linaro-dev June 2013

linaro-dev@lists.linaro.org

57 participants
67 discussions

Planned maintenance for review.android.git.linaro.org, Saturday 29th June

by Philip Colmer

This is to announce a small maintenance window for review.android.git.linaro.org where we will remove support for authentication via Launchpad, and only allow authentication via Linaro Login. As a result, you will not be required to enter the long OpenID URL that you have had to use since we migrated the server. The work will take place during Saturday morning and another email will be sent when the work has been completed. Regards Philip

10 years, 5 months

Test Result Summary of Linaro openembedded 13.06 release

by Soumya Basak

Linaro 13.06 Release (Calendar Week 26 2013): Here is the test result summary for 13.06 release Linaro openembedded images. [1] Linaro openembedded Minimal images (Column: Y) https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AgB-fT5LL31CdDhwR… please refer to the result spreadsheet. [2] Linaro openembedded Lamp images (Column: Y) https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AgB-fT5LL31CdDhwR… please refer to the result spreadsheet. Thank You. Best Regards Soumya Basak.

10 years, 5 months

One HMP issue over TC2

by Lei Wen

Sorry for inconvenience, as my bad mail editor... Change the mail subject to the corrected one On Thu, Jun 27, 2013 at 5:34 PM, Lei Wen <adrian.wenl(a)gmail.com> wrote: > Hi Morten and list, > > I am current investigating HMP related stuff over TC2 platform, and > find a strange issue. > Here I have two module, both are creating fixed loaded thread, but one > module would > bound those thread to the cpu, while another don't do the bound operation. > > [I test with linaro 2013.05 release to get below result] > With setting A7/A15 to the same frequency, 1G, > The result is: > Bounded one: > Five thread finished with 137s/137s/137s/138s/138s > > unbounded one(With HMP related configuration enabled, which is the default one): > Five thread finished with 138s/275s/275s/275s/275s > > unbounded one(With HMP related configuration disabled): > Five thread finished with 228s/229s/229s/229s/231s > > So it seems to me, that current configuration don't make TC2 to run in > its full performance. > It worries me for it may downgrade the benchmark for somehow. > > I haven't see into details, just post the result here to get your > guys' feedback. > While I get more detailed analysis, I would also post it here. > > Thanks, > Lei

10 years, 5 months

cxie4@marvell.com, wwang27@marvell.com, ytang5@marvell.com, leiwen@marvell.com

by Lei Wen

Hi Morten and list, I am current investigating HMP related stuff over TC2 platform, and find a strange issue. Here I have two module, both are creating fixed loaded thread, but one module would bound those thread to the cpu, while another don't do the bound operation. [I test with linaro 2013.05 release to get below result] With setting A7/A15 to the same frequency, 1G, The result is: Bounded one: Five thread finished with 137s/137s/137s/138s/138s unbounded one(With HMP related configuration enabled, which is the default one): Five thread finished with 138s/275s/275s/275s/275s unbounded one(With HMP related configuration disabled): Five thread finished with 228s/229s/229s/229s/231s So it seems to me, that current configuration don't make TC2 to run in its full performance. It worries me for it may downgrade the benchmark for somehow. I haven't see into details, just post the result here to get your guys' feedback. While I get more detailed analysis, I would also post it here. Thanks, Lei

10 years, 5 months

sched: Load-balancing within HMP domain in the HMP scheduler

by J

Hi. I have been trying to apply Linaro's HMP scheduler to a big.LITTLE machine, which has 4 big and 4 LITTLE cores. However, I could only utilize two cores with the HMP scheduler. While hmp_select_faster_cpu() always selects the first core in the big cluster and hmp_select_slower_cpu() always selects the first core in the LITTLE cluster, I found that my kernel with the HMP scheduler does no load-balancing at all within each HMP domain(big/LITTLE cluster). The function rebalance_domains() does almost nothing, because SD_LOAD_BALANCE flags for the sched_domains are always false. It's the same in select_task_rq_fair(). I'm wondering if this is normal, although I believe it shouldn't be. If it's not, how can I get the HMP scheduler to perform load-balancing within an HMP domain? What might I be doing wrong here? Thanks. - Antonio

