3fc4c60de5
* refs/heads/tmp-9bf5904: Linux 4.19.37 kernel/sysctl.c: fix out-of-bounds access when setting file-max Revert "locking/lockdep: Add debug_locks check in __lock_downgrade()" i2c-hid: properly terminate i2c_hid_dmi_desc_override_table[] array ASoC: rockchip: add missing INTERLEAVED PCM attribute tools include: Adopt linux/bits.h percpu: stop printing kernel addresses ALSA: info: Fix racy addition/deletion of nodes mm/vmstat.c: fix /proc/vmstat format for CONFIG_DEBUG_TLBFLUSH=y CONFIG_SMP=n device_cgroup: fix RCU imbalance in error case sched/fair: Limit sched_cfs_period_timer() loop to avoid hard lockup Revert "kbuild: use -Oz instead of -Os when using clang" tpm: Fix the type of the return value in calc_tpm2_event_size() tpm/tpm_i2c_atmel: Return -E2BIG when the transfer is incomplete modpost: file2alias: check prototype of handler modpost: file2alias: go back to simple devtable lookup mmc: sdhci: Handle auto-command errors mmc: sdhci: Rename SDHCI_ACMD12_ERR and SDHCI_INT_ACMD12ERR mmc: sdhci: Fix data command CRC error handling nfit/ars: Avoid stale ARS results nfit/ars: Allow root to busy-poll the ARS state machine nfit/ars: Introduce scrub_flags nfit/ars: Remove ars_start_flags timers/sched_clock: Prevent generic sched_clock wrap caused by tick_freeze() x86/speculation: Prevent deadlock on ssb_state::lock perf/x86: Fix incorrect PEBS_REGS x86/cpu/bugs: Use __initconst for 'const' init data perf/x86/amd: Add event map for AMD Family 17h drm/amdgpu/gmc9: fix VM_L2_CNTL3 programming mac80211: do not call driver wake_tx_queue op during reconfig rt2x00: do not increment sequence number while re-transmitting kprobes: Fix error check when reusing optimized probes kprobes: Mark ftrace mcount handler functions nokprobe x86/kprobes: Verify stack frame on kretprobe arm64: futex: Restore oldval initialization to work around buggy compilers drm/ttm: fix out-of-bounds read in ttm_put_pages() v2 crypto: x86/poly1305 - fix overflow during partial reduction ipmi: fix sleep-in-atomic in free_user at cleanup SRCU user->release_barrier coredump: fix race condition between mmget_not_zero()/get_task_mm() and core dumping Revert "svm: Fix AVIC incomplete IPI emulation" Revert "scsi: fcoe: clear FC_RP_STARTED flags when receiving a LOGO" scsi: core: set result when the command cannot be dispatched vt: fix cursor when clearing the screen serial: sh-sci: Fix HSCIF RX sampling point calculation serial: sh-sci: Fix HSCIF RX sampling point adjustment Input: elan_i2c - add hardware ID for multiple Lenovo laptops ALSA: core: Fix card races between register and disconnect ALSA: hda/realtek - add two more pin configuration sets to quirk table staging: comedi: ni_usb6501: Fix possible double-free of ->usb_rx_buf staging: comedi: ni_usb6501: Fix use of uninitialized mutex staging: comedi: vmk80xx: Fix possible double-free of ->usb_rx_buf staging: comedi: vmk80xx: Fix use of uninitialized semaphore staging: most: core: use device description as name io: accel: kxcjk1013: restore the range after resume. iio: core: fix a possible circular locking dependency iio: adc: at91: disable adc channel interrupt in timeout case iio: Fix scan mask selection iio: dac: mcp4725: add missing powerdown bits in store eeprom iio: ad_sigma_delta: select channel when reading register iio: cros_ec: Fix the maths for gyro scale calculation iio:chemical:bme680: Fix SPI read interface iio:chemical:bme680: Fix, report temperature in millidegrees iio/gyro/bmg160: Use millidegrees for temperature scale iio: gyro: mpu3050: fix chip ID reading staging: iio: ad7192: Fix ad7193 channel address Staging: iio: meter: fixed typo KVM: x86: svm: make sure NMI is injected after nmi_singlestep KVM: x86: Don't clear EFER during SMM transitions for 32-bit vCPU cifs: fix handle leak in smb2_query_symlink() cifs: Fix use-after-free in SMB2_read cifs: Fix use-after-free in SMB2_write CIFS: keep FileInfo handle live during oplock break net: IP6 defrag: use rbtrees in nf_conntrack_reasm.c