summaryrefslogtreecommitdiff
path: root/drivers/cpufreq
AgeCommit message (Collapse)Author
2014-03-26cpufreq: Make cpufreq_notify_transition & cpufreq_notify_post_transition staticViresh Kumar
cpufreq_notify_transition() and cpufreq_notify_post_transition() shouldn't be called directly by cpufreq drivers anymore and so these should be marked static. Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-26cpufreq: Convert existing drivers to use cpufreq_freq_transition_{begin|end}Viresh Kumar
CPUFreq core has new infrastructure that would guarantee serialized calls to target() or target_index() callbacks. These are called cpufreq_freq_transition_begin() and cpufreq_freq_transition_end(). This patch converts existing drivers to use these new set of routines. Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-26cpufreq: Make sure frequency transitions are serializedSrivatsa S. Bhat
Whenever we change the frequency of a CPU, we call the PRECHANGE and POSTCHANGE notifiers. They must be serialized, i.e. PRECHANGE and POSTCHANGE notifiers should strictly alternate, thereby preventing two different sets of PRECHANGE or POSTCHANGE notifiers from interleaving arbitrarily. The following examples illustrate why this is important: Scenario 1: ----------- A thread reading the value of cpuinfo_cur_freq, will call __cpufreq_cpu_get()->cpufreq_out_of_sync()->cpufreq_notify_transition() The ondemand governor can decide to change the frequency of the CPU at the same time and hence it can end up sending the notifications via ->target(). If the notifiers are not serialized, the following sequence can occur: - PRECHANGE Notification for freq A (from cpuinfo_cur_freq) - PRECHANGE Notification for freq B (from target()) - Freq changed by target() to B - POSTCHANGE Notification for freq B - POSTCHANGE Notification for freq A We can see from the above that the last POSTCHANGE Notification happens for freq A but the hardware is set to run at freq B. Where would we break then?: adjust_jiffies() in cpufreq.c & cpufreq_callback() in arch/arm/kernel/smp.c (which also adjusts the jiffies). All the loops_per_jiffy calculations will get messed up. Scenario 2: ----------- The governor calls __cpufreq_driver_target() to change the frequency. At the same time, if we change scaling_{min|max}_freq from sysfs, it will end up calling the governor's CPUFREQ_GOV_LIMITS notification, which will also call __cpufreq_driver_target(). And hence we end up issuing concurrent calls to ->target(). Typically, platforms have the following logic in their ->target() routines: (Eg: cpufreq-cpu0, omap, exynos, etc) A. If new freq is more than old: Increase voltage B. Change freq C. If new freq is less than old: decrease voltage Now, if the two concurrent calls to ->target() are X and Y, where X is trying to increase the freq and Y is trying to decrease it, we get the following race condition: X.A: voltage gets increased for larger freq Y.A: nothing happens Y.B: freq gets decreased Y.C: voltage gets decreased X.B: freq gets increased X.C: nothing happens Thus we can end up setting a freq which is not supported by the voltage we have set. That will probably make the clock to the CPU unstable and the system might not work properly anymore. This patch introduces a set of synchronization primitives to serialize frequency transitions, which are to be used as shown below: cpufreq_freq_transition_begin(); //Perform the frequency change cpufreq_freq_transition_end(); The _begin() call sends the PRECHANGE notification whereas the _end() call sends the POSTCHANGE notification. Also, all the necessary synchronization is handled within these calls. In particular, even drivers which set the ASYNC_NOTIFICATION flag can also use these APIs for performing frequency transitions (ie., you can call _begin() from one task, and call the corresponding _end() from a different task). The actual synchronization underneath is not that complicated: The key challenge is to allow drivers to begin the transition from one thread and end it in a completely different thread (this is to enable drivers that do asynchronous POSTCHANGE notification from bottom-halves, to also use the same interface). To achieve this, a 'transition_ongoing' flag, a 'transition_lock' spinlock and a wait-queue are added per-policy. The flag and the wait-queue are used in conjunction to create an "uninterrupted flow" from _begin() to _end(). The spinlock is used to ensure that only one such "flow" is in flight at any given time. Put together, this provides us all the necessary synchronization. Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-26intel_pstate: Use del_timer_sync in intel_pstate_cpu_stopDirk Brandewie
