a new branch in git, and/or update the patches attached to LMG-415
09:41:27 <yuxing_> bero, sorry
09:41:43 <yuxing_> What did you do?
09:41:52 <yuxing_> 1、on 512MB memory mobile , config different disksize to test the zramfs, set the disksize: 100MB、200MB、300MB、400MB
09:42:01 <yuxing_> 2、reading the zram driver code
09:42:09 <yuxing_> What are you going to do?
09:42:18 <yuxing_> 1、continue do the test
09:42:24 <yuxing_> 2、reading the zram driver code
09:42:32 <yspan> bero|2, I will paste it to LMG-415 first, and ask IT for repo in dev-private-git
09:42:36 <yuxing_> what problems do you have?
09:42:44 <yuxing_> none
09:43:09 <yspan> eric____, for dlopen idea, you can also reference https://github.com/koukaipan/fake-lib
09:43:35 <eric____> yspan: is it similar one?
09:44:40 <yspan> yes, the url demos an example.
09:44:47 <sunao> yspan, a small question, do you build the libjpeg for both 32bit and 64 bit version?  thank you
09:45:30 <yspan> sunao, yes. hmmm I'm responsible for libpng in fact.
09:45:49 <sunao> oh, sorry, yspan
09:46:04 <yspan> never mind :)
09:47:17 <eric____> yspan: thanks, but if I get the libpng patch, please let me know ASAP
09:47:30 <bero|2> initially the focus is on 32bit devices given we're unlikely to see 64bit devices with 512 MB RAM -- but of course we need to make sure we don't break 64bit stuff, given that's likely the next target
09:47:44 <eric____> if I can get the libpng patch somewhere
09:47:51 <eric____> let me know
09:48:29 <yspan> eric____, sure
09:48:54 <eric____> yspan: good, thanks!
09:50:22 <eric____> yspan: If you can document it, it would be better, :)
09:53:28 <bero|2> ok, I think we're almost done with the official meeting...
09:54:16 <bero|2> One thing for those who left after the meeting yesterday: Please think about further ideas to optimize the system. I'm sure everyone can come up with different ideas given we all have different backgrounds
09:54:31 <bero|2> We'll collect all the ideas in Tuesday's meeting, so take some time to think about it
09:54:39 <bero|2> does anyone want to bring up anything else?
09:55:23 <sunao> bero|2, where is the build instruction for hikey?
09:55:52 <yspan> eric____, the latest patch is updated in LMG-951 which is a follow-up of LMG-415
09:56:38 <sunao> get the code from github/96boards or linaro?
09:56:43 <eric____> yspan: I am afraid I have no permission to access
09:57:59 <bero|2> sunao: Not sure if there's any updated build instructions, but it comes down to:
09:58:16 <bero|2> repo init -u http://android-git.linaro.org/git/platform/manifest.git -b android-6.0.0_r26
09:58:22 <bero|2> cd .repo
09:58:25 <eric____> yspan: but if you can put it on git, it become much easier
09:58:35 <bero|2> git clone https://github.com/96boards/android_manifest local_manifests
09:58:36 <bero|2> cd ..
09:58:37 <bero|2> repo sync
09:58:46 <bero|2> lunch hikey-userdebug
09:58:50 <bero|2> make droidcore
09:59:09 <sunao> Thank you bero|2, I will have a try.
09:59:10 <eric____> message to me?
09:59:20 <bero|2> eric____: let me try to give you access
09:59:54 <eric____> bero|2: cool, thanks
10:00:26 <bero|2> eric____: try again now
10:01:40 <yuxing_> bero: We will optimize zram on how much memory of mobile phone?
10:01:44 <bero|2> ok, I think we can close the official part of the meeting -- feel free to ask anything at any time though, the only difference is that what's said outside of meeting time doesn't get logged and not everyone is required to be around
10:02:06 <bero|2> yuxing_: 512 MB is the first target, will probably also do 1 GB for the slightly higher end phones
10:02:08 <bero|2> #endmeeting
10:02:57 <yuxing_> ok, Thanks
10:03:12 <eric____> bero|2: still failed
10:06:58 <bero|2> eric____: That's odd, you show up in "shared visibility"... I've copied the patch off jira now: http://people.linaro.org/~bernhard.rosenkranzer/0001-libpng-separate-encoding-to-libpng-enc-to-limit-what.patch
10:07:40 <bero|2> wait, something went wrong there as well
10:07:41 <bero|2> fixing
10:08:08 <bero|2> should be ok now
11:31:56 <tixy> bhoj, does this look like out of space on system partition? https://android-build.linaro.org/jenkins/job/linaro-android_lcr-reference-juno-m/30/console
11:36:05 <bhoj> tixy, yes.
11:36:20 <bhoj> tixy, i'll have a look at it.
