aboutsummaryrefslogtreecommitdiffstats
path: root/arch/x86
Commit message (Collapse)AuthorAgeFilesLines
* Merge branch 'x86-platform-for-linus' of ↵Linus Torvalds2009-09-1855-907/+755
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (38 commits) x86: Move get/set_wallclock to x86_platform_ops x86: platform: Fix section annotations x86: apic namespace cleanup x86: Distangle ioapic and i8259 x86: Add Moorestown early detection x86: Add hardware_subarch ID for Moorestown x86: Add early platform detection x86: Move tsc_init to late_time_init x86: Move tsc_calibration to x86_init_ops x86: Replace the now identical time_32/64.c by time.c x86: time_32/64.c unify profile_pc x86: Move calibrate_cpu to tsc.c x86: Make timer setup and global variables the same in time_32/64.c x86: Remove mca bus ifdef from timer interrupt x86: Simplify timer_ack magic in time_32.c x86: Prepare unification of time_32/64.c x86: Remove do_timer hook x86: Add timer_init to x86_init_ops x86: Move percpu clockevents setup to x86_init_ops x86: Move xen_post_allocator_init into xen_pagetable_setup_done ... Fix up conflicts in arch/x86/include/asm/io_apic.h
| * x86: Move get/set_wallclock to x86_platform_opsFeng Tang2009-09-1612-83/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | get/set_wallclock() have already a set of platform dependent implementations (default, EFI, paravirt). MRST will add another variant. Moving them to platform ops simplifies the existing code and minimizes the effort to integrate new variants. Signed-off-by: Feng Tang <feng.tang@intel.com> LKML-Reference: <new-submission> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: platform: Fix section annotationsThomas Gleixner2009-09-163-4/+4
| | | | | | | | | | | | | | | | | | init_IRQ() and x86_late_time_init() are missing __init annotations. The x86 platform ops variables are annotated, but the annotation needs to be put between the variable name and the "=" of the initializer. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: apic namespace cleanupThomas Gleixner2009-08-315-12/+12
| | | | | | | | | | | | | | boot_cpu_physical_apicid is a global variable and used as function argument as well. Rename the function arguments to avoid confusion. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Distangle ioapic and i8259Thomas Gleixner2009-08-312-12/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The proposed Moorestown support patches use an extra feature flag mechanism to make the ioapic work w/o an i8259. There is a much simpler solution. Most i8259 specific functions are already called dependend on the irq number less than NR_IRQS_LEGACY. Replacing that constant by a read_mostly variable which can be set to 0 by the platform setup code allows us to achieve the same without any special feature flags. That trivial change allows us to proceed with MRST w/o doing a full blown overhaul of the ioapic code which would delay MRST unduly. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add Moorestown early detectionThomas Gleixner2009-08-315-0/+47
| | | | | | | | | | | | | | | | | | | | | | Moorestown MID devices need to be detected early in the boot process to setup and do not call x86_default_early_setup as there is no EBDA region to reserve. [ Copied the minimal code from Jacobs latest MRST series ] Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Jacob Pan <jacob.jun.pan@intel.com>
| * x86: Add hardware_subarch ID for MoorestownPan, Jacob jun2009-08-312-0/+11
| | | | | | | | | | | | | | | | | | | | x86 bootprotocol 2.07 has introduced hardware_subarch ID in the boot parameters provided by FW. We use it to identify Moorestown platforms. [ tglx: Cleanup and paravirt fix ] Signed-off-by: Jacob Pan <jacob.jun.pan@intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add early platform detectionThomas Gleixner2009-08-314-9/+20
| | | | | | | | | | | | | | | | | | | | | | | | Platforms like Moorestown require early setup and want to avoid the call to reserve_ebda_region. The x86_init override is too late when the MRST detection happens in setup_arch. Move the default i386 x86_init overrides and the call to reserve_ebda_region into a separate function which is called as the default of a switch case depending on the hardware_subarch id in boot params. This allows us to add a case for MRST and let MRST have its own early setup function. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move tsc_init to late_time_initThomas Gleixner2009-08-311-1/+1
| | | | | | | | | | | | | | | | | | | | We do not need the TSC before late_time_init. Move the tsc_init to the late time init code so we can also utilize HPET for calibration (which we claimed to do but never did except in some older kernel version). This also helps Moorestown to calibrate the TSC with the AHBT timer which needs to be initialized in late_time_init like HPET. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move tsc_calibration to x86_init_opsThomas Gleixner2009-08-3117-41/+48
