public inbox for ltp@lists.linux.it
 help / color / mirror / Atom feed
From: Avinesh Kumar <akumar@suse.de>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH v2] mmap04.c: Avoid vma merging
Date: Wed, 24 Jan 2024 14:26:02 +0100	[thread overview]
Message-ID: <20240124132603.16199-1-akumar@suse.de> (raw)
In-Reply-To: <325e7294-f6b1-4e27-a51b-8a8e146bf5bc@suse.cz>

We hit a scenario where new mapping was merged with existing mapping of
same permission and the return address from the mmap was hidden in the
merged mapping in /proc/self/maps, causing the test to fail.
To avoid this, we first  create a 2-page mapping with the different
permissions, and then remap the 2nd page with the perms being tested.

Signed-off-by: Avinesh Kumar <akumar@suse.de>
Reported-by: Martin Doucha <mdoucha@suse.cz>
Signed-off-by: Avinesh Kumar <akumar@suse.de>
---
 testcases/kernel/syscalls/mmap/mmap04.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/testcases/kernel/syscalls/mmap/mmap04.c b/testcases/kernel/syscalls/mmap/mmap04.c
index f6f4f7c98..fa85deed1 100644
--- a/testcases/kernel/syscalls/mmap/mmap04.c
+++ b/testcases/kernel/syscalls/mmap/mmap04.c
@@ -17,8 +17,8 @@
 #include "tst_test.h"
 #include <stdio.h>
 
-#define MMAPSIZE 1024
-static char *addr;
+static char *addr1;
+static char *addr2;
 
 static struct tcase {
 	int prot;
@@ -44,14 +44,23 @@ static struct tcase {
 static void run(unsigned int i)
 {
 	struct tcase *tc = &tcases[i];
-	char addr_str[20];
 	char perms[8];
 	char fmt[1024];
+	unsigned int pagesize;
+	int flag;
 
-	addr = SAFE_MMAP(NULL, MMAPSIZE, tc->prot, tc->flags, -1, 0);
+	pagesize = SAFE_SYSCONF(_SC_PAGESIZE);
 
-	sprintf(addr_str, "%" PRIxPTR, (uintptr_t)addr);
-	sprintf(fmt, "%s-%%*x %%s", addr_str);
+	/* To avoid new mapping getting merged with existing mappings, we first
+	 * create a 2-page mapping with the different permissions, and then remap
+	 * the 2nd page with the perms being tested.
+	 */
+	flag = (tc->flags & MAP_PRIVATE) ? MAP_SHARED : MAP_PRIVATE;
+	addr1 = SAFE_MMAP(NULL, pagesize * 2, PROT_NONE, MAP_ANONYMOUS | flag, -1, 0);
+
+	addr2 = SAFE_MMAP(addr1 + pagesize, pagesize, tc->prot, tc->flags | MAP_FIXED, -1, 0);
+
+	sprintf(fmt, "%" PRIxPTR "-%%*x %%s", (uintptr_t)addr2);
 	SAFE_FILE_LINES_SCANF("/proc/self/maps", fmt, perms);
 
 	if (!strcmp(perms, tc->exp_perms)) {
@@ -61,7 +70,7 @@ static void run(unsigned int i)
 						tc->exp_perms, perms);
 	}
 
-	SAFE_MUNMAP(addr, MMAPSIZE);
+	SAFE_MUNMAP(addr1, pagesize * 2);
 }
 
 static struct tst_test test = {
-- 
2.43.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

  reply	other threads:[~2024-01-24 13:26 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-23 16:55 [LTP] [PATCH] mmap04.c: Avoid vma merging Avinesh Kumar
2024-01-24 11:56 ` Martin Doucha
2024-01-24 13:26   ` Avinesh Kumar [this message]
2024-01-24 16:23     ` [LTP] [PATCH v2] " Martin Doucha
2024-01-24 17:05       ` Petr Vorel
2024-01-25  8:14         ` Avinesh Kumar
2024-01-25  8:25           ` Petr Vorel
2024-01-24 14:36   ` [LTP] [PATCH] " Avinesh Kumar

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=20240124132603.16199-1-akumar@suse.de \
    --to=akumar@suse.de \
    --cc=ltp@lists.linux.it \
    /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