linaro-dev

linaro-dev@lists.linaro.org

4332 discussions

[PATCH v6] ARM: imx: Add basic imx6q thermal driver

by Robert Lee

Based on v3.5-rc4 This patch add basic thermal driver support for the imx6q platform. This was implemented by hooking into the linux thermal framework which provides sysfs temperature readings and other data and a method to shutdown the system if it gets too hot. A major change in this revision is that cooling device implementation was removed as it is dependent on cpufreq and no imx6q cpufreq exist upstream and there is no recent activity on efforts to do so. Other changes in v6. 1. Fixed several problems that were pointed out in v5 review. 2. Re-worked the sensor reading code to better handle suspend/resume case 3. Made various naming changes to make it more specific to the anatop peripheral instead of the imx6q platform. This should reduce the changes required to use this driver in future platforms. 4. Removed test support for imx6q parts without calibrated temperature sensor fuses. These devices will simply exit thermal driver initialization with a warning message. 5. Made the thermal driver and the anatopmfd driver defaulted to be enabled in the Kconfig files which will still dependend on imx6q_soc being enabled. Not sure if this is the right place to do this or if it should be done in the defconfig file. Known future work will be to add back in cooling device support once cpufreq is supported. Improvements to coincide with that work may involve adding the active cooling trip points into device tree for hardware application specific settings. link to previous submissions: v5" https://lkml.org/lkml/2012/6/20/57 v4: http://comments.gmane.org/gmane.linux.acpi.devel/51779 v3: http://www.spinics.net/lists/arm-kernel/msg155955.html v2: http://www.spinics.net/lists/arm-kernel/msg155790.html v1: http://www.spinics.net/lists/arm-kernel/msg155111.html Changes in v5: 1. Modified to use anatop mfd driver for accessing anatop registers 2. Made necessary changes to work with latest generic CPU cooling code. 3. Added Config changes and functionality to allow testing on parts without without programmed temperature sensor calibration values. 4. General cleanup and addition of comments. Changes in v4: 1. Removed bad suspend/resume assignment into thermal class. After further examination and discussion with SoC designers, a sequence is now used for making measurements that is is unaffected by system suspendresumes. Temp Sensor automatically powers off in hardware during the low power mode caused by a system suspend. 2. Moved some structures from static to dynamic allocation. 3. Added some noise handling to temperatuer sensor readings. Changes in v3: 1. Fixed the various issues pointed out in v2 2. Made other code cleanup and a bit of re-organizing 3. Removed unecessary platform driver and device. Changes in v2: 1. Cleaned up some style issues pointed out in v1 2. Made various other code cleanup and re-organizing 3. Added temperature sensor calibration 4. Created platform driver and device to hook into pm suspend. Performed some basic testing to ensure proper cooling operating. If you want to test this, full testing requires imx6q cpufreq implementation (not yet in v3.5). Robert Lee (1): ARM: imx: Add basic imx6q thermal driver arch/arm/boot/dts/imx6q.dtsi | 5 + drivers/mfd/Kconfig | 1 + drivers/thermal/Kconfig | 9 + drivers/thermal/Makefile | 1 + drivers/thermal/imx_anatop_thermal.c | 465 ++++++++++++++++++++++++++++++++++ 5 files changed, 481 insertions(+) create mode 100644 drivers/thermal/imx_anatop_thermal.c -- 1.7.10

13 years, 4 months

[PATCH] perf sched replay: fix event lookup

by Dmitry Antipov

Use new function trace_find_event_by_name to lookup events before looking through /sys files. This helps 'perf sched replay' to map event names to IDs correctly when processing perf.data recorded on another machine. Signed-off-by: Dmitry Antipov <dmitry.antipov(a)linaro.org> --- tools/perf/util/evlist.c | 18 ++++++++++++++++-- tools/perf/util/trace-event-parse.c | 4 ++++ tools/perf/util/trace-event.h | 1 + 3 files changed, 21 insertions(+), 2 deletions(-) diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index 4ac5f5a..7ebb9c5 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -17,6 +17,7 @@ #include <unistd.h> #include "parse-events.h" +#include "trace-event.h" #include <sys/mman.h> @@ -231,12 +232,25 @@ int perf_evlist__set_tracepoints_handlers(struct perf_evlist *evlist, const struct perf_evsel_str_handler *assocs, size_t nr_assocs) { + struct event_format *event; struct perf_evsel *evsel; + char *p, *sys, *name; int err; - size_t i; + size_t i, off; for (i = 0; i < nr_assocs; i++) { - err = trace_event__id(assocs[i].name); + err = -ENOENT; + p = strchr(assocs[i].name, ':'); + if (!p) + goto out; + off = p - assocs[i].name; + sys = malloc(off + 1); + memcpy(sys, assocs[i].name, off); + sys[off] = '\0'; + name = p + 1; + event = trace_find_event_by_name(sys, name); + err = event ? event->id : trace_event__id(assocs[i].name); + free(sys); if (err < 0) goto out; diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index df2fddb..44cbb40 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -176,6 +176,10 @@ struct event_format *trace_find_event(int type) return pevent_find_event(pevent, type); } +struct event_format *trace_find_event_by_name(const char *sys, const char *name) +{ + return pevent_find_event_by_name(pevent, sys, name); +} void print_trace_event(int cpu, void *data, int size) { diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h index 639852a..66f83a0 100644 --- a/tools/perf/util/trace-event.h +++ b/tools/perf/util/trace-event.h @@ -40,6 +40,7 @@ int parse_event_file(char *buf, unsigned long size, char *sys); struct pevent_record *trace_peek_data(int cpu); struct event_format *trace_find_event(int type); +struct event_format *trace_find_event_by_name(const char *sys, const char *name); unsigned long long raw_field_value(struct event_format *event, const char *name, void *data); -- 1.7.7.6

