summaryrefslogtreecommitdiffstats
path: root/kernel (follow)
Commit message (Collapse)AuthorAgeFilesLines
* sys_sgetmask/sys_ssetmask: add CONFIG_SGETMASK_SYSCALLFabian Frederick2014-06-052-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sys_sgetmask and sys_ssetmask are obsolete system calls no longer supported in libc. This patch replaces architecture related __ARCH_WANT_SYS_SGETMAX by expert mode configuration.That option is enabled by default for those architectures. Signed-off-by: Fabian Frederick <fabf@skynet.be> Cc: Steven Miao <realmz6@gmail.com> Cc: Mikael Starvik <starvik@axis.com> Cc: Jesper Nilsson <jesper.nilsson@axis.com> Cc: David Howells <dhowells@redhat.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Michal Simek <monstr@monstr.eu> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Helge Deller <deller@gmx.de> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Greg Ungerer <gerg@uclinux.org> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* mm: page_alloc: use jump labels to avoid checking number_of_cpusetsMel Gorman2014-06-051-10/+4
| | | | | | | | | | | | | | | | | | | | | If cpusets are not in use then we still check a global variable on every page allocation. Use jump labels to avoid the overhead. Signed-off-by: Mel Gorman <mgorman@suse.de> Reviewed-by: Rik van Riel <riel@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Jan Kara <jack@suse.cz> Cc: Michal Hocko <mhocko@suse.cz> Cc: Hugh Dickins <hughd@google.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Theodore Ts'o <tytso@mit.edu> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* memcg: optimize the "Search everything else" loop in mm_update_next_owner()Oleg Nesterov2014-06-051-3/+9
| | | | | | | | | | | | | | | | for_each_process_thread() is sub-optimal. All threads share the same ->mm, we can swicth to the next process once we found a thread with ->mm != NULL and ->mm != mm. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Peter Chiang <pchiang@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* memcg: mm_update_next_owner() should skip kthreadsOleg Nesterov2014-06-051-6/+4
| | | | | | | | | | | | | | | | | | | "Search through everything else" in mm_update_next_owner() can hit a kthread which adopted this "mm" via use_mm(), it should not be used as mm->owner. Add the PF_KTHREAD check. While at it, change this code to use for_each_process_thread() instead of deprecated do_each_thread/while_each_thread. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Reviewed-by: Michal Hocko <mhocko@suse.cz> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: Peter Chiang <pchiang@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* memcg: kill CONFIG_MM_OWNEROleg Nesterov2014-06-052-4/+4
| | | | | | | | | | | | CONFIG_MM_OWNER makes no sense. It is not user-selectable, it is only selected by CONFIG_MEMCG automatically. So we can kill this option in init/Kconfig and do s/CONFIG_MM_OWNER/CONFIG_MEMCG/ globally. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Michal Hocko <mhocko@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* mm: get rid of __GFP_KMEMCGVladimir Davydov2014-06-051-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently to allocate a page that should be charged to kmemcg (e.g. threadinfo), we pass __GFP_KMEMCG flag to the page allocator. The page allocated is then to be freed by free_memcg_kmem_pages. Apart from looking asymmetrical, this also requires intrusion to the general allocation path. So let's introduce separate functions that will alloc/free pages charged to kmemcg. The new functions are called alloc_kmem_pages and free_kmem_pages. They should be used when the caller actually would like to use kmalloc, but has to fall back to the page allocator for the allocation is large. They only differ from alloc_pages and free_pages in that besides allocating or freeing pages they also charge them to the kmem resource counter of the current memory cgroup. [sfr@canb.auug.org.au: export kmalloc_order() to modules] Signed-off-by: Vladimir Davydov <vdavydov@parallels.com> Acked-by: Greg Thelen <gthelen@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.cz> Cc: Glauber Costa <glommer@gmail.com> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* kthread: fix return value of kthread_create() upon SIGKILL.Tetsuo Handa2014-06-051-2/+2
