linaro-kernel

linaro-kernel@lists.linaro.org

10095 discussions

[RFC] Comparison of power-efficient scheduling patch sets

by Morten Rasmussen

Hi, A number of patch sets related to power-efficient scheduling have been posted over the last couple of months. Most of them do not have much data to back them up, so I decided to do some testing. Common for all of the patch sets that I have tested, except one, is that they attempt to pack tasks on as few cpus as possible to allow the remaining cpus to enter deeper sleep states - a strategy that should make sense on most platforms that support per-cpu power gating and multi-socket machines. Kernel: 3.9 Patch sets: rlb-v4: sched: use runnable load based balance (Alex Shi) <https://lkml.org/lkml/2013/4/27/13> pas-v7: sched: power aware scheduling (Alex Shi) <https://lkml.org/lkml/2013/4/3/732> pst-v3: sched: packing small tasks (Vincent Guittot) <https://lkml.org/lkml/2013/3/22/183> pst-v4: sched: packing small tasks (Vincent Guittot) <https://lkml.org/lkml/2013/4/25/396> Configuration: pas-v7: Set to "powersaving" mode. pst-v4: Set to "Full" packing mode. Platform: ARM TC2 (test-chip), 2xCortex-A15 + 3xCortex-A7. Cortex-A15s disabled. Measurement technique: Time spent non-idle (not in idle state) for each cpu based on cpuidle ftrace events. TC2 does not have per-core power-gating, so packing inside the A7 cluster does not lead to any significant power savings. Note that any product grade hardware (TC2 is a test-chip) will very likely have per-core power-gating, so in those cases packing will have an appreciable effect on power savings. Measuring non-idle time rather than power should give a more clear idea about the effect of the patch sets given that the idle back-end is highly implementation specific. Benchmarks: audio playback (Android): 30s mp3 file playback on Android. bbench+audio (Android): Web page rendering while doing mp3 playback. andebench_native (Android): Android benchmark running in native mode. cyclictest: Short periodic tasks. Results: Two runs for each patch set. audio playback (Android) SMP non-idle % cpu 0 cpu 1 cpu 2 3.9_1 11.96 2.86 2.48 3.9_2 12.64 2.81 1.88 rlb-v4_1 12.61 2.44 1.90 rlb-v4_2 12.45 2.44 1.90 pas-v7_1 16.17 0.03 0.24 pas-v7_2 16.08 0.28 0.07 pst-v3_1 15.18 2.76 1.70 pst-v3_2 15.13 0.80 0.38 pst-v4_1 16.14 0.05 0.00 pst-v4_2 16.34 0.06 0.00 bbench+audio (Android) SMP non-idle % cpu 0 cpu 1 cpu 2 render time 3.9_1 25.00 20.73 21.22 812 3.9_2 24.29 19.78 22.34 795 rlb-v4_1 23.84 19.36 22.74 782 rlb-v4_2 24.07 19.36 22.74 797 pas-v7_1 28.29 17.86 16.01 869 pas-v7_2 28.62 18.54 15.05 908 pst-v3_1 29.14 20.59 21.72 830 pst-v3_2 27.69 18.81 20.06 830 pst-v4_1 42.20 13.63 2.29 880 pst-v4_2 41.56 14.40 2.17 935 andebench_native (8 threads) (Android) SMP non-idle % cpu 0 cpu 1 cpu 2 Score 3.9_1 99.22 98.88 99.61 4139 3.9_2 99.56 99.31 99.46 4148 rlb-v4_1 99.49 99.61 99.53 4153 rlb-v4_2 99.56 99.61 99.53 4149 pas-v7_1 99.53 99.59 99.29 4149 pas-v7_2 99.42 99.63 99.48 4150 pst-v3_1 97.89 99.33 99.42 4097 pst-v3_2 99.16 99.62 99.42 4097 pst-v4_1 99.34 99.01 99.59 4146 pst-v4_2 99.49 99.52 99.20 4146 cyclictest SMP non-idle % cpu 0 cpu 1 cpu 2 3.9_1 9.13 8.88 8.41 3.9_2 10.27 8.02 6.30 rlb-v4_1 8.88 8.09 8.11 rlb-v4_2 8.49 8.09 8.11 pas-v7_1 10.20 0.02 11.50 pas-v7_2 7.86 14.31 0.02 pst-v3_1 20.44 8.68 7.97 pst-v3_2 20.41 0.78 1.00 pst-v4_1 21.32 0.21 0.05 pst-v4_2 21.56 0.21 0.04 Overall, pas-v7 seems to do a fairly good job at packing. The idle time distribution seems to be somewhere between pst-v3 and the more aggressive pst-v4 for all the benchmarks. pst-v4 manages to keep two cpus nearly idle (<0.25% non-idle) for both cyclictest and audio, which is better than both pst-v3 and pas-v7. pas-v7 fails to pack cyclictest. Packing does come at at cost which can be seen for bbench+audio, where pst-v3 and rlb-v4 get better render times than pas-v7 and pst-v4 which do more aggressive packing. rlb-v4 does not pack, it is only included for reference. >From a packing perspective pst-v4 seems to do the best job for the workloads that I have tested on ARM TC2. The less aggressive packing in pst-v3 may be a better choice for in terms of performance. I'm well aware that these tests are heavily focused on mobile workloads. I would therefore encourage people to share your test results for your workloads on your platforms to complete the picture. Comments are also welcome. Thanks, Morten