Ensure that no timer callback is running since we are about to free the timer structure. We cannot guarantee that the call back is called on the CPU where the timer is running. Reported-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-26cpufreq: resume drivers before enabling governorsViresh Kumar
During suspend, we first stop governors and then suspend cpufreq drivers and resume must be exactly opposite of that. i.e. resume drivers first and then start governors. But the current code in resume enables governors first and then resume drivers. Fix it be changing code sequence there. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-20intel_pstate: Set core to min P state during core offlineDirk Brandewie
Change to use the new ->stop_cpu() callback to do clean up during CPU hotplug. The requested P state for an offline core will be used by the hardware coordination function to select the package P state. If the core is under load when it is offlined it will fix the package P state floor to the requested P state of offline core. Reported-by: Patrick Marlier <patrick.marlier@gmail.com> Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-20cpufreq: Add stop CPU callback to cpufreq_driver interfaceDirk Brandewie
This callback allows the driver to do clean up before the CPU is completely down and its state cannot be modified. This is used by the intel_pstate driver to reduce the requested P state prior to the core going away. This is required because the requested P state of the offline core is used to select the package P state. This effectively sets the floor package P state to the requested P state on the offline core. Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> [rjw: Minor modifications] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-20cpufreq: Remove unnecessary bracesStratos Karafotis
Remove unnecessary braces from a single statement. Signed-off-by: Stratos Karafotis <stratosk@semaphore.gr> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-20cpufreq: Fix checkpatch errors and warningsStratos Karafotis
Fix 2 checkpatch errors about using assignment in if condition, 1 checkpatch error about a required space after comma and 3 warnings about line over 80 characters. Signed-off-by: Stratos Karafotis <stratosk@semaphore.gr> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-20cpufreq: powerpc: add cpufreq transition latency for FSL e500mc SoCsZhuoyu Zhang
According to the data provided by HW Team, at least 12 internal platform clock cycles are required to stabilize a DFS clock switch on FSL e500mc Socs. This patch replaces the CPUFREQ_ETERNAL with appropriate HW clock transition latency to make DFS governors work normally on Freescale e500mc boards. Signed-off-by: Zhuoyu Zhang <Zhuoyu.Zhang@freescale.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-19cpufreq: remove unused notifier: CPUFREQ_{SUSPENDCHANGE|RESUMECHANGE}Viresh Kumar
Two cpufreq notifiers CPUFREQ_RESUMECHANGE and CPUFREQ_SUSPENDCHANGE have not been used for some time, so remove them to clean up code a bit. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-19cpufreq: Do not allow ->setpolicy drivers to provide ->targetRafael J. Wysocki
cpufreq drivers that provide the ->setpolicy() callback are supposed to have integrated governors, so they don't need to set ->target() or ->target_index() and may confuse the core if any of these callbacks is present. For this reason, add a check preventing ->setpolicy cpufreq drivers from registering if they have non-NULL ->target or ->target_index. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
2014-03-19cpufreq: arm_big_little: set 'physical_cluster' for each CPUviresh kumar
