linux-kernel - [x86] Make split_page_count check for empty levels.

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>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160111170428.GA8547@codemonkey.org.uk>
Date:	Mon, 11 Jan 2016 12:04:28 -0500
From:	Dave Jones <davej@...emonkey.org.uk>
To:	x86@...nel.org
Cc:	Linux Kernel <linux-kernel@...r.kernel.org>
Subject: [x86] Make split_page_count check for empty levels.

In PAGEALLOC_DEBUG builds, we disable the 2M pages.
Unfortunatly when we split up mappings during boot,
split_page_count doesn't take this into account, and
starts decrementing an empty direct_pages_count level.

This results in /proc/meminfo showing crazy things like
DirectMap2M:    18446744073709543424 kB

Signed-off-by: Dave Jones <davej@...emonkey.org.uk>

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index a3137a4feed1..ff0e0c6c350e 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -66,6 +66,9 @@ void update_page_count(int level, unsigned long pages)

 static void split_page_count(int level)
 {
+	if (direct_pages_count[level] == 0)
+		return;
+
 	direct_pages_count[level]--;
 	direct_pages_count[level - 1] += PTRS_PER_PTE;
 }