WSL2-Linux-Kernel/arch/x86/kernel/tsc_sync.c

218 строки
5.5 KiB
C
Исходник Обычный вид История

/*
* check TSC synchronization.
*
* Copyright (C) 2006, Red Hat, Inc., Ingo Molnar
*
* We check whether all boot CPUs have their TSC's synchronized,
* print a warning if not and turn off the TSC clock-source.
*
* The warp-check is point-to-point between two CPUs, the CPU
* initiating the bootup is the 'source CPU', the freshly booting
* CPU is the 'target CPU'.
*
* Only two CPUs may participate - they can enter in any order.
* ( The serial nature of the boot logic and the CPU hotplug lock
* protects against more than 2 CPUs entering this code. )
*/
#include <linux/spinlock.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/smp.h>
#include <linux/nmi.h>
#include <asm/tsc.h>
/*
* Entry/exit counters that make sure that both CPUs
* run the measurement code at once:
*/
static __cpuinitdata atomic_t start_count;
static __cpuinitdata atomic_t stop_count;
/*
* We use a raw spinlock in this exceptional case, because
* we want to have the fastest, inlined, non-debug version
* of a critical section, to be able to prove TSC time-warps:
*/
static __cpuinitdata arch_spinlock_t sync_lock = __ARCH_SPIN_LOCK_UNLOCKED;
static __cpuinitdata cycles_t last_tsc;
static __cpuinitdata cycles_t max_warp;
static __cpuinitdata int nr_warps;
/*
* TSC-warp measurement loop running on both CPUs:
*/
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
static __cpuinit void check_tsc_warp(unsigned int timeout)
{
cycles_t start, now, prev, end;
int i;
rdtsc_barrier();
start = get_cycles();
rdtsc_barrier();
/*
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
* The measurement runs for 'timeout' msecs:
*/
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
end = start + (cycles_t) tsc_khz * timeout;
now = start;
for (i = 0; ; i++) {
/*
* We take the global lock, measure TSC, save the
* previous TSC that was measured (possibly on
* another CPU) and update the previous TSC timestamp.
*/
arch_spin_lock(&sync_lock);
prev = last_tsc;
rdtsc_barrier();
now = get_cycles();
rdtsc_barrier();
last_tsc = now;
arch_spin_unlock(&sync_lock);
/*
* Be nice every now and then (and also check whether
* measurement is done [we also insert a 10 million
* loops safety exit, so we dont lock up in case the
* TSC readout is totally broken]):
*/
if (unlikely(!(i & 7))) {
if (now > end || i > 10000000)
break;
cpu_relax();
touch_nmi_watchdog();
}
/*
* Outside the critical section we can now see whether
* we saw a time-warp of the TSC going backwards:
*/
if (unlikely(prev > now)) {
arch_spin_lock(&sync_lock);
max_warp = max(max_warp, prev - now);
nr_warps++;
arch_spin_unlock(&sync_lock);
}
}
WARN(!(now-start),
"Warning: zero tsc calibration delta: %Ld [max: %Ld]\n",
now-start, end-start);
}
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
/*
* If the target CPU coming online doesn't have any of its core-siblings
* online, a timeout of 20msec will be used for the TSC-warp measurement
* loop. Otherwise a smaller timeout of 2msec will be used, as we have some
* information about this socket already (and this information grows as we
* have more and more logical-siblings in that socket).
*
* Ideally we should be able to skip the TSC sync check on the other
* core-siblings, if the first logical CPU in a socket passed the sync test.
* But as the TSC is per-logical CPU and can potentially be modified wrongly
* by the bios, TSC sync test for smaller duration should be able
* to catch such errors. Also this will catch the condition where all the
* cores in the socket doesn't get reset at the same time.