13 years, 4 months

Qemu networking

by Simone Ferlin-Oliveira

I want to make network work with Qemu using the tap device. I tried this out using a debian-wheezy ( http://elinux.org/BeagleBoardDebian#Debian_Wheezy_snapshot) and Linaro (nano) (http://www.linaro.org/downloads/1203) for the beagleboard and following some instructions online to set up the tap device. In both cases it did not work and I only see an usb0 device set up but not configured/linked. The eth0 fails while booting. Debugging Qemu while loading I found the following problem. Could you help me with that? 1. tap0 device already set up: 2. Command line: sudo qemu-system-arm -M beagle -m 512 -drive file=beagle-wheezy.img,if=sd,cache=writeback -clock unix -netdev tap,id=mynet,ifname=tap0,script=no -usb -device usb-net,netdev=mynet -serial stdio 3. Debugging Qemu: Cause of crash: s->dev.config IS NULL Program terminated with signal 11, Segmentation fault. #0 0xb74af3f1 in is_rndis (s=0xb9c73c20) at /home/sferlin/qemu-linaro-1.1.0-2012.06/hw/usb/dev-network.c:650 650 return s->dev.config->bConfigurationValue == DEV_RNDIS_CONFIG_VALUE; (gdb) up #1 0xb74b0c24 in usbnet_can_receive (nc=0xb9c763c8) at /home/sferlin/qemu-linaro-1.1.0-2012.06/hw/usb/dev-network.c:1292 1292 if (is_rndis(s) && s->rndis_state != RNDIS_DATA_INITIALIZED) { (gdb) up #2 0xb741bc48 in qemu_can_send_packet (sender=0xb9bf4d20) at net.c:367 367 !sender->peer->info->can_receive(sender->peer)) { (gdb) up #3 0xb7423b4b in tap_can_send (opaque=0xb9bf4d20) at net/tap.c:175 175 return qemu_can_send_packet(&s->nc); (gdb) up #4 0xb738b1b5 in qemu_iohandler_fill (pnfds=0xb7bc3f00, readfds=0xb7bc3d80, writefds=0xb7bc3e00, xfds=0xb7bc3e80) at iohandler.c:101 101 ioh->fd_read_poll(ioh->opaque) != 0)) {

13 years, 4 months

Android Platform Team 2012/6/27 Agenda Posted

by Zach Pfeffer

See: https://wiki.linaro.org/Platform/Android/Meetings/2012-06-27 Please feel free to add to it. Android team, Please list your status. See everyone tomorrow on #linaro-meeting at 14:00 UTC (8 AM Texas time). We've got some cool stuff to go over. Come, enjoy! -- Zach Pfeffer Android Platform Team Lead, Linaro Platform Teams Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

13 years, 4 months

howto turn power management etc off?

by Tom Gall

Hi All, I've been mucking around a bit with linaro server kinds of things. I remember from the last connect Michael Hope had a bit of a checklist before he would collect data. I'd like to make sure I'm getting all I can out of the board. One of the elements of the list was turning off power management. With a little light searching I didn't see anything. More generally, it does seem like a wiki page with a pre-benchmark checklist with things to check/do would be of value. Thanks! -- Regards, Tom "Where's the kaboom!? There was supposed to be an earth-shattering kaboom!" Marvin Martian Linaro.org │ Open source software for ARM SoCs h) tom.gall att linaro.org w) tom_gall att vnet.ibm.com h) tom_gall att mac.com