| | | | | | | | | | | | | | Commit 786235eeba0e ("kthread: make kthread_create() killable") meant for allowing kthread_create() to abort as soon as killed by the OOM-killer. But returning -ENOMEM is wrong if killed by SIGKILL from userspace. Change kthread_create() to return -EINTR upon SIGKILL. Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: Oleg Nesterov <oleg@redhat.com> Acked-by: David Rientjes <rientjes@google.com> Cc: <stable@vger.kernel.org> [3.13+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge tag 'pm+acpi-3.16-rc1' of ↵Linus Torvalds2014-06-047-77/+130
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm into next Pull ACPI and power management updates from Rafael Wysocki: "ACPICA is the leader this time (63 commits), followed by cpufreq (28 commits), devfreq (15 commits), system suspend/hibernation (12 commits), ACPI video and ACPI device enumeration (10 commits each). We have no major new features this time, but there are a few significant changes of how things work. The most visible one will probably be that we are now going to create platform devices rather than PNP devices by default for ACPI device objects with _HID. That was long overdue and will be really necessary to be able to use the same drivers for the same hardware blocks on ACPI and DT-based systems going forward. We're not expecting fallout from this one (as usual), but it's something to watch nevertheless. The second change having a chance to be visible is that ACPI video will now default to using native backlight rather than the ACPI backlight interface which should generally help systems with broken Win8 BIOSes. We're hoping that all problems with the native backlight handling that we had previously have been addressed and we are in a good enough shape to flip the default, but this change should be easy enough to revert if need be. In addition to that, the system suspend core has a new mechanism to allow runtime-suspended devices to stay suspended throughout system suspend/resume transitions if some extra conditions are met (generally, they are related to coordination within device hierarchy). However, enabling this feature requires cooperation from the bus type layer and for now it has only been implemented for the ACPI PM domain (used by ACPI-enumerated platform devices mostly today). Also, the acpidump utility that was previously shipped as a separate tool will now be provided by the upstream ACPICA along with the rest of ACPICA code, which will allow it to be more up to date and better supported, and we have one new cpuidle driver (ARM clps711x). The rest is improvements related to certain specific use cases, cleanups and fixes all over the place. Specifics: - ACPICA update to upstream version 20140424. That includes a number of fixes and improvements related to things like GPE handling, table loading, headers, memory mapping and unmapping, DSDT/SSDT overriding, and the Unload() operator. The acpidump utility from upstream ACPICA is included too. From Bob Moore, Lv Zheng, David Box, David Binderman, and Colin Ian King. - Fixes and cleanups related to ACPI video and backlight interfaces from Hans de Goede. That includes blacklist entries for some new machines and using native backlight by default. - ACPI device enumeration changes to create platform devices rather than PNP devices for ACPI device objects with _HID by default. PNP devices will still be created for the ACPI device object with device IDs corresponding to real PNP devices, so that change should not break things left and right, and we're expecting to see more and more ACPI-enumerated platform devices in the future. From Zhang Rui and Rafael J Wysocki. - Updates for the ACPI LPSS (Low-Power Subsystem) driver allowing it to handle system suspend/resume on Asus T100 correctly. From Heikki Krogerus and Rafael J Wysocki. - PM core update introducing a mechanism to allow runtime-suspended devices to stay suspended over system suspend/resume transitions if certain additional conditions related to coordination within device hierarchy are met. Related PM documentation update and ACPI PM domain support for the new feature. From Rafael J Wysocki. - Fixes and improvements related to the "freeze" sleep state. They affect several places including cpuidle, PM core, ACPI core, and the ACPI battery driver. From Rafael J Wysocki and Zhang Rui. - Miscellaneous fixes and updates of the ACPI core from Aaron Lu, Bjørn Mork, Hanjun Guo, Lan Tianyu, and Rafael J Wysocki. - Fixes and cleanups for the ACPI processor and ACPI PAD (Processor