mm/page_alloc: make the annotations of available memory more accurate

Not all systems use swap, so estimating available memory would help to
prevent swapping or OOM of system that not use swap.

And we need to reserve some page cache to prevent swapping or thrashing.
If somebody is accessing the pages in pagecache, and if too much would be
freed, most accesses might mean reading data from disk, i.e.  thrashing.

Link: https://lkml.kernel.org/r/20220623020833.972979-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang <yang.yang29@zte.com.cn>
Signed-off-by: CGEL ZTE <cgel.zte@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 52fd92b2c1fe3f83bb26b3b27416aca123f4556f..c9c02b23f02f3e90d768ff78c32dba020746024b 100644 (file)
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5800,14 +5800,14 @@ long si_mem_available(void)
 
        /*
         * Estimate the amount of memory available for userspace allocations,
-        * without causing swapping.
+        * without causing swapping or OOM.
         */
        available = global_zone_page_state(NR_FREE_PAGES) - totalreserve_pages;
 
        /*
         * Not all the page cache can be freed, otherwise the system will
-        * start swapping. Assume at least half of the page cache, or the
-        * low watermark worth of cache, needs to stay.
+        * start swapping or thrashing. Assume at least half of the page
+        * cache, or the low watermark worth of cache, needs to stay.
         */
        pagecache = pages[LRU_ACTIVE_FILE] + pages[LRU_INACTIVE_FILE];
        pagecache -= min(pagecache / 2, wmark_low);