linaro-dev February 2013

linaro-dev@lists.linaro.org

85 participants
82 discussions

[PATCH] ARM: EXYNOS: update secondary startup code to use 24 bits of MPIDR

by Chander Kashyap

Commit a0ae0240507... "ARM: kernel: add device tree init map function" started using 24 LSBs of MPIDR to calculate the cpu_logical_map. Hence update the same in exynos_secondary_startup code. Signed-off-by: Chander Kashyap <chander.kashyap(a)linaro.org> Signed-off-by: Inderpal Singh <inderpal.singh(a)linaro.org> --- arch/arm/mach-exynos/headsmp.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-exynos/headsmp.S b/arch/arm/mach-exynos/headsmp.S index 5364d4b..5c1c24a 100644 --- a/arch/arm/mach-exynos/headsmp.S +++ b/arch/arm/mach-exynos/headsmp.S @@ -22,7 +22,7 @@ */ ENTRY(exynos4_secondary_startup) mrc p15, 0, r0, c0, c0, 5 - and r0, r0, #15 + and r0, r0, #0xffffff adr r4, 1f ldmia r4, {r5, r6} sub r4, r4, r5 -- 1.7.9.5

11 years, 10 months

[powerdebug,1/2] documentation: add regulator related information

by Sanjay Singh Rawat

Add information about the regulator state and status field Signed-off-by: Sanjay Singh Rawat <sanjay.rawat(a)linaro.com> --- doc.txt | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 doc.txt diff --git a/doc.txt b/doc.txt new file mode 100644 index 0000000..36ffc46 --- /dev/null +++ b/doc.txt @@ -0,0 +1,20 @@ +Regulator related information +============================= +Regulator fields "state" and "status" give below information about the regulator. + +- state: field tell whether regulator is enabled or disabled. Following are the states + | - Enabled + | - Disabled + | - Unknown (if unable to determine) + | + | + ----> status: When enabled, depending on the regulator hardware capability(modes) and runtime requirement; regulators are set to specific modes. The status field gives a little + more specific information about the current "state" of the regulator, which operating modes it is in. Following can be the possible status: + - Idle + - Fast + - Normal + - Standby + - Bypass + - Error + - Undefined + - On/Off -- 1.7.9.5

11 years, 10 months

[PATCH 0/4] CPUFreq Fixes for 3.9

by Viresh Kumar

Hi Rafael, This is another unplanned patchset for all the platforms that i broke. :) Okay, there are two important fixes (1 & 4) and two general cleanups (2 & 3). I hope most of the issues would be resolved by these and we would be able to push clean cpufreq core into 3.9. I have pushed them in my for-rafael branch at: http://git.linaro.org/gitweb?p=people/vireshk/linux.git;a=shortlog;h=refs/h… @Artem & Valdis: Please test them and reply with your Tested-by's (in case they work :) ). Viresh Kumar (4): cpufreq: governors: Fix WARN_ON() for multi-policy platforms cpufreq: Remove unused HOTPLUG_CPU code cpufreq: Create a macro for unlock_policy_rwsem{read,write} cpufreq: Fix locking issues drivers/cpufreq/cpufreq.c | 126 ++++++++++++++++++------------------- drivers/cpufreq/cpufreq_governor.c | 32 ++++++---- 2 files changed, 79 insertions(+), 79 deletions(-) -- 1.7.12.rc2.18.g61b472e

11 years, 10 months

[PATCH 1/4] ACPI / idle: remove unused definition

by Daniel Lezcano

The different definitions are not used anywhere in the code. Signed-off-by: Daniel Lezcano <daniel.lezcano(a)linaro.org> --- drivers/acpi/processor_idle.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index e606e36..5cb5f24 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -64,11 +64,6 @@ #define ACPI_PROCESSOR_CLASS "processor" #define _COMPONENT ACPI_PROCESSOR_COMPONENT ACPI_MODULE_NAME("processor_idle"); -#define PM_TIMER_TICK_NS (1000000000ULL/ACPI_PM_TIMER_FREQUENCY) -#define C2_OVERHEAD 1 /* 1us */ -#define C3_OVERHEAD 1 /* 1us */ -#define PM_TIMER_TICKS_TO_US(p) \ - (((p) * 1000)/(ACPI_PM_TIMER_FREQUENCY/1000)) static unsigned int max_cstate __read_mostly = ACPI_PROCESSOR_MAX_POWER; module_param(max_cstate, uint, 0000); -- 1.7.9.5

