
* refs/heads/tmp-cf5b248: BACKPORT: FROMLIST: arm64: Remove logic to kill 32-bit tasks on 64-bit-only cores FROMLIST: arm64: Hook up cmdline parameter to allow mismatched 32-bit EL0 FROMLIST: arm64: Prevent offlining first CPU with 32-bit EL0 on mismatched system FROMLIST: arm64: exec: Adjust affinity for compat tasks with mismatched 32-bit EL0 FROMLIST: arm64: Implement task_cpu_possible_mask() FROMLIST: sched: Introduce force_compatible_cpus_allowed_ptr() to limit CPU affinity FROMLIST: sched: Reject CPU affinity changes based on task_cpu_possible_mask() BACKPORT: FROMLIST: cpuset: Honour task_cpu_possible_mask() in guarantee_online_cpus() FROMLIST: cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1 FROMLIST: sched: Introduce task_cpu_possible_mask() to limit fallback rq selection FROMLIST: arm64: Advertise CPUs capable of running 32-bit applications in sysfs BACKPORT: FROMLIST: arm64: Kill 32-bit applications scheduled on 64-bit-only CPUs FROMLIST: KVM: arm64: Kill 32-bit vCPUs on systems with mismatched EL0 support BACKPORT: FROMLIST: arm64: Allow mismatched 32-bit EL0 support FROMLIST: arm64: cpuinfo: Split AArch32 registers out into a separate struct Revert "ANDROID: arm64: Add support for asymmetric AArch32 EL0 configurations" Revert "ANDROID: arm64: Handle AArch32 tasks running on non AArch32 cpu" Revert "ANDROID: arm64: Disallow offlining the last aarch32 cpu" Revert "ANDROID: arm64: kvm: Hide asym aarch32 systems from KVM" Revert "ANDROID: arm64: Enable KVM for Asym AArch32" ANDROID: gki_defconfig: Remove CONFIG_ASYMMETRIC_AARCH32=y UPSTREAM: usb: pd: Reland VDO definitions of PD2.0 ANDROID: sched: Add PELT cmdline arg ANDROID: psci: use __pa_function for cpu_resume ANDROID: arm64: kernel: use __pa_function for secondary_entry ANDROID: arm64: add vendor hooks for kernel fault cases ANDROID: sched: add vendor hooks for bad scheduling ANDROID: power: add vendor hooks for try_to_freeze fail ANDROID: softlockup: add vendor hook for a softlockup task ANDROID: Fix sparse warning in __handle_speculative_fault caused by SPF ANDROID: mm, oom: Fix select_bad_process customization ANDROID: mm: sync rss in speculative page fault path ANDROID: GKI: Update ABI XML report ANDROID: dma-buf: heaps: Add a sysfs file to report total pool size. ANDROID: GKI: enable CONFIG_USB_DUMMY_HCD ANDROID: GKI: Update ABI XML report ANDROID: GKI: defconfig: Enable DMA-BUF sysfs stats ANDROID: zram: allow zram to allocate CMA pages Linux 5.10.13 vsock: fix the race conditions in multi-transport support tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN tcp: make TCP_USER_TIMEOUT accurate for zero window probes team: protect features update by RCU to avoid deadlock scsi: qla2xxx: Fix description for parameter ql2xenforce_iocb_limit ASoC: topology: Fix memory corruption in soc_tplg_denum_create_values() ASoC: topology: Properly unregister DAI on removal ASoC: mediatek: mt8183-mt6358: ignore TDM DAI link by default ASoC: mediatek: mt8183-da7219: ignore TDM DAI link by default NFC: fix possible resource leak NFC: fix resource leak when target index is invalid rxrpc: Fix memory leak in rxrpc_lookup_local selftests: forwarding: Specify interface when invoking mausezahn nvme-multipath: Early exit if no path is available iommu/vt-d: Correctly check addr alignment in qi_flush_dev_iotlb_pasid() iommu/amd: Use IVHD EFR for early initialization of IOMMU features of/device: Update dma_range_map only when dev has valid dma-ranges ACPI/IORT: Do not blindly trust DMA masks from firmware can: dev: prevent potential information leak in can_fill_info() net/mlx5: CT: Fix incorrect removal of tuple_nat_node from nat rhashtable net/mlx5e: Revert parameters on errors when changing MTU and LRO state without reset net/mlx5e: Revert parameters on errors when changing trust state without reset net/mlx5e: Correctly handle changing the number of queues when the interface is down net/mlx5e: Fix CT rule + encap slow path offload and deletion net/mlx5e: Disable hw-tc-offload when MLX5_CLS_ACT config is disabled net/mlx5: Maintain separate page trees for ECPF and PF functions net/mlx5e: Reduce tc unsupported key print level net/mlx5e: free page before return net/mlx5e: E-switch, Fix rate calculation for overflow net/mlx5: Fix memory leak on flow table creation error flow igc: fix link speed advertising i40e: acquire VSI pointer only after VF is initialized ice: Fix MSI-X vector fallback logic ice: Don't allow more channels than LAN MSI-X available ice: update dev_addr in ice_set_mac_address even if HW filter exists ice: Implement flow for IPv6 next header (extension header) ice: fix FDir IPv6 flexbyte mac80211: pause TX while changing interface type iwlwifi: pcie: reschedule in long-running memory reads iwlwifi: pcie: use jiffies for memory read spin time limit iwlwifi: pcie: set LTR on more devices iwlwifi: pnvm: don't try to load after failures iwlwifi: pnvm: don't skip everything when not reloading iwlwifi: pcie: avoid potential PNVM leaks ASoC: qcom: lpass: Fix out-of-bounds DAI ID lookup ASoC: SOF: Intel: soundwire: fix select/depend unmet dependencies pNFS/NFSv4: Update the layout barrier when we schedule a layoutreturn pNFS/NFSv4: Fix a layout segment