On 09/08/2016 03:45 AM, Peter Zijlstra wrote: > On Wed, Sep 07, 2016 at 07:44:17AM -0400, Prarit Bhargava wrote: >> The information in /sys/devices/system/cpu/cpuX/topology >> directory is useful for userspace monitoring applications and in-tree >> utilities like cpupower & turbostat. >> >> When down'ing a CPU the /sys/devices/system/cpu/cpuX/topology directory is >> removed during the CPU_DEAD hotplug callback in the kernel. The problem >> with this model is that the CPU has not been physically removed and the >> data in the topology directory is still valid and the cpu's location is >> now lost to userspace. >> >> This patch adds CONFIG_PERMANENT_CPU_TOPOLOGY, and is Y by default for >> x86, an N for all other arches. When enabled the kernel is modified so >> that the topology directory is added to the core cpu sysfs files so that >> the topology directory exists while the CPU is physically present. When >> disabled, the behavior of the current kernel is maintained (that is, the >> topology directory is removed on a soft down and added on an soft up). >> Adding CONFIG_PERMANENT_CPU_TOPOLOGY may require additional architecture >> so that the cpumask data the CPU's topology is not cleared during a CPU >> down. > > So how do you physically remove things and how does the information then > get removed? >
Physical hotremove was done with a toggle switch enabled test box. The topology directory gets removed when either common_cpu_attr_groups (!CONFIG_HOTPLUG_CPU) or hotplugable_cpu_attr_groups (CONFIG_HOTPLUG_CPU) are destroyed. The cpu device's attributes are set in register_cpu() and the device is destroyed in unregister_cpu(). IOW, the patch expands the lifetime of the topology directories to match that of the device. > Also, why is that an x86 only feature? The only arch that I could test on was x86 and I wasn't able to test to see if topology fields are valid for other arches on a soft remove. So x86 by default for now and other arches can make their decisions. P.
