From: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> To: oleg@redhat.com, rjw@rjwysocki.net, rusty@rustcorp.com.au, tglx@linutronix.de, akpm@linux-foundation.org Cc: paulmck@linux.vnet.ibm.com, tj@kernel.org, walken@google.com, ego@linux.vnet.ibm.com, linux@arm.linux.org.uk, linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, linux-pm@vger.kernel.org, linuxppc-dev@ozlabs.org, srivatsa.bhat@linux.vnet.ibm.com, Peter Zijlstra <peterz@infradead.org>, Paul Mackerras <paulus@samba.org>, Ingo Molnar <mingo@kernel.org>, Arnaldo Carvalho de Melo <acme@ghostprotocols.net>"Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Subject: [PATCH v3 04/52] CPU hotplug, perf: Fix CPU hotplug callback registration Date: Tue, 11 Mar 2014 02:04:39 +0530 [thread overview] Message-ID: <20140310203438.10746.24505.stgit@srivatsabhat.in.ibm.com> (raw) In-Reply-To: <20140310203312.10746.310.stgit@srivatsabhat.in.ibm.com> Subsystems that want to register CPU hotplug callbacks, as well as perform initialization for the CPUs that are already online, often do it as shown below: get_online_cpus(); for_each_online_cpu(cpu) init_cpu(cpu); register_cpu_notifier(&foobar_cpu_notifier); put_online_cpus(); This is wrong, since it is prone to ABBA deadlocks involving the cpu_add_remove_lock and the cpu_hotplug.lock (when running concurrently with CPU hotplug operations). Instead, the correct and race-free way of performing the callback registration is: cpu_notifier_register_begin(); for_each_online_cpu(cpu) init_cpu(cpu); /* Note the use of the double underscored version of the API */ __register_cpu_notifier(&foobar_cpu_notifier); cpu_notifier_register_done(); Fix the perf subsystem's hotplug notifier by using this latter form of callback registration. Also provide a bare-bones version of perf_cpu_notifier() that doesn't invoke the notifiers for the already online CPUs. This would be useful for subsystems that need to perform a different set of initialization for the already online CPUs, or don't need the initialization altogether. Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> --- include/linux/perf_event.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index e56b07f..3356abc 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -835,6 +835,8 @@ do { \ { .notifier_call = fn, .priority = CPU_PRI_PERF }; \ unsigned long cpu = smp_processor_id(); \ unsigned long flags; \ + \ + cpu_notifier_register_begin(); \ fn(&fn##_nb, (unsigned long)CPU_UP_PREPARE, \ (void *)(unsigned long)cpu); \ local_irq_save(flags); \ @@ -843,9 +845,21 @@ do { \ local_irq_restore(flags); \ fn(&fn##_nb, (unsigned long)CPU_ONLINE, \ (void *)(unsigned long)cpu); \ - register_cpu_notifier(&fn##_nb); \ + __register_cpu_notifier(&fn##_nb); \ + cpu_notifier_register_done(); \ } while (0) +/* + * Bare-bones version of perf_cpu_notifier(), which doesn't invoke the + * callback for already online CPUs. + */ +#define __perf_cpu_notifier(fn) \ +do { \ + static struct notifier_block fn##_nb = \ + { .notifier_call = fn, .priority = CPU_PRI_PERF }; \ + \ + __register_cpu_notifier(&fn##_nb); \ +} while (0) struct perf_pmu_events_attr { struct device_attribute attr;
WARNING: multiple messages have this Message-ID (diff)