Aggregator Device) drivers from Baoquan He, Manuel Schölling, Tony Camuso, and Toshi Kani. - System suspend/resume optimization in the ACPI battery driver from Lan Tianyu. - OPP (Operating Performance Points) subsystem updates from Chander Kashyap, Mark Brown, and Nishanth Menon. - cpufreq core fixes, updates and cleanups from Srivatsa S Bhat, Stratos Karafotis, and Viresh Kumar. - Updates, fixes and cleanups for the Tegra, powernow-k8, imx6q, s5pv210, nforce2, and powernv cpufreq drivers from Brian Norris, Jingoo Han, Paul Bolle, Philipp Zabel, Stratos Karafotis, and Viresh Kumar. - intel_pstate driver fixes and cleanups from Dirk Brandewie, Doug Smythies, and Stratos Karafotis. - Enabling the big.LITTLE cpufreq driver on arm64 from Mark Brown. - Fix for the cpuidle menu governor from Chander Kashyap. - New ARM clps711x cpuidle driver from Alexander Shiyan. - Hibernate core fixes and cleanups from Chen Gang, Dan Carpenter, Fabian Frederick, Pali Rohár, and Sebastian Capella. - Intel RAPL (Running Average Power Limit) driver updates from Jacob Pan. - PNP subsystem updates from Bjorn Helgaas and Fabian Frederick. - devfreq core updates from Chanwoo Choi and Paul Bolle. - devfreq updates for exynos4 and exynos5 from Chanwoo Choi and Bartlomiej Zolnierkiewicz. - turbostat tool fix from Jean Delvare. - cpupower tool updates from Prarit Bhargava, Ramkumar Ramachandra and Thomas Renninger. - New ACPI ec_access.c tool for poking at the EC in a safe way from Thomas Renninger" * tag 'pm+acpi-3.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (187 commits) ACPICA: Namespace: Remove _PRP method support. intel_pstate: Improve initial busy calculation intel_pstate: add sample time scaling intel_pstate: Correct rounding in busy calculation intel_pstate: Remove C0 tracking PM / hibernate: fixed typo in comment ACPI: Fix x86 regression related to early mapping size limitation ACPICA: Tables: Add mechanism to control early table checksum verification. ACPI / scan: use platform bus type by default for _HID enumeration ACPI / scan: always register ACPI LPSS scan handler ACPI / scan: always register memory hotplug scan handler ACPI / scan: always register container scan handler ACPI / scan: Change the meaning of missing .attach() in scan handlers ACPI / scan: introduce platform_id device PNP type flag ACPI / scan: drop unsupported serial IDs from PNP ACPI scan handler ID list ACPI / scan: drop IDs that do not comply with the ACPI PNP ID rule ACPI / PNP: use device ID list for PNPACPI device enumeration ACPI / scan: .match() callback for ACPI scan handlers ACPI / battery: wakeup the system only when necessary power_supply: allow power supply devices registered w/o wakeup source ...
| *---. Merge branches 'pnp', 'powercap', 'pm-runtime' and 'pm-opp'Rafael J. Wysocki2014-06-031-2/+1
| |\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * pnp: MAINTAINERS: Remove Bjorn Helgaas as PNP maintainer PNP / resources: remove positive test on unsigned values * powercap: powercap / RAPL: add new CPU IDs powercap / RAPL: further relax energy counter checks * pm-runtime: PM / runtime: Update documentation to reflect the current code flow * pm-opp: PM / OPP: discard duplicate OPPs PM / OPP: Make OPP invisible to users in Kconfig PM / OPP: fix incorrect OPP count handling in of_init_opp_table
| | | | * PM / OPP: Make OPP invisible to users in KconfigMark Brown2014-05-191-2/+1
| | | |/ | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The OPP code is an in kernel library selected by its users, there is no no architecture code required to implement it and enabling it without a user just increases the kernel size. Since the users select rather than depend on it just remove the ability to directly set the option from Kconfig. Signed-off-by: Mark Brown <broonie@linaro.org> Acked-by: Nishanth Menon <nm@ti.com> Acked-by: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| * | | Merge branch 'acpi-pm'Rafael J. Wysocki2014-06-031-0/+15
| |\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * acpi-pm: ACPI / PM: Export rest of the subsys PM callbacks ACPI / PM: Avoid resuming devices in ACPI PM domain during system suspend ACPI / PM: Hold ACPI scan lock over the "freeze" sleep state ACPI / PM: Export acpi_target_system_state() to modules
| | * \ \ Merge branch 'pm-sleep' into acpi-pmRafael J. Wysocki2014-05-203-22/+27
| | |\ \ \
| | * | | | ACPI / PM: Hold ACPI scan lock over the "freeze" sleep stateRafael J. Wysocki2014-05-161-0/+15