We have a per-CPU variable for managing which cluster a CPU belongs to. Currently, physical_cluster is set for policy->cpu only which leads to the following on some SoC's: - There are two clusters: - Cluster 0 has four ARM Cortex A7 CPUs (slower ones): 0,1,2,3 - Cluster 1 has four ARM Cortex A15 CPUs (faster ones): 4,5,6,7 - CPUs are booted in order 0,1..7 and so initially policy->cpu for A7 cluster would be 0 and for A15 cluster would be 4. - Now CPU4 (i.e. A15_0) is hotplugged out and so policy->cpu for A15 cluster becomes 5 (i.e. A15_1). - But physical cluster is only set for CPU0 and CPU4 in ARM big LITTLE driver and isn't updated. - Now freq change request comes for A15 cluster and we would try to update freq of physical_cluster of CPU5, i.e. A15_1. And it is currently set to zero (default value of uninitialized global variables). - And so we actually try to change freq of A7 cluster instead of A15. - This also results in kernel crash as sometimes we might request freq above A7's limit and CPU may behave badly.. Fix this by initializing physical_cluster for all CPUs of a policy. Based on previous work by Xin Wang. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-19cpufreq: arm_big_little: make vexpress driver depend on bL core driverviresh kumar
Currently vexpress big LITTLE driver selects ARM_BIG_LITTLE_CPUFREQ, so if CONFIG_BIG_LITTLE isn't enabled and CONFIG_ARM_VEXPRESS_SPC_CPUFREQ is enabled, we get the following build warnings: warning: (ARM_VEXPRESS_SPC_CPUFREQ) selects ARM_BIG_LITTLE_CPUFREQ which has unmet direct dependencies (ARCH_HAS_CPUFREQ && CPU_FREQ && (ARM || ARM64) && ARM && BIG_LITTLE && ARM_CPU_TOPOLOGY && HAVE_CLK) To fix this, make ARM_VEXPRESS_SPC_CPUFREQ depend on ARM_BIG_LITTLE_CPUFREQ instead of selecting it. This also moves the entry for ARM_VEXPRESS_SPC_CPUFREQ along with other big LITTLE config entries. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-17Merge back earlier 'pm-cpufreq' material.Rafael J. Wysocki
2014-03-13cpufreq: Skip current frequency initialization for ->setpolicy driversRafael J. Wysocki
After commit da60ce9f2fac (cpufreq: call cpufreq_driver->get() after calling ->init()) __cpufreq_add_dev() sometimes fails for CPUs handled by intel_pstate, because that driver may return 0 from its ->get() callback if it has not run long enough to collect enough samples on the given CPU. That didn't happen before commit da60ce9f2fac which added policy->cur initialization to __cpufreq_add_dev() to help reduce code duplication in other cpufreq drivers. However, the code added by commit da60ce9f2fac need not be executed for cpufreq drivers having the ->setpolicy callback defined, because the subsequent invocation of cpufreq_set_policy() will use that callback to initialize the policy anyway and it doesn't need policy->cur to be initialized upfront. The analogous code in cpufreq_update_policy() is also unnecessary for cpufreq drivers having ->setpolicy set and may be skipped for them as well. Since intel_pstate provides ->setpolicy, skipping the upfront policy->cur initialization for cpufreq drivers with that callback set will cover intel_pstate and the problem it's been having after commit da60ce9f2fac will be addressed. Fixes: da60ce9f2fac (cpufreq: call cpufreq_driver->get() after calling ->init()) References: https://bugzilla.kernel.org/show_bug.cgi?id=71931 Reported-and-tested-by: Patrik Lundquist <patrik.lundquist@gmail.com> Acked-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Cc: 3.13+ <stable@vger.kernel.org> # 3.13+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-12cpufreq: SPEAr: Instantiate as platform_driverViresh Kumar
As multiplatform build is being adopted by more and more ARM platforms, initcall function should be used very carefully. For example, when SPEAr cpufreq driver is enabled on a kernel booted on a non-SPEAr board, we will get following boot time error: spear_cpufreq: Invalid cpufreq_tbl To eliminate this undesired the effect, the patch changes SPEAr driver to have it instantiated as a platform_driver. Then it will only run on platforms that create the platform_device "spear-cpufreq". This patch also creates platform node for SPEAr13xx boards. Reported-by: Josh Cartwright <joshc@codeaurora.org> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-12cpufreq: Remove unnecessary variable/parameter 'frozen'Viresh Kumar
We have used 'frozen' variable/function parameter at many places to distinguish between CPU offline/online on suspend/resume vs sysfs removals. We now have another variable cpufreq_suspended which can be used in these cases, so we can get rid of all those variables or function parameters. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-12cpufreq: Remove cpufreq_generic_exit()Viresh Kumar
