Commit 560a036b authored Jan 10, 2012 by Johannes Weiner Committed by Linus Torvalds Jan 10, 2012

mm: bootmem: drop superfluous range check when freeing pages in bulk



The area node_bootmem_map represents is aligned to BITS_PER_LONG, and all
bits in any aligned word of that map valid.  When the represented area
extends beyond the end of the node, the non-existant pages will be marked
as reserved.

As a result, when freeing a page block, doing an explicit range check for
whether that block is within the node's range is redundant as the bitmap
is consulted anyway to see whether all pages in the block are unreserved.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

parent c3993076

mm/bootmem.c

+1 −1

Original line number	Diff line number	Diff line
		@@ -197,7 +197,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
		idx = start - bdata->node_min_pfn;
		vec = ~map[idx / BITS_PER_LONG];

		if (aligned && vec == ~0UL && start + BITS_PER_LONG <= end) {
		if (aligned && vec == ~0UL) {
		int order = ilog2(BITS_PER_LONG);

		__free_pages_bootmem(pfn_to_page(start), order);

Admin message