10 years, 5 months

Perf test giving strange results

by Andy Green

Hi - Trying perf test on llct-based 3.10-rc6 isn't entirely happy. root@aa9-eb:~# perf test 1: vmlinux symtab matches kallsyms : FAILED! 2: detect open syscall event : Ok 3: detect open syscall event on all cpus : Ok 4: read samples using the mmap interface : Ok 5: parse events tests : Warning: function sizeof not defined Warning: function sizeof not defined Warning: function scsi_trace_parse_cdb not defined Warning: function scsi_trace_parse_cdb not defined Warning: function scsi_trace_parse_cdb not defined Warning: function scsi_trace_parse_cdb not defined Warning: function jiffies_to_msecs not defined Warning: function jiffies_to_msecs not defined Warning: bad op token { Warning: bad op token { Warning: bad op token { Warning: Error: expected type 4 but read 0 Warning: bad op token { Warning: Error: expected type 4 but read 0 Warning: bad op token { Warning: bad op token { Warning: bad op token { Warning: bad op token { Warning: bad op token { Warning: unknown op '{' Warning: unknown op '{' Ok 6: Validate PERF_RECORD_* events & perf_sample fields : FAILED! 7: Test perf pmu format parsing : Ok 8: Test dso data interface : Ok 9: roundtrip evsel->name check : Ok 10: Check parsing of sched tracepoints fields : FAILED! 11: Generate and check syscalls:sys_enter_open event fields: Ok 12: struct perf_event_attr setup : (omitted) Ok 13: Test matching and linking mutliple hists : Ok 14: Try 'use perf' in python, checking link problems : FAILED! 15: Test breakpoint overflow signal handler : FAILED! 16: Test breakpoint overflow sampling : FAILED! 17: Test number of exit event of a simple workload : FAILED! 18: Test software clock events have valid period values : Ok I started tracing the first failure... it stops because it can't find "_stext" in /proc/kallsyms. But it is there root@aa9-eb:~# cat /proc/kallsyms | grep _stext 800081c0 T _stext test 5 (parse event tests) seems to be badly broken. Is this just our problem or is it broken for everyone? -Andy

10 years, 5 months

(no subject)

by Harigopal Gollamudi

-- Thanks and Regards, HariGopal G Android Engineer

10 years, 5 months

[meta-linaro-toolchain][PATCH] external-linaro-toolchain-versions: Fix NameError: global name 'CmdError' is not defined

by Martin Jansa

* fails with: ERROR: Execution of event handler 'external_linaro_toolchain_version_handler' failed Traceback (most recent call last): File "external_linaro_toolchain_version_handler(e)", line 12, in external_linaro_toolchain_version_handler(e=<bb.event.ConfigParsed object at 0x7759a50>) File "external-linaro-toolchain-versions.inc", line 4, in elt_get_gdb_version(d=<bb.data_smart.DataSmart object at 0x24b58d0>) NameError: global name 'CmdError' is not defined Signed-off-by: Martin Jansa <Martin.Jansa(a)gmail.com> --- .../conf/distro/include/external-linaro-toolchain-versions.inc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta-linaro-toolchain/conf/distro/include/external-linaro-toolchain-versions.inc b/meta-linaro-toolchain/conf/distro/include/external-linaro-toolchain-versions.inc index 92862af..0254895 100644 --- a/meta-linaro-toolchain/conf/distro/include/external-linaro-toolchain-versions.inc +++ b/meta-linaro-toolchain/conf/distro/include/external-linaro-toolchain-versions.inc @@ -86,7 +86,7 @@ def elt_get_kernel_version(d): def elt_get_gdb_version(d): try: stdout, stderr = elt_run(d, 'gdb', '-v') - except CmdError: + except bb.process.CmdError: return 'UNKNOWN' else: first_line = stdout.splitlines()[0] -- 1.8.2.1

10 years, 5 months

Muti touch device android failed to recognise

by dani maoz

Hi I have created a virtual multi touch virtual driver, I see that Event hub Detect the device but when i transmit events , no movement occurred on the screen (No X or Y changed) I tried to use protocol A pr Protocol B same result. why the events not transmitted to framework Any Idea See below the logs (Pandaboard Linaro 4.1.2) EventHub /EventHub( 1775): New device: id=7, fd=195, path='/dev/input/event3', name='test_mt_touch', classes=0x14, configuration='', keyLayout='', keyCharacterMap='', builtinKeyboard=false, usingSuspendBlockIoctl=false, usingClockIoctl=false I/InputReader( 1775): Device reconfigured: id=7, name='test_mt_touch', size 1280x720, orientation 0, mode 3, display id 0 I/InputReader( 1775): Device added: id=7, name='test_mt_touch', sources=0x00002002 Event transmitted from driver using getevent -lt /dev/input/event3 [ 3859.719026] EV_ABS ABS_MT_POSITION_X 000000c8 [ 3859.719056] EV_ABS ABS_MT_POSITION_Y 000000c8 [ 3859.719056] EV_ABS ABS_MT_WIDTH_MAJOR 0000002c [ 3859.719056] EV_ABS ABS_MT_TOUCH_MAJOR 00000000 [ 3859.719056] EV_SYN SYN_MT_REPORT 00000000 [ 3859.722932] EV_SYN SYN_REPORT 00000000 [ 3863.303383] EV_ABS ABS_MT_POSITION_X 000000c8 [ 3863.303383] EV_ABS ABS_MT_POSITION_Y 000000c8 [ 3863.303414] EV_ABS ABS_MT_WIDTH_MAJOR 0000002c [ 3863.303414] EV_ABS ABS_MT_TOUCH_MAJOR 00000028 [ 3863.303414] EV_SYN SYN_MT_REPORT 00000000 [ 3863.303475] EV_SYN SYN_REPORT 00000000