| | | |/ / | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The "freeze" sleep state suffers from the same issue that was addressed by commit ad07277e82de (ACPI / PM: Hold acpi_scan_lock over system PM transitions) for ACPI sleep states, that is, things break if ->remove() is called for devices whose system resume callbacks haven't been executed yet. It also can be addressed in the same way, by holding the ACPI scan lock over the "freeze" sleep state and PM transitions to and from that state, but ->begin() and ->end() platform operations for the "freeze" sleep state are needed for this purpose. This change has been tested on Acer Aspire S5 with Thunderbolt. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| * | | | Merge branch 'pm-sleep'Rafael J. Wysocki2014-06-036-75/+114
| |\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * pm-sleep: PM / hibernate: fixed typo in comment PM / sleep: unregister wakeup source when disabling device wakeup PM / sleep: Introduce command line argument for sleep state enumeration PM / sleep: Use valid_state() for platform-dependent sleep states only PM / sleep: Add state field to pm_states[] entries PM / sleep: Update device PM documentation to cover direct_complete PM / sleep: Mechanism to avoid resuming runtime-suspended devices unnecessarily PM / hibernate: Fix memory corruption in resumedelay_setup() PM / hibernate: convert simple_strtoul to kstrtoul PM / hibernate: Documentation: Fix script for unswapping PM / hibernate: no kernel_power_off when pm_power_off NULL PM / hibernate: use unsigned local variables in swsusp_show_speed()
| | * | | | PM / hibernate: fixed typo in commentNiv Yehezkel2014-06-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a trivial comment typo (s/mam/map) in kernel/power/swap.c. Signed-off-by: Niv Yehezkel <executerx@gmail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | | PM / sleep: Introduce command line argument for sleep state enumerationRafael J. Wysocki2014-05-262-8/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On some systems the platform doesn't support neither PM_SUSPEND_MEM nor PM_SUSPEND_STANDBY, so PM_SUSPEND_FREEZE is the only available system sleep state. However, some user space frameworks only use the "mem" and (sometimes) "standby" sleep state labels, so the users of those systems need to modify user space in order to be able to use system suspend at all and that is not always possible. For this reason, add a new kernel command line argument, relative_sleep_states, allowing the users of those systems to change the way in which the kernel assigns labels to system sleep states. Namely, for relative_sleep_states=1, the "mem", "standby" and "freeze" labels will enumerate the available system sleem states from the deepest to the shallowest, respectively, so that "mem" is always present in /sys/power/state and the other state strings may or may not be presend depending on what is supported by the platform. Update system sleep states documentation to reflect this change. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | | PM / sleep: Use valid_state() for platform-dependent sleep states onlyRafael J. Wysocki2014-05-264-37/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use the observation that, for platform-dependent sleep states (PM_SUSPEND_STANDBY, PM_SUSPEND_MEM), a given state is either always supported or always unsupported and store that information in pm_states[] instead of calling valid_state() every time we need to check it. Also do not use valid_state() for PM_SUSPEND_FREEZE, which is always valid, and move the pm_test_level validity check for PM_SUSPEND_FREEZE directly into enter_state(). Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | | PM / sleep: Add state field to pm_states[] entriesRafael J. Wysocki2014-05-264-27/+30
