ACPI: NUMA: Process hotpluggable memblocks when !CONFIG_MEMORY_HOTPLUG
authorVitaly Kuznetsov <vkuznets@redhat.com>
Thu, 16 Dec 2021 09:58:26 +0000 (10:58 +0100)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Fri, 17 Dec 2021 18:10:55 +0000 (19:10 +0100)
commitbcf6f1759adff0ec3f73ab7b4a873405ee26ff5c
tree6de7ed992c61dd55079aec3aec368bf14e75d8c3
parent2585cf9dfaaddf00b069673f27bb3f8530e2039c
ACPI: NUMA: Process hotpluggable memblocks when !CONFIG_MEMORY_HOTPLUG

Some systems (e.g. Hyper-V guests) have all their memory marked as
hotpluggable in SRAT. acpi_numa_memory_affinity_init(), however,
ignores all such regions when !CONFIG_MEMORY_HOTPLUG and this is
unfortunate as memory affinity (NUMA) information gets lost.

'Hot Pluggable' flag in SRAT only means that "system hardware supports
hot-add and hot-remove of this memory region", it doesn't prevent
memory from being cold-plugged there.

Ignore 'Hot Pluggable' bit instead of skipping the whole memory
affinity information when !CONFIG_MEMORY_HOTPLUG.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
drivers/acpi/numa/srat.c