cpufreq_generic_exit() is empty now and can be deleted. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-12cpufreq: add 'freq_table' in struct cpufreq_policyViresh Kumar
freq table is not per CPU but per policy, so it makes more sense to keep it within struct cpufreq_policy instead of a per-cpu variable. This patch does it. Over that, there is no need to set policy->freq_table to NULL in ->exit(), as policy structure is going to be freed soon. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-12cpufreq: Reformat printk() statementsJoe Perches
- Add missing newlines - Coalesce format fragments - Convert printks to pr_<level> - Align arguments Based-on-patch-by: Sören Brinkmann <soren.brinkmann@xilinx.com> Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: Tegra: Use cpufreq_generic_suspend()Viresh Kumar
The cpufreq core now supports suspending and resuming of cpufreq drivers and governors during systems suspend and resume, so use the common infrastructure instead of defining special PM notifiers for the same thing. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Tested-by: Stephen Warren <swarren@nvidia.com> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: s5pv210: Use cpufreq_generic_suspend()Viresh Kumar
The cpufreq core now supports suspending and resuming of cpufreq drivers and governors during systems suspend and resume, so use the common infrastructure instead of defining special PM notifiers for the same thing. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: exynos: Use cpufreq_generic_suspend()Viresh Kumar
The cpufreq core now supports suspending and resuming of cpufreq drivers and governors during systems suspend and resume, so use the common infrastructure instead of defining special PM notifiers for the same thing. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: Implement cpufreq_generic_suspend()Viresh Kumar
Multiple platforms need to set CPUs to a particular frequency before suspending the system, so provide a common infrastructure for them. Those platforms only need to point their ->suspend callback pointers to the generic routine. Tested-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: suspend governors on system suspend/hibernateViresh Kumar
This patch adds cpufreq suspend/resume calls to dpm_{suspend|resume}() for handling suspend/resume of cpufreq governors. Lan Tianyu (Intel) & Jinhyuk Choi (Broadcom) found an issue where the tunables configuration for clusters/sockets with non-boot CPUs was lost after system suspend/resume, as we were notifying governors with CPUFREQ_GOV_POLICY_EXIT on removal of the last CPU for that policy which caused the tunables memory to be freed. This is fixed by preventing any governor operations from being carried out between the device suspend and device resume stages of system suspend and resume, respectively. We could have added these callbacks at dpm_{suspend|resume}_noirq() level, but there is an additional problem that the majority of I/O devices is already suspended at that point and if cpufreq drivers want to change the frequency before suspending, then that not be possible on some platforms (which depend on peripherals like i2c, regulators, etc). Reported-and-tested-by: Lan Tianyu <tianyu.lan@intel.com> Reported-by: Jinhyuk Choi <jinchoi@broadcom.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: move call to __find_governor() to cpufreq_init_policy()viresh kumar
We call __find_governor() during the addition of the first CPU of each policy from __cpufreq_add_dev() to find the last governor used for this CPU before it was hot-removed. After that we call cpufreq_parse_governor() in cpufreq_init_policy(), either with this governor, or with the default governor. Right after that policy->governor is set to NULL. While that code is not functionally problematic, the structure of it is suboptimal, because some of the code required in cpufreq_init_policy() is being executed by its caller, __cpufreq_add_dev(). So, it would make more sense to get all of it together in a single place to make code more readable. Accordingly, move the code needed for policy initialization to cpufreq_init_policy() and initialize policy->governor to NULL at the beginning. In order to clean up the code a bit more, some of the #ifdefs for CONFIG_HOTPLUG_CPU are dropped too. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06Merge back earlier 'pm-cpufreq' material.Rafael J. Wysocki
