From: Michal Hocko <mhocko@xxxxxxxx> Subject: mm/sparse.c: do not waste pre allocated memmap space Vincent has noticed [1] that there is something unusual with the memmap allocations going on on his platform : I noticed this because on my ARM64 platform, with 1 GiB of memory the : first [and only] section is allocated from the zeroing path while with : 2 GiB of memory the first 1 GiB section is allocated from the : non-zeroing path. The underlying problem is that although sparse_buffer_init allocates enough memory for all sections on the node sparse_buffer_alloc is not able to consume them due to mismatch in the expected allocation alignement. While sparse_buffer_init preallocation uses the PAGE_SIZE alignment the real memmap has to be aligned to section_map_size() this results in a wasted initial chunk of the preallocated memmap and unnecessary fallback allocation for a section. While we are at it also change __populate_section_memmap to align to the requested size because at least VMEMMAP has constrains to have memmap properly aligned. [1] http://lkml.kernel.org/r/20191030131122.8256-1-vincent.whitchurch@xxxxxxxx [akpm@xxxxxxxxxxxxxxxxxxxx: tweak layout, per David] Link: http://lkml.kernel.org/r/20191119092642.31799-1-mhocko@xxxxxxxxxx Fixes: 35fd1eb1e821 ("mm/sparse: abstract sparse buffer allocations") Signed-off-by: Michal Hocko <mhocko@xxxxxxxx> Reported-by: Vincent Whitchurch <vincent.whitchurch@xxxxxxxx> Debugged-by: Vincent Whitchurch <vincent.whitchurch@xxxxxxxx> Acked-by: David Hildenbrand <david@xxxxxxxxxx> Cc: Pavel Tatashin <pasha.tatashin@xxxxxxxxxx> Cc: Oscar Salvador <OSalvador@xxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/sparse.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) --- a/mm/sparse.c~mm-sparse-do-not-waste-pre-allocated-memmap-space +++ a/mm/sparse.c @@ -458,8 +458,7 @@ struct page __init *__populate_section_m if (map) return map; - map = memblock_alloc_try_nid_raw(size, - PAGE_SIZE, addr, + map = memblock_alloc_try_nid_raw(size, size, addr, MEMBLOCK_ALLOC_ACCESSIBLE, nid); if (!map) panic("%s: Failed to allocate %lu bytes align=0x%lx nid=%d from=%pa\n", @@ -482,10 +481,13 @@ static void __init sparse_buffer_init(un { phys_addr_t addr = __pa(MAX_DMA_ADDRESS); WARN_ON(sparsemap_buf); /* forgot to call sparse_buffer_fini()? */ - sparsemap_buf = - memblock_alloc_try_nid_raw(size, PAGE_SIZE, - addr, - MEMBLOCK_ALLOC_ACCESSIBLE, nid); + /* + * Pre-allocated buffer is mainly used by __populate_section_memmap + * and we want it to be properly aligned to the section size - this is + * especially the case for VMEMMAP which maps memmap to PMDs + */ + sparsemap_buf = memblock_alloc_try_nid_raw(size, section_map_size(), + addr, MEMBLOCK_ALLOC_ACCESSIBLE, nid); sparsemap_buf_end = sparsemap_buf + size; } _