| | | | | | | | | | | | | | | | | | | | TSC calibration is modified by the vmware hypervisor and paravirt by separate means. Moorestown wants to add its own calibration routine as well. So make calibrate_tsc a proper x86_init_ops function and override it by paravirt or by the early setup of the vmware hypervisor. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Replace the now identical time_32/64.c by time.cThomas Gleixner2009-08-313-122/+1
| | | | | | | | | | | | Remove the redundant copy. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: time_32/64.c unify profile_pcThomas Gleixner2009-08-312-10/+11
| | | | | | | | | | | | | | The code is identical except for the formatting and a useless #ifdef. Make it the same. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move calibrate_cpu to tsc.cThomas Gleixner2009-08-314-56/+55
| | | | | | | | | | | | | | Move the code where it's only user is. Also we need to look whether this hardwired hackery might interfere with perfcounters. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Make timer setup and global variables the same in time_32/64.cThomas Gleixner2009-08-312-11/+35
| | | | | | | | | | | | | | | | The timer and timer irq setup code is identical in 32 and 64 bit. Make it the same formatting as well. Also add the global variables under the necessary ifdefs to both files. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Remove mca bus ifdef from timer interruptThomas Gleixner2009-08-312-21/+6
| | | | | | | | | | | | | | MCA_bus is constant 0 when CONFIG_MCA=n. So the compiler removes that code w/o needing an extra #ifdef Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Simplify timer_ack magic in time_32.cThomas Gleixner2009-08-313-4/+21
| | | | | | | | | | | | | | | | | | Let the compiler optimize the timer_ack magic away in the 32bit timer interrupt and put the same code into time_64.c. It's optimized out for CONFIG_X86_IO_APIC on 32bit and for 64bit because timer_ack is const 0 in both cases. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Prepare unification of time_32/64.cThomas Gleixner2009-08-312-39/+18
| | | | | | | | | | | | Unify the top comment and the includes. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Remove do_timer hookThomas Gleixner2009-08-312-19/+4
| | | | | | | | | | | | | | This is a left over of the old x86 sub arch support. Remove it and open code it like we do in time_64.c Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add timer_init to x86_init_opsThomas Gleixner2009-08-3117-113/+53
| | | | | | | | | | | | | | | | | | | | | | | | The timer init code is convoluted with several quirks and the paravirt timer chooser. Figuring out which code path is actually taken is not for the faint hearted. Move the numaq TSC quirk to tsc_pre_init x86_init_ops function and replace the paravirt time chooser and the remaining x86 quirk with a simple x86_init_ops function. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move percpu clockevents setup to x86_init_opsThomas Gleixner2009-08-3111-28/+42
| | | | | | | | | | | | | | | | | | | | paravirt overrides the setup of the default apic timers as per cpu timers. Moorestown needs to override that as well. Move it to x86_init_ops setup and create a separate x86_cpuinit struct which holds the function for the secondary evtl. hotplugabble CPUs. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move xen_post_allocator_init into xen_pagetable_setup_doneThomas Gleixner2009-08-317-19/+4
| | | | | | | | | | | | | | | | | | We really do not need two paravirt/x86_init_ops functions which are called in two consecutive source lines. Move the only user of post_allocator_init into the already existing pagetable_setup_done function. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move paravirt pagetable_setup to x86_init_opsThomas Gleixner2009-08-3111-46/+32
| | | | | | | | | | | | Replace more paravirt hackery by proper x86_init_ops. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move paravirt banner printout to x86_init_opsThomas Gleixner2009-08-317-14/+12
| | | | | | | | | | | | | | | | Replace another obscure paravirt magic and move it to x86_init_ops. Such a hook is also useful for embedded and special hardware. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Replace ARCH_SETUP by a proper x86_init_opsThomas Gleixner2009-08-317-9/+15
| | | | | | | | | | | | | | ARCH_SETUP is a horrible leftover from the old arch/i386 mach support code. It still has a lonely user in xen. Move it to x86_init_ops. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move traps_init to x86_init_opsThomas Gleixner2009-08-317-27/+8
| | | | | | | | | | | | Replace the quirks by a simple x86_init_ops function. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move irq_init to x86_init_opsThomas Gleixner2009-08-3112-42/+11
