From: Andi Kleen <andi@firstfloor.org>
To: cpw@sgi.com
Cc: linux-numa@vger.kernel.org, Andi Kleen <ak@linux.intel.com>
Subject: [PATCH 7/8] memhog: add option to disable transparent huge pages
Date: Wed, 14 Dec 2011 11:38:37 -0800 [thread overview]
Message-ID: <1323891518-1493-7-git-send-email-andi@firstfloor.org> (raw)
In-Reply-To: <1323891518-1493-1-git-send-email-andi@firstfloor.org>
From: Andi Kleen <ak@linux.intel.com>
Transparent Hugepages confuse the page accounting in the
test suite. Add an option to memhog (-H) to disable them.
---
memhog.c | 9 +++++++++
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/memhog.c b/memhog.c
index 2bc4251..b22f097 100755
--- a/memhog.c
+++ b/memhog.c
@@ -20,6 +20,7 @@
#include <sys/mman.h>
#include <sys/fcntl.h>
#include <string.h>
+#include <stdbool.h>
#include "numa.h"
#include "numaif.h"
#include "util.h"
@@ -37,6 +38,7 @@ void usage(void)
{
printf("memhog [-rNUM] size[kmg] [policy [nodeset]]\n");
printf("-rNUM repeat memset NUM times\n");
+ printf("-H disable transparent hugepages\n");
print_policies();
exit(1);
}
@@ -66,6 +68,7 @@ int main(int ac, char **av)
int loose = 0;
int i;
int fd = -1;
+ bool disable_hugepage = false;
nodes = numa_allocate_nodemask();
gnodes = numa_allocate_nodemask();
@@ -80,6 +83,9 @@ int main(int ac, char **av)
case 'r':
repeat = atoi(av[1] + 2);
break;
+ case 'H':
+ disable_hugepage = true;
+ break;
default:
usage();
}
@@ -113,6 +119,9 @@ int main(int ac, char **av)
if (mbind(map, length, policy, nodes->maskp, nodes->size, 0) < 0)
terr("mbind");
+
+ if (disable_hugepage)
+ madvise(map, length, MADV_NOHUGEPAGE);
gpolicy = -1;
if (get_mempolicy(&gpolicy, gnodes->maskp, gnodes->size, map, MPOL_F_ADDR) < 0)
--
1.7.5.4
next prev parent reply other threads:[~2011-12-14 19:38 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-12-14 19:38 [PATCH 1/8] Add support for a "same" numamask to numactl Andi Kleen
2011-12-14 19:38 ` [PATCH 2/8] Use constructors for numa_init/exit Andi Kleen
2011-12-14 19:38 ` [PATCH 3/8] Use glibc syscall if possible Andi Kleen
2011-12-14 19:38 ` [PATCH 4/8] Run full test suite before exiting Andi Kleen
2011-12-14 19:38 ` [PATCH 5/8] Add IO affinity support to libnuma Andi Kleen
2011-12-14 19:38 ` [PATCH 6/8] Clean regression test temp files Andi Kleen
2011-12-14 19:38 ` Andi Kleen [this message]
2011-12-14 19:38 ` [PATCH 8/8] Disable transparent huge pages in test suite Andi Kleen
2011-12-19 15:12 ` numactl/libnuma 2.0.8-rc3 Cliff Wickman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1323891518-1493-7-git-send-email-andi@firstfloor.org \
--to=andi@firstfloor.org \
--cc=ak@linux.intel.com \
--cc=cpw@sgi.com \
--cc=linux-numa@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).