11:36:53 <tixy> bhoj, thanks
12:15:55 <eric____> bero|2: it still show "permission violation", :(
14:29:14 <bero|2> eric____: This one? http://people.linaro.org/~bernhard.rosenkranzer/0001-libpng-separate-encoding-to-libpng-enc-to-limit-what.patch
14:31:26 <tixy> bhoj, thanks for the lmc fix, now that fabo has merged it will the Android builds pick up the new version? (I'm wondering if I should hit the 'build now' button for lcr-reference)
14:32:23 <fabo> tixy: it's deployed, yes
14:33:58 <tixy> fabo, thanks, so I've hit the build button and we'll know in about an hour... :-)
14:34:28 <fabo> tixy: :)
01:27:20 <eric____> bero|2: it looks OK on this webpage, thanks
04:04:46 <tgall_foo> let's use the Android channel for the meeting
04:04:51 <linarobot> tgall_foo: Error: Can't start another meeting, one is in progress.
04:04:55 <tgall_foo> #endmeeting
04:05:06 <tgall_foo> hmm looks like someone forgot to turn it off :-)
04:05:23 * tgall_foo heads over to #linaro-meeting
09:06:25 <eric____> hello, did the meeting canceled?
10:54:42 <bero|2> eric____: on Monday and Friday, the meeting is a few hours earlier (and run by Tom - he's in a different timezone)
10:54:57 <bero|2> eric____: but it looks like it didn't happen today anyway
10:55:04 <bero|2> eric____: which is fine, we'll just read the doc
11:21:33 <tixy> anone know the process for adding repos under android-git.linaro.org ? Is it something a person with the right Gerrit permissions can do, or does it need an IT request?
11:23:05 <broonie> You can definitely add repositories with the right permissions in gerrit, I don't know who has those in Linaro though.
11:24:13 <mingling> test
11:32:33 <bhoj> tixy, systems can do it. is this for gralloc ?
11:59:49 <tixy> bhoj, yes, for gralloc, who's 'systems'? I don't keep up on company reorgs.
12:02:09 <bhoj> tixy, it is doanac paul and team. I think I can create the repository. Let me know what repository would keep ARM happy ? hardware/arm/gralloc vs hardware/mali/gralloc vs vendor/arm/gralloc vs vendor/mali/gralloc
12:03:38 <tixy> bhoj, thanks, I'll check
12:07:12 <tixy> bhoj, the bug report asks for it under hardware/arm/gralloc (presumably in the source tree) so would also make sense for the git repo to be same path
12:07:31 <bhoj> tixy, okay.
12:17:59 <bhoj> tixy, https://android-review.linaro.org/#/admin/projects/hardware/arm/gralloc
12:55:56 <ajb-linaro> It's been a while since I last built Android and I notice there is a new build system now. Are we meant to run our own Jack instances on the hackbox server or just have per user ones?
14:39:22 <ajb-linaro> bhoj: have you managed to get the new Jack server to honour SERVER_PORT_* settings?
14:40:25 <bhoj> ajb-linaro, I haven't been successful with JACK much. I build with "export ANDROID_COMPILE_WITH_JACK=false"  on hackbox.
14:44:55 <ajb-linaro> bhoj: I'll try that, thanks
14:45:14 <ajb-linaro> bhoj: funnily enough my jack fails to start because you have one running ;-)
14:47:49 <bhoj> ajb-linaro, yes i have seen this happening to me. do you want me to kill my instance ?
14:48:31 <ajb-linaro> bhoj: no - it's ok, I'll try without for now... is the jack source code in the AOSP tree? maybe I should see how it picks the port
19:44:52 <robher> jstultz: android config changes: https://android.googlesource.com/device/generic/brillo/+/997086a5690742e1b1d721ec6b94bc0e7d7597fe
19:45:27 <jstultz> robher: sweet..  i should start asking for more things :)
19:46:03 <robher> jstultz: like puppies?
19:49:12 <jstultz> robher: hrm.. though looking closer at the commit, it sort of looks like more local optimization rather then a larger cleanup overhaul..  wish the DDOC link was public.. :)
19:51:18 <robher> jstultz: yes. maybe it is something to build on.
19:51:25 <jstultz> robher: but i guess its missing the example config file in there.. so maybe it will help.. it just seems like its keeping all the crazy make file variables, but just allowing for assignment through config values..
19:52:26 <robher> jstultz: here it is in use: https://android.googlesource.com/product/google/example-ledflasher/+/master/config/
19:52:47 <jstultz> robher: uuuuh.
19:52:53 <jstultz> its one value per file..
19:53:04 <robher> everything is a file.
19:53:26 <jstultz> apparently..
19:53:39 <jstultz> i guess its not one value per file, but one variable per file..
19:53:57 <jstultz> robher: what was the make variable count you had w/ your debug tool?
19:54:59 <robher> It was 20MB worth. many are generated though.