13 years, 1 month

[PATCH] ACPI / processor: Remove unused macros in processor_driver.c

by Hanjun Guo

ACPI_PROCESSOR_FILE_INFO, ACPI_PROCESSOR_FILE_THROTTLING and ACPI_PROCESSOR_FILE_LIMIT are used for procfs, but this feature was removed in commit d09fe555 (ACPI processor: remove deprecated ACPI procfs I/F) long ago. So, these macros should also be removed. ACPI_PROCESSOR_LIMIT_USER and ACPI_PROCESSOR_LIMIT_THERMAL are not used by any code, remove them too. Signed-off-by: Hanjun Guo <hanjun.guo(a)linaro.org> --- drivers/acpi/processor_driver.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c index d93963f..823be116 100644 --- a/drivers/acpi/processor_driver.c +++ b/drivers/acpi/processor_driver.c @@ -43,16 +43,10 @@ #define PREFIX "ACPI: " -#define ACPI_PROCESSOR_FILE_INFO "info" -#define ACPI_PROCESSOR_FILE_THROTTLING "throttling" -#define ACPI_PROCESSOR_FILE_LIMIT "limit" #define ACPI_PROCESSOR_NOTIFY_PERFORMANCE 0x80 #define ACPI_PROCESSOR_NOTIFY_POWER 0x81 #define ACPI_PROCESSOR_NOTIFY_THROTTLING 0x82 -#define ACPI_PROCESSOR_LIMIT_USER 0 -#define ACPI_PROCESSOR_LIMIT_THERMAL 1 - #define _COMPONENT ACPI_PROCESSOR_COMPONENT ACPI_MODULE_NAME("processor_driver"); -- 1.7.9.5

13 years, 1 month

[GIT PULL] ARM: CPUFreq updates for 3.11

by Viresh Kumar

Hi Rafael, This is part of my patchset "CPUFreq: Fix {PRE|POST}CHANGE notification sequence". It contains changes specific to cpufreq core and ARM drivers. I have left changes to non-ARM cpufreq drivers from this pull request. See if you can take them for 3.11. The following changes since commit 9e895ace5d82df8929b16f58e9f515f6d54ab82d: Linux 3.10-rc7 (2013-06-22 09:47:31 -1000) are available in the git repository at: git://git.linaro.org/people/vireshk/linux.git cpufreq-fix-notification-arm for you to fetch changes up to f2d65dd0d573e6961165f0be35bac84997b9e707: cpufreq: tegra: call CPUFREQ_POSTCHANGE notfier in error cases (2013-06-24 11:14:24 +0530) ---------------------------------------------------------------- Viresh Kumar (9): cpufreq: make sure frequency transitions are serialized cpufreq: arm-big-little: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: davinci: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: dbx500: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: exynos: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: imx6q: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: omap: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: s3c64xx: call CPUFREQ_POSTCHANGE notfier in error cases cpufreq: tegra: call CPUFREQ_POSTCHANGE notfier in error cases drivers/cpufreq/arm_big_little.c | 4 +--- drivers/cpufreq/cpufreq.c | 9 +++++++++ drivers/cpufreq/davinci-cpufreq.c | 3 +++ drivers/cpufreq/dbx500-cpufreq.c | 4 ++-- drivers/cpufreq/exynos-cpufreq.c | 10 ++++++++-- drivers/cpufreq/imx6q-cpufreq.c | 17 +++++++++++------ drivers/cpufreq/omap-cpufreq.c | 6 +++--- drivers/cpufreq/s3c64xx-cpufreq.c | 8 ++++++-- drivers/cpufreq/tegra-cpufreq.c | 4 ++-- 9 files changed, 45 insertions(+), 20 deletions(-)

