From 0b45143b4b9440579e7fa889708cfc4bc7fdb9a3 Mon Sep 17 00:00:00 2001 From: Georges Aureau Date: Thu, 11 Mar 2021 09:10:28 -0600 Subject: x86/platform/uv: Add more to secondary CPU kdump info Add call to run_crash_ipi_callback() to gather more info of what the secondary CPUs were doing to help with failure analysis. Excerpt from Georges: 'It is only changing where crash secondaries will be stalling after having taken care of properly laying down "crash note regs". Please note that "crash note regs" are a key piece of data used by crash dump debuggers to provide a reliable backtrace of running processors.' Secondary change pursuant to a5f526ecb075 ("CodingStyle: Inclusive Terminology"): change master/slave to main/secondary. [ bp: Massage commit message. ] Signed-off-by: Georges Aureau Signed-off-by: Mike Travis Signed-off-by: Borislav Petkov Reviewed-by: Steve Wahl Link: https://lkml.kernel.org/r/20210311151028.82678-1-mike.travis@hpe.com --- arch/x86/platform/uv/uv_nmi.c | 39 ++++++++++++++++++++++++--------------- 1 file changed, 24 insertions(+), 15 deletions(-) (limited to 'arch/x86/platform') diff --git a/arch/x86/platform/uv/uv_nmi.c b/arch/x86/platform/uv/uv_nmi.c index eafc530c8767..f83810f7bcc2 100644 --- a/arch/x86/platform/uv/uv_nmi.c +++ b/arch/x86/platform/uv/uv_nmi.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -834,34 +835,42 @@ static void uv_nmi_touch_watchdogs(void) touch_nmi_watchdog(); } -static atomic_t uv_nmi_kexec_failed; - #if defined(CONFIG_KEXEC_CORE) -static void uv_nmi_kdump(int cpu, int master, struct pt_regs *regs) +static atomic_t uv_nmi_kexec_failed; +static void uv_nmi_kdump(int cpu, int main, struct pt_regs *regs) { + /* Check if kdump kernel loaded for both main and secondary CPUs */ + if (!kexec_crash_image) { + if (main) + pr_err("UV: NMI error: kdump kernel not loaded\n"); + return; + } + /* Call crash to dump system state */ - if (master) { + if (main) { pr_emerg("UV: NMI executing crash_kexec on CPU%d\n", cpu); crash_kexec(regs); - pr_emerg("UV: crash_kexec unexpectedly returned, "); + pr_emerg("UV: crash_kexec unexpectedly returned\n"); atomic_set(&uv_nmi_kexec_failed, 1); - if (!kexec_crash_image) { - pr_cont("crash kernel not loaded\n"); - return; + + } else { /* secondary */ + + /* If kdump kernel fails, secondaries will exit this loop */ + while (atomic_read(&uv_nmi_kexec_failed) == 0) { + + /* Once shootdown cpus starts, they do not return */ + run_crash_ipi_callback(regs); + + mdelay(10); } - pr_cont("kexec busy, stalling cpus while waiting\n"); } - - /* If crash exec fails the slaves should return, otherwise stall */ - while (atomic_read(&uv_nmi_kexec_failed) == 0) - mdelay(10); } #else /* !CONFIG_KEXEC_CORE */ -static inline void uv_nmi_kdump(int cpu, int master, struct pt_regs *regs) +static inline void uv_nmi_kdump(int cpu, int main, struct pt_regs *regs) { - if (master) + if (main) pr_err("UV: NMI kdump: KEXEC not supported in this kernel\n"); atomic_set(&uv_nmi_kexec_failed, 1); } -- cgit v1.2.3 From 27743f01e391ee1d80e3be2a09237507b965f91b Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 20 Apr 2021 10:03:02 +0200 Subject: x86/platform/uv: Remove dead !CONFIG_KEXEC_CORE code The !CONFIG_KEXEC_CORE code in arch/x86/platform/uv/uv_nmi.c was unused, untested and didn't even build for 7 years. Since we fixed this by requiring X86_UV to depend on CONFIG_KEXEC_CORE, remove the (now) dead code. Also move the uv_nmi_kexec_failed definition back up to where the other file-scope global variables are defined. Signed-off-by: Ingo Molnar Cc: Mike Travis Cc: linux-kernel@vger.kernel.org --- arch/x86/platform/uv/uv_nmi.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) (limited to 'arch/x86/platform') diff --git a/arch/x86/platform/uv/uv_nmi.c b/arch/x86/platform/uv/uv_nmi.c index f83810f7bcc2..1556108ad29a 100644 --- a/arch/x86/platform/uv/uv_nmi.c +++ b/arch/x86/platform/uv/uv_nmi.c @@ -92,6 +92,8 @@ static atomic_t uv_nmi_cpus_in_nmi = ATOMIC_INIT(-1); static atomic_t uv_nmi_slave_continue; static cpumask_var_t uv_nmi_cpu_mask; +static atomic_t uv_nmi_kexec_failed; + /* Values for uv_nmi_slave_continue */ #define SLAVE_CLEAR 0 #define SLAVE_CONTINUE 1 @@ -835,8 +837,6 @@ static void uv_nmi_touch_watchdogs(void) touch_nmi_watchdog(); } -#if defined(CONFIG_KEXEC_CORE) -static atomic_t uv_nmi_kexec_failed; static void uv_nmi_kdump(int cpu, int main, struct pt_regs *regs) { /* Check if kdump kernel loaded for both main and secondary CPUs */ @@ -867,15 +867,6 @@ static void uv_nmi_kdump(int cpu, int main, struct pt_regs *regs) } } -#else /* !CONFIG_KEXEC_CORE */ -static inline void uv_nmi_kdump(int cpu, int main, struct pt_regs *regs) -{ - if (main) - pr_err("UV: NMI kdump: KEXEC not supported in this kernel\n"); - atomic_set(&uv_nmi_kexec_failed, 1); -} -#endif /* !CONFIG_KEXEC_CORE */ - #ifdef CONFIG_KGDB #ifdef CONFIG_KGDB_KDB static inline int uv_nmi_kdb_reason(void) -- cgit v1.2.3