- 15 Jul, 2016 7 commits
-
-
Richard Cochran authored
As part of the hotplug cleanup, the CPU_STARTING/DYING actions are going away soon. This driver needlessly uses those two macro, and so this patch replaces that code with something more sensible. Commit: 8da83734 ("ACPI / processor: Fix STARTING/DYING action in acpi_cpu_soft_notify()") added checks for those two actions, because the notification callback can sleep, causing a hung CPU. This patch instead checks for the ONLINE/DEAD actions, which are the ones that are handled by the driver in the first place. Signed-off-by:
Richard Cochran <rcochran@linutronix.de> Signed-off-by:
Anna-Maria Gleixner <anna-maria@linutronix.de> Reviewed-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Len Brown <lenb@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-acpi@vger.kernel.org Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153335.964962885@linutronix.deSigned-off-by:
Ingo Molnar <mingo@kernel.org>
-
Thomas Gleixner authored
Install the callbacks via the state machine. The core won't invoke the callbacks on already online CPUs. Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Radim Krcmar <rkrcmar@redhat.com> Cc: kvm@vger.kernel.org Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153335.886159080@linutronix.deSigned-off-by:
-
Thomas Gleixner authored
Straight forward conversion w/o bells and whistles. Signed-off-by:
-
Thomas Gleixner authored
Straight forward conversion plus commentary why code which is executed in hotplug callbacks needs to be invoked before installing them. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine. There is no setup just one teardown callback. Remove the silly comment about the workqueue up dependency. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. We assumed that the priority ordering was ment to invoke the online callback as the last step. In the original code this also invoked the down prepare callback as the last step. With the symmetric state machine the down prepare callback is now the first step. Signed-off-by:
-
Anna-Maria Gleixner authored
Since the following commit: 1cf4f629 ("cpu/hotplug: Move online calls to hotplugged cpu") ... the CPU_ONLINE and CPU_DOWN_PREPARE notifiers are always run on the hot plugged CPU, and as of commit: 3b9d6da6 ("cpu/hotplug: Fix rollback during error-out in __cpu_disable()") the CPU_DOWN_FAILED notifier also runs on the hot plugged CPU. This patch converts the SMP functional calls into direct calls. smp_function_call_single() executes the function with interrupts disabled. This calling convention is not preserved because there is no reason to do so. Signed-off-by:
-
- 14 Jul, 2016 29 commits
-
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Get rid of the prio ordering of the separate notifiers and use a proper state callback pair. Signed-off-by:
Tejun Heo <tj@kernel.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Lai Jiangshan <jiangshanlai@gmail.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Nicolas Iooss <nicolas.iooss_linux@m4x.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153335.197083890@linutronix.deSigned-off-by:
-
Anna-Maria Gleixner authored
Install the callbacks via the state machine. Signed-off-by:
-
Thomas Gleixner authored
All users converted to state machine callbacks. Signed-off-by:
-
Anna-Maria Gleixner authored
To simplify the hotplug mechanism move the starting callback to online. There is no functional requirement that the cpumask bit has to be set in the starting callback. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Install the callback via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Thomas Gleixner authored
Convert the notifiers to state machine states and let the core code do the setup for the already online CPUs. This notifier has a completely undocumented ordering requirement versus perf hardcoded in the notifier priority. This odering is only required for CPU down, so that hardware migration happens before the core is notified about the outgoing CPU. Signed-off-by:
-
Thomas Gleixner authored
Replace the perf_notifier() install mechanism, which invokes magically the callback on the current CPU. Convert the hardware specific callbacks which are invoked from the x86 perf core to return proper error codes instead of totally pointless NOTIFY_BAD return values. Signed-off-by:
-
Thomas Gleixner authored
Actually a nice symmetric startup/teardown pair which fits properly into the state machine concept. In the long run we should be able to invoke the startup callback for the boot CPU via the state machine and get rid of the init function which invokes it on the boot CPU. Note: This comes actually before the perf hardware callbacks. In the notifier model the hardware callbacks have a higher priority than the core callback. But that's solely for CPU offline so that hardware migration of events happens before the core is notified about the outgoing CPU. With the symetric state array model we have the following ordering: UP: core -> hardware DOWN: hardware -> core Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine. Signed-off-by:
Jason Cooper <jason@lakedaemon.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153333.416260485@linutronix.deSigned-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine. Signed-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
-
Richard Cochran authored
Install the callbacks via the state machine. Signed-off-by:
-
Richard Cochran authored
More or less straightforward, although this driver sports some very interesting SMP setup code. Regarding the callback ordering, this deleted comment is interesting: ... the GIC needs to be up before the ARM generic timers. That comment is half baken as the same requirement is true for perf. Signed-off-by:
-
Sebastian Andrzej Siewior authored
Install the callbacks via the state machine and let the core invoke the callbacks on the already online CPUs. Signed-off-by:
Andy Lutomirski <luto@amacapital.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: rt@linutronix.de Link: http://lkml.kernel.org/r/20160713153332.987560239@linutronix.deSigned-off-by:
-
Thomas Gleixner authored
We switched the hotplug machinery to smpboot threads. Early registration of hotplug callbacks, i.e. from do_pre_smp_initcalls(), happens before the threads are initialized. Instead of moving the thread init, we simply handle it in the hotplug code itself and invoke the function directly. Signed-off-by:
-
- 13 Jul, 2016 4 commits
-
-
Thomas Gleixner authored
-
Thomas Gleixner authored
-
Thomas Gleixner authored
-
Thomas Gleixner authored
Xiaolong Ye reported lock debug warnings triggered by the following commit: 8de4a0066106 ("perf/x86: Convert the core to the hotplug state machine") The bug is the following: the cpuhp_bp_states[] array is cut short when CONFIG_SMP=n, but the dynamically registered callbacks are stored nevertheless and happily scribble outside of the array bounds... We need to store them in case that the state is unregistered so we can invoke the teardown function. That's independent of CONFIG_SMP. Make sure the array is large enough. Reported-by:kernel test robot <xiaolong.ye@intel.com> Signed-off-by:
-