leak in pnfs_layout_process() powerpc/64s: prevent recursive replay_soft_interrupts causing superfluous interrupt ASoC: Intel: Skylake: skl-topology: Fix OOPs ib skl_tplg_complete spi: altera: Fix memory leak on error path ASoC: qcom: lpass-ipq806x: fix bitwidth regmap field ASoC: qcom: Fix broken support to MI2S TERTIARY and QUATERNARY ASoC: qcom: Fix incorrect volatile registers ASoC: dt-bindings: lpass: Fix and common up lpass dai ids RDMA/cxgb4: Fix the reported max_recv_sge value firmware: imx: select SOC_BUS to fix firmware build arm64: dts: imx8mp: Correct the gpio ranges of gpio3 ARM: dts: imx6qdl-sr-som: fix some cubox-i platforms ARM: dts: imx6qdl-kontron-samx6i: fix i2c_lcd/cam default status ARM: imx: fix imx8m dependencies arm64: dts: ls1028a: fix the offset of the reset register xfrm: Fix wraparound in xfrm_policy_addr_delta() selftests: xfrm: fix test return value override issue in xfrm_policy.sh xfrm: fix disable_xfrm sysctl when used on xfrm interfaces xfrm: Fix oops in xfrm_replay_advance_bmp Revert "block: simplify set_init_blocksize" to regain lost performance Revert "RDMA/mlx5: Fix devlink deadlock on net namespace deletion" netfilter: nft_dynset: add timeout extension to template ARM: zImage: atags_to_fdt: Fix node names on added root nodes ARM: imx: build suspend-imx6.S with arm instruction set clk: qcom: gcc-sm250: Use floor ops for sdcc clks clk: mmp2: fix build without CONFIG_PM clk: imx: fix Kconfig warning for i.MX SCU clk blk-mq: test QUEUE_FLAG_HCTX_ACTIVE for sbitmap_shared in hctx_may_queue xen-blkfront: allow discard-* nodes to be optional tee: optee: replace might_sleep with cond_resched KVM: Documentation: Fix spec for KVM_CAP_ENABLE_CAP_VM uapi: fix big endian definition of ipv6_rpl_sr_hdr drm/i915/selftest: Fix potential memory leak drm/i915: Check for all subplatform bits drm/nouveau/dispnv50: Restore pushing of all data. drm/vc4: Correct POS1_SCL for hvs5 drm/vc4: Correct lbm size and calculation drm/nouveau/svm: fail NOUVEAU_SVM_INIT ioctl on unsupported devices ARM: dts: imx6qdl-kontron-samx6i: fix pwms for lcd-backlight net/mlx5e: Fix IPSEC stats drm/i915/pmu: Don't grab wakeref when enabling events drm/i915/gt: Clear CACHE_MODE prior to clearing residuals iwlwifi: Fix IWL_SUBDEVICE_NO_160 macro to use the correct bit. mt7601u: fix rx buffer refcounting mt76: mt7663s: fix rx buffer refcounting mt7601u: fix kernel crash unplugging the device arm64: dts: broadcom: Fix USB DMA address translation for Stingray leds: trigger: fix potential deadlock with libata xen: Fix XenStore initialisation for XS_LOCAL io_uring: fix wqe->lock/completion_lock deadlock KVM: Forbid the use of tagged userspace addresses for memslots KVM: x86: get smi pending status correctly KVM: nVMX: Sync unsync'd vmcs02 state to vmcs12 on migration KVM: x86: allow KVM_REQ_GET_NESTED_STATE_PAGES outside guest mode for VMX KVM: nSVM: cancel KVM_REQ_GET_NESTED_STATE_PAGES on nested vmexit KVM: arm64: Filter out v8.1+ events on v8.0 HW KVM: x86/pmu: Fix UBSAN shift-out-of-bounds warning in intel_pmu_refresh() KVM: x86/pmu: Fix HW_REF_CPU_CYCLES event pseudo-encoding in intel_arch_events[] btrfs: fix possible free space tree corruption with online conversion btrfs: fix lockdep warning due to seqcount_mutex on 32bit arch drivers: soc: atmel: add null entry at the end of at91_soc_allowed_list[] drivers: soc: atmel: Avoid calling at91_soc_init on non AT91 SoCs crypto: marvel/cesa - Fix tdma descriptor on 64-bit efi/apple-properties: Reinstate support for boolean properties x86/entry: Emit a symbol for register restoring thunk PM: hibernate: flush swap writer after marking s390/vfio-ap: No need to disable IRQ after queue reset s390: uv: Fix sysfs max number of VCPUs reporting net: usb: qmi_wwan: added support for Thales Cinterion PLSx3 modem family bcache: only check feature sets when sb->version >= BCACHE_SB_VERSION_CDEV_WITH_FEATURES drivers/nouveau/kms/nv50-: Reject format modifiers for cursor planes drm/i915/gt: Always try to reserve GGTT address 0x0 drm/i915: Always flush the active worker before returning from the wait drm/nouveau/kms/gk104-gp1xx: Fix > 64x64 cursors Revert "drm/amdgpu/swsmu: drop set_fan_speed_percent (v2)" ASoC: AMD Renoir - refine DMI entries for some Lenovo products x86/xen: avoid warning in Xen pv guest with CONFIG_AMD_MEM_ENCRYPT enabled wext: fix NULL-ptr-dereference with cfg80211's lack of commit() ARM: dts: imx6qdl-gw52xx: fix duplicate regulator naming ARM: dts: ux500: Reserve memory carveouts ARM: dts: tbs2910: rename MMC node aliases media: rc: ensure that uevent can be read directly after rc device register media: rc: ite-cir: fix min_timeout calculation media: rc: fix timeout handling after switch to microsecond durations media: hantro: Fix reset_raw_fmt initialization media: cedrus: Fix H264 decoding media: cec: add stm32 driver parisc: Enable -mlong-calls gcc option by default when !CONFIG_MODULES ALSA: hda/via: Apply the workaround generically for Clevo machines ALSA: hda/realtek: Enable headset of ASUS B1400CEPE with ALC256 kernel: kexec: remove the lock operation of system_transition_mutex ACPI: thermal: Do not call acpi_thermal_check() directly ACPI: sysfs: Prefer "compatible" modalias