19:55:07 <jstultz> robher: it was a big number if i recall.. so joy.. now we'll have a ton of tiny files..
19:55:41 <robher> jstultz: just don't use 64KB page tables.
19:56:21 <jstultz> I never really considered kconfig to be that amazing..  but i feel like i've grown much more respect for it as i see other projects. :)
19:56:52 <jstultz> robher: we'll have to introduce kernel optimizations to pack multiple files (w/ same perms) into the same page
19:57:19 <jstultz> robher: vastly improves build times!
07:34:17 <liuyq> bhoj, hi
08:05:55 <bhoj> liuyq, hi
08:08:41 <liuyq> bhoj, when I boot hikey, the grub menu is not displayed, I only got the grub shell, do you have any idea what's the problem?
08:09:53 <liuyq> bhoj, it's OK on Friday, but today I got this problem
08:10:13 <bhoj> liuyq, latest UEFI is not supportted on AOSP. Please use  http://builds.96boards.org/snapshots/hikey/linaro/uefi/79/
08:13:43 <liuyq> bhoj, ok, will try with this version
08:13:46 <liuyq> btw, have talked the hikey failure on LAVA with lava team?
08:14:59 <bhoj> liuyq, I have fixed template to use boot_fat.uefi.img now but with recent changes in UEFI i am not sure if it will break LAVA again.
11:30:45 <ajb-linaro> https://bpaste.net/show/f454aa4ef3ba
11:31:15 <ajb-linaro> I'm having problems building the current AOSP due to what look like some sort of Java lint thing
11:31:29 <ajb-linaro> "121 unresolved references to classes or interfaces"
11:31:40 <ajb-linaro> is this something I can turn off or is the upstream build broken?
11:33:28 <bhoj> ajb-linaro, emulator build from today morning is working. boots to UI.
11:34:33 <ajb-linaro> bhoj: :-/ I can't even get it to build
11:35:14 <bhoj> ajb-linaro, is it again on hackbox ?
11:44:43 <ajb-linaro> bhoj: yeah
13:07:13 <tgall> morning
14:33:42 <stooart> bhoj: Hello! I finally managed to persuade my Juno that booting was in its best interest. However, I'm not sure the vendor.tar.gz file/kernel are matching.
14:33:48 <stooart> see: file /dev/mali0 is not of a compatible version (user 10.1, kernel 8.0)
14:34:15 <stooart> (I'm on the lor_marshmallow branch)
14:34:55 <stooart> (of the local_manifest)
14:36:20 <bhoj> stooart, did you use http://snapshots.linaro.org/android/binaries/arm/20151111/  version ?
14:36:56 <stooart> bhoj: Yes. Althought I can double-check.
14:38:06 <bhoj> stooart, the builds using the same binaries are working in automation https://validation.linaro.org/scheduler/job/595245/log_file. Let me check the kernel that is being used in the local manifest
14:38:36 <stooart> Yes, I do have that version.
14:40:50 <bhoj> stooart, did you use juno_prebuilt target or something ?
14:41:10 <bhoj> stooart, found the issue.
14:41:25 <bhoj> stooart, kernel tag is wrong in the local_manifest. fixing it now.
14:49:17 <stooart> Ah, great, thanks vishal. I thought the version differnece might have been an issue.
14:56:14 <stooart> bhoj: Please let me know when you are done, and I can have a go.
15:14:34 <bhoj> stooart, https://dev-private-review.linaro.org/404  it is 404 but link should be found :P
15:21:40 <stooart> bhoj: Understood :-)
15:42:31 <liuyq> ryanharkin, ping
15:43:05 <ryanharkin> liuyq, hi
15:43:15 <liuyq> ryanharkin, is it possible to check if the latest firmware is applied for the juno from lava log?
15:44:21 <liuyq> ryanharkin, like from the log here: https://validation.linaro.org/scheduler/job/597635/log_file#L_51
15:44:37 <ryanharkin> liuyq, i don't know, depends how verbose copying the firmware is
15:45:08 * ryanharkin is looking
15:45:15 <liuyq> ryanharkin, no version information will be displayed on the console?