13 years, 4 months

[PATCH] [CPUFREQ] VCPUFREQ: HR timer based generic cycle stealer

by Pantelis Antoniou

It turns out that hr timers have a nice propery always been executed on the cpu started on. So it is possible to have a generic bigLITTLE simulator based on generic code. Many thanks to Paul E. McKenney that pointed this out. --- drivers/cpufreq/Kconfig | 4 + drivers/cpufreq/Makefile | 1 + drivers/cpufreq/vcpufreq-hrtimer.c | 211 ++++++++++++++++++++++++++++++++++++ 3 files changed, 216 insertions(+), 0 deletions(-) create mode 100644 drivers/cpufreq/vcpufreq-hrtimer.c diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index 1fef0ad..3fa3523 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -211,6 +211,10 @@ if VCPUFREQ choice prompt "VCPUFREQ Platform glue Layer" +config VCPUFREQ_HRTIMER + bool "HRTIMER based VCPUFREQ driver" + depends on HIGH_RES_TIMERS + config VCPUFREQ_OMAP2PLUS bool "OMAP VCPUFREQ driver" depends on ARCH_OMAP2PLUS diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile index 97d3011..128ea25 100644 --- a/drivers/cpufreq/Makefile +++ b/drivers/cpufreq/Makefile @@ -57,3 +57,4 @@ obj-$(CONFIG_CPU_FREQ_MAPLE) += maple-cpufreq.o # Virtual driver obj-$(CONFIG_VCPUFREQ) += vcpufreq.o obj-$(CONFIG_VCPUFREQ_OMAP2PLUS) += vcpufreq-omap.o +obj-$(CONFIG_VCPUFREQ_HRTIMER) += vcpufreq-hrtimer.o diff --git a/drivers/cpufreq/vcpufreq-hrtimer.c b/drivers/cpufreq/vcpufreq-hrtimer.c new file mode 100644 index 0000000..98616a9 --- /dev/null +++ b/drivers/cpufreq/vcpufreq-hrtimer.c @@ -0,0 +1,211 @@ +/* + * Copyright 2012 Pantelis Antoniou <panto(a)antoniou-consulting.com> + * + * Virtual CPUFreq glue driver for OMAP2PLUS + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#define pr_fmt(fmt) "cpufreq: " fmt + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/cpufreq.h> +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/interrupt.h> +#include <linux/irq.h> +#include <linux/math64.h> +#include <linux/delay.h> + +#include <asm/smp_plat.h> +#include <asm/cpu.h> +#include <linux/hrtimer.h> + +#include "vcpufreq.h" + +struct hrtimer_info { + unsigned int cpu; + struct hrtimer hrtimer; + ktime_t interval; + unsigned int irq; + uint64_t counter; + unsigned int timer_rate; + unsigned int hogtime; +}; + +static DEFINE_PER_CPU(struct hrtimer_info, hr_timer); + +static enum hrtimer_restart vcpufreq_hrtimer_handler(struct hrtimer *hrtimer) +{ + struct hrtimer_info __percpu *hti = container_of(hrtimer, struct hrtimer_info, hrtimer); + + BUG_ON(hti == NULL); + + /* we rely on being executed on the proper CPU */ + BUG_ON(hti->cpu != smp_processor_id()); + + hti->counter++; + + /* + * udelay is not always accurate for this; + * unfortunately that is all we have + */ + udelay(hti->hogtime); + + hrtimer_forward_now(hrtimer, hti->interval); + + return HRTIMER_RESTART; +} + +struct vpcufreq_hr_timer_set_freq_info { + unsigned int cpu; + unsigned int new_freq; + unsigned int old_freq; + int ret; +}; + +static void hr_timer_glue_set_freq(void *data) +{ + struct vpcufreq_hr_timer_set_freq_info *info = data; + struct hrtimer_info __percpu *hti; + int ret = 0; + unsigned int hog_timer_rate; + unsigned int freq = vcpufreq_get_maxspeed(); + unsigned int hogtime = vcpufreq_get_hogtime(); + u32 div, rem; + + BUG_ON(smp_processor_id() != info->cpu); + + hti = &per_cpu(hr_timer, info->cpu); + + /* should never happen; checked before */ + BUG_ON(info->new_freq == info->old_freq); + + /* max freq; stop the timer */ + if (info->new_freq == freq) { + pr_debug("#%d: shut down timer\n", info->cpu); + /* no error */ + ret = 0; + goto hr_stop_timer; + + } + + /* timer was stopped, we should start it */ + if (info->old_freq != freq) + hrtimer_cancel(&hti->hrtimer); + + hog_timer_rate = div_u64((u64)(freq - info->new_freq) * 1000000, freq * hogtime); + pr_debug("#%d: hog timer rate = %u\n", info->cpu, hog_timer_rate); + + div = div_u64_rem(1000000000 / hog_timer_rate, 1000000000, &rem); + hti->interval = ktime_set(div, rem); + hti->hogtime = hogtime; + + pr_debug("#%d: internal set to %usec %unsec\n", info->cpu, div, rem); + + /* timer bound to this CPU please */ + ret = hrtimer_start(&hti->hrtimer, hti->interval, HRTIMER_MODE_PINNED); + if (ret != 0) { + pr_err("#%d: failed to start hrtimer (%d)\n", info->cpu, ret); + goto hr_stop_timer; + } + + pr_debug("#%d: setting freq to %u\n", info->cpu, info->new_freq); + + vcpufreq_set_speed(info->cpu, info->new_freq); + ret = 0; + goto out; + +hr_stop_timer: + /* clear everything */ + + hrtimer_cancel(&hti->hrtimer); + + /* always return to max speed here */ + vcpufreq_set_speed(info->cpu, freq); +out: + info->ret = ret; +} + +int vcpufreq_glue_set_freq(unsigned int cpu, unsigned int new_freq, + unsigned int old_freq) +{ + struct vpcufreq_hr_timer_set_freq_info info; + int ret; + + memset(&info, 0, sizeof(info)); + + info.cpu = cpu; + info.new_freq = new_freq; + info.old_freq = old_freq; + info.ret = -EINVAL; + + ret = smp_call_function_single(cpu, hr_timer_glue_set_freq, &info, 1); + if (ret != 0) { + pr_err("#%d: failed to call per CPU glue set freq\n", cpu); + return ret; + } + + if (info.ret != 0) { + pr_err("#%d: failed to set freq\n", cpu); + return info.ret; + } + + return 0; +} + +static void hr_timer_glue_init_cpu(void *info) +{ + struct hrtimer_info __percpu *hti; + struct cpufreq_policy *policy = info; + + BUG_ON(policy == NULL); + + BUG_ON(smp_processor_id() != policy->cpu); + + hti = &per_cpu(hr_timer, policy->cpu); + + hrtimer_init(&hti->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_PINNED); + hti->hrtimer.function = vcpufreq_hrtimer_handler; + + pr_info("#%d: %s\n", policy->cpu, __func__); +} + +int vcpufreq_glue_init(struct cpufreq_policy *policy, int *freq) +{ + struct hrtimer_info __percpu *hti; + int ret = 0; + + BUG_ON(freq == NULL); + + if (*freq == 0) { + *freq = 1000000; /* simulated 1GHz */ + } + + /* initialize per cpu structure */ + hti = &per_cpu(hr_timer, policy->cpu); + memset(hti, 0, sizeof(*hti)); + hti->cpu = policy->cpu; + + ret = smp_call_function_single(policy->cpu, hr_timer_glue_init_cpu, policy, 1); + if (ret != 0) { + pr_err("#%d: failed to call per CPU glue init\n", policy->cpu); + goto error_out; + } + + ret = 0; + +error_out: + return ret; +} + +int vcpufreq_glue_exit(struct cpufreq_policy *policy) +{ + return 0; +} -- 1.7.1