tty: avoid using vfs_iocb_iter_write() for redirected console writes nbd: freeze the queue while we're adding connections iwlwifi: provide gso_type to GSO packets ANDROID: Add filp_open_block() for zram UPSTREAM: usb: pd: Update VDO definitions UPSTREAM: xhci: fix bounce buffer usage for non-sg list case UPSTREAM: usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720 UPSTREAM: usb: xhci-mtk: break loop when find the endpoint to drop UPSTREAM: usb: typec: Return void in typec_partner_set_pd_revision ANDROID: GKI: Update ABI with virtual_device symbols ANDROID: make per-cgroup PSI tracking configurable BACKPORT: FROMLIST: dmabuf: Add the capability to expose DMA-BUF stats in sysfs UPSTREAM: usb: typec: tcpci_maxim: Enable data path when partner is USB Comm capable UPSTREAM: usb: typec: tcpci: Add Callback to Usb Communication capable partner UPSTREAM: usb: typec: tcpm: Add Callback to Usb Communication capable partner UPSTREAM: usb: typec: tcpm: Set in_ams flag when Source caps have been received UPSTREAM: usb: typec: tcpm: Handle vbus shutoff when in source mode ANDROID: GKI: Update virtual_device symbol list ANDROID: timer: Add vendor hook for timer calc index ANDROID: Make vsock virtio packet buff size configurable ANDROID: ipi: Add function to return nr_ipi and ipi_desc UPSTREAM: usb: typec: Add typec_partner_set_pd_revision UPSTREAM: usb: typec: Provide PD Specification Revision for cable and partner UPSTREAM: usb: typec: Standardize PD Revision format with Type-C Revision UPSTREAM: usb: xhci-mtk: skip dropping bandwidth of unchecked endpoints UPSTREAM: modpost: turn static exports into error Revert "FROMLIST: modpost: Make static exports fatal" UPSTREAM: modpost: turn section mismatches to error from fatal() UPSTREAM: modpost: change license incompatibility to error() from fatal() UPSTREAM: modpost: turn missing MODULE_LICENSE() into error UPSTREAM: modpost: refactor error handling and clarify error/fatal difference UPSTREAM: modpost: rename merror() to error() Revert "ANDROID: GKI: bring WPAN into GKI" ANDROID: GKI: update .xml file ANDROID: GKI: bring WPAN into GKI ANDROID: db845c_gki.fragment: Drop CONFIG_USB_XHCI_HCD Revert "ANDROID: Update db845c KMI symbol list for DWC3 changes" Revert "Revert "ANDROID: GKI: Enable CONFIG_USB_XHCI_HCD"" Revert "Revert "ANDROID: db845c_gki.fragment: Drop CONFIG_USB_DWC3 from config frament"" Revert "Revert "ANDROID: GKI: enable CONFIG_USB_DWC3 to be build in"" Conflicts: arch/arm64/Kconfig init/Kconfig Change-Id: I931bd41521892079722e6b8211d4864a685d3cb8 Signed-off-by: Ivaylo Georgiev <irgeorgiev@codeaurora.org>
785 lines
21 KiB
C
785 lines
21 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* Detect hard and soft lockups on a system
|
|
*
|
|
* started by Don Zickus, Copyright (C) 2010 Red Hat, Inc.
|
|
*
|
|
* Note: Most of this code is borrowed heavily from the original softlockup
|
|
* detector, so thanks to Ingo for the initial implementation.
|
|
* Some chunks also taken from the old x86-specific nmi watchdog code, thanks
|
|
* to those contributors as well.
|
|
*/
|
|
|
|
#define pr_fmt(fmt) "watchdog: " fmt
|
|
|
|
#include <linux/mm.h>
|
|
#include <linux/cpu.h>
|
|
#include <linux/device.h>
|
|
#include <linux/nmi.h>
|
|
#include <linux/init.h>
|
|
#include <linux/module.h>
|
|
#include <linux/sysctl.h>
|
|
#include <linux/tick.h>
|
|
#include <linux/sched/clock.h>
|
|
#include <linux/sched/debug.h>
|
|
#include <linux/sched/isolation.h>
|
|
#include <linux/stop_machine.h>
|
|
|
|
#include <asm/irq_regs.h>
|
|
#include <linux/kvm_para.h>
|
|
|
|
#include <trace/hooks/softlockup.h>
|
|
|
|
static DEFINE_MUTEX(watchdog_mutex);
|
|
|
|
#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG)
|
|
# define WATCHDOG_DEFAULT (SOFT_WATCHDOG_ENABLED | NMI_WATCHDOG_ENABLED)
|
|
# define NMI_WATCHDOG_DEFAULT 1
|
|
#else
|
|
# define WATCHDOG_DEFAULT (SOFT_WATCHDOG_ENABLED)
|
|
# define NMI_WATCHDOG_DEFAULT 0
|
|
#endif
|
|
|
|
unsigned long __read_mostly watchdog_enabled;
|
|
int __read_mostly watchdog_user_enabled = 1;
|
|
int __read_mostly nmi_watchdog_user_enabled = NMI_WATCHDOG_DEFAULT;
|
|
int __read_mostly soft_watchdog_user_enabled = 1;
|
|
int __read_mostly watchdog_thresh = 10;
|
|
static int __read_mostly nmi_watchdog_available;
|
|
|
|
struct cpumask watchdog_cpumask __read_mostly;
|
|
unsigned long *watchdog_cpumask_bits = cpumask_bits(&watchdog_cpumask);
|
|
|
|
#ifdef CONFIG_HARDLOCKUP_DETECTOR
|
|
|
|
# ifdef CONFIG_SMP
|
|
int __read_mostly sysctl_hardlockup_all_cpu_backtrace;
|
|
# endif /* CONFIG_SMP */
|
|
|
|
/*
|
|
* Should we panic when a soft-lockup or hard-lockup occurs:
|
|
*/
|
|
unsigned int __read_mostly hardlockup_panic =
|
|
CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE;
|
|
/*
|
|
* We may not want to enable hard lockup detection by default in all cases,
|
|
* for example when running the kernel as a guest on a hypervisor. In these
|
|
* cases this function can be called to disable hard lockup detection. This
|
|
* function should only be executed once by the boot processor before the
|
|
* kernel command line parameters are parsed, because otherwise it is not
|
|
* possible to override this in hardlockup_panic_setup().