net: IP6 defrag: use rbtrees for IPv6 defrag net: IP defrag: encapsulate rbtree defrag code into callable functions sch_cake: Simplify logic in cake_select_tin() nfp: flower: remove vlan CFI bit from push vlan action nfp: flower: replace CFI with vlan present sch_cake: Make sure we can write the IP header before changing DSCP bits sch_cake: Use tc_skb_protocol() helper for getting packet protocol route: Avoid crash from dereferencing NULL rt->from net/mlx5: FPGA, tls, idr remove on flow delete net/tls: prevent bad memory access in tls_is_sk_tx_device_offloaded() net/mlx5: FPGA, tls, hold rcu read lock a bit longer net: thunderx: don't allow jumbo frames with XDP net: thunderx: raise XDP MTU to 1508 ipv4: ensure rcu_read_lock() in ipv4_link_failure() ipv4: recompile ip options in ipv4_link_failure vhost: reject zero size iova range tipc: missing entries in name table of publications team: set slave to promisc if team is already in promisc mode tcp: tcp_grow_window() needs to respect tcp_space() net: fou: do not use guehdr after iptunnel_pull_offloads in gue_udp_recv net: Fix missing meta data in skb with vlan packet net: bridge: multicast: use rcu to access port list from br_multicast_start_querier net: bridge: fix per-port af_packet sockets net: atm: Fix potential Spectre v1 vulnerabilities failover: allow name change on IFF_UP slave interfaces bonding: fix event handling for stacked bonds ANDROID: cuttlefish_defconfig: Enable CONFIG_XFRM_STATISTICS Conflicts: drivers/mmc/host/sdhci.c drivers/mmc/host/sdhci.h Change-Id: Iff88e7f663de91949ddf065ec1d3b69c2fd3a9a9 Signed-off-by: Ivaylo Georgiev <irgeorgiev@codeaurora.org>
319 lines
8.6 KiB
C
319 lines
8.6 KiB
C
/*
|
|
* sched_clock.c: Generic sched_clock() support, to extend low level
|
|
* hardware time counters to full 64-bit ns values.
|
|
*
|
|
* This program is free software; you can redistribute it and/or modify
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
* published by the Free Software Foundation.
|
|
*/
|
|
#include <linux/clocksource.h>
|
|
#include <linux/init.h>
|
|
#include <linux/jiffies.h>
|
|
#include <linux/ktime.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/moduleparam.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/sched/clock.h>
|
|
#include <linux/syscore_ops.h>
|
|
#include <linux/hrtimer.h>
|
|
#include <linux/sched_clock.h>
|
|
#include <linux/seqlock.h>
|
|
#include <linux/bitops.h>
|
|
|
|
/**
|
|
* struct clock_read_data - data required to read from sched_clock()
|
|
*
|
|
* @epoch_ns: sched_clock() value at last update
|
|
* @epoch_cyc: Clock cycle value at last update.
|
|
* @sched_clock_mask: Bitmask for two's complement subtraction of non 64bit
|
|
* clocks.
|
|
* @read_sched_clock: Current clock source (or dummy source when suspended).
|
|
* @mult: Multipler for scaled math conversion.
|
|
* @shift: Shift value for scaled math conversion.
|
|
*
|
|
* Care must be taken when updating this structure; it is read by
|
|
* some very hot code paths. It occupies <=40 bytes and, when combined
|
|
* with the seqcount used to synchronize access, comfortably fits into
|
|
* a 64 byte cache line.
|
|
*/
|
|
struct clock_read_data {
|
|
u64 epoch_ns;
|
|
u64 epoch_cyc;
|
|
u64 sched_clock_mask;
|
|
u64 (*read_sched_clock)(void);
|
|
u32 mult;
|
|
u32 shift;
|
|
};
|
|
|
|
/**
|
|
* struct clock_data - all data needed for sched_clock() (including
|
|
* registration of a new clock source)
|
|
*
|
|
* @seq: Sequence counter for protecting updates. The lowest
|
|
* bit is the index for @read_data.
|
|
* @read_data: Data required to read from sched_clock.
|
|
* @wrap_kt: Duration for which clock can run before wrapping.
|
|
* @rate: Tick rate of the registered clock.
|
|
* @actual_read_sched_clock: Registered hardware level clock read function.
|
|
*
|
|
* The ordering of this structure has been chosen to optimize cache
|
|
* performance. In particular 'seq' and 'read_data[0]' (combined) should fit
|
|
* into a single 64-byte cache line.