13 years, 4 months

git.linaro.org ?

by Dechesne, Nicolas

hi there, git.linaro.org in my browser is sending me to Launchpad authentication page (https://linaro-private.git.linaro.org/openid/+login). And I can't access the public git trees. is that a known issue? thx, nico

13 years, 4 months

Unable to build linux-linaro-tracking (3.4) with CPU_IDLE

by Ricardo Salveti

Hey Andrey, While looking on publishing the kernel packages for Ubuntu based on llt, we noticed that it's unable to build in case CPU_IDLE is enabled. This is basically because it seems one patch is applied twice, and both adding the same content at the cpuidle.h file: $ git blame drivers/cpuidle/cpuidle.h 78bb15d5 (Colin Cross 2012-05-03 10:23:24 +0800 35) #ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED ... b667844d (Colin Cross 2012-05-03 10:37:35 +0800 65) #ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED Looking further, it seems that the culprit is b667844d, as you can see at http://git.linaro.org/gitweb?p=kernel/linux-linaro-tracking.git;a=commitdif… Guess that we should not only revert it, but maybe fix on the tree that actually pulled the patch in. Thanks, -- Ricardo Salveti de Araujo

13 years, 4 months

Google IO Starts Tomorrow, Sessions Streamed

by Zach Pfeffer

Take a look: https://developers.google.com/events/io/ Not only are there some great sessions, but the way the sessions are done may help us plan our future Connects. -- Zach Pfeffer Android Platform Team Lead, Linaro Platform Teams Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

13 years, 4 months

panda linaro kernel source code link

by maqsood khan

Hello Please send me the panda linaro kernel source code link, If possible to send source code please sende it otherwise send link only. Regards Maqsood khan

13 years, 4 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-dev