|
|
*/
|
|
void __init hardlockup_detector_disable(void)
|
|
{
|
|
nmi_watchdog_user_enabled = 0;
|
|
}
|
|
|
|
static int __init hardlockup_panic_setup(char *str)
|
|
{
|
|
if (!strncmp(str, "panic", 5))
|
|
hardlockup_panic = 1;
|
|
else if (!strncmp(str, "nopanic", 7))
|
|
hardlockup_panic = 0;
|
|
else if (!strncmp(str, "0", 1))
|
|
nmi_watchdog_user_enabled = 0;
|
|
else if (!strncmp(str, "1", 1))
|
|
nmi_watchdog_user_enabled = 1;
|
|
return 1;
|
|
}
|
|
__setup("nmi_watchdog=", hardlockup_panic_setup);
|
|
|
|
#endif /* CONFIG_HARDLOCKUP_DETECTOR */
|
|
|
|
/*
|
|
* These functions can be overridden if an architecture implements its
|
|
* own hardlockup detector.
|
|
*
|
|
* watchdog_nmi_enable/disable can be implemented to start and stop when
|
|
* softlockup watchdog threads start and stop. The arch must select the
|
|
* SOFTLOCKUP_DETECTOR Kconfig.
|
|
*/
|
|
int __weak watchdog_nmi_enable(unsigned int cpu)
|
|
{
|
|
hardlockup_detector_perf_enable();
|
|
return 0;
|
|
}
|
|
|
|
void __weak watchdog_nmi_disable(unsigned int cpu)
|
|
{
|
|
hardlockup_detector_perf_disable();
|
|
}
|
|
|
|
/* Return 0, if a NMI watchdog is available. Error code otherwise */
|
|
int __weak __init watchdog_nmi_probe(void)
|
|
{
|
|
return hardlockup_detector_perf_init();
|
|
}
|
|
|
|
/**
|
|
* watchdog_nmi_stop - Stop the watchdog for reconfiguration
|
|
*
|
|
* The reconfiguration steps are:
|
|
* watchdog_nmi_stop();
|
|
* update_variables();
|
|
* watchdog_nmi_start();
|
|
*/
|
|
void __weak watchdog_nmi_stop(void) { }
|
|
|
|
/**
|
|
* watchdog_nmi_start - Start the watchdog after reconfiguration
|
|
*
|
|
* Counterpart to watchdog_nmi_stop().
|
|
*
|
|
* The following variables have been updated in update_variables() and
|
|
* contain the currently valid configuration:
|
|
* - watchdog_enabled
|
|
* - watchdog_thresh
|
|
* - watchdog_cpumask
|
|
*/
|
|
void __weak watchdog_nmi_start(void) { }
|
|
|
|
/**
|
|
* lockup_detector_update_enable - Update the sysctl enable bit
|
|
*
|
|
* Caller needs to make sure that the NMI/perf watchdogs are off, so this
|
|
* can't race with watchdog_nmi_disable().
|
|
*/
|
|
static void lockup_detector_update_enable(void)
|
|
{
|
|
watchdog_enabled = 0;
|
|
if (!watchdog_user_enabled)
|
|
return;
|
|
if (nmi_watchdog_available && nmi_watchdog_user_enabled)
|
|
watchdog_enabled |= NMI_WATCHDOG_ENABLED;
|
|
if (soft_watchdog_user_enabled)
|
|
watchdog_enabled |= SOFT_WATCHDOG_ENABLED;
|
|
}
|
|
|
|
#ifdef CONFIG_SOFTLOCKUP_DETECTOR
|
|
|
|
#define SOFTLOCKUP_RESET ULONG_MAX
|
|
|
|
#ifdef CONFIG_SMP
|
|
int __read_mostly sysctl_softlockup_all_cpu_backtrace;
|
|
#endif
|
|
|
|
static struct cpumask watchdog_allowed_mask __read_mostly;
|
|
|
|
/* Global variables, exported for sysctl */
|
|
unsigned int __read_mostly softlockup_panic =
|
|
CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE;
|
|
|
|
static bool softlockup_initialized __read_mostly;
|
|
static u64 __read_mostly sample_period;
|
|
|
|
static DEFINE_PER_CPU(unsigned long, watchdog_touch_ts);
|
|
static DEFINE_PER_CPU(struct hrtimer, watchdog_hrtimer);
|
|
static DEFINE_PER_CPU(unsigned int, watchdog_en);
|
|
static DEFINE_PER_CPU(bool, softlockup_touch_sync);
|
|
static DEFINE_PER_CPU(bool, soft_watchdog_warn);
|
|
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts);
|
|
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved);
|
|
static unsigned long soft_lockup_nmi_warn;
|
|
|
|
static int __init nowatchdog_setup(char *str)
|
|
{
|
|
watchdog_user_enabled = 0;
|
|
return 1;
|
|
}
|
|
__setup("nowatchdog", nowatchdog_setup);
|
|
|
|
static int __init nosoftlockup_setup(char *str)
|
|
{
|
|
soft_watchdog_user_enabled = 0;
|
|
return 1;
|
|
}
|
|
__setup("nosoftlockup", nosoftlockup_setup);
|
|
|
|
static int __init watchdog_thresh_setup(char *str)
|
|
{
|
|
get_option(&str, &watchdog_thresh);
|
|
return 1;
|
|
}
|
|
__setup("watchdog_thresh=", watchdog_thresh_setup);
|
|
|
|
static void __lockup_detector_cleanup(void);
|
|
|
|
/*
|
|
* Hard-lockup warnings should be triggered after just a few seconds. Soft-
|
|
* lockups can have false positives under extreme conditions. So we generally
|
|
* want a higher threshold for soft lockups than for hard lockups. So we couple
|
|
* the thresholds with a factor: we make the soft threshold twice the amount of
|
|
* time the hard threshold is.