13 years, 1 month

[PATCH 1/2] cpufreq: Fix minor formatting issues

by Viresh Kumar

There were few noticeable formatting issues present in core cpufreq code. This patch tries to fix them to make code look better. These are: - Removing few extra blank lines - Adding few blank lines - replacing spaces with tabs - removing unnecessary spaces - rearranging code a bit - Fixing multiline comments - Break lines longer than 80 columns into multi-line. At the end it also updates copyright information in cpufreq.c to include my name for 2013. Signed-off-by: Viresh Kumar <viresh.kumar(a)linaro.org> --- drivers/cpufreq/cpufreq.c | 61 +++++++++++++++-------------------- drivers/cpufreq/cpufreq_governor.h | 4 +-- drivers/cpufreq/cpufreq_performance.c | 4 --- drivers/cpufreq/cpufreq_powersave.c | 6 ++-- drivers/cpufreq/cpufreq_stats.c | 4 +-- drivers/cpufreq/cpufreq_userspace.c | 4 --- include/linux/cpufreq.h | 41 ++++++++++++----------- 7 files changed, 52 insertions(+), 72 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 2d53f47..7c38842 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -3,6 +3,7 @@ * * Copyright (C) 2001 Russell King * (C) 2002 - 2003 Dominik Brodowski <linux(a)brodo.de> + * (C) 2013 Viresh Kumar <viresh.kumar(a)linaro.org> * * Oct 2005 - Ashok Raj <ashok.raj(a)intel.com> * Added handling for CPU hotplug @@ -12,7 +13,6 @@ * 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) KBUILD_MODNAME ": " fmt @@ -41,11 +41,12 @@ */ static struct cpufreq_driver *cpufreq_driver; static DEFINE_PER_CPU(struct cpufreq_policy *, cpufreq_cpu_data); +static DEFINE_RWLOCK(cpufreq_driver_lock); + #ifdef CONFIG_HOTPLUG_CPU /* This one keeps track of the previously set governor of a removed CPU */ static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor); #endif -static DEFINE_RWLOCK(cpufreq_driver_lock); /* * cpu_policy_rwsem is a per CPU reader-writer semaphore designed to cure @@ -150,7 +151,6 @@ static struct cpufreq_policy *__cpufreq_cpu_get(unsigned int cpu, bool sysfs) if (!try_module_get(cpufreq_driver->owner)) goto err_out_unlock; - /* get the CPU */ data = per_cpu(cpufreq_cpu_data, cpu); @@ -220,7 +220,7 @@ static void cpufreq_cpu_put_sysfs(struct cpufreq_policy *data) */ #ifndef CONFIG_SMP static unsigned long l_p_j_ref; -static unsigned int l_p_j_ref_freq; +static unsigned int l_p_j_ref_freq; static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) { @@ -233,7 +233,7 @@ static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) pr_debug("saving %lu as reference value for loops_per_jiffy; " "freq is %u kHz\n", l_p_j_ref, l_p_j_ref_freq); } - if ((val == CPUFREQ_POSTCHANGE && ci->old != ci->new) || + if ((val == CPUFREQ_POSTCHANGE && ci->old != ci->new) || (val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) { loops_per_jiffy = cpufreq_scale(l_p_j_ref, l_p_j_ref_freq, ci->new); @@ -248,7 +248,6 @@ static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) } #endif - void __cpufreq_notify_transition(struct cpufreq_policy *policy, struct cpufreq_freqs *freqs, unsigned int state) { @@ -294,6 +293,7 @@ void __cpufreq_notify_transition(struct cpufreq_policy *policy, break; } } + /** * cpufreq_notify_transition - call notifier chain and adjust_jiffies * on frequency transition. @@ -311,7 +311,6 @@ void cpufreq_notify_transition(struct cpufreq_policy *policy, EXPORT_SYMBOL_GPL(cpufreq_notify_transition); - /********************************************************************* * SYSFS INTERFACE * *********************************************************************/ @@ -376,7 +375,6 @@ out: return err; } - /** * cpufreq_per_cpu_attr_read() / show_##file_name() - * print out cpufreq information @@ -441,7 +439,6 @@ static ssize_t show_cpuinfo_cur_freq(struct cpufreq_policy *policy, return sprintf(buf, "%u\n", cur_freq); } - /** * show_scaling_governor - show the current policy for the specified CPU */ @@ -457,7 +454,6 @@ static ssize_t show_scaling_governor(struct cpufreq_policy *policy, char *buf) return -EINVAL; } - /** * store_scaling_governor - store policy for the specified CPU */ @@ -480,8 +476,10 @@ static ssize_t store_scaling_governor(struct cpufreq_policy *policy, &new_policy.governor)) return -EINVAL; - /* Do not use cpufreq_set_policy here or the user_policy.max - will be wrongly overridden */ + /* + * Do not use cpufreq_set_policy here or the user_policy.max + * will be wrongly overridden + */ ret = __cpufreq_set_policy(policy, &new_policy); policy->user_policy.policy = policy->policy; @@ -1005,7 +1003,8 @@ static void update_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu) * Caller should already have policy_rwsem in write mode for this CPU. * This routine frees the rwsem before returning. */ -static int __cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) +static int __cpufreq_remove_dev(struct device *dev, + struct subsys_interface *sif) { unsigned int cpu = dev->id, ret, cpus; unsigned long flags; @@ -1112,7 +1111,6 @@ static int __cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif return 0; } - static int cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) { unsigned int cpu = dev->id; @@ -1125,7 +1123,6 @@ static int cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) return retval; } - static void handle_update(struct work_struct *work) { struct cpufreq_policy *policy = @@ -1136,7 +1133,8 @@ static void handle_update(struct work_struct *work) } /** - * cpufreq_out_of_sync - If actual and saved CPU frequency differs, we're in deep trouble. + * cpufreq_out_of_sync - If actual and saved CPU frequency differs, we're + * in deep trouble. * @cpu: cpu number * @old_freq: CPU frequency the kernel thinks the CPU runs at * @new_freq: CPU frequency the CPU actually runs at @@ -1151,7 +1149,6 @@ static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, struct cpufreq_freqs freqs; unsigned long flags; - pr_debug("Warning: CPU frequency out of sync: cpufreq and timing " "core thinks of %u, is %u kHz.\n", old_freq, new_freq); @@ -1166,7 +1163,6 @@ static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE); } - /** * cpufreq_quick_get - get the CPU frequency (in kHz) from policy->cur * @cpu: CPU number @@ -1212,7 +1208,6 @@ unsigned int cpufreq_quick_get_max(unsigned int cpu) } EXPORT_SYMBOL(cpufreq_quick_get_max); - static unsigned int __cpufreq_get(unsigned int cpu) { struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); @@ -1271,7 +1266,6 @@ static struct subsys_interface cpufreq_interface = { .remove_dev = cpufreq_remove_dev, }; - /** * cpufreq_bp_suspend - Prepare the boot CPU for system suspend. * @@ -1408,11 +1402,10 @@ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) } EXPORT_SYMBOL(cpufreq_register_notifier); - /** * cpufreq_unregister_notifier - unregister a driver with cpufreq * @nb: notifier block to be unregistered - * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER + * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER * * Remove a driver from the CPU frequency notifier list. * @@ -1448,7 +1441,6 @@ EXPORT_SYMBOL(cpufreq_unregister_notifier); * GOVERNORS * *********************************************************************/ - int __cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) @@ -1581,7 +1573,6 @@ static int __cpufreq_governor(struct cpufreq_policy *policy, return ret; } - int cpufreq_register_governor(struct cpufreq_governor *governor) { int err; @@ -1606,7 +1597,6 @@ int cpufreq_register_governor(struct cpufreq_governor *governor) } EXPORT_SYMBOL_GPL(cpufreq_register_governor); - void cpufreq_unregister_governor(struct cpufreq_governor *governor) { #ifdef CONFIG_HOTPLUG_CPU @@ -1636,7 +1626,6 @@ void cpufreq_unregister_governor(struct cpufreq_governor *governor) EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); - /********************************************************************* * POLICY INTERFACE * *********************************************************************/ @@ -1665,7 +1654,6 @@ int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu) } EXPORT_SYMBOL(cpufreq_get_policy); - /* * data : current policy. * policy : policy to be set. @@ -1699,8 +1687,10 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, blocking_notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_INCOMPATIBLE, policy); - /* verify the cpu speed can be set within this limit, - which might be different to the first one */ + /* + * verify the cpu speed can be set within this limit, which might be + * different to the first one + */ ret = cpufreq_driver->verify(policy); if (ret) goto error_out; @@ -1802,8 +1792,10 @@ int cpufreq_update_policy(unsigned int cpu) policy.policy = data->user_policy.policy; policy.governor = data->user_policy.governor; - /* BIOS might change freq behind our back - -> ask driver for current freq and notify governors about a change */ + /* + * BIOS might change freq behind our back + * -> ask driver for current freq and notify governors about a change + */ if (cpufreq_driver->get) { policy.cur = cpufreq_driver->get(cpu); if (!data->cur) { @@ -1852,7 +1844,7 @@ static int __cpuinit cpufreq_cpu_callback(struct notifier_block *nfb, } static struct notifier_block __refdata cpufreq_cpu_notifier = { - .notifier_call = cpufreq_cpu_callback, + .notifier_call = cpufreq_cpu_callback, }; /********************************************************************* @@ -1864,7 +1856,7 @@ static struct notifier_block __refdata cpufreq_cpu_notifier = { * @driver_data: A struct cpufreq_driver containing the values# * submitted