lkml.org 
[lkml]   [2018]   [Sep]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Subject[PATCH v5 0/4] Address issues slowing persistent memory initialization
From
Date
This patch set is meant to be a v5 of my earlier submission with the same
title[1].

The main changes from the previous version are that I have added a new
patch to address an issue that had disabled deferred memory init on my
system due to recent config changes related to CONFIG_NO_BOOTMEM. In
addition I dropped the original patches 4 and 5 from the previous set as
that is going to need to be a separate set of patches.

The main thing this patch set achieves is that it allows us to initialize
each node worth of persistent memory independently. As a result we reduce
page init time by about 2 minutes because instead of taking 30 to 40 seconds
per node and going through each node one at a time, we process all 4 nodes
in parallel in the case of a 12TB persistent memory setup spread evenly over
4 nodes.

[1]: https://lkml.org/lkml/2018/9/21/4
---

Alexander Duyck (4):
mm: Remove now defunct NO_BOOTMEM from depends list for deferred init
mm: Provide kernel parameter to allow disabling page init poisoning
mm: Create non-atomic version of SetPageReserved for init use
mm: Defer ZONE_DEVICE page initialization to the point where we init pgmap


Documentation/admin-guide/kernel-parameters.txt | 12 +++
arch/csky/Kconfig | 1
include/linux/mm.h | 2
include/linux/page-flags.h | 9 ++
kernel/memremap.c | 24 ++---
mm/Kconfig | 1
mm/debug.c | 46 ++++++++++
mm/hmm.c | 12 ++-
mm/memblock.c | 5 -
mm/page_alloc.c | 101 ++++++++++++++++++++++-
mm/sparse.c | 4 -
11 files changed, 184 insertions(+), 33 deletions(-)

--

\
 
 \ /
  Last update: 2018-09-25 22:19    [W:0.283 / U:72.780 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site