lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20160504143628.GU2858@techsingularity.net>
Date:	Wed, 4 May 2016 15:36:28 +0100
From:	Mel Gorman <mgorman@...hsingularity.net>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	Vlastimil Babka <vbabka@...e.cz>,
	Jesper Dangaard Brouer <brouer@...hat.com>,
	Linux-MM <linux-mm@...ck.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: [PATCH] mm, page_alloc: Restore the original nodemask if the fast
 path allocation failed

(Andrew, this is on top of the pagealloc optimisation series in mmotm. It
 could be classed as a fix-up but it's subtle enough that it deserves its
 own changelog. "Clever" fixes had other consequences.)

The page allocator fast path uses either the requested nodemask or
cpuset_current_mems_allowed if cpusets are enabled. If the allocation
context allows watermarks to be ignored then it can also ignore memory
policies. However, on entering the allocator slowpath the nodemask may
still be cpuset_current_mems_allowed and the policies are enforced.
This patch resets the nodemask appropriately before entering the slowpath.

Signed-off-by: Vlastimil Babka <vbabka@...e.cz>
Signed-off-by: Mel Gorman <mgorman@...hsingularity.net>

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 79100583b9de..ec5155ab1482 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3637,6 +3637,12 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
 	alloc_mask = memalloc_noio_flags(gfp_mask);
 	ac.spread_dirty_pages = false;
 
+	/*
+	 * Restore the original nodemask if it was potentially replaced with
+	 * &cpuset_current_mems_allowed to optimize the fast-path attempt.
+	 */
+	if (cpusets_enabled())
+		ac.nodemask = nodemask;
 	page = __alloc_pages_slowpath(alloc_mask, order, &ac);
 
 no_zone:

-- 
Mel Gorman
SUSE Labs

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