by the CPU Frequency driver. * - * Registers a CPU Frequency driver to this core code. This code + * Registers a CPU Frequency driver to this core code. This code * returns zero on success, -EBUSY when another driver got here first * (and isn't unregistered in the meantime). * @@ -1931,11 +1923,10 @@ err_null_driver: } EXPORT_SYMBOL_GPL(cpufreq_register_driver); - /** * cpufreq_unregister_driver - unregister the current CPUFreq driver * - * Unregister the current CPUFreq driver. Only call this if you have + * Unregister the current CPUFreq driver. Only call this if you have * the right to do so, i.e. if you have succeeded in initialising before! * Returns zero if successful, and -EINVAL if the cpufreq_driver is * currently not initialised. diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h index e16a961..72e7994 100644 --- a/drivers/cpufreq/cpufreq_governor.h +++ b/drivers/cpufreq/cpufreq_governor.h @@ -81,7 +81,7 @@ static ssize_t show_##file_name##_gov_sys \ return sprintf(buf, "%u\n", tuners->file_name); \ } \ \ -static ssize_t show_##file_name##_gov_pol \ +static ssize_t show_##file_name##_gov_pol \ (struct cpufreq_policy *policy, char *buf) \ { \ struct dbs_data *dbs_data = policy->governor_data; \ @@ -91,7 +91,7 @@ static ssize_t show_##file_name##_gov_pol \ #define store_one(_gov, file_name) \ static ssize_t store_##file_name##_gov_sys \ -(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) \ +(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) \ { \ struct dbs_data *dbs_data = _gov##_dbs_cdata.gdbs_data; \ return store_##file_name(dbs_data, buf, count); \ diff --git a/drivers/cpufreq/cpufreq_performance.c b/drivers/cpufreq/cpufreq_performance.c index ceee068..9fef7d6 100644 --- a/drivers/cpufreq/cpufreq_performance.c +++ b/drivers/cpufreq/cpufreq_performance.c @@ -17,7 +17,6 @@ #include <linux/cpufreq.h> #include <linux/init.h> - static int cpufreq_governor_performance(struct cpufreq_policy *policy, unsigned int event) { @@ -44,19 +43,16 @@ struct cpufreq_governor cpufreq_gov_performance = { .owner = THIS_MODULE, }; - static int __init cpufreq_gov_performance_init(void) { return cpufreq_register_governor(&cpufreq_gov_performance); } - static void __exit cpufreq_gov_performance_exit(void) { cpufreq_unregister_governor(&cpufreq_gov_performance); } - MODULE_AUTHOR("Dominik Brodowski <linux(a)brodo.de>"); MODULE_DESCRIPTION("CPUfreq policy governor 'performance'"); MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/cpufreq_powersave.c b/drivers/cpufreq/cpufreq_powersave.c index 2d948a1..32109a1 100644 --- a/drivers/cpufreq/cpufreq_powersave.c +++ b/drivers/cpufreq/cpufreq_powersave.c @@ -1,7 +1,7 @@ /* - * linux/drivers/cpufreq/cpufreq_powersave.c + * linux/drivers/cpufreq/cpufreq_powersave.c * - * Copyright (C) 2002 - 2003 Dominik Brodowski <linux(a)brodo.de> + * Copyright (C) 2002 - 2003 Dominik Brodowski <linux(a)brodo.de> * * * This program is free software; you can redistribute it and/or modify @@ -48,13 +48,11 @@ static int __init cpufreq_gov_powersave_init(void) return cpufreq_register_governor(&cpufreq_gov_powersave); } - static void __exit cpufreq_gov_powersave_exit(void) { cpufreq_unregister_governor(&cpufreq_gov_powersave); } - MODULE_AUTHOR("Dominik Brodowski <linux(a)brodo.de>"); MODULE_DESCRIPTION("CPUfreq policy governor 'powersave'"); MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c index fb65dec..6d35caa 100644 --- a/drivers/cpufreq/cpufreq_stats.c +++ b/drivers/cpufreq/cpufreq_stats.c @@ -27,7 +27,7 @@ static spinlock_t cpufreq_stats_lock; struct cpufreq_stats { unsigned int cpu; unsigned int total_trans; - unsigned long long last_time; + unsigned long long last_time; unsigned int max_state; unsigned int state_num; unsigned int last_index; @@ -116,7 +116,7 @@ static ssize_t show_trans_table(struct cpufreq_policy *policy, char *buf) len += snprintf(buf + len, PAGE_SIZE - len, "%9u: ", stat->freq_table[i]); - for (j = 0; j < stat->state_num; j++) { + for (j = 0; j < stat->state_num; j++) { if (len >= PAGE_SIZE) break; len += snprintf(buf + len, PAGE_SIZE - len, "%9u ", diff --git a/drivers/cpufreq/cpufreq_userspace.c b/drivers/cpufreq/cpufreq_userspace.c index bbeb9c0..c4f2dc3 100644 --- a/drivers/cpufreq/cpufreq_userspace.c +++ b/drivers/cpufreq/cpufreq_userspace.c @@ -104,7 +104,6 @@ static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq) return ret; } - static ssize_t show_speed(struct cpufreq_policy *policy, char *buf) { return sprintf(buf, "%u\n", per_cpu(cpu_cur_freq, policy->cpu)); @@ -185,7 +184,6 @@ static int cpufreq_governor_userspace(struct cpufreq_policy *policy, return rc; } - #ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE static #endif @@ -202,13 +200,11 @@ static int __init cpufreq_gov_userspace_init(void) return cpufreq_register_governor(&cpufreq_gov_userspace); } - static void __exit cpufreq_gov_userspace_exit(void) { cpufreq_unregister_governor(&cpufreq_gov_userspace); } - MODULE_AUTHOR("Dominik Brodowski <linux(a)brodo.de>, " "Russell King <rmk(a)arm.linux.org.uk>"); MODULE_DESCRIPTION("CPUfreq policy governor 'userspace'"); diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 037d36a..dc204c3 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -1,8 +1,8 @@ /* - * linux/include/linux/cpufreq.h + * linux/include/linux/cpufreq.h * - * Copyright (C) 2001 Russell King - * (C) 2002 - 2003 Dominik Brodowski <linux(a)brodo.de> + * Copyright (C) 2001 Russell King + * (C) 2002 - 2003 Dominik Brodowski <linux(a)brodo.de> * * 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 @@ -26,7 +26,6 @@ /* Print length for names. Extra 1 space for accomodating '\n' in prints */ #define CPUFREQ_NAME_PLEN (CPUFREQ_NAME_LEN + 1) - /********************************************************************* * CPUFREQ NOTIFIER INTERFACE * *********************************************************************/ @@ -148,17 +147,18 @@ struct cpufreq_freqs { u8 flags; /* flags of cpufreq_driver, see below. */ }; - /** - * cpufreq_scale - "old * mult / div" calculation for large values (32-bit-arch safe) + * cpufreq_scale - "old * mult / div" calculation for large values (32-bit-arch + * safe) * @old: old value * @div: divisor * @mult: multiplier * * - * new = old * mult / div + * new = old * mult / div */ -static inline unsigned long cpufreq_scale(unsigned long old, u_int div, u_int mult) +static inline unsigned long cpufreq_scale(unsigned long old, u_int div, + u_int mult) { #if BITS_PER_LONG == 32 @@ -211,14 +211,12 @@ extern int __cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation); - extern int __cpufreq_driver_getavg(struct cpufreq_policy *policy, unsigned int cpu); int cpufreq_register_governor(struct cpufreq_governor *governor); void cpufreq_unregister_governor(struct cpufreq_governor *governor); - /********************************************************************* * CPUFREQ DRIVER INTERFACE * *********************************************************************/ @@ -229,7 +227,7 @@ void cpufreq_unregister_governor(struct cpufreq_governor *governor); struct freq_attr; struct cpufreq_driver { - struct module *owner; + struct module *owner; char name[CPUFREQ_NAME_LEN]; u8 flags; /* @@ -277,11 +275,11 @@ struct cpufreq_driver { int cpufreq_register_driver(struct cpufreq_driver *driver_data); int cpufreq_unregister_driver(struct cpufreq_driver *driver_data); - void cpufreq_notify_transition(struct cpufreq_policy *policy, struct cpufreq_freqs *freqs, unsigned int state); -static inline void cpufreq_verify_within_limits(struct cpufreq_policy *policy, unsigned int min, unsigned int max) +static inline void cpufreq_verify_within_limits(struct cpufreq_policy *policy, + unsigned int min, unsigned int max) { if (policy->min < min) policy->min = min; @@ -342,7 +340,9 @@ int cpufreq_update_policy(unsigned int cpu); bool have_governor_per_policy(void); #ifdef CONFIG_CPU_FREQ -/* query the current CPU frequency (in kHz). If zero, cpufreq couldn't detect it */ +/* + * query the current CPU frequency (in kHz). If zero, cpufreq couldn't detect it + */ unsigned int cpufreq_get(unsigned int cpu); #else static inline unsigned int cpufreq_get(unsigned int cpu) @@ -351,7 +351,9 @@ static inline unsigned int cpufreq_get(unsigned int cpu) } #endif -/* query the last known CPU freq (in kHz). If zero, cpufreq couldn't detect it */ +/* + * query the last known CPU freq (in kHz). If zero, cpufreq couldn't detect it + */ #ifdef CONFIG_CPU_FREQ unsigned int cpufreq_quick_get(unsigned int cpu); unsigned int cpufreq_quick_get_max(unsigned int cpu); @@ -366,16 +368,14 @@ static inline unsigned int cpufreq_quick_get_max(unsigned int cpu) } #endif - /********************************************************************* * CPUFREQ DEFAULT GOVERNOR * *********************************************************************/ - /* - Performance governor is fallback governor if any other gov failed to - auto load due latency restrictions -*/ + * Performance governor is fallback governor if any other gov failed to auto + * load due latency restrictions + */ #ifdef CONFIG_CPU_FREQ_GOV_PERFORMANCE extern struct cpufreq_governor cpufreq_gov_performance; #endif @@ -395,7 +395,6 @@ extern struct cpufreq_governor cpufreq_gov_conservative; #define CPUFREQ_DEFAULT_GOVERNOR (&cpufreq_gov_conservative) #endif - /********************************************************************* * FREQUENCY TABLE HELPERS * *********************************************************************/ -- 1.7.12.rc2.18.g61b472e