15:45:43 * liuyq tried to find string like 0.11.6-linaro1, but not find
15:46:12 <ryanharkin> liuyq, no, the SCP binaries don't output their version on the main console, they output to another serial port
15:48:33 <ryanharkin> liuyq, actually, the best way to tell is to check the trusted firmware message, they are not explicit, but they imply the correct versions from their build stamps
15:48:48 * ryanharkin is checking what is the correct build stamp
15:49:41 <ryanharkin> first off, this line tells me that you're using 0.11.6-linaro1:
15:49:59 <ryanharkin> NOTICE:  BL1: v1.1(debug):4266228
15:50:40 <ryanharkin> because "4266228" is the commit id of the latest ARM-TF repo I used - it was never published before
15:50:53 <liuyq> ryanharkin, there is a kernel panic occurred during this job: https://validation.linaro.org/scheduler/job/597635/log_file#L_65_147
15:51:52 <liuyq> ryanharkin, not sure if that has relation with the firmware
15:52:08 <ryanharkin> liuyq, and I can see that the build date matches 0.11.6 exactly also:  "Built : 13:32:39, Nov 11 2015"
15:53:47 <ryanharkin> liuyq, that bug looks related to SCPI code in the kernel, which is of course related to the SCP firmware update.  I would expect it to me more like a kernel bug than an SCP bug, but you never know
15:58:43 <tixy> ryanharkin, so the firmware being used is the latest we got from ARM?
16:00:09 <liuyq> ryanharkin, tixy same problem here: https://validation.linaro.org/scheduler/job/597629/log_file#L_65_92
16:00:37 <ryanharkin> tixy,  it looks like it - it has the latest commit id and date stamp for the BL1 I built, which implied they are running the matching FIP
16:04:30 <tixy> ryanharkin, liuyq that BUG was code I added to detect firmware lockup issues, and avoid kernel data corruption leading to obscure crashes, i.e. bugs 1766 and 1841 Guess I need to investigate more, perhaps I added a new failure mechanism
16:05:12 <ryanharkin> tixy, ok, thanks
16:08:48 <tixy> liuyq, any idea what tests are being run when this happens? (Kernel timestamp in logs shows it's after > 1hour, how I can make it reliably reproducible in less time)
16:09:31 <liuyq> tixy, wait a sec
16:18:48 <liuyq> tixy, for this job https://validation.linaro.org/scheduler/job/597635/log_file#L_65_147, it ran the CTS-opengl test plan,  and it only ran for about 30 minutes.
16:19:12 <tixy> liuyq, thanks
16:19:42 <liuyq> tixy, get the cts package from here: http://testdata.validation.linaro.org/cts/android-cts-6.0_r1.zip, and run "run cts --plan CTS-opengl --disable-reboot" on the cts prompt
16:21:34 <tixy> liuyq, you read my mind, thanks :-)
16:22:06 <liuyq> tixy, :)
19:07:41 <pundir> jstultz, hi!
19:07:53 <pundir> for 4.4-rc1 based LLCT tree
19:08:06 <pundir> i�m planning to take it up as a probable experimental/android-4.4 tree and will prepare a rebased tree instead.
19:08:29 <pundir> instead of a merged tree
19:08:35 <pundir> like we do regularly
19:08:38 <pundir> sounds good?
19:28:54 <jstultz> pundir: eh.. hrm..
19:29:31 <jstultz> pundir:  so i'm not sure what the odds of an experimental/android-4.4 showing up soon is.
19:30:01 <jstultz> pundir: i agree it would be good to have a rebased tree...
19:30:47 <jstultz> pundir: but for llct i'm not sure if that's the right call, since if it doesn't show up, we'll have to pull items from experimental/android-4.1 or even an official andorid-4.1 tree
19:33:05 <jstultz> so yea.. i don't know what the best approach is right off.
19:58:22 <A6358-Guest> I have some screenshots of recipes that I would like to put on 4x6 cards but still be able to read the words. Is there a way to do this or am I going to have to print them out on full sheets of paper? They are not on a computer they are on a tablet not going to type/write them out. Even if they where on a computer my question is still the same. All
19:58:22 <A6358-Guest> I want to know is if I can print them on 4x6 index cards and have them readable. http://tinypic.com/r/2pt4f1s/9
01:57:17 <zhizhoutian_> mingling:
08:33:43 <xruxa> morning all
08:33:50 <linarobot> xruxa: Error: Can't start another meeting, one is in progress.
08:33:51 <Qian_> hi all
08:34:05 <xruxa> ok, room blocked, lets try linaro-meeting
08:34:07 <sunao> hi all
08:35:11 <bero2> hi, who is there?
08:37:36 * liuyq is here
08:37:39 * zhizhoutian__ is here
08:38:47 <linarobot> bero2: Error: Can't start another meeting, one is in progress.
08:39:23 <bero2> liuyq: do you want to go first?
08:40:17 <liuyq> bero2, here  or the linaro-meeting?
08:40:33 <zhizhoutian__> bero2: btw, ming.ling will take part in this meeting
08:41:22 <liuyq> bero, the meeting is in progress in #linaro-meeting.
08:41:48 <liuyq> bero2, ^^^
08:44:12 <yspan> bero2, would you try endmeeting with 'bero' I remember you lost connection in last meeting ?
08:58:19 <liuyq> sunao, I use hdmi2vga converter, and it works without usb mouse and keyboard
08:58:48 <liuyq> sunao, if I plugged the usb mouse, it will cause display problem
09:03:52 <bero> #endmeeting