10 years, 5 months

big.LITTLE MP status June 18, 2013

by Serge Broslavsky

https://wiki.linaro.org/projects/big.LITTLE.MP Work Items for functional completion: https://wiki.linaro.org/projects/big.LITTLE.MP/WorkItems Main Project page: https://launchpad.net/linaro-big-little-system Development ============ 1. big.LITTLE IKS IKS snapshot: git://git.linaro.org/landing-teams/working/arm/kernel.git More information: http://lists.linaro.org/pipermail/linaro-dev/2013-May/015882.html Also the latest Linaro 13.05 release contains big.LITTLE release with IKS code integrated. By default MP is used (IKS is off). To enable IKS: echo 1 > /sys/kernel/bL_switcher/active Same is planned for Linaro 12.06 release. 2. big.LITTLE MP There are two branches, which are functionally equivalent - they tweak the scheduler to select the right processes to move to the A15 and try to decrease power consumption while maintaining performance. * A Reference big.LITTLE MP patchset (based on ARM Beta patchset), merged with linux-linaro: https://git.linaro.org/gitweb?p=arm/big.LITTLE/mp.git;a=shortlog;h=refs/hea… Recommended to be used for testing. * And an upstream (work in progress) big.LITTLE MP patchset: https://git.linaro.org/gitweb?p=arm/big.LITTLE/mp.git;a=shortlog;h=refs/hea… Being worked on in order to achieve the performance/power numbers of the reference one. Upstream discussion is ongoing. * Documentation: Code flow is already covered by several LWN articles. See https://wiki.linaro.org/WorkingGroups/PowerManagement/Doc/ in the Articles section. We will provide a tuning guide once we have reached agreement upstream. * Automated benchmarking code is ready: - Golden Board is to be used for running daily tests. - Both branches - reference and upstream to be tested in the future. Blueprint Updates ------------------ * https://blueprints.launchpad.net/linaro-big-little-system/+spec/big-little-… Ongoing work. * https://blueprints.launchpad.net/linaro-big-little-system/+spec/small-task-… - Big variation previously seen is not visible on the Golden Board. It might be caused by an USB stick with insufficiently high speed - working on this hypothesis. - Latest patchset (not yet published upstream) shows low delta with the reference patch set results. * https://blueprints.launchpad.net/linaro-big-little-system/+spec/sched-coope… - Postponed. Priorities are to be re-evaluated. QA == big.LITTLE MP LAVA Dashboard http://validation.linaro.org/lava-server/dashboard/image-reports/linaro-and… Testing (week 25): sched_tests.git #Cases #Run Pass Fail Absolute pass rate Regression 20 19 19 0 100 % mpbasicsuite 14 10 10 0 100 % mpcoresuite 10 6 6 0 84 % [1][2] mpextendedsuite 20 10 7 3 70 % [3] mploadbalance 2 2 2 0 100 % --- [1] https://bugs.launchpad.net/linaro-big-little-system/+bug/1188655 [2] core_test_scn02.1 reported failed in full execution and passed independently: https://pastebin.linaro.org/view/21cf80f3 [3] https://bugs.launchpad.net/linaro-big-little-system/+bug/1183769 Bugs: https://bugs.launchpad.net/linaro-big-little-system/+bugs?field.tag=bl-mp Highlight: Bug 1166246 - Lockup on post 3.9 IKS kernel - can be hardware or software bug, investigation continues. - seems to be related to MMC being in use (switching to USB stick can be used as a reliable workaround). Highlight: Bug 1183821 - perf failed to count events on A15 cluster - existing perf code is at the end of it's lifecycle. - a rewrite is ongoing. - will stay unfixed for Linaro 13.6 release. -- Best Regards, Serge Broslavsky <serge.broslavsky(a)linaro.org> Project Manager, Linaro M: +37129426328 IRC: ototo Skype: serge.broslavsky http://linaro.org | Open source software for ARM SoCs

10 years, 5 months

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-dev June 2013