From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ozlabs.org (bilbo.ozlabs.org [203.11.71.1]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 40bdqg2m0tzF2ST for ; Wed, 2 May 2018 23:07:59 +1000 (AEST) From: Michael Ellerman To: linuxppc-dev@ozlabs.org Cc: npiggin@gmail.com Subject: [RFC PATCH 3/4] powerpc/64s: Wire up arch_trigger_cpumask_backtrace() Date: Wed, 2 May 2018 23:07:28 +1000 Message-Id: <20180502130729.24077-3-mpe@ellerman.id.au> In-Reply-To: <20180502130729.24077-1-mpe@ellerman.id.au> References: <20180502130729.24077-1-mpe@ellerman.id.au> List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , This allows eg. the RCU stall detector, or the soft/hardlockup detectors to trigger a backtrace on all CPUs. We implement this by sending a "safe" NMI, which will actually only send an IPI. Unfortunately the generic code prints "NMI", so that's a little confusing but we can probably live with it. If one of the CPUs doesn't respond to the IPI, we then print some info from it's paca and do a backtrace based on its saved_r1. Example output: INFO: rcu_sched detected stalls on CPUs/tasks: 2-...0: (0 ticks this GP) idle=1be/1/4611686018427387904 softirq=1055/1055 fqs=25735 (detected by 4, t=58847 jiffies, g=58, c=57, q=1258) Sending NMI from CPU 4 to CPUs 2: CPU 2 didn't respond to backtrace IPI, inspecting paca. irq_soft_mask: 0x01 in_mce: 0 in_nmi: 0 current: 3623 (bash) Back trace of paca->saved_r1 (0xc0000000e1c83ba0) (possibly stale): Call Trace: [c0000000e1c83ba0] [0000000000000014] 0x14 (unreliable) [c0000000e1c83bc0] [c000000000765798] lkdtm_do_action+0x48/0x80 [c0000000e1c83bf0] [c000000000765a40] direct_entry+0x110/0x1b0 [c0000000e1c83c90] [c00000000058e650] full_proxy_write+0x90/0xe0 [c0000000e1c83ce0] [c0000000003aae3c] __vfs_write+0x6c/0x1f0 [c0000000e1c83d80] [c0000000003ab214] vfs_write+0xd4/0x240 [c0000000e1c83dd0] [c0000000003ab5cc] ksys_write+0x6c/0x110 [c0000000e1c83e30] [c00000000000b860] system_call+0x58/0x6c Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/nmi.h | 4 ++++ arch/powerpc/kernel/stacktrace.c | 51 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 55 insertions(+) diff --git a/arch/powerpc/include/asm/nmi.h b/arch/powerpc/include/asm/nmi.h index 9c80939b4d14..e97f58689ca7 100644 --- a/arch/powerpc/include/asm/nmi.h +++ b/arch/powerpc/include/asm/nmi.h @@ -4,6 +4,10 @@ #ifdef CONFIG_PPC_WATCHDOG extern void arch_touch_nmi_watchdog(void); +extern void arch_trigger_cpumask_backtrace(const cpumask_t *mask, + bool exclude_self); +#define arch_trigger_cpumask_backtrace arch_trigger_cpumask_backtrace + #else static inline void arch_touch_nmi_watchdog(void) {} #endif diff --git a/arch/powerpc/kernel/stacktrace.c b/arch/powerpc/kernel/stacktrace.c index d534ed901538..cf4652d5df80 100644 --- a/arch/powerpc/kernel/stacktrace.c +++ b/arch/powerpc/kernel/stacktrace.c @@ -11,12 +11,15 @@ */ #include +#include #include #include #include #include #include +#include + /* * Save stack-backtrace addresses into a stack_trace buffer. */ @@ -76,3 +79,51 @@ save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace) save_context_stack(trace, regs->gpr[1], current, 0); } EXPORT_SYMBOL_GPL(save_stack_trace_regs); + +#ifdef CONFIG_PPC_BOOK3S_64 +static void handle_backtrace_ipi(struct pt_regs *regs) +{ + nmi_cpu_backtrace(regs); +} + +static void raise_backtrace_ipi(cpumask_t *mask) +{ + unsigned int cpu; + + for_each_cpu(cpu, mask) { + if (cpu == smp_processor_id()) + handle_backtrace_ipi(NULL); + else + smp_send_safe_nmi_ipi(cpu, handle_backtrace_ipi, 5 * USEC_PER_SEC); + } + + for_each_cpu(cpu, mask) { + struct paca_struct *p = paca_ptrs[cpu]; + + cpumask_clear_cpu(cpu, mask); + + pr_warn("CPU %d didn't respond to backtrace IPI, inspecting paca.\n", cpu); + if (!virt_addr_valid(p)) { + pr_warn("paca pointer appears corrupt? (%px)\n", p); + continue; + } + + pr_warn("irq_soft_mask: 0x%02x in_mce: %d in_nmi: %d", + p->irq_soft_mask, p->in_mce, p->in_nmi); + + if (virt_addr_valid(p->__current)) + pr_cont(" current: %d (%s)\n", p->__current->pid, + p->__current->comm); + else + pr_cont(" current pointer corrupt? (%px)\n", p->__current); + + pr_warn("Back trace of paca->saved_r1 (0x%016llx) (possibly stale):\n", p->saved_r1); + show_stack(p->__current, (unsigned long *)p->saved_r1); + } +} + +void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self) +{ + nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_backtrace_ipi); +} +#endif /* CONFIG_PPC64 */ -- 2.14.1