11 years, 10 months

[PATCH 0/4][V2] davinci: cpuidle - some cleanups

by Daniel Lezcano

This patchset does some cleanup. It could have been folded in a single patch but the review would have been less clean than splitting it into small and trivial patches. The main purpose of this patch is to remove the usage of the driver_data field from the state_usage structure. Len Brown is doing this cleanup in the intel_idle.c file. With this patchset, the processor_idle.c file will be the last user of this field. Also, the patchset simplify the code and makes it a bit more clear to read. Thanks to Sekhar for testing the patchset. Changelog: V2: * removed global static variable initialization to 'false' * added Acked-by: Sekhar Nori <nsekhar(a)ti.com> V1: initial post Daniel Lezcano (4): davinci: cpuidle - use global variable for ddr2 flag davinci: cpuidle - move code to prevent forward declaration davinci: cpuidle - remove the ops davinci: cpuidle - remove useless initialization arch/arm/mach-davinci/cpuidle.c | 84 ++++++++++++--------------------------- 1 file changed, 25 insertions(+), 59 deletions(-) -- 1.7.9.5

11 years, 10 months

big.LITTLE MP Scheduler test summary-Week-6

by Naresh Kamboju

Hi, Calendar Week 6, 2013: Here is test result summary for big.LITTLE MP Scheduler test on TC2 platform with Android image sched_tests.git No of Test Cases Tests Run Tests Pass Tests Fail Absolute pass rate (%) Failure Analysis/Comments Regression 20 20 20 0 100 - mpbasicsuite 10 10 10 0 100 - mpcoresuite 4 4 4 0 100 - mpextendedsuite 6 6 4 2 66.7 - mploadbalance 2 2 2 0 100 - Android Build: https://android-build.linaro.org/builds/~linaro-android-restricted/vexpress… Linux kernel: Linux version 3.8.0-rc6-00860-gabb16ac (jenkins-build@ip-10-40-91-94) (gcc version 4.7.3 20130102 (prerelease) Results spread sheet: https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0Ai9ggMs8bsGJdER3a… mpbasicsuite: Deprected 4 test cases "FT0005" "FT0006" "FT0009" and "FT0010" mpcoresuite: Deprecated 4 test cases due to change in behaviour due to global load balancing patch The list of deprecated tests are - core_test_scn02.2, core_test_scn03.1, core_test_scn03.2, core_test_scn05.1 mpextendedsuite: Deprecated 8 test cases due to change in behaviour due to global load balancing patch The list of deprecated tests are - extd_test_scn01.2, extd_test_scn01.3, extd_test_scn02.1, extd_test_scn02.2, extd_test_scn03.2, extd_test_scn03.3, extd_test_scn07.1, extd_test_scn08.1 Here two test case reported FAILURE: extd_test_scn04.1 and extd_test_scn04.2 these needs to be analysed. Best regards Naresh Kamboju

11 years, 10 months

[PATCH] sched: fix wrong rq's runnable_avg update with rt task

by Vincent Guittot

When a RT task is scheduled on an idle CPU, the update of the rq's load is not done because CFS's functions are not called. Then, the idle_balance, which is called just before entering the idle function, updates the rq's load and makes the assumption that the elapsed time since the last update, was only running time. The rq's load of a CPU that only runs a periodic RT task, is close to LOAD_AVG_MAX whatever the running duration of the RT task is. A new idle_exit function is called when the prev task is the idle function so the elapsed time will be accounted as idle time in the rq's load. Signed-off-by: Vincent Guittot <vincent.guittot(a)linaro.org> --- kernel/sched/core.c | 3 +++ kernel/sched/fair.c | 10 ++++++++++ kernel/sched/sched.h | 5 +++++ 3 files changed, 18 insertions(+) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 26058d0..592e06c 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2927,6 +2927,9 @@ need_resched: pre_schedule(rq, prev); + if (unlikely(prev == rq->idle)) + idle_exit(cpu, rq); + if (unlikely(!rq->nr_running)) idle_balance(cpu, rq); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 5eea870..520fe55 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -1562,6 +1562,16 @@ static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq, se->avg.decay_count = atomic64_read(&cfs_rq->decay_counter); } /* migrations, e.g. sleep=0 leave decay_count == 0 */ } + +/* + * Update the rq's load with the elapsed idle time before a task is + * scheduled. if the newly scheduled task is not a CFS task, idle_exit will + * be the only way to update the runnable statistic. + */ +void idle_exit(int this_cpu, struct rq *this_rq) +{ + update_rq_runnable_avg(this_rq, 0); +} #else static inline void update_entity_load_avg(struct sched_entity *se, int update_cfs_rq) {} diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index fc88644..9707092 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -877,6 +877,7 @@ extern const struct sched_class idle_sched_class; extern void trigger_load_balance(struct rq *rq, int cpu); extern void idle_balance(int this_cpu, struct rq *this_rq); +extern void idle_exit(int this_cpu, struct rq *this_rq); #else /* CONFIG_SMP */ @@ -884,6 +885,10 @@ static inline void idle_balance(int cpu, struct rq *rq) { } +static inline void idle_exit(int this_cpu, struct rq *this_rq) +{ +} + #endif extern void sysrq_sched_debug_show(void); -- 1.7.9.5

