[LTP] [PATCH] thp/thp03: new testcase

[LTP] [PATCH] thp/thp03: new testcase

[cxie]

From: Madper Xie <cxie@redhat.com>

Modified form a reproducer for
           https://patchwork.kernel.org/patch/1358441/
pmd_present would return the wrong value on PROT_NONE ranges or in case
of a non reproducible race with split_huge_page.

The system will crash when this test failed.

Signed-off-by: Madper Xie <cxie@redhat.com>
---
 testcases/kernel/mem/thp/thp03.c | 117 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 117 insertions(+)
 create mode 100644 testcases/kernel/mem/thp/thp03.c

diff --git a/testcases/kernel/mem/thp/thp03.c b/testcases/kernel/mem/thp/thp03.c
new file mode 100644
index 0000000..ee6fa1e
--- /dev/null
+++ b/testcases/kernel/mem/thp/thp03.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright (C) 2012  Red Hat, Inc.
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ * Further, this software is distributed without any warranty that it
+ * is free of the rightful claim of any third person regarding
+ * infringement or the like.  Any license provided herein, whether
+ * implied or otherwise, applies only to this software file.  Patent
+ * licenses, if any, provided herein do not apply to combinations of
+ * this program with other software, or any other product whatsoever.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ *
+ * thp03 - Case for spliting unaligned memory.
+ *       - System will panic if failed.
+ *
+ * There was a bug in THP, will crash happened due to the following
+ * reason according to developers:
+ *
+ * most VM places are using pmd_none but a few are still using
+ * pmd_present. The meaning is about the same for the pmd. However
+ * pmd_present would return the wrong value on PROT_NONE ranges or in
+ * case of a non reproducible race with split_huge_page.
+ * When the code using pmd_present gets a false negative, the kernel will
+ * crash. It's just an annoying DoS with a BUG_ON triggering: no memory
+ * corruption and no data corruption (nor userland nor kernel).
+ */
+
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include "mem.h"
+#include "safe_macros.h"
+#include "test.h"
+#include "usctest.h"
+
+char *TCID = "thp03";
+
+static void thp_test(void);
+
+static long hugepage_size;
+static long unaligned_size;
+static long page_size;
+
+int main(int argc, char **argv)
+{
+	int lc;
+	char *msg;
+
+	msg = parse_opts(argc, argv, NULL, NULL);
+	if (msg != NULL)
+		tst_brkm(TBROK, NULL, "OPTION PARSING ERROR - %s", msg);
+
+	setup();
+
+	for (lc = 0; TEST_LOOPING(lc); lc++) {
+		Tst_count = 0;
+
+		thp_test();
+	}
+	tst_resm(TPASS, "system didn't crash, pass.");
+	cleanup();
+	tst_exit();
+}
+
+static void thp_test(void)
+{
+	void *p;
+
+	p = mmap(NULL, unaligned_size, PROT_READ | PROT_WRITE,
+		MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
+	if (p == MAP_FAILED)
+		tst_brkm(TBROK|TERRNO, cleanup, "mmap");
+
+	memset(p, 0x00, unaligned_size);
+	if (mprotect(p, unaligned_size, PROT_NONE) == -1)
+		tst_brkm(TBROK|TERRNO, cleanup, "mprotect");
+	if (madvise(p + hugepage_size, page_size, MADV_MERGEABLE) == -1)
+		tst_brkm(TBROK|TERRNO, cleanup, "madvise");
+
+	switch (fork()) {
+	case -1:
+		tst_brkm(TBROK|TERRNO, cleanup, "fork");
+	case 0:
+		exit(0);
+	default:
+		if (waitpid(-1, NULL, 0) == -1)
+			tst_brkm(TBROK|TERRNO, cleanup, "waitpid");
+	}
+}
+
+void setup(void)
+{
+	hugepage_size = read_meminfo("Hugepagesize:") * KB;
+	unaligned_size = hugepage_size * 4 - 1;
+	page_size = SAFE_SYSCONF(NULL, _SC_PAGESIZE);
+
+	tst_sig(FORK, DEF_HANDLER, cleanup);
+	TEST_PAUSE;
+}
+
+void cleanup(void)
+{
+	TEST_CLEANUP;
+}
-- 
1.8.0


------------------------------------------------------------------------------
The Windows 8 Center - In partnership with Sourceforge
Your idea - your app - 30 days.
Get started!
http://windows8center.sourceforge.net/
what-html-developers-need-to-know-about-coding-windows-8-metro-style-apps/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] [PATCH] thp/thp03: new testcase

[Caspar Zhang]

On 10/29/2012 05:29 PM, cxie@redhat.com wrote:
> From: Madper Xie <cxie@redhat.com>
>
> Modified form a reproducer for
>             https://patchwork.kernel.org/patch/1358441/
> pmd_present would return the wrong value on PROT_NONE ranges or in case
> of a non reproducible race with split_huge_page.
>
> The system will crash when this test failed.
>
> Signed-off-by: Madper Xie <cxie@redhat.com>

Please also attach the runtest entry.