*/
static inline unsigned int loop_timeout(int cpu)
{
return (cpumask_weight(cpu_core_mask(cpu)) > 1) ? 2 : 20;
}
/*
* Source CPU calls into this - it waits for the freshly booted
* target CPU to arrive and then starts the measurement:
*/
void __cpuinit check_tsc_sync_source(int cpu)
{
int cpus = 2;
/*
* No need to check if we already know that the TSC is not
* synchronized:
*/
if (unsynchronized_tsc())
return;
if (tsc_clocksource_reliable) {
if (cpu == (nr_cpu_ids-1) || system_state != SYSTEM_BOOTING)
pr_info(
"Skipped synchronization checks as TSC is reliable.\n");
return;
}
/*
* Reset it - in case this is a second bootup:
*/
atomic_set(&stop_count, 0);
/*
* Wait for the target to arrive:
*/
while (atomic_read(&start_count) != cpus-1)
cpu_relax();
/*
* Trigger the target to continue into the measurement too:
*/
atomic_inc(&start_count);
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
check_tsc_warp(loop_timeout(cpu));
while (atomic_read(&stop_count) != cpus-1)
cpu_relax();
if (nr_warps) {
pr_warning("TSC synchronization [CPU#%d -> CPU#%d]:\n",
smp_processor_id(), cpu);
pr_warning("Measured %Ld cycles TSC warp between CPUs, "
"turning off TSC clock.\n", max_warp);
mark_tsc_unstable("check_tsc_sync_source failed");
} else {
pr_debug("TSC synchronization [CPU#%d -> CPU#%d]: passed\n",
smp_processor_id(), cpu);
}
x86: fix: s2ram + P4 + tsc = annoyance s2ram recently became useful here, except for the kernel's annoying habit of disabling my P4's perfectly good TSC. [ 107.894470] CPU 1 is now offline [ 107.894474] SMP alternatives: switching to UP code [ 107.895832] CPU0 attaching sched-domain: [ 107.895836] domain 0: span 1 [ 107.895838] groups: 1 [ 107.896097] CPU1 is down [ 3.726156] Intel machine check architecture supported. [ 3.726165] Intel machine check reporting enabled on CPU#0. [ 3.726167] CPU0: Intel P4/Xeon Extended MCE MSRs (12) available [ 3.726170] CPU0: Thermal monitoring enabled [ 3.726175] Back to C! [ 3.726708] Force enabled HPET at resume [ 3.726775] Enabling non-boot CPUs ... [ 3.727049] CPU0 attaching NULL sched-domain. [ 3.727165] SMP alternatives: switching to SMP code [ 3.727858] Booting processor 1/1 eip 3000 [ 3.727862] CPU 1 irqstacks, hard=b042f000 soft=b042d000 [ 3.738173] Initializing CPU#1 [ 3.798912] Calibrating delay using timer specific routine.. 5986.12 BogoMIPS (lpj=2993061) [ 3.798920] CPU: After generic identify, caps: bfebfbff 00000000 00000000 00000000 00004400 00000000 00000000 00000000 [ 3.798931] CPU: Trace cache: 12K uops, L1 D cache: 8K [ 3.798934] CPU: L2 cache: 512K [ 3.798936] CPU: Physical Processor ID: 0 [ 3.798938] CPU: After all inits, caps: bfebfbff 00000000 00000000 0000b080 00004400 00000000 00000000 00000000 [ 3.798946] Intel machine check architecture supported. [ 3.798952] Intel machine check reporting enabled on CPU#1. [ 3.798955] CPU1: Intel P4/Xeon Extended MCE MSRs (12) available [ 3.798959] CPU1: Thermal monitoring enabled [ 3.799161] CPU1: Intel(R) Pentium(R) 4 CPU 3.00GHz stepping 09 [ 3.799187] checking TSC synchronization [CPU#0 -> CPU#1]: [ 3.819181] Measured 63588552840 cycles TSC warp between CPUs, turning off TSC clock. [ 3.819184] Marking TSC unstable due to: check_tsc_sync_source failed. If check_tsc_warp() is called after initial boot, and the TSC has in the meantime been set (BIOS, user, silicon, elves) to a value lower than the last stored/stale value, we blame the TSC. Reset to pristine condition after every test. Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-01-30 15:30:04 +03:00
/*
* Reset it - just in case we boot another CPU later:
*/
atomic_set(&start_count, 0);
nr_warps = 0;
max_warp = 0;
last_tsc = 0;
/*
* Let the target continue with the bootup:
*/
atomic_inc(&stop_count);
}
/*
* Freshly booted CPUs call into this:
*/
void __cpuinit check_tsc_sync_target(void)
{
int cpus = 2;
if (unsynchronized_tsc() || tsc_clocksource_reliable)
return;
/*
* Register this CPU's participation and wait for the
* source CPU to start the measurement:
*/
atomic_inc(&start_count);
while (atomic_read(&start_count) != cpus)
cpu_relax();
x86/tsc: Reduce the TSC sync check time for core-siblings For each logical CPU that is coming online, we spend 20msec for checking the TSC synchronization. And as this is done sequentially for each logical CPU boot, this time gets added up depending on the number of logical CPU's supported by the platform. Minimize this by using the socket topology information. If the target CPU coming online doesn't have any of its core-siblings online, a timeout of 20msec will be used for the TSC-warp measurement loop. Otherwise a smaller timeout of 2msec will be used, as we have some information about this socket already (and this information grows as we have more and more logical-siblings in that socket). Ideally we should be able to skip the TSC sync check on the other core-siblings, if the first logical CPU in a socket passed the sync test. But as the TSC is per-logical CPU and can potentially be modified wrongly by the bios before the OS boot, TSC sync test for smaller duration should be able to catch such errors. Also this will catch the condition where all the cores in the socket doesn't get reset at the same time. For example, with this modification, time spent in TSC sync checks on a 4 socket 10-core with HT system gets reduced from 1580msec to 212msec. Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jack Steiner <steiner@sgi.com> Cc: venki@google.com Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/1328581940.29790.20.camel@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-07 06:32:20 +04:00
check_tsc_warp(loop_timeout(smp_processor_id()));
/*
* Ok, we are done:
*/
atomic_inc(&stop_count);
/*
* Wait for the source CPU to print stuff:
*/
while (atomic_read(&stop_count) != cpus)
cpu_relax();
}