Messages in this thread Patch in this message | | | Date | Fri, 20 Feb 2015 13:45:41 -0800 | From | Andrew Morton <> | Subject | Re: [PATCH] mm: hide per-cpu lists in output of show_mem() |
| |
On Fri, 20 Feb 2015 17:39:42 +0300 Konstantin Khlebnikov <khlebnikov@yandex-team.ru> wrote:
> This makes show_mem() much less verbose at huge machines. Instead of > huge and almost useless dump of counters for each per-zone per-cpu > lists this patch prints sum of these counters for each zone (free_pcp) > and size of per-cpu list for current cpu (local_pcp). > > Flag SHOW_MEM_PERCPU_LISTS reverts old verbose mode.
Forgot to update the comment:
--- a/mm/page_alloc.c~mm-hide-per-cpu-lists-in-output-of-show_mem-fix +++ a/mm/page_alloc.c @@ -3243,8 +3243,11 @@ static void show_migration_types(unsigne * Show free area list (used inside shift_scroll-lock stuff) * We also calculate the percentage fragmentation. We do this by counting the * memory on each free list with the exception of the first item on the list. - * Suppresses nodes that are not allowed by current's cpuset if - * SHOW_MEM_FILTER_NODES is passed. + * + * Bits in @filter: + * SHOW_MEM_FILTER_NODES: suppress nodes that are not allowed by current's + * cpuset. + * SHOW_MEM_PERCPU_LISTS: display full per-node per-cpu pcp lists */ void show_free_areas(unsigned int filter) {
Is there really any point in having SHOW_MEM_PERCPU_LISTS? There isn't presently a way of setting it(?).
| |