From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751972AbdAYQf4 (ORCPT ); Wed, 25 Jan 2017 11:35:56 -0500 Received: from relay1.sgi.com ([192.48.180.66]:59844 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751030AbdAYQfS (ORCPT ); Wed, 25 Jan 2017 11:35:18 -0500 Message-Id: <20170125163517.756690240@asylum.americas.sgi.com> References: <20170125163517.162577002@asylum.americas.sgi.com> User-Agent: quilt/0.46-1 Date: Wed, 25 Jan 2017 10:35:20 -0600 From: "'Mike Travis" , '@sgi.com To: Ingo Molnar , Thomas Gleixner , "H. Peter Anvin" , Russ Anderson Cc: x86@kernel.org, linux-kernel@vger.kernel.org, Dimitri Sivanich Subject: [PATCH 3/7] x86/platform/UV: Add basic CPU NMI health check Content-Disposition: inline; filename=uv4_add-health-check Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Add a low impact health check triggered by the system NMI command that essentially checks which CPUs are responding to external NMI's. Signed-off-by: Mike Travis Acked-by: Dimitri Sivanich Reviewed-by: Russ Anderson Reviewed-by: Alex Thorlton --- arch/x86/platform/uv/uv_nmi.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) --- linux.orig/arch/x86/platform/uv/uv_nmi.c +++ linux/arch/x86/platform/uv/uv_nmi.c @@ -176,6 +176,7 @@ module_param_named(debug, uv_nmi_debug, * "kdump" - do crash dump * "kdb" - enter KDB (default) * "kgdb" - enter KGDB + * "health" - check if CPUs respond to NMI */ static char uv_nmi_action[8] = "kdb"; module_param_string(action, uv_nmi_action, sizeof(uv_nmi_action), 0644); @@ -571,6 +572,22 @@ static void uv_nmi_sync_exit(int master) } } +/* Current "health" check is to check which CPU's are responsive */ +static void uv_nmi_action_health(int cpu, struct pt_regs *regs, int master) +{ + if (master) { + int in = atomic_read(&uv_nmi_cpus_in_nmi); + int out = num_online_cpus() - in; + + pr_alert("UV: NMI CPU health check (non-responding:%d)\n", out); + atomic_set(&uv_nmi_slave_continue, SLAVE_EXIT); + } else { + while (!atomic_read(&uv_nmi_slave_continue)) + cpu_relax(); + } + uv_nmi_sync_exit(master); +} + /* Walk through cpu list and dump state of each */ static void uv_nmi_dump_state(int cpu, struct pt_regs *regs, int master) { @@ -747,7 +764,9 @@ int uv_handle_nmi(unsigned int reason, s uv_nmi_wait(master); /* Process actions other than "kdump": */ - if (uv_nmi_action_is("ips") || uv_nmi_action_is("dump")) { + if (uv_nmi_action_is("health")) { + uv_nmi_action_health(cpu, regs, master); + } else if (uv_nmi_action_is("ips") || uv_nmi_action_is("dump")) { uv_nmi_dump_state(cpu, regs, master); } else if (uv_nmi_action_is("kdb") || uv_nmi_action_is("kgdb")) { uv_call_kgdb_kdb(cpu, regs, master); --