| | | |/ / | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To allow sleep states corresponding to the "mem", "standby" and "freeze" lables to be different from the pm_states[] indexes of those strings, introduce struct pm_sleep_state, consisting of a string label and a state number, and turn pm_states[] into an array of objects of that type. This modification should not lead to any functional changes. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | PM / hibernate: Fix memory corruption in resumedelay_setup()Dan Carpenter2014-05-161-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the original code "resume_delay" is an int so on 64 bits, the call to kstrtoul() will cause memory corruption. We may as well fix a style issue here as well and make "resume_delay" unsigned int, since that's what we pass to ssleep(). Fixes: 317cf7e5e85e (PM / hibernate: convert simple_strtoul to kstrtoul) Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | PM / hibernate: convert simple_strtoul to kstrtoulFabian Frederick2014-05-091-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace obsolete function. Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | Merge branch 'pm-cpuidle' into pm-sleepRafael J. Wysocki2014-05-072-13/+9
| | |\ \ \
| | * | | | PM / hibernate: no kernel_power_off when pm_power_off NULLSebastian Capella2014-05-011-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Reboot logic in kernel/reboot will avoid calling kernel_power_off when pm_power_off is null, and instead uses kernel_halt. Change hibernate's power_down to follow the behavior in the reboot call. Calling the notifier twice (once for SYS_POWER_OFF and again for SYS_HALT) causes a panic during hibernation on Kirkwood Openblocks A6 board. Signed-off-by: Sebastian Capella <sebastian.capella@linaro.org> Reported-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com> Reviewed-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| | * | | | PM / hibernate: use unsigned local variables in swsusp_show_speed()Chen Gang2014-04-281-5/+9
| | | |_|/ | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | do_div() needs 'u64' type, or it reports warning. And negative number is meaningless for "speed", so change all signed to unsigned within swsusp_show_speed(). The related warning (with allmodconfig for unicore32): CC kernel/power/hibernate.o kernel/power/hibernate.c: In function ‘swsusp_show_speed’: kernel/power/hibernate.c:237: warning: comparison of distinct pointer types lacks a cast Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com> [rjw: Subject] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| * | | | Merge branch 'pm-cpuidle'Rafael J. Wysocki2014-06-032-13/+9
| |\ \ \ \ | | | |/ / | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * pm-cpuidle: PM / suspend: Always use deepest C-state in the "freeze" sleep state cpuidle / menu: move repeated correction factor check to init cpuidle / menu: Return (-1) if there are no suitable states cpuidle: Combine cpuidle_enabled() with cpuidle_select() ARM: clps711x: Add cpuidle driver
* | | | | Merge branch 'sched-core-for-linus' of ↵Linus Torvalds2014-06-0311-339/+522
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into next Pull scheduler updates from Ingo Molnar: "The main scheduling related changes in this cycle were: - various sched/numa updates, for better performance - tree wide cleanup of open coded nice levels - nohz fix related to rq->nr_running use - cpuidle changes and continued consolidation to improve the kernel/sched/idle.c high level idle scheduling logic. As part of this effort I pulled cpuidle driver changes from Rafael as well. - standardized idle polling amongst architectures - continued work on preparing better power/energy aware scheduling - sched/rt updates - misc fixlets and cleanups" * 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (49 commits) sched/numa: Decay ->wakee_flips instead of zeroing sched/numa: Update migrate_improves/degrades_locality() sched/numa: Allow task switch if load imbalance improves sched/rt: Fix 'struct sched_dl_entity' and dl_task_time() comments, to match the current upstream code sched: Consolidate open coded implementations of nice level frobbing into nice_to_rlimit() and rlimit_to_nice() sched: Initialize rq->age_stamp on processor start sched, nohz: Change rq->nr_running to always use wrappers sched: Fix the rq->next_balance logic in rebalance_domains() and idle_balance() sched: Use clamp() and clamp_val() to make sys_nice() more readable sched: Do not zero sg->cpumask and sg->sgp->power in build_sched_groups() sched/numa: Fix initialization of sched_domain_topology for NUMA sched: Call select_idle_sibling() when not affine_sd sched: Simplify return logic in sched_read_attr() sched: Simplify return logic in sched_copy_attr() sched: Fix exec_start/task_hot on migrated tasks arm64: Remove TIF_POLLING_NRFLAG metag: Remove TIF_POLLING_NRFLAG sched/idle: Make cpuidle_idle_call() void sched/idle: Reflow cpuidle_idle_call() sched/idle: Delay clearing the polling bit ...