From: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> To: paulus@samba.org, oleg@redhat.com, mingo@kernel.org, rjw@rjwysocki.net, rusty@rustcorp.com.au, peterz@infradead.org, tglx@linutronix.de, akpm@linux-foundation.org Cc: paulmck@linux.vnet.ibm.com, tj@kernel.org, walken@google.com, ego@linux.vnet.ibm.com, linux@arm.linux.org.uk, linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, linux-pm@vger.kernel.org, linuxppc-dev@ozlabs.org, srivatsa.bhat@linux.vnet.ibm.com, Arnaldo Carvalho de Melo <acme@ghostprotocols.net>"Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com> Subject: [PATCH v3 04/52] CPU hotplug, perf: Fix CPU hotplug callback registration Date: Tue, 11 Mar 2014 02:04:39 +0530 [thread overview] Message-ID: <20140310203438.10746.24505.stgit@srivatsabhat.in.ibm.com> (raw) Message-ID: <20140310203439.Ra9V8oTqa3i9k5tq9LESf6YKR9x0PJyxk9HfMC8vh5k@z> (raw) In-Reply-To: <20140310203312.10746.310.stgit@srivatsabhat.in.ibm.com> Subsystems that want to register CPU hotplug callbacks, as well as perform initialization for the CPUs that are already online, often do it as shown below: get_online_cpus(); for_each_online_cpu(cpu) init_cpu(cpu); register_cpu_notifier(&foobar_cpu_notifier); put_online_cpus(); This is wrong, since it is prone to ABBA deadlocks involving the cpu_add_remove_lock and the cpu_hotplug.lock (when running concurrently with CPU hotplug operations). Instead, the correct and race-free way of performing the callback registration is: cpu_notifier_register_begin(); for_each_online_cpu(cpu) init_cpu(cpu); /* Note the use of the double underscored version of the API */ __register_cpu_notifier(&foobar_cpu_notifier); cpu_notifier_register_done(); Fix the perf subsystem's hotplug notifier by using this latter form of callback registration. Also provide a bare-bones version of perf_cpu_notifier() that doesn't invoke the notifiers for the already online CPUs. This would be useful for subsystems that need to perform a different set of initialization for the already online CPUs, or don't need the initialization altogether. Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> --- include/linux/perf_event.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index e56b07f..3356abc 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -835,6 +835,8 @@ do { \ { .notifier_call = fn, .priority = CPU_PRI_PERF }; \ unsigned long cpu = smp_processor_id(); \ unsigned long flags; \ + \ + cpu_notifier_register_begin(); \ fn(&fn##_nb, (unsigned long)CPU_UP_PREPARE, \ (void *)(unsigned long)cpu); \ local_irq_save(flags); \ @@ -843,9 +845,21 @@ do { \ local_irq_restore(flags); \ fn(&fn##_nb, (unsigned long)CPU_ONLINE, \ (void *)(unsigned long)cpu); \ - register_cpu_notifier(&fn##_nb); \ + __register_cpu_notifier(&fn##_nb); \ + cpu_notifier_register_done(); \ } while (0) +/* + * Bare-bones version of perf_cpu_notifier(), which doesn't invoke the + * callback for already online CPUs. + */ +#define __perf_cpu_notifier(fn) \ +do { \ + static struct notifier_block fn##_nb = \ + { .notifier_call = fn, .priority = CPU_PRI_PERF }; \ + \ + __register_cpu_notifier(&fn##_nb); \ +} while (0) struct perf_pmu_events_attr { struct device_attribute attr;
