linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] x86/mm: use max memory block size with unaligned memory end
@ 2020-06-04  3:54 Daniel Jordan
  2020-06-04  7:22 ` David Hildenbrand
  0 siblings, 1 reply; 10+ messages in thread
From: Daniel Jordan @ 2020-06-04  3:54 UTC (permalink / raw)
  To: linux-mm, linux-kernel
  Cc: Andrew Morton, Andy Lutomirski, Dave Hansen, David Hildenbrand,
	Michal Hocko, Pavel Tatashin, Peter Zijlstra, Steven Sistare,
	Daniel Jordan

Some of our servers spend 14 out of the 21 seconds of kernel boot
initializing memory block sysfs directories and then creating symlinks
between them and the corresponding nodes.  The slowness happens because
the machines get stuck with the smallest supported memory block size on
x86 (128M), which results in 16,288 directories to cover the 2T of
installed RAM, and each of these paths does a linear search of the
memory blocks for every block id, with atomic ops at each step.

Commit 078eb6aa50dc ("x86/mm/memory_hotplug: determine block size based
on the end of boot memory") chooses the block size based on alignment
with memory end.  That addresses hotplug failures in qemu guests, but
for bare metal systems whose memory end isn't aligned to the smallest
size, it leaves them at 128M.

For such systems, use the largest supported size (2G) to minimize
overhead on big machines.  That saves nearly all of the 14 seconds so
the kernel boots 3x faster.

There are some simple ways to avoid the linear searches, but for now it
makes no difference with a 2G block.

Signed-off-by: Daniel Jordan <daniel.m.jordan@oracle.com>
---
 arch/x86/mm/init_64.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 8b5f73f5e207c..d388127d1b519 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -1390,6 +1390,15 @@ static unsigned long probe_memory_block_size(void)
 		goto done;
 	}
 
+	/*
+	 * Memory end isn't aligned to any allowed block size, so default to
+	 * the largest to minimize overhead on large memory systems.
+	 */
+	if (!IS_ALIGNED(boot_mem_end, MIN_MEMORY_BLOCK_SIZE)) {
+		bz = MAX_BLOCK_SIZE;
+		goto done;
+	}
+
 	/* Find the largest allowed block size that aligns to memory end */
 	for (bz = MAX_BLOCK_SIZE; bz > MIN_MEMORY_BLOCK_SIZE; bz >>= 1) {
 		if (IS_ALIGNED(boot_mem_end, bz))

base-commit: 3d77e6a8804abcc0504c904bd6e5cdf3a5cf8162
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2020-06-05  7:45 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-06-04  3:54 [PATCH] x86/mm: use max memory block size with unaligned memory end Daniel Jordan
2020-06-04  7:22 ` David Hildenbrand
2020-06-04 17:22   ` Daniel Jordan
2020-06-04 17:45     ` David Hildenbrand
2020-06-04 18:12       ` Daniel Jordan
2020-06-04 18:55         ` David Hildenbrand
2020-06-04 22:24           ` Daniel Jordan
2020-06-04 20:00         ` Dave Hansen
2020-06-04 22:27           ` Daniel Jordan
2020-06-05  7:44           ` David Hildenbrand

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).