|
|
*/
|
|
static int get_softlockup_thresh(void)
|
|
{
|
|
return watchdog_thresh * 2;
|
|
}
|
|
|
|
/*
|
|
* Returns seconds, approximately. We don't need nanosecond
|
|
* resolution, and we don't need to waste time with a big divide when
|
|
* 2^30ns == 1.074s.
|
|
*/
|
|
static unsigned long get_timestamp(void)
|
|
{
|
|
return running_clock() >> 30LL; /* 2^30 ~= 10^9 */
|
|
}
|
|
|
|
static void set_sample_period(void)
|
|
{
|
|
/*
|
|
* convert watchdog_thresh from seconds to ns
|
|
* the divide by 5 is to give hrtimer several chances (two
|
|
* or three with the current relation between the soft
|
|
* and hard thresholds) to increment before the
|
|
* hardlockup detector generates a warning
|
|
*/
|
|
sample_period = get_softlockup_thresh() * ((u64)NSEC_PER_SEC / 5);
|
|
watchdog_update_hrtimer_threshold(sample_period);
|
|
}
|
|
|
|
/* Commands for resetting the watchdog */
|
|
static void __touch_watchdog(void)
|
|
{
|
|
__this_cpu_write(watchdog_touch_ts, get_timestamp());
|
|
}
|
|
|
|
/**
|
|
* touch_softlockup_watchdog_sched - touch watchdog on scheduler stalls
|
|
*
|
|
* Call when the scheduler may have stalled for legitimate reasons
|
|
* preventing the watchdog task from executing - e.g. the scheduler
|
|
* entering idle state. This should only be used for scheduler events.
|
|
* Use touch_softlockup_watchdog() for everything else.
|
|
*/
|
|
notrace void touch_softlockup_watchdog_sched(void)
|
|
{
|
|
/*
|
|
* Preemption can be enabled. It doesn't matter which CPU's timestamp
|
|
* gets zeroed here, so use the raw_ operation.
|
|
*/
|
|
raw_cpu_write(watchdog_touch_ts, SOFTLOCKUP_RESET);
|
|
}
|
|
|
|
notrace void touch_softlockup_watchdog(void)
|
|
{
|
|
touch_softlockup_watchdog_sched();
|
|
wq_watchdog_touch(raw_smp_processor_id());
|
|
}
|
|
EXPORT_SYMBOL(touch_softlockup_watchdog);
|
|
|
|
void touch_all_softlockup_watchdogs(void)
|
|
{
|
|
int cpu;
|
|
|
|
/*
|
|
* watchdog_mutex cannpt be taken here, as this might be called
|
|
* from (soft)interrupt context, so the access to
|
|
* watchdog_allowed_cpumask might race with a concurrent update.
|
|
*
|
|
* The watchdog time stamp can race against a concurrent real
|
|
* update as well, the only side effect might be a cycle delay for
|
|
* the softlockup check.
|
|
*/
|
|
for_each_cpu(cpu, &watchdog_allowed_mask)
|
|
per_cpu(watchdog_touch_ts, cpu) = SOFTLOCKUP_RESET;
|
|
wq_watchdog_touch(-1);
|
|
}
|
|
|
|
void touch_softlockup_watchdog_sync(void)
|
|
{
|
|
__this_cpu_write(softlockup_touch_sync, true);
|
|
__this_cpu_write(watchdog_touch_ts, SOFTLOCKUP_RESET);
|
|
}
|
|
|
|
static int is_softlockup(unsigned long touch_ts)
|
|
{
|
|
unsigned long now = get_timestamp();
|
|
|
|
if ((watchdog_enabled & SOFT_WATCHDOG_ENABLED) && watchdog_thresh){
|
|
/* Warn about unreasonable delays. */
|
|
if (time_after(now, touch_ts + get_softlockup_thresh()))
|
|
return now - touch_ts;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
/* watchdog detector functions */
|
|
bool is_hardlockup(void)
|
|
{
|
|
unsigned long hrint = __this_cpu_read(hrtimer_interrupts);
|
|
|
|
if (__this_cpu_read(hrtimer_interrupts_saved) == hrint)
|
|
return true;
|
|
|
|
__this_cpu_write(hrtimer_interrupts_saved, hrint);
|
|
return false;
|
|
}
|
|
|
|
static void watchdog_interrupt_count(void)
|
|
{
|
|
__this_cpu_inc(hrtimer_interrupts);
|
|
}
|
|
|
|
static DEFINE_PER_CPU(struct completion, softlockup_completion);
|
|
static DEFINE_PER_CPU(struct cpu_stop_work, softlockup_stop_work);
|
|
|
|
/*
|
|
* The watchdog thread function - touches the timestamp.
