linux_dsm_epyc7002/arch/x86/kernel/smp.c
Don Zickus 3603a2512f x86, reboot: Use NMI instead of REBOOT_VECTOR to stop cpus
A recent discussion started talking about the locking on the
pstore fs and how it relates to the kmsg infrastructure.  We
noticed it was possible for userspace to r/w to the pstore fs
(grabbing the locks in the process) and block the panic path
from r/w to the same fs.

The reason was the cpu with the lock could be doing work while
the crashing cpu is panic'ing.  Busting those spinlocks might
cause those cpus to step on each other's data.  Fine, fair
enough.

It was suggested it would be nice to serialize the panic path
(ie stop the other cpus) and have only one cpu running.  This
would allow us to bust the spinlocks and not worry about another
cpu stepping on the data.

Of course, smp_send_stop() does this in the panic case.
kmsg_dump() would have to be moved to be called after it.  Easy
enough.

The only problem is on x86 the smp_send_stop() function calls
the REBOOT_VECTOR.  Any cpu with irqs disabled (which pstore and
its backend ERST would do), block this IPI and thus do not stop.
 This makes it difficult to reliably log data to the pstore fs.

The patch below switches from the REBOOT_VECTOR to NMI (and
mimics what kdump does).  Switching to NMI allows us to deliver
the IPI when irqs are disabled, increasing the reliability of
this function.

However, Andi carefully noted that on some machines this
approach does not work because of broken BIOSes or whatever.

To help accomodate this, the next couple of patches will run a
selftest and provide a knob to disable.

V2:
  uses atomic ops to serialize the cpu that shuts everyone down
V3:
  comment cleanup

Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Robert Richter <robert.richter@amd.com>
Cc: seiji.aguchi@hds.com
Cc: vgoyal@redhat.com
Cc: mjg@redhat.com
Cc: tony.luck@intel.com
Cc: gong.chen@intel.com
Cc: satoru.moriya@hds.com
Cc: avi@redhat.com
Cc: Andi Kleen <andi@firstfloor.org>
Link: http://lkml.kernel.org/r/1318533267-18880-2-git-send-email-dzickus@redhat.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2011-12-05 12:00:14 +01:00