2014-03-06cpufreq: Initialize governor for a new policy under policy->rwsemViresh Kumar
policy->rwsem is used to lock access to all parts of code modifying struct cpufreq_policy, but it's not used on a new policy created by __cpufreq_add_dev(). Because of that, if cpufreq_update_policy() is called in a tight loop on one CPU in parallel with offline/online of another CPU, then the following crash can be triggered: Unable to handle kernel NULL pointer dereference at virtual address 00000020 pgd = c0003000 [00000020] *pgd=80000000004003, *pmd=00000000 Internal error: Oops: 206 [#1] PREEMPT SMP ARM PC is at __cpufreq_governor+0x10/0x1ac LR is at cpufreq_update_policy+0x114/0x150 ---[ end trace f23a8defea6cd706 ]--- Kernel panic - not syncing: Fatal exception CPU0: stopping CPU: 0 PID: 7136 Comm: mpdecision Tainted: G D W 3.10.0-gd727407-00074-g979ede8 #396 [<c0afe180>] (notifier_call_chain+0x40/0x68) from [<c02a23ac>] (__blocking_notifier_call_chain+0x40/0x58) [<c02a23ac>] (__blocking_notifier_call_chain+0x40/0x58) from [<c02a23d8>] (blocking_notifier_call_chain+0x14/0x1c) [<c02a23d8>] (blocking_notifier_call_chain+0x14/0x1c) from [<c0803c68>] (cpufreq_set_policy+0xd4/0x2b8) [<c0803c68>] (cpufreq_set_policy+0xd4/0x2b8) from [<c0803e7c>] (cpufreq_init_policy+0x30/0x98) [<c0803e7c>] (cpufreq_init_policy+0x30/0x98) from [<c0805a18>] (__cpufreq_add_dev.isra.17+0x4dc/0x7a4) [<c0805a18>] (__cpufreq_add_dev.isra.17+0x4dc/0x7a4) from [<c0805d38>] (cpufreq_cpu_callback+0x58/0x84) [<c0805d38>] (cpufreq_cpu_callback+0x58/0x84) from [<c0afe180>] (notifier_call_chain+0x40/0x68) [<c0afe180>] (notifier_call_chain+0x40/0x68) from [<c02812dc>] (__cpu_notify+0x28/0x44) [<c02812dc>] (__cpu_notify+0x28/0x44) from [<c0aeed90>] (_cpu_up+0xf4/0x1dc) [<c0aeed90>] (_cpu_up+0xf4/0x1dc) from [<c0aeeed4>] (cpu_up+0x5c/0x78) [<c0aeeed4>] (cpu_up+0x5c/0x78) from [<c0aec808>] (store_online+0x44/0x74) [<c0aec808>] (store_online+0x44/0x74) from [<c03a40f4>] (sysfs_write_file+0x108/0x14c) [<c03a40f4>] (sysfs_write_file+0x108/0x14c) from [<c03517d4>] (vfs_write+0xd0/0x180) [<c03517d4>] (vfs_write+0xd0/0x180) from [<c0351ca8>] (SyS_write+0x38/0x68) [<c0351ca8>] (SyS_write+0x38/0x68) from [<c0205de0>] (ret_fast_syscall+0x0/0x30) Fix that by taking locks at appropriate places in __cpufreq_add_dev() as well. Reported-by: Saravana Kannan <skannan@codeaurora.org> Suggested-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: Initialize policy before making it available for others to useViresh Kumar
Policy must be fully initialized before it is being made available for use by others. Otherwise cpufreq_cpu_get() would be able to grab a half initialized policy structure that might not have affected_cpus (for example) populated. Then, anybody accessing those fields will get a wrong value and that will lead to unpredictable results. In order to fix this, do all the necessary initialization before we make the policy structure available via cpufreq_cpu_get(). That will guarantee that any code accessing fields of the policy will get correct data from them. Reported-by: Saravana Kannan <skannan@codeaurora.org> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [rjw: Changelog] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-06cpufreq: use cpufreq_cpu_get() to avoid cpufreq_get() race conditionsAaron Plattner
If a module calls cpufreq_get while cpufreq is initializing, it's possible for it to be called after cpufreq_driver is set but before cpufreq_cpu_data is written during subsys_interface_register. This happens because cpufreq_get doesn't take the cpufreq_driver_lock around its use of cpufreq_cpu_data. Fix this by using cpufreq_cpu_get(cpu) to look up the policy rather than reading it out of cpufreq_cpu_data directly. cpufreq_cpu_get() takes the appropriate locks to prevent this race from happening. Since it's possible for policy to be NULL if the caller passes in an invalid CPU number or calls the function before cpufreq is initialized, delete the BUG_ON(!policy) and simply return 0. Don't try to return -ENOENT because that's negative and the function returns an unsigned integer. References: https://bbs.archlinux.org/viewtopic.php?id=177934 Signed-off-by: Aaron Plattner <aplattner@nvidia.com> Cc: 3.13+ <stable@vger.kernel.org> # 3.13+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-02cpufreq: stats: Refactor common code into __cpufreq_stats_create_table()Saravana Kannan