13 years, 1 month

[PATCH v2] Documentation / cpu hotplug: Rephrase the outdated description for MADT entries

by Hanjun Guo

More than 256 entries in ACPI MADT is supported from ACPI 3.0, so the information should be updated. -v2: Rephrase the information instead of simply removing it according to Rafael's advice. Suggested-by: Rafael J. Wysocki <rjw(a)sisk.pl> Signed-off-by: Hanjun Guo <hanjun.guo(a)linaro.org> --- Documentation/cpu-hotplug.txt | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt index 9f40135..67f733d 100644 --- a/Documentation/cpu-hotplug.txt +++ b/Documentation/cpu-hotplug.txt @@ -370,8 +370,10 @@ A: There is no clear spec defined way from ACPI that can give us that CPUs in MADT as hotpluggable CPUS. In the case there are no disabled CPUS we assume 1/2 the number of CPUs currently present can be hotplugged. - Caveat: Today's ACPI MADT can only provide 256 entries since the apicid field - in MADT is only 8 bits. + Caveat: ACPI MADT can only provide 256 entries in systems with only ACPI 2.0c + or earlier ACPI version supported, because the apicid field in MADT is only + 8 bits. From ACPI 3.0, this limitation was removed since the apicid field + was extended to 32 bits with x2APIC introduced. User Space Notification -- 1.7.9.5

13 years, 1 month

[GIT PULL] CPUFreq ARM Fixes for 3.11

by Viresh Kumar

Hi Rafael, The following changes since commit 9e895ace5d82df8929b16f58e9f515f6d54ab82d: Linux 3.10-rc7 (2013-06-22 09:47:31 -1000) are available in the git repository at: git://git.linaro.org/people/vireshk/linux.git cpufreq-next for you to fetch changes up to 166b9addd83aaf6eb22d9db7dc015f81913c9a0e: cpufreq: s3c2416: fix forgotten driver_data conversions (2013-06-24 11:22:33 +0530) ---------------------------------------------------------------- Heiko Stübner (1): cpufreq: s3c2416: fix forgotten driver_data conversions drivers/cpufreq/s3c2416-cpufreq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