|
|
*
|
|
* It only runs once every sample_period seconds (4 seconds by
|
|
* default) to reset the softlockup timestamp. If this gets delayed
|
|
* for more than 2*watchdog_thresh seconds then the debug-printout
|
|
* triggers in watchdog_timer_fn().
|
|
*/
|
|
static int softlockup_fn(void *data)
|
|
{
|
|
__touch_watchdog();
|
|
complete(this_cpu_ptr(&softlockup_completion));
|
|
|
|
return 0;
|
|
}
|
|
|
|
/* watchdog kicker functions */
|
|
static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
|
|
{
|
|
unsigned long touch_ts = __this_cpu_read(watchdog_touch_ts);
|
|
struct pt_regs *regs = get_irq_regs();
|
|
int duration;
|
|
int softlockup_all_cpu_backtrace = sysctl_softlockup_all_cpu_backtrace;
|
|
|
|
if (!watchdog_enabled)
|
|
return HRTIMER_NORESTART;
|
|
|
|
/* kick the hardlockup detector */
|
|
watchdog_interrupt_count();
|
|
|
|
/* kick the softlockup detector */
|
|
if (completion_done(this_cpu_ptr(&softlockup_completion))) {
|
|
reinit_completion(this_cpu_ptr(&softlockup_completion));
|
|
stop_one_cpu_nowait(smp_processor_id(),
|
|
softlockup_fn, NULL,
|
|
this_cpu_ptr(&softlockup_stop_work));
|
|
}
|
|
|
|
/* .. and repeat */
|
|
hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period));
|
|
|
|
if (touch_ts == SOFTLOCKUP_RESET) {
|
|
if (unlikely(__this_cpu_read(softlockup_touch_sync))) {
|
|
/*
|
|
* If the time stamp was touched atomically
|
|
* make sure the scheduler tick is up to date.
|
|
*/
|
|
__this_cpu_write(softlockup_touch_sync, false);
|
|
sched_clock_tick();
|
|
}
|
|
|
|
/* Clear the guest paused flag on watchdog reset */
|
|
kvm_check_and_clear_guest_paused();
|
|
__touch_watchdog();
|
|
return HRTIMER_RESTART;
|
|
}
|
|
|
|
/* check for a softlockup
|
|
* This is done by making sure a high priority task is
|
|
* being scheduled. The task touches the watchdog to
|
|
* indicate it is getting cpu time. If it hasn't then
|
|
* this is a good indication some task is hogging the cpu
|
|
*/
|
|
duration = is_softlockup(touch_ts);
|
|
if (unlikely(duration)) {
|
|
/*
|
|
* If a virtual machine is stopped by the host it can look to
|
|
* the watchdog like a soft lockup, check to see if the host
|
|
* stopped the vm before we issue the warning
|
|
*/
|
|
if (kvm_check_and_clear_guest_paused())
|
|
return HRTIMER_RESTART;
|
|
|
|
/* only warn once */
|
|
if (__this_cpu_read(soft_watchdog_warn) == true)
|
|
return HRTIMER_RESTART;
|
|
|
|
if (softlockup_all_cpu_backtrace) {
|
|
/* Prevent multiple soft-lockup reports if one cpu is already
|
|
* engaged in dumping cpu back traces
|
|
*/
|
|
if (test_and_set_bit(0, &soft_lockup_nmi_warn)) {
|
|
/* Someone else will report us. Let's give up */
|
|
__this_cpu_write(soft_watchdog_warn, true);
|
|
return HRTIMER_RESTART;
|
|
}
|
|
}
|
|
|
|
pr_emerg("BUG: soft lockup - CPU#%d stuck for %us! [%s:%d]\n",
|
|
smp_processor_id(), duration,
|
|
current->comm, task_pid_nr(current));
|
|
print_modules();
|
|
print_irqtrace_events(current);
|
|
if (regs)
|
|
show_regs(regs);
|
|
else
|
|
dump_stack();
|
|
|
|
if (softlockup_all_cpu_backtrace) {
|
|
/* Avoid generating two back traces for current
|
|
* given that one is already made above
|
|
*/
|
|
trigger_allbutself_cpu_backtrace();
|
|
|
|
clear_bit(0, &soft_lockup_nmi_warn);
|
|
/* Barrier to sync with other cpus */
|
|
smp_mb__after_atomic();
|
|
}
|
|
|
|
trace_android_vh_watchdog_timer_softlockup(duration, regs, !!softlockup_panic);
|
|
add_taint(TAINT_SOFTLOCKUP, LOCKDEP_STILL_OK);
|
|
if (softlockup_panic)
|
|
panic("softlockup: hung tasks");
|
|
__this_cpu_write(soft_watchdog_warn, true);
|
|
} else
|
|
__this_cpu_write(soft_watchdog_warn, false);
|
|
|
|
return HRTIMER_RESTART;
|
|
}
|
|
|
|
void watchdog_enable(unsigned int cpu)
|
|
{
|
|
struct hrtimer *hrtimer = this_cpu_ptr(&watchdog_hrtimer);
|
|
struct completion *done = this_cpu_ptr(&softlockup_completion);
|
|
unsigned int *enabled = this_cpu_ptr(&watchdog_en);
|
|
|
|
WARN_ON_ONCE(cpu != smp_processor_id());
|
|
|
|
init_completion(done);
|
|
complete(done);
|
|
|
|
if (*enabled)
|
|
return;
|
|
|
|
/*
|
|
* Start the timer first to prevent the NMI watchdog triggering
|
|
* before the timer has a chance to fire.