11 years, 10 months

Linaro Release 13.01 Postmortem Summary

by David Zinman

Postmortem and lessons learned for Linaro's release 2013.01 https://wiki.linaro.org/Cycles/1301/Release/Review Highlights and Key Successes ============================ http://www.linaro.org/downloads/1301 http://wiki.linaro.org/Cycles/1301/Release#Release_Information For the first release of the year Linaro has some nice new technologies to augment the regular platforms that have been released. The Developer platform Team has enabled 64bit HipHop VM development in OpenEmbedded, continued to merge ARMv8 support into the OpenEmbedded platform and upstream, engaged initial support for the Arndale board and released Linux Linaro 3.8-rc4 2013.01. The first stable Juice builds have been created and integrated into Linaro's automated testing infrastructure LAVA. The builds allow early Android development to begin on upcoming ARMv8 Models. The builds have been tested against 40 manual and 60 automated tests that cover all aspects of the platform including core Android feature testing and platform use. The Linaro Enterprise Group (LEG) has released Linaro UEFI 2013.01. The highlights include fixes for native building of UEFI on ARM platforms and improvements to flashing images on Arndale board and enable native building for PandaBoard. The Power Management Group hosted a big.LITTLE sprint in Cambridge led by Amit Kucheria. A meeting of minds between ARM, Linaro and the community succeeded in achieving enhanced focus and a more defined roadmap for the big.LITTLE story. Topics included clearly defined benchmarking to determine the effectiveness of the platform, enhanced verification to augment the traditional kernel power/performance testing, and a well defined roadmap for big.LITTLE on Android. In related big.LITTLE MP development, version 14 of the big.LITTLE MP tree has been released. https://wiki.linaro.org/Cycles/1301/Release#Release_Information Blueprints ========= The number of high or essential priority blueprints that missed the cycle: Android 1 out of 7 Developer Platform 3 out of 4 Infrastructure 1 out of 3 Lava 3 out of 7 QA 1 out of 7 Total 9 out of 28 33% of high or essential priority blueprints scheduled for this cycle were not delivered. Total blueprints: 29 out of 61 missed the cycle. High priority missed blueprints recap: 12.05: 19 out of 48, 39% 12.06: 13 out of 31, 42% 12.07: 14 out of 31, 45% 12.08: 6 out of 26, 23% 12.09: 9 out of 28, 32% 12.10: 15 out of 38, 39% 12.11: 7 out of 19, 37%** 12.12: not available * Not included is data from working groups and landing teams Source: https://docs.google.com/spreadsheet/ccc?key=0AjEaTwrvj1bidE5ZeFpsV05id0tRNm… -- David Zinman, Project Manager Linaro.org | Open source software for ARM SoCs

11 years, 10 months

Question about per-cpu workload calculation of interactive governor

by zhoujie wu

Hi Dear Poynor/All cpufreq owner, I studied the cpufreq interactive governor recently, and there is per cpu timer for each cpu, that means the workload will be queried and calculated based on each cpu's workload when timer expires. I am considering about the case of bad application then workload balance is not well issued, may lead to one core is busy, the other is idle on SMP system. This may lead to one core want to increase the core frequency as it is busy, the other core want to decrease the core frequency as it is idle. But SMP system two cores always runs at the same frequency, and the behavior will lead to performance drop. How does interactive handle this case? I know that ondemand governor has opportunity to consider the max workload of all online cpu and make frequency change decision based on the max workload(policy->cpus set to all cpu_mask). Whether interactive will consider the same way? Could you please help to give some suggestion or let me know if i have some misunderstanding of this part? Thanks. -- Zhoujie Wu

11 years, 10 months

Foundation_v8 ssh/scp problem

by Henry Cheang

Hello, I am running Foundation_v8 with images from Linaro as follows: ./Foundation_v8 --image=img-foundation.axf --block-device=vexpress64-openembedded_sdk-armv8_20130127-242.img --network=nat --network-nat-ports=8022=22 I noticed that often when I ssd or scp into the target (ssh -p 8022 root@localhost ), the Foundation_v8 would spike to 100% usage on the host for a few minute before ssh would returns which makes the process very slow. Doing "top" in the target shows that 98% of cpu is being used servicing sirq (soft irq). Has anybody seen this and is there a solution? Thanks.

11 years, 10 months

← Newer
1
2
3
4
5
6
7
8
9
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-dev February 2013