13 years, 1 month

[PATCH] arm64: Add support to pass earlyprintk argument via device tree

by Pranavkumar Sawargaonkar

This patch adds support for defining and passing earlyprintk related information i.e. device and address information via device tree by adding it inside "chosen" node. This will help user to just specify "earlyprintk" from bootargs without actually knowing the address and device to enable earlyprintk. Mechanism: One can just append earlyprintk=device-type,address (same as we pass through command line) in "/chosen" node to notify kernel which is the earlyprintk device and what is its address. Backward Compatibility: This patch also allows existing method of specifying earlyprintk parameter via bootargs. Existing method i.e. passing via bootargs will still have precedence over device tree i.e. if one specifies earlyprintk=device-type,address in bootargs then kernel will use information from bootargs instead of device tree. If user just specifies earlyprintk (without =...) then kernel will look for device tree earlyprintk parameter. Signed-off-by: Pranavkumar Sawargaonkar <pranavkumar(a)linaro.org> Signed-off-by: Anup Patel <anup.patel(a)linaro.org> --- arch/arm64/kernel/early_printk.c | 7 +++++++ arch/arm64/kernel/setup.c | 21 ++++++++++++++++++++- 2 files changed, 27 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kernel/early_printk.c b/arch/arm64/kernel/early_printk.c index fbb6e18..4e6f845 100644 --- a/arch/arm64/kernel/early_printk.c +++ b/arch/arm64/kernel/early_printk.c @@ -29,6 +29,8 @@ static void __iomem *early_base; static void (*printch)(char ch); +extern char *earlyprintk_dt_args; + /* * PL011 single character TX. */ @@ -116,6 +118,11 @@ static int __init setup_early_printk(char *buf) phys_addr_t paddr = 0; if (!buf) { + /* Try to check if Device Tree has this argument or not ? */ + buf = earlyprintk_dt_args; + } + + if (!buf) { pr_warning("No earlyprintk arguments passed.\n"); return 0; } diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 6a9a532..fb2d56f 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -60,6 +60,8 @@ EXPORT_SYMBOL(processor_id); unsigned int elf_hwcap __read_mostly; EXPORT_SYMBOL_GPL(elf_hwcap); +char *earlyprintk_dt_args; + static const char *cpu_name; static const char *machine_name; phys_addr_t __fdt_pointer __initdata; @@ -122,6 +124,23 @@ static void __init setup_processor(void) elf_hwcap = 0; } +int __init early_init_dt_scan_chosen_arm64(unsigned long node, + const char *uname, + int depth, void *data) +{ + char *prop; + + /* Check if this is chosen node */ + if (early_init_dt_scan_chosen(node, uname, depth, data) == 0) + return 0; + + prop = of_get_flat_dt_prop(node, "earlyprintk", NULL); + if (prop) + earlyprintk_dt_args = prop; + + return 1; +} + static void __init setup_machine_fdt(phys_addr_t dt_phys) { struct boot_param_header *devtree; @@ -165,7 +184,7 @@ static void __init setup_machine_fdt(phys_addr_t dt_phys) pr_info("Machine: %s\n", machine_name); /* Retrieve various information from the /chosen node */ - of_scan_flat_dt(early_init_dt_scan_chosen, boot_command_line); + of_scan_flat_dt(early_init_dt_scan_chosen_arm64, boot_command_line); /* Initialize {size,address}-cells info */ of_scan_flat_dt(early_init_dt_scan_root, NULL); /* Setup memory, calling early_init_dt_add_memory_arch */ -- 1.7.9.5

13 years, 1 month

[RFC patch 0/2] ARM: at91: cpuidle: move driver to drivers/cpuidle

by Daniel Lezcano

The number of cpuidle drivers is increasing more and more. Today we have in total 24 drivers. A lot of the code is duplicated, at least for the initialization. A work of consolidation has been done during this year: * a lot of code cleanup in all the drivers * time keeping is now part of the framework * timer broadcast is now part of the framework * a WFI state function for ARM is defined and used in the drivers * an init function has been proposed to factor out the initialization across the drivers (patchset pending) What has been observed is a lot of code duplicationis, modification made in the framework takes awhile before reaching the driver which uses an old API, duplicate routine and bugs, etc ... It appears the drivers are belonging to different trees, the cpuidle framework is under linux-pm, the drivers are per SoC tree. The communication is made difficult because of the different mailing lists where the cpuidle are submitted. After this work, it is time to prevent all these problems to occur again. I propose to move the cpuidle drivers to the drivers/cpuidle directory, hence having one single submission path for cpuidle in order to have the cpuidle framework and the different drivers synced. This series move the AT91 cpuidle driver under drivers/cpuidle. That does not change the rule to have the patches acked-by the author of the driver. Note the calxeda and kirkwood drivers are now in drivers/cpuidle. Daniel Lezcano (2): ARM: at91: cpuidle: encapsulate the standby code ARM: at91: cpuidle: move the driver to drivers/cpuidle directory arch/arm/mach-at91/Makefile | 1 - arch/arm/mach-at91/cpuidle.c | 66 ------------------------------------------ arch/arm/mach-at91/pm.c | 8 ++++- drivers/cpuidle/Makefile | 1 + drivers/cpuidle/at91.c | 55 +++++++++++++++++++++++++++++++++++ 5 files changed, 63 insertions(+), 68 deletions(-) delete mode 100644 arch/arm/mach-at91/cpuidle.c create mode 100644 drivers/cpuidle/at91.c -- 1.7.9.5