| * | | | | sched/numa: Decay ->wakee_flips instead of zeroingRik van Riel2014-05-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Affine wakeups have the potential to interfere with NUMA placement. If a task wakes up too many other tasks, affine wakeups will get disabled. However, regardless of how many other tasks it wakes up, it gets re-enabled once a second, potentially interfering with NUMA placement of other tasks. By decaying wakee_wakes in half instead of zeroing it, we can avoid that problem for some workloads. Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: chegu_vinod@hp.com Cc: umgwanakikbuti@gmail.com Link: http://lkml.kernel.org/r/20140516001332.67f91af2@annuminas.surriel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched/numa: Update migrate_improves/degrades_locality()Rik van Riel2014-05-221-13/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Update the migrate_improves/degrades_locality() functions with knowledge of pseudo-interleaving. Do not consider moving tasks around within the set of group's active nodes as improving or degrading locality. Instead, leave the load balancer free to balance the load between a numa_group's active nodes. Also, switch from the group/task_weight functions to the group/task_fault functions. The "weight" functions involve a division, but both calls use the same divisor, so there's no point in doing that from these functions. On a 4 node (x10 core) system, performance of SPECjbb2005 seems unaffected, though the number of migrations with 2 8-warehouse wide instances seems to have almost halved, due to the scheduler running each instance on a single node. Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: mgorman@suse.de Cc: chegu_vinod@hp.com Link: http://lkml.kernel.org/r/20140515130306.61aae7db@cuia.bos.redhat.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched/numa: Allow task switch if load imbalance improvesRik van Riel2014-05-221-10/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the NUMA balancing code only allows moving tasks between NUMA nodes when the load on both nodes is in balance. This breaks down when the load was imbalanced to begin with. Allow tasks to be moved between NUMA nodes if the imbalance is small, or if the new imbalance is be smaller than the original one. Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: mgorman@suse.de Cc: chegu_vinod@hp.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: http://lkml.kernel.org/r/20140514132221.274b3463@annuminas.surriel.com
| * | | | | sched/rt: Fix 'struct sched_dl_entity' and dl_task_time() comments, to match ↵xiaofeng.yan2014-05-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the current upstream code Signed-off-by: xiaofeng.yan <xiaofeng.yan@huawei.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1399605687-18094-1-git-send-email-xiaofeng.yan@huawei.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Consolidate open coded implementations of nice level frobbing into ↵Dongsheng Yang2014-05-222-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | nice_to_rlimit() and rlimit_to_nice() Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/a568a1e3cc8e78648f41b5035fa5e381d36274da.1399532322.git.yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Initialize rq->age_stamp on processor startCorey Minyard2014-05-221-0/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the sched_clock time starts at a large value, the kernel will spin in sched_avg_update for a long time while rq->age_stamp catches up with rq->clock. The comment in kernel/sched/clock.c says that there is no strict promise that it starts at zero. So initialize rq->age_stamp when a cpu starts up to avoid this. I was seeing long delays on a simulator that didn't start the clock at zero. This might also be an issue on reboots on processors that don't re-initialize the timer to zero on reset, and when using kexec. Signed-off-by: Corey Minyard <cminyard@mvista.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1399574859-11714-1-git-send-email-minyard@acm.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched, nohz: Change rq->nr_running to always use wrappersKirill Tkhai2014-05-225-15/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sometimes ->nr_running may cross 2 but interrupt is not being sent to rq's cpu. In this case we don't reenable the timer. Looks like this may be the reason for rare unexpected effects, if nohz is enabled. Patch replaces all places of direct changing of nr_running and makes add_nr_running() caring about crossing border. Signed-off-by: Kirill Tkhai <tkhai@yandex.ru> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20140508225830.2469.97461.stgit@localhost Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Fix the rq->next_balance logic in rebalance_domains() and idle_balance()Jason Low2014-05-221-23/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, in idle_balance(), we update rq->next_balance when we pull_tasks. However, it is also important to update this in the !pulled_tasks case too. When the CPU is "busy" (the CPU isn't idle), rq->next_balance gets computed using sd->busy_factor (so we increase the balance interval when the CPU is busy). However, when the CPU goes idle, rq->next_balance could still be set to a large value that was computed with the sd->busy_factor. Thus, we need to also update rq->next_balance in idle_balance() in the cases where !pulled_tasks too, so that rq->next_balance gets updated without taking the busy_factor into account when the CPU is about to go idle. This patch makes rq->next_balance get updated independently of whether or not we pulled_task. Also, we add logic to ensure that we always traverse at least 1 of the sched domains to get a proper next_balance value for updating rq->next_balance. Additionally, since load_balance() modifies the sd->balance_interval, we need to re-obtain the sched domain's interval after the call to load_balance() in rebalance_domains() before we update rq->next_balance. This patch adds and uses 2 new helper functions, update_next_balance() and get_sd_balance_interval() to update next_balance and obtain the sched domain's balance_interval. Signed-off-by: Jason Low <jason.low2@hp.com> Reviewed-by: Preeti U Murthy <preeti@linux.vnet.ibm.