admin-guide/mm/ksm.rst

8 KSM is a memory-saving de-duplication feature, enabled by CONFIG_KSM=y,
19 content which can be replaced by a single write-protected page (which
26 KSM's merged pages were originally locked into kernel memory, but can now
46 unmerges whatever it merged in that range.  Note: this unmerging call
47 may suddenly require more memory than is available - possibly failing
48 with EAGAIN, but more probably arousing the Out-Of-Memory killer.
60 will exceed ``vm.max_map_count`` (see Documentation/admin-guide/sysctl/vm.rst).
68 restricting its use to areas likely to benefit.  KSM's scans may use a lot
92         specifies if pages from different NUMA nodes can be merged.
109         * set to 0 to stop ksmd from running but keep merged pages,
111         * set to 2 to stop ksmd and unmerge all pages currently merged, but
120         empty pages are merged with the kernel zero page(s) instead of
128         effective for pages merged after the change.
155         scan. It's a noop if not a single KSM page hit the
201 save each scanned page's brief rmap information. Some of these pages may
202 be merged, but some may not be abled to be merged after being checked
205 1) How to determine whether KSM save memory or consume memory in system-wide
208 	general_profit =~ ksm_saved_pages * sizeof(page) - (all_rmap_items) *
219 	process_profit =~ ksm_saved_pages * sizeof(page) -
229 ``ksm_merging_pages`` means a bad madvise-applied policy, so developers or
231 for reference, a page's size is usually 4K, and the rmap_item's size is
232 separately 32B on 32-bit CPU architecture and 64B on 64-bit CPU architecture.
233 so if the ``ksm_rmap_items/ksm_merging_pages`` ratio exceeds 64 on 64-bit CPU
234 or exceeds 128 on 32-bit CPU, then the app's madvise policy should be dropped,
241 KSM might help save memory, it's a tradeoff by may suffering delay on KSM COW
253 	cannot do all the locking needed to reconstitute a cross-anon_vma KSM page.
255 --