|
|
*/
|
|
struct clock_data {
|
|
seqcount_t seq;
|
|
struct clock_read_data read_data[2];
|
|
ktime_t wrap_kt;
|
|
unsigned long rate;
|
|
|
|
u64 (*actual_read_sched_clock)(void);
|
|
};
|
|
|
|
static struct hrtimer sched_clock_timer;
|
|
static int irqtime = -1;
|
|
static u64 suspend_ns;
|
|
static u64 suspend_cycles;
|
|
static u64 resume_cycles;
|
|
|
|
core_param(irqtime, irqtime, int, 0400);
|
|
|
|
static u64 notrace jiffy_sched_clock_read(void)
|
|
{
|
|
/*
|
|
* We don't need to use get_jiffies_64 on 32-bit arches here
|
|
* because we register with BITS_PER_LONG
|
|
*/
|
|
return (u64)(jiffies - INITIAL_JIFFIES);
|
|
}
|
|
|
|
static struct clock_data cd ____cacheline_aligned = {
|
|
.read_data[0] = { .mult = NSEC_PER_SEC / HZ,
|
|
.read_sched_clock = jiffy_sched_clock_read, },
|
|
.actual_read_sched_clock = jiffy_sched_clock_read,
|
|
};
|
|
|
|
static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift)
|
|
{
|
|
return (cyc * mult) >> shift;
|
|
}
|
|
|
|
unsigned long long notrace sched_clock(void)
|
|
{
|
|
u64 cyc, res;
|
|
unsigned long seq;
|
|
struct clock_read_data *rd;
|
|
|
|
do {
|
|
seq = raw_read_seqcount(&cd.seq);
|
|
rd = cd.read_data + (seq & 1);
|
|
|
|
cyc = (rd->read_sched_clock() - rd->epoch_cyc) &
|
|
rd->sched_clock_mask;
|
|
res = rd->epoch_ns + cyc_to_ns(cyc, rd->mult, rd->shift);
|
|
} while (read_seqcount_retry(&cd.seq, seq));
|
|
|
|
return res;
|
|
}
|
|
|
|
/*
|
|
* Updating the data required to read the clock.
|
|
*
|
|
* sched_clock() will never observe mis-matched data even if called from
|
|
* an NMI. We do this by maintaining an odd/even copy of the data and
|
|
* steering sched_clock() to one or the other using a sequence counter.
|
|
* In order to preserve the data cache profile of sched_clock() as much
|
|
* as possible the system reverts back to the even copy when the update
|
|
* completes; the odd copy is used *only* during an update.
|
|
*/
|
|
static void update_clock_read_data(struct clock_read_data *rd)
|
|
{
|
|
/* update the backup (odd) copy with the new data */
|
|
cd.read_data[1] = *rd;
|
|
|
|
/* steer readers towards the odd copy */
|
|
raw_write_seqcount_latch(&cd.seq);
|
|
|
|
/* now its safe for us to update the normal (even) copy */
|
|
cd.read_data[0] = *rd;
|
|
|
|
/* switch readers back to the even copy */
|
|
raw_write_seqcount_latch(&cd.seq);
|
|
}
|
|
|
|
/*
|
|
* Atomically update the sched_clock() epoch.
|
|
*/
|
|
static void update_sched_clock(void)
|
|
{
|
|
u64 cyc;
|
|
u64 ns;
|
|
struct clock_read_data rd;
|
|
|
|
rd = cd.read_data[0];
|
|
|
|
cyc = cd.actual_read_sched_clock();
|
|
ns = rd.epoch_ns + cyc_to_ns((cyc - rd.epoch_cyc) & rd.sched_clock_mask, rd.mult, rd.shift);
|
|
|
|
rd.epoch_ns = ns;
|
|
rd.epoch_cyc = cyc;
|
|
|
|
update_clock_read_data(&rd);
|
|
}
|
|
|
|
static enum hrtimer_restart sched_clock_poll(struct hrtimer *hrt)
|
|
{
|
|
update_sched_clock();
|
|
hrtimer_forward_now(hrt, cd.wrap_kt);
|
|
|
|
return HRTIMER_RESTART;
|
|
}
|
|
|
|
void __init
|
|
sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
|
|
{
|
|
u64 res, wrap, new_mask, new_epoch, cyc, ns;
|
|
u32 new_mult, new_shift;
|
|
unsigned long r;
|
|
char r_unit;
|
|
struct clock_read_data rd;
|
|
|
|
if (cd.rate > rate)
|
|
return;
|
|
|
|
WARN_ON(!irqs_disabled());
|
|
|
|
/* Calculate the mult/shift to convert counter ticks to ns. */
|
|
clocks_calc_mult_shift(&new_mult, &new_shift, rate, NSEC_PER_SEC, 3600);
|
|
|
|
new_mask = CLOCKSOURCE_MASK(bits);
|
|
cd.rate = rate;
|
|
|
|
/* Calculate how many nanosecs until we risk wrapping */
|
|
wrap = clocks_calc_max_nsecs(new_mult, new_shift, 0, new_mask, NULL);
|
|
cd.wrap_kt = ns_to_ktime(wrap);