Reviewed-by: Caspar Zhang <caspar@casparzhang.com>

> ---
>   testcases/kernel/mem/thp/thp03.c | 117 +++++++++++++++++++++++++++++++++++++++
>   1 file changed, 117 insertions(+)
>   create mode 100644 testcases/kernel/mem/thp/thp03.c
>
> diff --git a/testcases/kernel/mem/thp/thp03.c b/testcases/kernel/mem/thp/thp03.c
> new file mode 100644
> index 0000000..ee6fa1e
> --- /dev/null
> +++ b/testcases/kernel/mem/thp/thp03.c
> @@ -0,0 +1,117 @@
> +/*
> + * Copyright (C) 2012  Red Hat, Inc.
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of version 2 of the GNU General Public
> + * License as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope that it would be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
> + *
> + * Further, this software is distributed without any warranty that it
> + * is free of the rightful claim of any third person regarding
> + * infringement or the like.  Any license provided herein, whether
> + * implied or otherwise, applies only to this software file.  Patent
> + * licenses, if any, provided herein do not apply to combinations of
> + * this program with other software, or any other product whatsoever.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
> + * 02110-1301, USA.
> + *
> + * thp03 - Case for spliting unaligned memory.
> + *       - System will panic if failed.
> + *
> + * There was a bug in THP, will crash happened due to the following
> + * reason according to developers:
> + *
> + * most VM places are using pmd_none but a few are still using
> + * pmd_present. The meaning is about the same for the pmd. However
> + * pmd_present would return the wrong value on PROT_NONE ranges or in
> + * case of a non reproducible race with split_huge_page.
> + * When the code using pmd_present gets a false negative, the kernel will
> + * crash. It's just an annoying DoS with a BUG_ON triggering: no memory
> + * corruption and no data corruption (nor userland nor kernel).
> + */
> +
> +#include <sys/mman.h>
> +#include <sys/types.h>
> +#include <sys/wait.h>
> +#include <fcntl.h>
> +#include <stdlib.h>
> +#include <string.h>
> +#include "mem.h"
> +#include "safe_macros.h"
> +#include "test.h"
> +#include "usctest.h"
> +
> +char *TCID = "thp03";
> +
> +static void thp_test(void);
> +
> +static long hugepage_size;
> +static long unaligned_size;
> +static long page_size;
> +
> +int main(int argc, char **argv)
> +{
> +	int lc;
> +	char *msg;
> +
> +	msg = parse_opts(argc, argv, NULL, NULL);
> +	if (msg != NULL)
> +		tst_brkm(TBROK, NULL, "OPTION PARSING ERROR - %s", msg);
> +
> +	setup();
> +
> +	for (lc = 0; TEST_LOOPING(lc); lc++) {
> +		Tst_count = 0;
> +
> +		thp_test();
> +	}
> +	tst_resm(TPASS, "system didn't crash, pass.");
> +	cleanup();
> +	tst_exit();
> +}
> +
> +static void thp_test(void)
> +{
> +	void *p;
> +
> +	p = mmap(NULL, unaligned_size, PROT_READ | PROT_WRITE,
> +		MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
> +	if (p == MAP_FAILED)
> +		tst_brkm(TBROK|TERRNO, cleanup, "mmap");
> +
> +	memset(p, 0x00, unaligned_size);
> +	if (mprotect(p, unaligned_size, PROT_NONE) == -1)
> +		tst_brkm(TBROK|TERRNO, cleanup, "mprotect");
> +	if (madvise(p + hugepage_size, page_size, MADV_MERGEABLE) == -1)
> +		tst_brkm(TBROK|TERRNO, cleanup, "madvise");
> +
> +	switch (fork()) {
> +	case -1:
> +		tst_brkm(TBROK|TERRNO, cleanup, "fork");
> +	case 0:
> +		exit(0);
> +	default:
> +		if (waitpid(-1, NULL, 0) == -1)
> +			tst_brkm(TBROK|TERRNO, cleanup, "waitpid");
> +	}
> +}
> +
> +void setup(void)
> +{
> +	hugepage_size = read_meminfo("Hugepagesize:") * KB;
> +	unaligned_size = hugepage_size * 4 - 1;
> +	page_size = SAFE_SYSCONF(NULL, _SC_PAGESIZE);
> +
> +	tst_sig(FORK, DEF_HANDLER, cleanup);
> +	TEST_PAUSE;
> +}
> +
> +void cleanup(void)
> +{
> +	TEST_CLEANUP;
> +}
>


------------------------------------------------------------------------------
The Windows 8 Center - In partnership with Sourceforge
Your idea - your app - 30 days.
Get started!
http://windows8center.sourceforge.net/
what-html-developers-need-to-know-about-coding-windows-8-metro-style-apps/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] [PATCH] thp/thp03: new testcase

[chrubis]

Hi!
> Modified form a reproducer for
>            https://patchwork.kernel.org/patch/1358441/

I would add this link as well as kernel commit hash to the
comment with testcase description too. Just in case.

> +char *TCID = "thp03";

It's missing:

int TST_TOTAL = 1;

So that test framework knows the number of executed tests.

(I'm not sure if it's currently used, but all the other
 tests defined this)

> +static void thp_test(void);
> +
> +static long hugepage_size;
> +static long unaligned_size;
> +static long page_size;

The rests seems okay.

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
The Windows 8 Center - In partnership with Sourceforge
Your idea - your app - 30 days.
Get started!
http://windows8center.sourceforge.net/
what-html-developers-need-to-know-about-coding-windows-8-metro-style-apps/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list