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: daniel.lezcano@linaro.org Cc: alex.shi@linaro.org Cc: efault@gmx.de Cc: vincent.guittot@linaro.org Cc: morten.rasmussen@arm.com Cc: aswin@hp.com Link: http://lkml.kernel.org/r/1399596562.2200.7.camel@j-VirtualBox Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Use clamp() and clamp_val() to make sys_nice() more readableDongsheng Yang2014-05-221-9/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Suggested-by: Kees Cook <keescook@chromium.org> Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1399541715-19568-1-git-send-email-yangds.fnst@cn.fujitsu.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Do not zero sg->cpumask and sg->sgp->power in build_sched_groups()Dietmar Eggemann2014-05-221-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is no need to zero struct sched_group member cpumask and struct sched_group_power member power since both structures are already allocated as zeroed memory in __sdt_alloc(). This patch has been tested with BUG_ON(!cpumask_empty(sched_group_cpus(sg))); and BUG_ON(sg->sgp->power); in build_sched_groups() on ARM TC2 and INTEL i5 M520 platform including CPU hotplug scenarios. Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1398865178-12577-1-git-send-email-dietmar.eggemann@arm.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched/numa: Fix initialization of sched_domain_topology for NUMAVincent Guittot2014-05-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Jet Chen has reported a kernel panics when booting qemu-system-x86_64 with kvm64 cpu. A panic occured while building the sched_domain. In sched_init_numa, we create a new topology table in which both default levels and numa levels are copied. The last row of the table must have a null pointer in the mask field. The current implementation doesn't add this last row in the computation of the table size. So we add 1 row in the allocation size that will be used as the last row of the table. The kzalloc will ensure that the mask field is NULL. Reported-by: Jet Chen <jet.chen@intel.com> Tested-by: Jet Chen <jet.chen@intel.com> Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: fengguang.wu@intel.com Link: http://lkml.kernel.org/r/1399972261-25693-1-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Call select_idle_sibling() when not affine_sdRik van Riel2014-05-221-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On smaller systems, the top level sched domain will be an affine domain, and select_idle_sibling is invoked for every SD_WAKE_AFFINE wakeup. This seems to be working well. On larger systems, with the node distance between far away NUMA nodes being > RECLAIM_DISTANCE, select_idle_sibling is only called if the waker and the wakee are on nodes less than RECLAIM_DISTANCE apart. This patch leaves in place the policy of not pulling the task across nodes on such systems, while fixing the issue that select_idle_sibling is not called at all in certain circumstances. The code will look for an idle CPU in the same CPU package as the CPU where the task ran previously. Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: morten.rasmussen@arm.com Cc: george.mccollister@gmail.com Cc: ktkhai@parallels.com Cc: Mel Gorman <mgorman@suse.de> Cc: Mike Galbraith <umgwanakikbuti@gmail.com> Link: http://lkml.kernel.org/r/20140514114037.2d93266f@annuminas.surriel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Simplify return logic in sched_read_attr()Michael Kerrisk2014-05-221-7/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Gotos are chained pointlessly here, and the 'out' label can be dispensed with. Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/536CEC29.9090503@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Simplify return logic in sched_copy_attr()Michael Kerrisk2014-05-221-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The logic in this function is a little contorted, clean it up: * Rather than having chained gotos for the -EFBIG case, just return -EFBIG directly. * Now, the label 'out' is no longer needed, and 'ret' must be zero zero by the time we fall through to this point, so just return 0. Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/536CEC24.9080201@gmail.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | sched: Fix exec_start/task_hot on migrated tasksBen Segall2014-05-221-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | task_hot checks exec_start on any runnable task, but if it has been migrated since the it last ran, then exec_start is a clock_task from another cpu. If the old cpu's clock_task was sufficiently far ahead of this cpu's then the task will not be considered for another migration until it has run. Instead reset exec_start whenever a task is migrated, since it is presumably no longer hot anyway. Signed-off-by: Ben Segall <bsegall@google.com> [ Made it compile. ] Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/r/20140515225920.7179.13924.stgit@sword-of-the-dawn.mtv.corp.google.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | Merge branch 'sched/urgent' into sched/core to avoid conflicts with upcoming ↵Ingo Molnar2014-05-226-31/+78