300 lines
8.1 KiB
C

/*
* Intel SMP support routines.
*
* (c) 1995 Alan Cox, Building #3 <alan@lxorguk.ukuu.org.uk>
* (c) 1998-99, 2000, 2009 Ingo Molnar <mingo@redhat.com>
* (c) 2002,2003 Andi Kleen, SuSE Labs.
*
* i386 and x86_64 integration by Glauber Costa <gcosta@redhat.com>
*
* This code is released under the GNU General Public License version 2 or
* later.
*/
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/delay.h>
#include <linux/spinlock.h>
#include <linux/export.h>
#include <linux/kernel_stat.h>
#include <linux/mc146818rtc.h>
#include <linux/cache.h>
#include <linux/interrupt.h>
#include <linux/cpu.h>
#include <linux/gfp.h>
#include <asm/mtrr.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>
#include <asm/proto.h>
#include <asm/apic.h>
#include <asm/nmi.h>
/*
* Some notes on x86 processor bugs affecting SMP operation:
*
* Pentium, Pentium Pro, II, III (and all CPUs) have bugs.
* The Linux implications for SMP are handled as follows:
*
* Pentium III / [Xeon]
* None of the E1AP-E3AP errata are visible to the user.
*
* E1AP. see PII A1AP
* E2AP. see PII A2AP
* E3AP. see PII A3AP
*
* Pentium II / [Xeon]
* None of the A1AP-A3AP errata are visible to the user.
*
* A1AP. see PPro 1AP
* A2AP. see PPro 2AP
* A3AP. see PPro 7AP
*
* Pentium Pro
* None of 1AP-9AP errata are visible to the normal user,
* except occasional delivery of 'spurious interrupt' as trap #15.
* This is very rare and a non-problem.
*
* 1AP. Linux maps APIC as non-cacheable
* 2AP. worked around in hardware
* 3AP. fixed in C0 and above steppings microcode update.
* Linux does not use excessive STARTUP_IPIs.
* 4AP. worked around in hardware
* 5AP. symmetric IO mode (normal Linux operation) not affected.
* 'noapic' mode has vector 0xf filled out properly.
* 6AP. 'noapic' mode might be affected - fixed in later steppings
* 7AP. We do not assume writes to the LVT deassering IRQs
* 8AP. We do not enable low power mode (deep sleep) during MP bootup
* 9AP. We do not use mixed mode
*
* Pentium
* There is a marginal case where REP MOVS on 100MHz SMP
* machines with B stepping processors can fail. XXX should provide
* an L1cache=Writethrough or L1cache=off option.
*
* B stepping CPUs may hang. There are hardware work arounds
* for this. We warn about it in case your board doesn't have the work
* arounds. Basically that's so I can tell anyone with a B stepping
* CPU and SMP problems "tough".
*
* Specific items [From Pentium Processor Specification Update]
*
* 1AP. Linux doesn't use remote read
* 2AP. Linux doesn't trust APIC errors
* 3AP. We work around this
* 4AP. Linux never generated 3 interrupts of the same priority
* to cause a lost local interrupt.
* 5AP. Remote read is never used
* 6AP. not affected - worked around in hardware
* 7AP. not affected - worked around in hardware
* 8AP. worked around in hardware - we get explicit CS errors if not
* 9AP. only 'noapic' mode affected. Might generate spurious
* interrupts, we log only the first one and count the
* rest silently.
* 10AP. not affected - worked around in hardware
* 11AP. Linux reads the APIC between writes to avoid this, as per
* the documentation. Make sure you preserve this as it affects
* the C stepping chips too.
* 12AP. not affected - worked around in hardware
* 13AP. not affected - worked around in hardware
* 14AP. we always deassert INIT during bootup
* 15AP. not affected - worked around in hardware
* 16AP. not affected - worked around in hardware
* 17AP. not affected - worked around in hardware
* 18AP. not affected - worked around in hardware
* 19AP. not affected - worked around in BIOS
*
* If this sounds worrying believe me these bugs are either ___RARE___,
* or are signal timing bugs worked around in hardware and there's
* about nothing of note with C stepping upwards.
*/
/*
* this function sends a 'reschedule' IPI to another CPU.
* it goes straight through and wastes no time serializing
* anything. Worst case is that we lose a reschedule ...
*/
static void native_smp_send_reschedule(int cpu)
{
if (unlikely(cpu_is_offline(cpu))) {
WARN_ON(1);
return;
}
apic->send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);
}
void native_send_call_func_single_ipi(int cpu)
{
apic->send_IPI_mask(cpumask_of(cpu), CALL_FUNCTION_SINGLE_VECTOR);
}
void native_send_call_func_ipi(const struct cpumask *mask)
{
cpumask_var_t allbutself;
if (!alloc_cpumask_var(&allbutself, GFP_ATOMIC)) {
apic->send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
return;
}
cpumask_copy(allbutself, cpu_online_mask);
cpumask_clear_cpu(smp_processor_id(), allbutself);
if (cpumask_equal(mask, allbutself) &&
cpumask_equal(cpu_online_mask, cpu_callout_mask))
apic->send_IPI_allbutself(CALL_FUNCTION_VECTOR);
else
apic->send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
free_cpumask_var(allbutself);
}
static atomic_t stopping_cpu = ATOMIC_INIT(-1);
static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
{
/* We are registered on stopping cpu too, avoid spurious NMI */
if (raw_smp_processor_id() == atomic_read(&stopping_cpu))
return NMI_HANDLED;
stop_this_cpu(NULL);
return NMI_HANDLED;
}
static void native_nmi_stop_other_cpus(int wait)
{
unsigned long flags;
unsigned long timeout;
if (reboot_force)
return;
/*
* Use an own vector here because smp_call_function
* does lots of things not suitable in a panic situation.
*/
if (num_online_cpus() > 1) {
/* did someone beat us here? */
if (atomic_cmpxchg(&stopping_cpu, -1, safe_smp_processor_id() != -1))
return;
if (register_nmi_handler(NMI_LOCAL, smp_stop_nmi_callback,
NMI_FLAG_FIRST, "smp_stop"))
/* Note: we ignore failures here */
return;
/* sync above data before sending NMI */
wmb();
apic->send_IPI_allbutself(NMI_VECTOR);
/*
* Don't wait longer than a second if the caller
* didn't ask us to wait.
*/
timeout = USEC_PER_SEC;
while (num_online_cpus() > 1 && (wait || timeout--))
udelay(1);
}
local_irq_save(flags);
disable_local_APIC();
local_irq_restore(flags);
}
/*
* this function calls the 'stop' function on all other CPUs in the system.
*/
asmlinkage void smp_reboot_interrupt(void)
{
ack_APIC_irq();
irq_enter();
stop_this_cpu(NULL);
irq_exit();
}
static void native_irq_stop_other_cpus(int wait)
{
unsigned long flags;
unsigned long timeout;
if (reboot_force)
return;
/*
* Use an own vector here because smp_call_function
* does lots of things not suitable in a panic situation.
* On most systems we could also use an NMI here,
* but there are a few systems around where NMI
* is problematic so stay with an non NMI for now
* (this implies we cannot stop CPUs spinning with irq off
* currently)
*/
if (num_online_cpus() > 1) {
apic->send_IPI_allbutself(REBOOT_VECTOR);
/*
* Don't wait longer than a second if the caller
* didn't ask us to wait.
*/
timeout = USEC_PER_SEC;
while (num_online_cpus() > 1 && (wait || timeout--))
udelay(1);
}
local_irq_save(flags);
disable_local_APIC();
local_irq_restore(flags);
}
/*
* Reschedule call back.
*/
void smp_reschedule_interrupt(struct pt_regs *regs)
{
ack_APIC_irq();
inc_irq_stat(irq_resched_count);
scheduler_ipi();
/*
* KVM uses this interrupt to force a cpu out of guest mode
*/
}
void smp_call_function_interrupt(struct pt_regs *regs)
{
ack_APIC_irq();
irq_enter();
generic_smp_call_function_interrupt();
inc_irq_stat(irq_call_count);
irq_exit();
}
void smp_call_function_single_interrupt(struct pt_regs *regs)
{
ack_APIC_irq();
irq_enter();
generic_smp_call_function_single_interrupt();
inc_irq_stat(irq_call_count);
irq_exit();
}
struct smp_ops smp_ops = {
.smp_prepare_boot_cpu = native_smp_prepare_boot_cpu,
.smp_prepare_cpus = native_smp_prepare_cpus,
.smp_cpus_done = native_smp_cpus_done,
.stop_other_cpus = native_nmi_stop_other_cpus,
.smp_send_reschedule = native_smp_send_reschedule,
.cpu_up = native_cpu_up,
.cpu_die = native_cpu_die,
.cpu_disable = native_cpu_disable,
.play_dead = native_play_dead,
.send_call_func_ipi = native_send_call_func_ipi,
.send_call_func_single_ipi = native_send_call_func_single_ipi,
};
EXPORT_SYMBOL_GPL(smp_ops);