next prev parent reply other threads:[~2014-03-10 20:34 UTC|newest] Thread overview: 123+ messages / expand[flat|nested] mbox.gz Atom feed top 2014-03-10 20:33 [PATCH v3 00/52] CPU hotplug: Fix issues with callback registration Srivatsa S. Bhat 2014-03-10 20:33 ` Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 01/52] CPU hotplug: Add lockdep annotations to get/put_online_cpus() Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 02/52] CPU hotplug: Provide lockless versions of callback registration functions Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 03/52] Doc/cpu-hotplug: Specify race-free way to register CPU hotplug callbacks Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat [this message] 2014-03-10 20:34 ` [PATCH v3 04/52] CPU hotplug, perf: Fix CPU hotplug callback registration Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 05/52] ia64, salinfo: Fix " Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 06/52] ia64, palinfo: Fix CPU " Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:34 ` [PATCH v3 07/52] ia64, topology: " Srivatsa S. Bhat 2014-03-10 20:34 ` Srivatsa S. Bhat 2014-03-10 20:35 ` [PATCH v3 08/52] ia64, err-inject: " Srivatsa S. Bhat 2014-03-10 20:35 ` Srivatsa S. Bhat 2014-03-10 20:35 ` [PATCH v3 09/52] arm, hw-breakpoint: " Srivatsa S. Bhat 2014-03-10 20:35 ` Srivatsa S. Bhat 2014-03-10 20:35 ` [PATCH v3 10/52] arm, kvm: " Srivatsa S. Bhat 2014-03-10 20:35 ` Srivatsa S. Bhat 2014-03-12 23:21 ` Christoffer Dall 2014-03-12 23:21 ` Christoffer Dall 2014-03-14 5:43 ` Srivatsa S. Bhat 2014-03-14 5:43 ` Srivatsa S. Bhat 2014-03-14 19:10 ` Christoffer Dall 2014-03-14 19:10 ` Christoffer Dall 2014-03-18 10:23 ` [UPDATED PATCH " Srivatsa S. Bhat 2014-03-18 22:08 ` Christoffer Dall 2014-03-18 22:08 ` Christoffer Dall 2014-03-10 20:35 ` [PATCH v3 11/52] s390, cacheinfo: " Srivatsa S. Bhat 2014-03-10 20:35 ` Srivatsa S. Bhat 2014-03-10 20:35 ` [PATCH v3 12/52] s390, smp: " Srivatsa S. Bhat 2014-03-10 20:35 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 13/52] sparc, sysfs: " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 14/52] powerpc, " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 15/52] x86, msr: " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 16/52] x86, cpuid: " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 17/52] x86, vsyscall: " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:36 ` [PATCH v3 18/52] x86, intel, uncore: " Srivatsa S. Bhat 2014-03-10 20:36 ` Srivatsa S. Bhat 2014-03-10 20:37 ` [PATCH v3 19/52] x86, mce: " Srivatsa S. Bhat 2014-03-10 20:37 ` Srivatsa S. Bhat 2014-03-10 20:37 ` [PATCH v3 20/52] x86, therm_throt.c: " Srivatsa S. Bhat 2014-03-10 20:37 ` Srivatsa S. Bhat 2014-03-10 20:37 ` [PATCH v3 21/52] x86, therm_throt.c: Remove unused therm_cpu_lock Srivatsa S. Bhat 2014-03-10 20:37 ` Srivatsa S. Bhat 2014-03-10 20:37 ` [PATCH v3 22/52] x86, amd, ibs: Fix CPU hotplug callback registration Srivatsa S. Bhat 2014-03-10 20:37 ` Srivatsa S. Bhat 2014-03-10 20:37 ` [PATCH v3 23/52] x86, intel, cacheinfo: " Srivatsa S. Bhat 2014-03-10 20:37 ` Srivatsa S. Bhat 2014-03-10 20:38 ` [PATCH v3 24/52] x86, intel, rapl: " Srivatsa S. Bhat 2014-03-10 20:38 ` Srivatsa S. Bhat 2014-03-10 20:38 ` [PATCH v3 25/52] x86, amd, uncore: " Srivatsa S. Bhat 2014-03-10 20:38 ` Srivatsa S. Bhat 2014-03-10 20:38 ` [PATCH v3 26/52] x86, hpet: " Srivatsa S. Bhat 2014-03-10 20:38 ` Srivatsa S. Bhat 2014-03-10 20:38 ` [PATCH v3 27/52] x86, pci, amd-bus: " Srivatsa S. Bhat 2014-03-10 20:38 ` Srivatsa S. Bhat 2014-03-10 20:38 ` [PATCH v3 28/52] x86, oprofile, nmi: " Srivatsa S. Bhat 2014-03-10 20:38 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 29/52] x86, kvm: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 30/52] arm64, hw_breakpoint.c: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 31/52] arm64, debug-monitors: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 32/52] powercap, intel-rapl: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-12 22:27 ` Jacob Pan 2014-03-10 20:39 ` [PATCH v3 33/52] scsi, bnx2i: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 34/52] scsi, bnx2fc: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 35/52] scsi, fcoe: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-10 20:39 ` [PATCH v3 36/52] zsmalloc: " Srivatsa S. Bhat 2014-03-10 20:39 ` Srivatsa S. Bhat 2014-03-14 6:41 ` Minchan Kim 2014-03-10 20:40 ` [PATCH v3 37/52] acpi-cpufreq: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 38/52] drivers/base/topology.c: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 39/52] clocksource, dummy-timer: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 40/52] intel-idle: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 41/52] oprofile, nmi-timer: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 42/52] octeon, watchdog: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:40 ` [PATCH v3 43/52] thermal, x86-pkg-temp: " Srivatsa S. Bhat 2014-03-10 20:40 ` Srivatsa S. Bhat 2014-03-10 20:41 ` [PATCH v3 44/52] hwmon, coretemp: " Srivatsa S. Bhat 2014-03-10 20:41 ` Srivatsa S. Bhat 2014-03-10 20:41 ` [PATCH v3 45/52] hwmon, via-cputemp: " Srivatsa S. Bhat 2014-03-10 20:41 ` Srivatsa S. Bhat 2014-03-10 20:41 ` [PATCH v3 46/52] xen, balloon: " Srivatsa S. Bhat 2014-03-10 20:41 ` Srivatsa S. Bhat 2014-03-10 20:41 ` [PATCH v3 47/52] trace, ring-buffer: " Srivatsa S. Bhat 2014-03-10 20:41 ` Srivatsa S. Bhat 2014-03-10 20:42 ` [PATCH v3 48/52] profile: " Srivatsa S. Bhat 2014-03-10 20:42 ` Srivatsa S. Bhat 2014-03-10 20:42 ` [PATCH v3 49/52] mm, vmstat: " Srivatsa S. Bhat 2014-03-10 20:42 ` Srivatsa S. Bhat 2014-03-10 20:42 ` [PATCH v3 50/52] mm, zswap: " Srivatsa S. Bhat 2014-03-10 20:42 ` Srivatsa S. Bhat 2014-03-10 20:42 ` [PATCH v3 51/52] net/core/flow.c: " Srivatsa S. Bhat 2014-03-10 20:42 ` Srivatsa S. Bhat 2014-03-10 20:42 ` [PATCH v3 52/52] net/iucv/iucv.c: " Srivatsa S. Bhat 2014-03-10 20:42 ` Srivatsa S. Bhat 2014-03-11 0:31 ` [PATCH v3 00/52] CPU hotplug: Fix issues with " Rafael J. Wysocki 2014-03-11 0:31 ` Rafael J. Wysocki 2014-03-11 22:07 ` Andrew Morton 2014-03-11 22:07 ` Andrew Morton 2014-03-12 20:48 ` Srivatsa S. Bhat 2014-03-12 20:48 ` Srivatsa S. Bhat
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20140310203438.10746.24505.stgit@srivatsabhat.in.ibm.com \ --to=srivatsa.bhat@linux.vnet.ibm.com \ --cc=acme@ghostprotocols.net \ --cc=akpm@linux-foundation.org \ --cc=ego@linux.vnet.ibm.com \ --cc=linux-arch@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-pm@vger.kernel.org \ --cc=linux@arm.linux.org.uk \ --cc=linuxppc-dev@ozlabs.org \ --cc=mingo@kernel.org \ --cc=oleg@redhat.com \ --cc=paulmck@linux.vnet.ibm.com \ --cc=paulus@samba.org \ --cc=peterz@infradead.org \ --cc=rjw@rjwysocki.net \ --cc=rusty@rustcorp.com.au \ --cc=tglx@linutronix.de \ --cc=tj@kernel.org \ --cc=walken@google.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).