]> git.baikalelectronics.ru Git - kernel.git/commit
mm/page_alloc: skip non present sections on zone initialization
authorKirill A. Shutemov <kirill@shutemov.name>
Fri, 31 Jan 2020 06:13:57 +0000 (22:13 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 31 Jan 2020 18:30:38 +0000 (10:30 -0800)
commit75a9b40a507d427c2f84dc3a88ba53424fe5d642
treebba5464134ecf526046af1dfbf3e539335fff6ea
parent1c0e4502c5ff056d6d8297ccf4e09591a00e1833
mm/page_alloc: skip non present sections on zone initialization

memmap_init_zone() can be called on the ranges with holes during the
boot.  It will skip any non-valid PFNs one-by-one.  It works fine as
long as holes are not too big.

But huge holes in the memory map causes a problem.  It takes over 20
seconds to walk 32TiB hole.  x86-64 with 5-level paging allows for much
larger holes in the memory map which would practically hang the system.

Deferred struct page init doesn't help here.  It only works on the
present ranges.

Skipping non-present sections would fix the issue.

Link: http://lkml.kernel.org/r/20191230093828.24613-1-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Cc: "Jin, Zhi" <zhi.jin@intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Pavel Tatashin <pasha.tatashin@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c