|
|
|
|
rd = cd.read_data[0];
|
|
|
|
/* Update epoch for new counter and update 'epoch_ns' from old counter*/
|
|
new_epoch = read();
|
|
cyc = cd.actual_read_sched_clock();
|
|
ns = rd.epoch_ns + cyc_to_ns((cyc - rd.epoch_cyc) & rd.sched_clock_mask, rd.mult, rd.shift);
|
|
cd.actual_read_sched_clock = read;
|
|
|
|
rd.read_sched_clock = read;
|
|
rd.sched_clock_mask = new_mask;
|
|
rd.mult = new_mult;
|
|
rd.shift = new_shift;
|
|
rd.epoch_cyc = new_epoch;
|
|
rd.epoch_ns = ns;
|
|
|
|
update_clock_read_data(&rd);
|
|
|
|
if (sched_clock_timer.function != NULL) {
|
|
/* update timeout for clock wrap */
|
|
hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL);
|
|
}
|
|
|
|
r = rate;
|
|
if (r >= 4000000) {
|
|
r /= 1000000;
|
|
r_unit = 'M';
|
|
} else {
|
|
if (r >= 1000) {
|
|
r /= 1000;
|
|
r_unit = 'k';
|
|
} else {
|
|
r_unit = ' ';
|
|
}
|
|
}
|
|
|
|
/* Calculate the ns resolution of this counter */
|
|
res = cyc_to_ns(1ULL, new_mult, new_shift);
|
|
|
|
pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lluns\n",
|
|
bits, r, r_unit, res, wrap);
|
|
|
|
/* Enable IRQ time accounting if we have a fast enough sched_clock() */
|
|
if (irqtime > 0 || (irqtime == -1 && rate >= 1000000))
|
|
enable_sched_clock_irqtime();
|
|
|
|
pr_debug("Registered %pF as sched_clock source\n", read);
|
|
}
|
|
|
|
void __init generic_sched_clock_init(void)
|
|
{
|
|
/*
|
|
* If no sched_clock() function has been provided at that point,
|
|
* make it the final one one.
|
|
*/
|
|
if (cd.actual_read_sched_clock == jiffy_sched_clock_read)
|
|
sched_clock_register(jiffy_sched_clock_read, BITS_PER_LONG, HZ);
|
|
|
|
update_sched_clock();
|
|
|
|
/*
|
|
* Start the timer to keep sched_clock() properly updated and
|
|
* sets the initial epoch.
|
|
*/
|
|
hrtimer_init(&sched_clock_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
|
|
sched_clock_timer.function = sched_clock_poll;
|
|
hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL);
|
|
}
|
|
|
|
/*
|
|
* Clock read function for use when the clock is suspended.
|
|
*
|
|
* This function makes it appear to sched_clock() as if the clock
|
|
* stopped counting at its last update.
|
|
*
|
|
* This function must only be called from the critical
|
|
* section in sched_clock(). It relies on the read_seqcount_retry()
|
|
* at the end of the critical section to be sure we observe the
|
|
* correct copy of 'epoch_cyc'.
|
|
*/
|
|
static u64 notrace suspended_sched_clock_read(void)
|
|
{
|
|
unsigned long seq = raw_read_seqcount(&cd.seq);
|
|
|
|
return cd.read_data[seq & 1].epoch_cyc;
|
|
}
|
|
|
|
int sched_clock_suspend(void)
|
|
{
|
|
struct clock_read_data *rd = &cd.read_data[0];
|
|
|
|
update_sched_clock();
|
|
|
|
suspend_ns = rd->epoch_ns;
|
|
suspend_cycles = rd->epoch_cyc;
|
|
pr_info("suspend ns:%17llu suspend cycles:%17llu\n",
|
|
rd->epoch_ns, rd->epoch_cyc);
|
|
hrtimer_cancel(&sched_clock_timer);
|
|
rd->read_sched_clock = suspended_sched_clock_read;
|
|
|
|
return 0;
|
|
}
|
|
|
|
void sched_clock_resume(void)
|
|
{
|
|
struct clock_read_data *rd = &cd.read_data[0];
|
|
|
|
rd->epoch_cyc = cd.actual_read_sched_clock();
|
|
resume_cycles = rd->epoch_cyc;
|
|
pr_info("resume cycles:%17llu\n", rd->epoch_cyc);
|
|
hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL);
|
|
rd->read_sched_clock = cd.actual_read_sched_clock;
|
|
}
|
|
|
|
static struct syscore_ops sched_clock_ops = {
|
|
.suspend = sched_clock_suspend,
|
|
.resume = sched_clock_resume,
|
|
};
|
|
|
|
static int __init sched_clock_syscore_init(void)
|
|
{
|
|
register_syscore_ops(&sched_clock_ops);
|
|
|
|
return 0;
|
|
}
|
|
device_initcall(sched_clock_syscore_init);
|