| | | | | | | | | | | | | | | | | | irq_init is overridden by x86_quirks and by paravirts. Unify the whole mess and make it an unconditional x86_init_ops function which defaults to the standard function and can be overridden by the early platform code. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move pre_intr_init to x86_init_opsThomas Gleixner2009-08-317-31/+22
| | | | | | | | | | | | | | | | | | Replace the quirk machinery by a x86_init_ops function which defaults to the standard implementation. This is also a preparatory patch for Moorestown support which needs to replace the default init_ISA_irqs as well. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move get/find_smp_config to x86_init_opsThomas Gleixner2009-08-318-54/+42
| | | | | | | | | | | | | | Replace the quirk machinery by a x86_init_ops function which defaults to the standard implementation. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move oem_bus_info to x86_init_opsThomas Gleixner2009-08-276-10/+18
| | | | | | | | Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move mpc_oem_pci_bus to x86_init_opsThomas Gleixner2009-08-274-4/+6
| | | | | | | | Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move smp_read_mpc_oem to x86_init_ops.Thomas Gleixner2009-08-276-10/+13
| | | | | | | | | | | | Move smp_read_mpc_oem from quirks to x86_init. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move mpc_apic_id to x86_init_opsThomas Gleixner2009-08-276-7/+15
| | | | | | | | | | | | | | The mpc_apic_id setup is handled by a x86_quirk. Make it a x86_init_ops function with a default implementation. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move ioapic_ids_setup to x86_init_opsThomas Gleixner2009-08-277-16/+13
| | | | | | | | | | | | | | | | 32bit and also the numaq code have special requirements on the ioapic_id setup. Convert it to a x86_init_ops function and get rid of the quirks and #ifdefs Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Sanitize smp_record and move it to x86_init_opsThomas Gleixner2009-08-275-9/+34
| | | | | | | | | | | | | | | | | | | | | | The x86 quirkification introduced an extra ugly hackery with a variable pointer in the mpparse code. If the pointer is initialized then it is dereferenced and the variable set to 0 or incremented. Create a x86_init_ops function and let the affected numaq code hold the function. Default init is a setup noop. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Move memory_setup to x86_init_opsThomas Gleixner2009-08-2711-32/+11
| | | | | | | | | | | | | | | | | | memory_setup is overridden by x86_quirks and by paravirts with weak functions and quirks. Unify the whole mess and make it an unconditional x86_init_ops function which defaults to the standard function and can be overridden by the early platform code. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add reserve_ebda_region to x86_init_opsThomas Gleixner2009-08-274-4/+7
| | | | | | | | | | | | | | | | reserve_ebda_region needs to be called befor start_kernel. Moorestown needs to override it. Make it a x86_init_ops function and initialize it with the default reserve_ebda_region. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add request_standard_resources to x86_initThomas Gleixner2009-08-275-13/+25
| | | | | | | | | | | | | | | | | | | | | | | | The 32bit and the 64bit code are slighty different in the reservation of standard resources. Also the upcoming Moorestown support needs its own version of that. Add it to x86_init_ops and initialize it with the 64bit default. 32bit overrides it in early boot. Now moorestown can add it's own override w/o sprinkling the code with more #ifdefs Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add probe_roms to x86_initThomas Gleixner2009-08-274-3/+18
| | | | | | | | | | | | | | | | | | | | probe_roms is only used on 32bit. Add it to the x86_init ops and remove the #ifdefs. Default initializer is x86_init_noop() which is overridden in the 32bit boot code. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * x86: Add x86_init infrastructureThomas Gleixner2009-08-274-1/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The upcoming Moorestown support brings the embedded world to x86. The setup code of x86 has already a couple of hooks which are either x86_quirks or paravirt ops. Some of those setup hooks are pretty convoluted like the timer setup and the tsc calibration code. But there are other places which could do with a cleanup. Instead of having inline functions/macros which are modified at compile time I decided to introduce x86_init ops which are unconditional in the code and make it clear that they can be changed either during compile time or in the early boot process. The function pointers are initialized by default functions which can be noops so that the pointer can be called unconditionally in the most cases. This also allows us to remove 32bit/64bit, paravirt and other #ifdeffery. paravirt guests are just a hardware platform in the setup code, so we should treat them as such and not hide all behind multiple layers of indirection and compile time dependencies. It's more obvious that x86_init.timers.timer_init() is a function pointer than the late_time_init = choose_time_init() obscurity. It's also way simpler to grep for x86_init.timers.timer_init and find all the places which modify that function pointer instead of analyzing weak functions, macros and paravirt indirections. Note. This is not a general paravirt_ops replacement. It just will move setup related hooks which are potentially useful for other platform setup purposes as well out of the paravirt domain. Add the base infrastructure without any functionality. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * Merge branch 'sched/clock' into x86/cleanupsThomas Gleixner2009-08-2710-108/+81
| |\ | | | | | | | | | | | | | | | | | | Reason: The tsc init cleanup depends on sched_clock_init moving past late_time_init. Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * \ Merge branch 'x86/paravirt' into x86/cleanupsThomas Gleixner2009-08-272-710/+721
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | Reason: The setup cleanups conflict with the paravirt cleanups. Avoid a rather large merge conflict Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* | \ \ Merge branch 'next' of ↵Linus Torvalds2009-09-182-32/+33
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq * 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq: [CPUFREQ] Fix NULL ptr regression in powernow-k8 [CPUFREQ] Create a blacklist for processors that should not load the acpi-cpufreq module. [CPUFREQ] Powernow-k8: Enable more than 2 low P-states [CPUFREQ] remove rwsem lock from CPUFREQ_GOV_STOP call (second call site) [CPUFREQ] ondemand - Use global sysfs dir for tuning settings [CPUFREQ] Introduce global, not per core: /sys/devices/system/cpu/cpufreq [CPUFREQ] Bail out of cpufreq_add_dev if the link for a managed CPU got created [CPUFREQ] Factor out policy setting from cpufreq_add_dev [CPUFREQ] Factor out interface creation from cpufreq_add_dev [CPUFREQ] Factor out symlink creation from cpufreq_add_dev [CPUFREQ] cleanup up -ENOMEM handling in cpufreq_add_dev [CPUFREQ] Reduce scope of cpu_sys_dev in cpufreq_add_dev [CPUFREQ] update Doc for cpuinfo_cur_freq and scaling_cur_freq
| * | | | [CPUFREQ] Fix NULL ptr regression in powernow-k8Kurt Roeckx2009-09-161-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes bugzilla #13780 From: Kurt Roeckx <kurt@roeckx.be> Signed-off-by: Dave Jones <davej@redhat.com>
| * | | | [CPUFREQ] Create a blacklist for processors that should not load the ↵Prarit Bhargava2009-09-011-0/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | acpi-cpufreq module. Create a blacklist for processors that should not load the acpi-cpufreq module. The initial entry in the blacklist function is the Intel 0f68 processor. It's specification update mentions errata AL30 which implies that cpufreq should not run on this processor. Signed-off-by: Prarit Bhargava <prarit@redhat.com> Signed-off-by: Dave Jones <davej@redhat.com>
| * | | | [CPUFREQ] Powernow-k8: Enable more than 2 low P-statesMark Langsdorf2009-09-011-25/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove an obsolete check that used to prevent there being more than 2 low P-states. Now that low-to-low P-states changes are enabled, it prevents otherwise workable configurations with multiple low P-states. Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com> Tested-by: Krists Krilovs <pow@pow.za.net> Signed-off-by: Dave Jones <davej@redhat.com>
* | | | | Merge branch 'timers-for-linus' of ↵Linus Torvalds2009-09-186-27/+52
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (34 commits) time: Prevent 32 bit overflow with set_normalized_timespec() clocksource: Delay clocksource down rating to late boot clocksource: clocksource_select must be called with mutex locked clocksource: Resolve cpu hotplug dead lock with TSC unstable, fix crash timers: Drop a function prototype clocksource: Resolve cpu hotplug dead lock with TSC unstable timer.c: Fix S/390 comments timekeeping: Fix invalid getboottime() value timekeeping: Fix up read_persistent_clock() breakage on sh timekeeping: Increase granularity of read_persistent_clock(), build fix time: Introduce CLOCK_REALTIME_COARSE x86: Do not unregister PIT clocksource on PIT oneshot setup/shutdown clocksource: Avoid clocksource watchdog circular locking dependency clocksource: Protect the watchdog rating changes with clocksource_mutex clocksource: Call clocksource_change_rating() outside of watchdog_lock timekeeping: Introduce read_boot_clock timekeeping: Increase granularity of read_persistent_clock() timekeeping: Update clocksource with stop_machine timekeeping: Add timekeeper read_clock helper functions timekeeping: Move NTP adjusted clock multiplier to struct timekeeper ... Fix trivial conflict due to MIPS lemote -> loongson renaming.