|
|
*/
|
|
hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
|
|
hrtimer->function = watchdog_timer_fn;
|
|
hrtimer_start(hrtimer, ns_to_ktime(sample_period),
|
|
HRTIMER_MODE_REL_PINNED_HARD);
|
|
|
|
/* Initialize timestamp */
|
|
__touch_watchdog();
|
|
/* Enable the perf event */
|
|
if (watchdog_enabled & NMI_WATCHDOG_ENABLED)
|
|
watchdog_nmi_enable(cpu);
|
|
|
|
/*
|
|
* Need to ensure above operations are observed by other CPUs before
|
|
* indicating that timer is enabled. This is to synchronize core
|
|
* isolation and hotplug. Core isolation will wait for this flag to be
|
|
* set.
|
|
*/
|
|
mb();
|
|
*enabled = 1;
|
|
}
|
|
|
|
void watchdog_disable(unsigned int cpu)
|
|
{
|
|
struct hrtimer *hrtimer = per_cpu_ptr(&watchdog_hrtimer, cpu);
|
|
unsigned int *enabled = per_cpu_ptr(&watchdog_en, cpu);
|
|
|
|
if (!*enabled)
|
|
return;
|
|
|
|
/*
|
|
* Disable the perf event first. That prevents that a large delay
|
|
* between disabling the timer and disabling the perf event causes
|
|
* the perf NMI to detect a false positive.
|
|
*/
|
|
watchdog_nmi_disable(cpu);
|
|
hrtimer_cancel(hrtimer);
|
|
wait_for_completion(per_cpu_ptr(&softlockup_completion, cpu));
|
|
|
|
/*
|
|
* No need for barrier here since disabling the watchdog is
|
|
* synchronized with hotplug lock
|
|
*/
|
|
*enabled = 0;
|
|
}
|
|
|
|
bool watchdog_configured(unsigned int cpu)
|
|
{
|
|
return *per_cpu_ptr(&watchdog_en, cpu);
|
|
}
|
|
|
|
static int softlockup_stop_fn(void *data)
|
|
{
|
|
watchdog_disable(smp_processor_id());
|
|
return 0;
|
|
}
|
|
|
|
static void softlockup_stop_all(void)
|
|
{
|
|
int cpu;
|
|
|
|
if (!softlockup_initialized)
|
|
return;
|
|
|
|
for_each_cpu(cpu, &watchdog_allowed_mask)
|
|
smp_call_on_cpu(cpu, softlockup_stop_fn, NULL, false);
|
|
|
|
cpumask_clear(&watchdog_allowed_mask);
|
|
}
|
|
|
|
static int softlockup_start_fn(void *data)
|
|
{
|
|
watchdog_enable(smp_processor_id());
|
|
return 0;
|
|
}
|
|
|
|
static void softlockup_start_all(void)
|
|
{
|
|
int cpu;
|
|
|
|
cpumask_copy(&watchdog_allowed_mask, &watchdog_cpumask);
|
|
for_each_cpu(cpu, &watchdog_allowed_mask)
|
|
smp_call_on_cpu(cpu, softlockup_start_fn, NULL, false);
|
|
}
|
|
|
|
int lockup_detector_online_cpu(unsigned int cpu)
|
|
{
|
|
if (cpumask_test_cpu(cpu, &watchdog_allowed_mask))
|
|
watchdog_enable(cpu);
|
|
return 0;
|
|
}
|
|
|
|
int lockup_detector_offline_cpu(unsigned int cpu)
|
|
{
|
|
if (cpumask_test_cpu(cpu, &watchdog_allowed_mask))
|
|
watchdog_disable(cpu);
|
|
return 0;
|
|
}
|
|
|
|
static void lockup_detector_reconfigure(void)
|
|
{
|
|
cpus_read_lock();
|
|
watchdog_nmi_stop();
|
|
|
|
softlockup_stop_all();
|
|
set_sample_period();
|
|
lockup_detector_update_enable();
|
|
if (watchdog_enabled && watchdog_thresh)
|
|
softlockup_start_all();
|
|
|
|
watchdog_nmi_start();
|
|
cpus_read_unlock();
|
|
/*
|
|
* Must be called outside the cpus locked section to prevent
|
|
* recursive locking in the perf code.
|
|
*/
|
|
__lockup_detector_cleanup();
|
|
}
|
|
|
|
/*
|
|
* Create the watchdog thread infrastructure and configure the detector(s).
|
|
*
|
|
* The threads are not unparked as watchdog_allowed_mask is empty. When
|
|
* the threads are successfully initialized, take the proper locks and
|
|
* unpark the threads in the watchdog_cpumask if the watchdog is enabled.
|
|
*/
|
|
static __init void lockup_detector_setup(void)
|
|
{
|
|
/*
|
|
* If sysctl is off and watchdog got disabled on the command line,
|
|
* nothing to do here.
|
|
*/
|
|
lockup_detector_update_enable();
|
|
|
|
if (!IS_ENABLED(CONFIG_SYSCTL) &&
|
|
!(watchdog_enabled && watchdog_thresh))
|
|
return;
|
|
|
|
mutex_lock(&watchdog_mutex);
|
|
lockup_detector_reconfigure();
|
|
softlockup_initialized = true;
|
|
mutex_unlock(&watchdog_mutex);
|
|
}
|
|
|
|
#else /* CONFIG_SOFTLOCKUP_DETECTOR */
|
|
static void lockup_detector_reconfigure(void)
|
|
{
|
|
cpus_read_lock();
|
|
watchdog_nmi_stop();
|
|
lockup_detector_update_enable();
|
|
watchdog_nmi_start();
|
|
cpus_read_unlock();
|
|
}
|
|
static inline void lockup_detector_setup(void)
|
|
{
|
|
lockup_detector_reconfigure();
|
|
}
|
|
#endif /* !CONFIG_SOFTLOCKUP_DETECTOR */
|
|
|
|
static void __lockup_detector_cleanup(void)
|
|
{
|
|
lockdep_assert_held(&watchdog_mutex);
|
|
hardlockup_detector_perf_cleanup();
|
|
}
|
|
|
|
/**
|
|
* lockup_detector_cleanup - Cleanup after cpu hotplug or sysctl changes
|
|
*
|
|
* Caller must not hold the cpu hotplug rwsem.