cpufreq_frequency_get_table() is called from all callers of __cpufreq_stats_create_table(). So, move it inside. Suggested-by: Viresh Kumar <viresh.kumar@linaro.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Saravana Kannan <skannan@codeaurora.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-02cpufreq: stats: Fix error handling in __cpufreq_stats_create_table()Saravana Kannan
Remove sysfs group if __cpufreq_stats_create_table() fails after creating one. Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Saravana Kannan <skannan@codeaurora.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-02cpufreq: stats: Remove redundant cpufreq_cpu_get() callSaravana Kannan
__cpufreq_stats_create_table always gets pass the valid and real policy struct. So, there's no need to call cpufreq_cpu_get() to get the policy again. Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Saravana Kannan <skannan@codeaurora.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-02intel_pstate: fix pid_reset to use fixed point valuesDirk Brandewie
commit d253d2a526 (Improve accuracy by not truncating until final result), changed internal variables of the PID to be fixed point numbers. Update the pid_reset() to reflect this change. Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-02intel_pstate: remove unneeded sample buffersDirk Brandewie
Remove unneeded sample buffers, intel_pstate operates on the most recent sample only. This save some memory and make the code more readable. Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-01cpufreq: Return error if ->get() failed in cpufreq_update_policy()Viresh Kumar
cpufreq_update_policy() calls cpufreq_driver->get() to get current frequency of a CPU and it is not supposed to fail or return zero. Return error in case that happens. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-03-01cpufreq: enable ARM drivers on arm64Rob Herring
Enable cpufreq and power kconfig menus on arm64 along with arm cpufreq drivers. The power menu is needed for OPP support. At least on Calxeda systems, the same cpufreq driver is used for arm and arm64 based systems. Signed-off-by: Rob Herring <rob.herring@calxeda.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Mark Brown <broonie@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-27cpufreq: Mark function as static in cpufreq.cRashika Kheria
Mark function as static in cpufreq.c because it is not used outside this file. This eliminates the following warning in cpufreq.c: drivers/cpufreq/cpufreq.c:355:9: warning: no previous prototype for ‘show_boost’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by: Josh Triplett <josh@joshtriplett.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-27Merge back earlier 'pm-cpufreq' material.Rafael J. Wysocki
2014-02-26intel_pstate: Change busy calculation to use fixed point math.Dirk Brandewie
Commit fcb6a15c2e (intel_pstate: Take core C0 time into account for core busy calculation) introduced a regression on some processor SKUs supported by intel_pstate. This was due to the truncation caused by using integer math to calculate core busy and C0 percentages. On a i7-4770K processor operating at 800Mhz going to 100% utilization the percent busy of the CPU using integer math is 22%, but it actually is 22.85%. This value scaled to the current frequency returned 97 which the PID interpreted as no error and did not adjust the P state. Tested on i7-4770K, i7-2600, i5-3230M. Fixes: fcb6a15c2e7e (intel_pstate: Take core C0 time into account for core busy calculation) References: https://lkml.org/lkml/2014/2/19/626 References: https://bugzilla.kernel.org/show_bug.cgi?id=70941 Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-24cpufreq: don't call cpufreq_update_policy() on CPU additionViresh Kumar