| * | | | | clocksource: Resolve cpu hotplug dead lock with TSC unstableThomas Gleixner2009-08-281-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Martin Schwidefsky analyzed it: To register a clocksource the clocksource_mutex is acquired and if necessary timekeeping_notify is called to install the clocksource as the timekeeper clock. timekeeping_notify uses stop_machine which needs to take cpu_add_remove_lock mutex. Starting a new cpu is done with the cpu_add_remove_lock mutex held. native_cpu_up checks the tsc of the new cpu and if the tsc is no good clocksource_change_rating is called. Which needs the clocksource_mutex and the deadlock is complete. The solution is to replace the TSC via the clocksource watchdog mechanism. Mark the TSC as unstable and schedule the watchdog work so it gets removed in the watchdog thread context. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <new-submission> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: John Stultz <johnstul@us.ibm.com>
| * | | | | time: Introduce CLOCK_REALTIME_COARSEjohn stultz2009-08-213-3/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After talking with some application writers who want very fast, but not fine-grained timestamps, I decided to try to implement new clock_ids to clock_gettime(): CLOCK_REALTIME_COARSE and CLOCK_MONOTONIC_COARSE which returns the time at the last tick. This is very fast as we don't have to access any hardware (which can be very painful if you're using something like the acpi_pm clocksource), and we can even use the vdso clock_gettime() method to avoid the syscall. The only trade off is you only get low-res tick grained time resolution. This isn't a new idea, I know Ingo has a patch in the -rt tree that made the vsyscall gettimeofday() return coarse grained time when the vsyscall64 sysctrl was set to 2. However this affects all applications on a system. With this method, applications can choose the proper speed/granularity trade-off for themselves. Signed-off-by: John Stultz <johnstul@us.ibm.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: nikolag@ca.ibm.com Cc: Darren Hart <dvhltc@us.ibm.com> Cc: arjan@infradead.org Cc: jonathan@jonmasters.org LKML-Reference: <1250734414.6897.5.camel@localhost.localdomain> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * | | | | x86: Do not unregister PIT clocksource on PIT oneshot setup/shutdownThomas Gleixner2009-08-211-19/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This basically reverts commit 1a0c009ac (x86: unregister PIT clocksource when PIT is disabled) because the problem which was tried to address with that patch has been solved by commit 3f68535ada (clocksource: sanity check sysfs clocksource changes). The problem addressed by the original patch is that PIT could be selected as clocksource after the system switched the PIT off or set the PIT into one shot mode which would result in complete timekeeping wreckage. Now with the sysfs sanity check in place PIT cannot be selected again when the system is in oneshot mode. The system will not switch to one shot mode as long as PIT is installed because PIT is not suitable for one shot. The shutdown case which happens when the lapic timer is installed is covered by the fact that init_pit_clocksource() is called after the lapic timer take over and then does not install the PIT clocksource at all. We should have done the sanity checks back then, but ... This also solves the locking problem which was reported vs. the clocksource rework. LKML-Reference: <new-submission> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: john stultz <johnstul@us.ibm.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| * | | | | timekeeping: Increase granularity of read_persistent_clock()Martin Schwidefsky2009-08-151-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The persistent clock of some architectures (e.g. s390) have a better granularity than seconds. To reduce the delta between the host clock and the guest clock in a virtualized system change the read_persistent_clock function to return a struct timespec. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Ingo Molnar <mingo@elte.hu> Acked-by: John Stultz <johnstul@us.ibm.com> Cc: Daniel Walker <dwalker@fifo99.com> LKML-Reference: <20090814134811.013873340@de.ibm.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
generated by cgit v1.1 at 2024-12-26 21:19:11 +0000