|
|
*/
|
|
void lockup_detector_cleanup(void)
|
|
{
|
|
mutex_lock(&watchdog_mutex);
|
|
__lockup_detector_cleanup();
|
|
mutex_unlock(&watchdog_mutex);
|
|
}
|
|
|
|
/**
|
|
* lockup_detector_soft_poweroff - Interface to stop lockup detector(s)
|
|
*
|
|
* Special interface for parisc. It prevents lockup detector warnings from
|
|
* the default pm_poweroff() function which busy loops forever.
|
|
*/
|
|
void lockup_detector_soft_poweroff(void)
|
|
{
|
|
watchdog_enabled = 0;
|
|
}
|
|
|
|
#ifdef CONFIG_SYSCTL
|
|
|
|
/* Propagate any changes to the watchdog threads */
|
|
static void proc_watchdog_update(void)
|
|
{
|
|
/* Remove impossible cpus to keep sysctl output clean. */
|
|
cpumask_and(&watchdog_cpumask, &watchdog_cpumask, cpu_possible_mask);
|
|
lockup_detector_reconfigure();
|
|
}
|
|
|
|
/*
|
|
* common function for watchdog, nmi_watchdog and soft_watchdog parameter
|
|
*
|
|
* caller | table->data points to | 'which'
|
|
* -------------------|----------------------------|--------------------------
|
|
* proc_watchdog | watchdog_user_enabled | NMI_WATCHDOG_ENABLED |
|
|
* | | SOFT_WATCHDOG_ENABLED
|
|
* -------------------|----------------------------|--------------------------
|
|
* proc_nmi_watchdog | nmi_watchdog_user_enabled | NMI_WATCHDOG_ENABLED
|
|
* -------------------|----------------------------|--------------------------
|
|
* proc_soft_watchdog | soft_watchdog_user_enabled | SOFT_WATCHDOG_ENABLED
|
|
*/
|
|
static int proc_watchdog_common(int which, struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
int err, old, *param = table->data;
|
|
|
|
mutex_lock(&watchdog_mutex);
|
|
|
|
if (!write) {
|
|
/*
|
|
* On read synchronize the userspace interface. This is a
|
|
* racy snapshot.
|
|
*/
|
|
*param = (watchdog_enabled & which) != 0;
|
|
err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
|
|
} else {
|
|
old = READ_ONCE(*param);
|
|
err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
|
|
if (!err && old != READ_ONCE(*param))
|
|
proc_watchdog_update();
|
|
}
|
|
mutex_unlock(&watchdog_mutex);
|
|
return err;
|
|
}
|
|
|
|
/*
|
|
* /proc/sys/kernel/watchdog
|
|
*/
|
|
int proc_watchdog(struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
return proc_watchdog_common(NMI_WATCHDOG_ENABLED|SOFT_WATCHDOG_ENABLED,
|
|
table, write, buffer, lenp, ppos);
|
|
}
|
|
|
|
/*
|
|
* /proc/sys/kernel/nmi_watchdog
|
|
*/
|
|
int proc_nmi_watchdog(struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
if (!nmi_watchdog_available && write)
|
|
return -ENOTSUPP;
|
|
return proc_watchdog_common(NMI_WATCHDOG_ENABLED,
|
|
table, write, buffer, lenp, ppos);
|
|
}
|
|
|
|
/*
|
|
* /proc/sys/kernel/soft_watchdog
|
|
*/
|
|
int proc_soft_watchdog(struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
return proc_watchdog_common(SOFT_WATCHDOG_ENABLED,
|
|
table, write, buffer, lenp, ppos);
|
|
}
|
|
|
|
/*
|
|
* /proc/sys/kernel/watchdog_thresh
|
|
*/
|
|
int proc_watchdog_thresh(struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
int err, old;
|
|
|
|
mutex_lock(&watchdog_mutex);
|
|
|
|
old = READ_ONCE(watchdog_thresh);
|
|
err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
|
|
|
|
if (!err && write && old != READ_ONCE(watchdog_thresh))
|
|
proc_watchdog_update();
|
|
|
|
mutex_unlock(&watchdog_mutex);
|
|
return err;
|
|
}
|
|
|
|
/*
|
|
* The cpumask is the mask of possible cpus that the watchdog can run
|
|
* on, not the mask of cpus it is actually running on. This allows the
|
|
* user to specify a mask that will include cpus that have not yet
|
|
* been brought online, if desired.
|
|
*/
|
|
int proc_watchdog_cpumask(struct ctl_table *table, int write,
|
|
void *buffer, size_t *lenp, loff_t *ppos)
|
|
{
|
|
int err;
|
|
|
|
mutex_lock(&watchdog_mutex);
|
|
|
|
err = proc_do_large_bitmap(table, write, buffer, lenp, ppos);
|
|
if (!err && write)
|
|
proc_watchdog_update();
|
|
|
|
mutex_unlock(&watchdog_mutex);
|
|
return err;
|
|
}
|
|
#endif /* CONFIG_SYSCTL */
|
|
|
|
void __init lockup_detector_init(void)
|
|
{
|
|
if (tick_nohz_full_enabled())
|
|
pr_info("Disabling watchdog on nohz_full cores by default\n");
|
|
|
|
cpumask_copy(&watchdog_cpumask,
|
|
housekeeping_cpumask(HK_FLAG_TIMER));
|
|
|
|
if (!watchdog_nmi_probe())
|
|
nmi_watchdog_available = true;
|
|
lockup_detector_setup();
|
|
}
|