13 years, 1 month

[PATCH] tick: fix tick_broadcast_pending_mask not cleared

by Daniel Lezcano

The recent modification in the cpuidle framework consolidated the timer broadcast code across the different drivers by setting a new flag in the idle state. It tells the cpuidle core code to enter/exit to the broadcast mode for the cpu when entering a deep idle state. The broadcast timer enter/exit is no longer handled by the back-end driver. This change made the local interrupt to be enabled *before* calling CLOCK_EVENT_NOTIFY_EXIT. bad or not (see below) ? On a tegra114, a four cores system, when the flag has been introduced in the driver, the following warning appeared: [ 25.629559] WARNING: CPU: 2 PID: 0 at kernel/time/tick-broadcast.c:578 tick_broadcast_oneshot_control +0x1a4/0x1d0() [ 25.629565] Modules linked in: [ 25.629574] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 3.10.0-rc3-next-20130529+ #15 [ 25.629601] [<c00148f4>] (unwind_backtrace+0x0/0xf8) from [<c0011040>] (show_stack+0x10/0x14) [ 25.629616] [<c0011040>] (show_stack+0x10/0x14) from [<c0504248>] (dump_stack+0x80/0xc4) [ 25.629633] [<c0504248>] (dump_stack+0x80/0xc4) from [<c00231ac>] (warn_slowpath_common+0x64/0x88) [ 25.629646] [<c00231ac>] (warn_slowpath_common+0x64/0x88) from [<c00231ec>] (warn_slowpath_null+0x1c/0x24) [ 25.629657] [<c00231ec>] (warn_slowpath_null+0x1c/0x24) from [<c00667f8>] (tick_broadcast_oneshot_control+0x1a4/0x1d0) [ 25.629670] [<c00667f8>] (tick_broadcast_oneshot_control+0x1a4/0x1d0) from [<c0065cd0>] (tick_notify+0x240/0x40c) [ 25.629685] [<c0065cd0>] (tick_notify+0x240/0x40c) from [<c0044724>] (notifier_call_chain+0x44/0x84) [ 25.629699] [<c0044724>] (notifier_call_chain+0x44/0x84) from [<c0044828>] (raw_notifier_call_chain+0x18/0x20) [ 25.629712] [<c0044828>] (raw_notifier_call_chain+0x18/0x20) from [<c00650cc>] (clockevents_notify+0x28/0x170) [ 25.629726] [<c00650cc>] (clockevents_notify+0x28/0x170) from [<c033f1f0>] (cpuidle_idle_call+0x11c/0x168) [ 25.629739] [<c033f1f0>] (cpuidle_idle_call+0x11c/0x168) from [<c000ea94>] (arch_cpu_idle+0x8/0x38) [ 25.629755] [<c000ea94>] (arch_cpu_idle+0x8/0x38) from [<c005ea80>] (cpu_startup_entry+0x60/0x134) [ 25.629767] [<c005ea80>] (cpu_startup_entry+0x60/0x134) from [<804fe9a4>] (0x804fe9a4) [ 25.629772] ---[ end trace 5484e77e2531bccc ]--- I don't have the hardware, so I wasn't able to reproduce the warning but after looking a while in the code, I deduced the following: 1. the CPU2 enters a deep idle state and sets the broadcast timer 2. the timer expires, the tick_handle_oneshot_broadcast function is called, setting the tick_broadcast_pending_mask and waking up the idle cpu CPU2 3. the CPU2 exits idle and invokes tick_broadcast_oneshot_control with CLOCK_EVENT_NOTIFY_EXIT with the following code: [...] if (dev->next_event.tv64 == KTIME_MAX) goto out; if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_pending_mask)) goto out; [...] 4. if there is no next event planned for CPU2, we fulfil the first condition and we jump to the 'out' section without clearing the tick_broadcast_pending_mask 5. CPU2 goes to deep idle again and calls tick_broadcast_oneshot_control with CLOCK_NOTIFY_EVENT_ENTER but with the tick_broadcast_pending_mask set for CPU2, leading to the WARNING. Above, it is mentionned the change moved the CLOCK_EVENT_NOTIFY_EXIT after the local interrupt were enabled. If it is before, this warning does not occur. My hypothesis is the code path described before does not happen because when a broadcast timer expires, dev->next_event.tv64 is always different from KTIME_MAX because the timer handler did not set the value yet (local interrupt are still disabled). I don't see anywhere in the code, a clockevents_notify(ENTER/EXIT) block must be done with the local interrupt disabled in between, furthermore the function uses 'raw_spin_lock_irqsave' which make me think, we don't care about that. Invert the conditions and make the tick broadcast code immune from the local interrupts context. Signed-off-by: Daniel Lezcano <daniel.lezcano(a)linaro.org> Reported-by: Joseph Lo <josephl(a)nvidia.com> --- kernel/time/tick-broadcast.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index d067c01..58d88e8 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -610,8 +610,6 @@ void tick_broadcast_oneshot_control(unsigned long reason) } else { if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) { clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); - if (dev->next_event.tv64 == KTIME_MAX) - goto out; /* * The cpu which was handling the broadcast * timer marked this cpu in the broadcast @@ -625,6 +623,8 @@ void tick_broadcast_oneshot_control(unsigned long reason) tick_broadcast_pending_mask)) goto out; + if (dev->next_event.tv64 == KTIME_MAX) + goto out; /* * If the pending bit is not set, then we are * either the CPU handling the broadcast -- 1.7.9.5

13 years, 1 month

[PATCH] arm: l2x0: add support for prefetch and pwr control register configuration