cpufreq_update_policy() is called from two places currently. From a workqueue handled queued from cpufreq_bp_resume() for boot CPU and from cpufreq_cpu_callback() whenever a CPU is added. The first one makes sure that boot CPU is running on the frequency present in policy->cpu. But we don't really need a call from cpufreq_cpu_callback(), because we always call cpufreq_driver->init() (which will set policy->cur correctly) whenever first CPU of any policy is added back. And so every policy structure is guaranteed to have the right frequency in policy->cur. Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-24cpufreq: Refactor cpufreq_set_policy()Rafael J. Wysocki
Reduce the rampant usage of goto and the indentation level in cpufreq_set_policy() to improve the readability of that code. No functional changes should result from that. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
2014-02-21intel_pstate: Add support for Baytrail turbo P statesDirk Brandewie
A documentation update exposed the existance of the turbo ratio register. Update baytrail support to use the turbo range. Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Cc: 3.13+ <stable@vger.kernel.org> # 3.13+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-21intel_pstate: Use LFM bus ratio as min ratio/P stateDirk Brandewie
LFM (max efficiency ratio) is the max frequency at minimum voltage supported by the processor. Using LFM as the minimum P state increases performmance without affecting power. By not using P states below LFM we avoid using P states that are less power efficient. Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Cc: 3.13+ <stable@vger.kernel.org> # 3.13+ Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-19cpufreq: powernow-k8: Initialize per-cpu data-structures properlySrivatsa S. Bhat
The powernow-k8 driver maintains a per-cpu data-structure called powernow_data that is used to perform the frequency transitions. It initializes this data structure only for the policy->cpu. So, accesses to this data structure by other CPUs results in various problems because they would have been uninitialized. Specifically, if a cpu (!= policy->cpu) invokes the drivers' ->get() function, it returns 0 as the KHz value, since its per-cpu memory doesn't point to anything valid. This causes problems during suspend/resume since cpufreq_update_policy() tries to enforce this (0 KHz) as the current frequency of the CPU, and this madness gets propagated to adjust_jiffies() as well. Eventually, lots of things start breaking down, including the r8169 ethernet card, in one particularly interesting case reported by Pierre Ossman. Fix this by initializing the per-cpu data-structures of all the CPUs in the policy appropriately. References: https://bugzilla.kernel.org/show_bug.cgi?id=70311 Reported-by: Pierre Ossman <pierre@ossman.eu> Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Cc: All applicable <stable@vger.kernel.org> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-19cpufreq: remove sysfs link when a cpu != policy->cpu, is removedviresh kumar
Commit 42f921a (cpufreq: remove sysfs files for CPUs which failed to come back after resume) tried to do this but missed this piece of code to fix. Currently we are getting this on suspend/resume: ------------[ cut here ]------------ WARNING: CPU: 0 PID: 877 at fs/sysfs/dir.c:52 sysfs_warn_dup+0x68/0x84() sysfs: cannot create duplicate filename '/devices/system/cpu/cpu1/cpufreq' Modules linked in: brcmfmac brcmutil CPU: 0 PID: 877 Comm: test-rtc-resume Not tainted 3.14.0-rc2-00259-g9398a10cd964 #12 [<c0015bac>] (unwind_backtrace) from [<c0011850>] (show_stack+0x10/0x14) [<c0011850>] (show_stack) from [<c056e018>] (dump_stack+0x80/0xcc) [<c056e018>] (dump_stack) from [<c0025e44>] (warn_slowpath_common+0x64/0x88) [<c0025e44>] (warn_slowpath_common) from [<c0025efc>] (warn_slowpath_fmt+0x30/0x40) [<c0025efc>] (warn_slowpath_fmt) from [<c012776c>] (sysfs_warn_dup+0x68/0x84) [<c012776c>] (sysfs_warn_dup) from [<c0127a54>] (sysfs_do_create_link_sd+0xb0/0xb8) [<c0127a54>] (sysfs_do_create_link_sd) from [<c038ef64>] (__cpufreq_add_dev.isra.27+0x2a8/0x814) [<c038ef64>] (__cpufreq_add_dev.isra.27) from [<c038f548>] (cpufreq_cpu_callback+0x70/0x8c) [<c038f548>] (cpufreq_cpu_callback) from [<c0043864>] (notifier_call_chain+0x44/0x84) [<c0043864>] (notifier_call_chain) from [<c0025f60>] (__cpu_notify+0x28/0x44) [<c0025f60>] (__cpu_notify) from [<c00261e8>] (_cpu_up+0xf0/0x140) [<c00261e8>] (_cpu_up) from [<c0569eb8>] (enable_nonboot_cpus+0x68/0xb0) [<c0569eb8>] (enable_nonboot_cpus) from [<c006339c>] (suspend_devices_and_enter+0x198/0x2dc) [<c006339c>] (suspend_devices_and_enter) from [<c0063654>] (pm_suspend+0x174/0x1e8) [<c0063654>] (pm_suspend) from [<c00624e0>] (state_store+0x6c/0xbc) [<c00624e0>] (state_store) from [<c01fc200>] (kobj_attr_store+0x14/0x20) [<c01fc200>] (kobj_attr_store) from [<c0126e50>] (sysfs_kf_write+0x44/0x48) [<c0126e50>] (sysfs_kf_write) from [<c012a274>] (kernfs_fop_write+0xb4/0x14c) [<c012a274>] (kernfs_fop_write) from [<c00d4818>] (vfs_write+0xa8/0x180) [<c00d4818>] (vfs_write) from [<c00d4bb8>] (SyS_write+0x3c/0x70) [<c00d4bb8>] (SyS_write) from [<c000e620>] (ret_fast_syscall+0x0/0x30) ---[ end trace 76969904b614c18f ]--- Fix this by removing sysfs link for cpufreq directory when cpu removed isn't policy->cpu. Revamps: 42f921a (cpufreq: remove sysfs files for CPUs which failed to come back after resume) Reported-and-tested-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-13intel_pstate: Remove energy reporting from pstate_sample tracepointDirk Brandewie
Remove the reporting of energy since it does not provide any useful information about the state of the driver and will be a maintainance headache going forward since the RAPL energy units register is not architectural and subject to change between micro-architectures References: https://bugzilla.kernel.org/show_bug.cgi?id=69831 Fixes: b69880f9ccf7 (intel_pstate: Add trace point to report internal state.) Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-02-04intel_pstate: Take core C0 time into account for core busy calculationDirk Brandewie
Take non-idle time into account when calculating core busy time. This ensures that intel_pstate will notice a decrease in load. References: https://bugzilla.kernel.org/show_bug.cgi?id=66581 Cc: 3.10+ <stable@vger.kernel.org> # 3.10+ Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-01-31Merge tag 'pm+acpi-3.14-rc1-2' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm Pull ACPI and power management fixes and cleanups from Rafael Wysocki: - ACPI device hotplug fix preventing ACPI drivers from binding to device objects that acpi_bus_trim() has been called for and the devices represented by them may not be operational. - Recent cpufreq changes related to the "boost" (turbo) feature broke the acpi-cpufreq error code path causing a NULL pointer dereference to occur on some systems. Fix from Konrad Rzeszutek Wilk. - The log level of a CPU initialization error message added recently needs to be reduced, because the particular BIOS issue indicated by it turns out to be widespread and doesn't really matter for the majority of systems having it. From Jiang Liu. - The regulator API needs to be told to stay away from things on systems with ACPI BIOSes or it may conflict with the BIOS' own handling of voltage regulators. Fix from Mark Brown that works around a 3.13 regression in lm90 on PCs occuring if the regulator API is enabled. - Prevent the Exynos4 devfreq driver from being built on multiplatform, because it depends on things that aren't available during such builds. From Sachin Kamat. - Upstream ACPICA doesn't use the bool type as defined in the kernel, so modify the kernel's ACPICA code to follow the upstream in that respect (only one variable definition is affected) to reduce divergences between the two. From Lv Zheng. - Make the ACPI device PM code use ACPI_COMPANION() instead of its own routine doing the same thing (and invokng ACPI_COMPANION() in the process). - Modify some routines in the ACPI processor driver to follow the common convention and return negative integers on errors. From Hanjun Guo. * tag 'pm+acpi-3.14-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: ACPI / scan: Clear match_driver flag in acpi_bus_trim() ACPI / init: Flag use of ACPI and ACPI idioms for power supplies to regulator API acpi-cpufreq: De-register CPU notifier and free struct msr on error. ACPICA: Remove bool usage from ACPICA. PM / devfreq: Disable Exynos4 driver build on multiplatform ACPI / PM: Use ACPI_COMPANION() to get ACPI companions of devices ACPI / scan: reduce log level of "ACPI: \_PR_.CPU4: failed to get CPU APIC ID" ACPI / processor: Return specific error value when mapping lapic id