From 4038f901cf102a40715b900984ed7540a9fa637f Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 26 Sep 2006 10:52:27 +0200 Subject: [PATCH] [PATCH] i386/x86-64: Fix NMI watchdog suspend/resume Making NMI suspend/resume work with SMP. We use CPU hotplug to offline APs in SMP suspend/resume. Only BSP executes sysdev's .suspend/.resume method. APs should follow CPU hotplug code path. And: +From: Don Zickus Makes the start/stop paths of nmi watchdog more robust to handle the suspend/resume cases more gracefully. AK: I merged the two patches together Signed-off-by: Shaohua Li Signed-off-by: Andi Kleen Cc: Don Zickus Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/nmi.c | 33 ++++++++++++++++++++++++++------- arch/i386/kernel/smpboot.c | 3 ++- arch/x86_64/kernel/nmi.c | 33 ++++++++++++++++++++++++++------- arch/x86_64/kernel/smpboot.c | 2 ++ include/asm-i386/nmi.h | 1 + include/asm-x86_64/nmi.h | 1 + 6 files changed, 58 insertions(+), 15 deletions(-) diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 6241e4448cab..8e4ed930ce6b 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk { static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk); /* local prototypes */ -static void stop_apic_nmi_watchdog(void *unused); static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu); extern void show_registers(struct pt_regs *regs); @@ -341,15 +340,20 @@ static int nmi_pm_active; /* nmi_active before suspend */ static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state) { + /* only CPU0 goes here, other CPUs should be offline */ nmi_pm_active = atomic_read(&nmi_active); - disable_lapic_nmi_watchdog(); + stop_apic_nmi_watchdog(NULL); + BUG_ON(atomic_read(&nmi_active) != 0); return 0; } static int lapic_nmi_resume(struct sys_device *dev) { - if (nmi_pm_active > 0) - enable_lapic_nmi_watchdog(); + /* only CPU0 goes here, other CPUs should be offline */ + if (nmi_pm_active > 0) { + setup_apic_nmi_watchdog(NULL); + touch_nmi_watchdog(); + } return 0; } @@ -626,11 +630,21 @@ static void stop_p4_watchdog(void) void setup_apic_nmi_watchdog (void *unused) { + struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); + /* only support LOCAL and IO APICs for now */ if ((nmi_watchdog != NMI_LOCAL_APIC) && (nmi_watchdog != NMI_IO_APIC)) return; + if (wd->enabled == 1) + return; + + /* cheap hack to support suspend/resume */ + /* if cpu0 is not active neither should the other cpus */ + if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0)) + return; + if (nmi_watchdog == NMI_LOCAL_APIC) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -663,17 +677,22 @@ void setup_apic_nmi_watchdog (void *unused) return; } } - __get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1; + wd->enabled = 1; atomic_inc(&nmi_active); } -static void stop_apic_nmi_watchdog(void *unused) +void stop_apic_nmi_watchdog(void *unused) { + struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); + /* only support LOCAL and IO APICs for now */ if ((nmi_watchdog != NMI_LOCAL_APIC) && (nmi_watchdog != NMI_IO_APIC)) return; + if (wd->enabled == 0) + return; + if (nmi_watchdog == NMI_LOCAL_APIC) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -697,7 +716,7 @@ static void stop_apic_nmi_watchdog(void *unused) return; } } - __get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0; + wd->enabled = 0; atomic_dec(&nmi_active); } diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index efe07990e7fc..9367af76ce37 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1376,7 +1376,8 @@ int __cpu_disable(void) */ if (cpu == 0) return -EBUSY; - + if (nmi_watchdog == NMI_LOCAL_APIC) + stop_apic_nmi_watchdog(NULL); clear_local_APIC(); /* Allow any queued timer interrupts to get serviced */ local_irq_enable(); diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index dd57410dad51..5a35975e5763 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk { static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk); /* local prototypes */ -static void stop_apic_nmi_watchdog(void *unused); static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu); /* converts an msr to an appropriate reservation bit */ @@ -337,15 +336,20 @@ static int nmi_pm_active; /* nmi_active before suspend */ static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state) { + /* only CPU0 goes here, other CPUs should be offline */ nmi_pm_active = atomic_read(&nmi_active); - disable_lapic_nmi_watchdog(); + stop_apic_nmi_watchdog(NULL); + BUG_ON(atomic_read(&nmi_active) != 0); return 0; } static int lapic_nmi_resume(struct sys_device *dev) { - if (nmi_pm_active > 0) - enable_lapic_nmi_watchdog(); + /* only CPU0 goes here, other CPUs should be offline */ + if (nmi_pm_active > 0) { + setup_apic_nmi_watchdog(NULL); + touch_nmi_watchdog(); + } return 0; } @@ -561,11 +565,21 @@ static void stop_p4_watchdog(void) void setup_apic_nmi_watchdog(void *unused) { + struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); + /* only support LOCAL and IO APICs for now */ if ((nmi_watchdog != NMI_LOCAL_APIC) && (nmi_watchdog != NMI_IO_APIC)) return; + if (wd->enabled == 1) + return; + + /* cheap hack to support suspend/resume */ + /* if cpu0 is not active neither should the other cpus */ + if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0)) + return; + if (nmi_watchdog == NMI_LOCAL_APIC) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -582,17 +596,22 @@ void setup_apic_nmi_watchdog(void *unused) return; } } - __get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1; + wd->enabled = 1; atomic_inc(&nmi_active); } -static void stop_apic_nmi_watchdog(void *unused) +void stop_apic_nmi_watchdog(void *unused) { + struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); + /* only support LOCAL and IO APICs for now */ if ((nmi_watchdog != NMI_LOCAL_APIC) && (nmi_watchdog != NMI_IO_APIC)) return; + if (wd->enabled == 0) + return; + if (nmi_watchdog == NMI_LOCAL_APIC) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -607,7 +626,7 @@ static void stop_apic_nmi_watchdog(void *unused) return; } } - __get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0; + wd->enabled = 0; atomic_dec(&nmi_active); } diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 975380207b46..15555879ce95 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -1233,6 +1233,8 @@ int __cpu_disable(void) if (cpu == 0) return -EBUSY; + if (nmi_watchdog == NMI_LOCAL_APIC) + stop_apic_nmi_watchdog(NULL); clear_local_APIC(); /* diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h index 13b5d8311bf7..303bcd4592bb 100644 --- a/include/asm-i386/nmi.h +++ b/include/asm-i386/nmi.h @@ -23,6 +23,7 @@ extern int reserve_evntsel_nmi(unsigned int); extern void release_evntsel_nmi(unsigned int); extern void setup_apic_nmi_watchdog (void *); +extern void stop_apic_nmi_watchdog (void *); extern void disable_timer_nmi_watchdog(void); extern void enable_timer_nmi_watchdog(void); extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason); diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h index 2c23b0df87d2..578596494275 100644 --- a/include/asm-x86_64/nmi.h +++ b/include/asm-x86_64/nmi.h @@ -54,6 +54,7 @@ extern int reserve_evntsel_nmi(unsigned int); extern void release_evntsel_nmi(unsigned int); extern void setup_apic_nmi_watchdog (void *); +extern void stop_apic_nmi_watchdog (void *); extern void disable_timer_nmi_watchdog(void); extern void enable_timer_nmi_watchdog(void); extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);