by Chander Kashyap

This patch adds support to configure prefetch and pwr control registers of pl310 cache controllor. Signed-off-by: Chander Kashyap <chander.kashyap(a)linaro.org> --- Documentation/devicetree/bindings/arm/l2cc.txt | 6 ++++ arch/arm/include/asm/hardware/cache-l2x0.h | 23 +++++++++++++ arch/arm/mm/cache-l2x0.c | 44 ++++++++++++++++++++++++ 3 files changed, 73 insertions(+) diff --git a/Documentation/devicetree/bindings/arm/l2cc.txt b/Documentation/devicetree/bindings/arm/l2cc.txt index cbef09b..66876b6 100644 --- a/Documentation/devicetree/bindings/arm/l2cc.txt +++ b/Documentation/devicetree/bindings/arm/l2cc.txt @@ -34,6 +34,12 @@ Optional properties: - arm,filter-ranges : <start length> Starting address and length of window to filter. Addresses in the filter window are directed to the M1 port. Other addresses will go to the M0 port. +- arm,prefetch-control : Prefetch related configurations. Specifies 8 cells of + prefetch-offset, not-same-ID-on-exclusive-sequence-en, incr-double-Linefill-en, + prefetch-drop-en, data-prefetch-en, instruction-prefetch-en and + double-linefill-en. +- arm,pwr-control : Operting mode clock and power mode selection. Specifies two + cells of standby-mode-en and dynamic_clk_gating_en. - interrupts : 1 combined interrupt. - cache-id-part: cache id part number to be used if it is not present on hardware diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h index 3b2c40b..188f4da 100644 --- a/arch/arm/include/asm/hardware/cache-l2x0.h +++ b/arch/arm/include/asm/hardware/cache-l2x0.h @@ -106,6 +106,29 @@ #define L2X0_WAY_SIZE_SHIFT 3 +#define L2X0_PREFETCH_CTRL_OFFSET_SHIFT 0 +#define L2X0_PREFETCH_CTRL_NOT_SAME_ID_SHIFT 21 +#define L2X0_PREFETCH_CTRL_INC_DOUBLE_LINEFILL_SHIFT 23 +#define L2X0_PREFETCH_CTRL_PREFETCH_DROP_SHIFT 24 +#define L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_WR_SHIFT 27 +#define L2X0_PREFETCH_CTRL_DATA_PREFETCH_EN_SHIFT 28 +#define L2X0_PREFETCH_CTRL_INSTR_PREFETCH_EN_SHIFT 29 +#define L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_EN_SHIFT 30 + +#define PREFETCH_R2P0_MASK (~(1 << L2X0_PREFETCH_CTRL_OFFSET_SHIFT)) +#define PREFETCH_R3P0_MASK \ + (~((1 << L2X0_PREFETCH_CTRL_NOT_SAME_ID_SHIFT) | \ + (1 << L2X0_PREFETCH_CTRL_INC_DOUBLE_LINEFILL_SHIFT) | \ + (1 << L2X0_PREFETCH_CTRL_PREFETCH_DROP_SHIFT) | \ + (1 << L2X0_PREFETCH_CTRL_DATA_PREFETCH_EN_SHIFT) | \ + (1 << L2X0_PREFETCH_CTRL_INSTR_PREFETCH_EN_SHIFT) | \ + (1 << L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_EN_SHIFT))) +#define PREFETCH_R3P1_MASK \ + (~(1 << L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_WR_SHIFT)) + +#define L2X0_PWR_CTRL_STANDBY_MODE_EN_SHIFT 0 +#define L2X0_PWR_CTRL_DYNAMIC_CLK_GATE_EN_SHIFT 1 + #ifndef __ASSEMBLY__ extern void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask); #if defined(CONFIG_CACHE_L2X0) && defined(CONFIG_OF) diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c index c465fac..490e182 100644 --- a/arch/arm/mm/cache-l2x0.c +++ b/arch/arm/mm/cache-l2x0.c @@ -563,6 +563,11 @@ static void __init pl310_of_setup(const struct device_node *np, u32 data[3] = { 0, 0, 0 }; u32 tag[3] = { 0, 0, 0 }; u32 filter[2] = { 0, 0 }; + u32 prefetch[8] = { 0, 0, 0, 0, 0, 0, 0, 0 }; + u32 prefetch_val; + u32 pwr[2] = { 0, 0 }; + u32 l2x0_revision = readl_relaxed(l2x0_base + L2X0_CACHE_ID) & + L2X0_CACHE_ID_RTL_MASK; of_property_read_u32_array(np, "arm,tag-latency", tag, ARRAY_SIZE(tag)); if (tag[0] && tag[1] && tag[2]) @@ -589,6 +594,45 @@ static void __init pl310_of_setup(const struct device_node *np, writel_relaxed((filter[0] & ~(SZ_1M - 1)) | L2X0_ADDR_FILTER_EN, l2x0_base + L2X0_ADDR_FILTER_START); } + + if (l2x0_revision >= L2X0_CACHE_ID_RTL_R2P0) { + of_property_read_u32_array(np, "arm,prefetch-control", + prefetch, ARRAY_SIZE(prefetch)); + prefetch_val = readl_relaxed(l2x0_base + L2X0_PREFETCH_CTRL); + prefetch_val &= PREFETCH_R2P0_MASK; + prefetch_val |= prefetch[0] << L2X0_PREFETCH_CTRL_OFFSET_SHIFT; + if (l2x0_revision >= L2X0_CACHE_ID_RTL_R3P0) { + prefetch_val &= PREFETCH_R3P0_MASK; + prefetch_val |= + ((prefetch[1]) << + L2X0_PREFETCH_CTRL_NOT_SAME_ID_SHIFT) | + ((prefetch[2]) << + L2X0_PREFETCH_CTRL_INC_DOUBLE_LINEFILL_SHIFT) | + ((prefetch[3]) << + L2X0_PREFETCH_CTRL_PREFETCH_DROP_SHIFT) | + ((prefetch[5]) << + L2X0_PREFETCH_CTRL_DATA_PREFETCH_EN_SHIFT) | + ((prefetch[6]) << + L2X0_PREFETCH_CTRL_INSTR_PREFETCH_EN_SHIFT) | + ((prefetch[7]) << + L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_EN_SHIFT); + if (l2x0_revision >= L2X0_CACHE_ID_RTL_R3P1) { + prefetch_val &= PREFETCH_R3P1_MASK; + prefetch_val |= + ((prefetch[4]) << + L2X0_PREFETCH_CTRL_DOUBLE_LINEFILL_WR_SHIFT); + } + writel_relaxed(prefetch_val, + l2x0_base + L2X0_PREFETCH_CTRL); + + of_property_read_u32_array(np, "arm,pwr-control", + pwr, ARRAY_SIZE(pwr)); + writel_relaxed( + ((pwr[0]) << L2X0_PWR_CTRL_STANDBY_MODE_EN_SHIFT) | + ((pwr[1]) << L2X0_PWR_CTRL_DYNAMIC_CLK_GATE_EN_SHIFT), + l2x0_base + L2X0_POWER_CTRL); + } + } } static void __init pl310_save(void) -- 1.7.9.5

13 years, 2 months

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

linaro-kernel