| |\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | changes Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * \ \ \ \ \ Merge branch 'pm-cpuidle' of ↵Ingo Molnar2014-05-222-9/+6
| |\ \ \ \ \ \ | | | |_|/ / / | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm into sched/core Pull scheduling related CPU idle updates from Rafael J. Wysocki. Conflicts: kernel/sched/idle.c Signed-off-by: Ingo Molnar <mingo@kernel.org>
| | * | | | | PM / suspend: Always use deepest C-state in the "freeze" sleep stateRafael J. Wysocki2014-05-071-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If freeze_enter() is called, we want to bypass the current cpuidle governor and always use the deepest available (that is, not disabled) C-state, because we want to save as much energy as reasonably possible then and runtime latency constraints don't matter at that point, since the system is in a sleep state anyway. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Tested-by: Aubrey Li <aubrey.li@linux.intel.com>
| | * | | | | cpuidle: Combine cpuidle_enabled() with cpuidle_select()Rafael J. Wysocki2014-05-011-13/+7
| | | |_|/ / | | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since both cpuidle_enabled() and cpuidle_select() are only called by cpuidle_idle_call(), it is not really useful to keep them separate and combining them will help to avoid complicating cpuidle_idle_call() even further if governors are changed to return error codes sometimes. This code modification shouldn't lead to any functional changes. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
| * | | | | Merge tag 'v3.15-rc6' into sched/core, to pick up the latest fixesIngo Molnar2014-05-2219-137/+156
| |\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | | sched/idle: Make cpuidle_idle_call() voidRafael J. Wysocki2014-05-081-5/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The only value ever returned by cpuidle_idle_call() is 0 and its only caller ignores that value anyway, so make it void. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Daniel Lezcano <daniel.lezcano@linaro.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/4717784.WmVEpDoliM@vostro.rjw.lan Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | | sched/idle: Reflow cpuidle_idle_call()Peter Zijlstra2014-05-081-73/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Apply goto to reduce lines and nesting levels. Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Cc: Daniel Lezcano <daniel.lezcano@linaro.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/n/tip-cc6vb0snt3sr7op6rlbfeqfh@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | | sched/idle: Delay clearing the polling bitPeter Zijlstra2014-05-081-7/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the generic idle functions assuming !polling we should only clear the polling bit at the very last opportunity in order to avoid spurious IPIs. Ideally we'd flip the default to polling, but that means auditing all arch idle functions. Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org> Cc: Daniel Lezcano <daniel.lezcano@linaro.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/n/tip-vq7719foqzf6z5h4j7eh7f9e@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | | sched/idle: Avoid spurious wakeup IPIsPeter Zijlstra2014-05-081-5/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Because mwait_idle_with_hints() gets called from !idle context it must call current_clr_polling(). This however means that resched_task() is very likely to send an IPI even when we were polling: CPU0 CPU1 if (current_set_polling_and_test()) goto out; __monitor(&ti->flags); if (!need_resched()) __mwait(eax, ecx); set_tsk_need_resched(p); smp_mb(); out: current_clr_polling(); if (!tsk_is_polling(p)) smp_send_reschedule(cpu); So while it is correct (extra IPIs aren't a problem, whereas a missed IPI would be) it is a performance problem (for some). Avoid this issue by using fetch_or() to atomically set NEED_RESCHED and test if POLLING_NRFLAG is set. Since a CPU stuck in mwait is unlikely to modify the flags word, contention on the cmpxchg is unlikely and thus we should mostly succeed in a single go. Signed-off-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Nicolas Pitre <nico@linaro.org> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/n/tip-kf5suce6njh5xf5d3od13rr0@git.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
| * | | | | | sched/fair: Stop searching for tasks in newidle balance if there are ↵Jason Low2014-05-071-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | runnable tasks It was found that when running some workloads (such as AIM7) on large systems with many cores, CPUs do not remain idle for long. Thus, tasks can wake/get enqueued while doing idle balancing. In this patch, while traversing the domains in idle balance, in addition to checking for pulled_task, we add an extra check for this_rq->nr_running for determining if we should stop searching for tasks to pull. If there are runnable tasks on this rq, then we will stop traversing the domains. This reduces the chance that idle balance delays a task from running. This patch resulted in approximately a 6% performance improvement when running a Java Server workload on an 8 socket machine. Signed-off-by: Jason Low <jason.low2@hp.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: daniel.lezcano@linaro.org Cc: alex.shi@linaro.org Cc: preeti@linux.vnet.ibm.com Cc: efault@gmx.de Cc: vincent.guittot@linaro.org Cc: morten.rasmussen@arm.com Cc: aswin@hp.com Cc: chegu_vinod@hp.com Link: http://lkml.kernel.org/r/1398303035-18255-4-git-send-email-jason.low2@hp.com Signed-off-by: Ingo Molnar <mingo@kernel.org>