public inbox for ltp@lists.linux.it
 help / color / mirror / Atom feed
* [LTP] [PATCH v2] mbind01, get_mempolicy01: dont use hardcoded node 0
@ 2012-05-24 14:44 Jan Stancek
  2012-05-25  0:41 ` Wanlong Gao
  0 siblings, 1 reply; 2+ messages in thread
From: Jan Stancek @ 2012-05-24 14:44 UTC (permalink / raw)
  To: ltp-list; +Cc: Jeffrey Burke

[-- Attachment #1: Type: text/plain, Size: 887 bytes --]


These tests are failing on following setup:
  # numactl -H
  available: 2 nodes (0-1)
  node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
  23 24 25 26 27 28 29 30 31
  node 0 size: 0 MB
  node 0 free: 0 MB
  node 1 cpus:
  node 1 size: 12288 MB
  node 1 free: 9998 MB
  node distances:
  node   0   1
    0:  10  40
    1:  40  10

Avoid hardcoded node 0 by picking first available node obtained using
get_mempolicy(..., MPOL_F_MEMS_ALLOWED);

Using whole nodemask doesn't work with MPOL_PREFERRED, because set_mempolicy
will use only first node and subsequent check for bitmask equality may fail.

Signed-off-by: Jan Stancek <jstancek@redhat.com>
---
 .../syscalls/get_mempolicy/get_mempolicy01.c       |   19 +++++++++++++++++--
 testcases/kernel/syscalls/mbind/mbind01.c          |   19 +++++++++++++++++--
 2 files changed, 34 insertions(+), 4 deletions(-)



[-- Attachment #2: 0001-mbind01-get_mempolicy01-dont-use-hardcoded-node-0.patch --]
[-- Type: text/x-patch, Size: 2796 bytes --]

diff --git a/testcases/kernel/syscalls/get_mempolicy/get_mempolicy01.c b/testcases/kernel/syscalls/get_mempolicy/get_mempolicy01.c
index 0565d6f..8069df1 100644
--- a/testcases/kernel/syscalls/get_mempolicy/get_mempolicy01.c
+++ b/testcases/kernel/syscalls/get_mempolicy/get_mempolicy01.c
@@ -231,15 +231,30 @@ static int do_test(struct test_case *tc)
 #endif
 	char *p = NULL;
 	unsigned long len = MEM_LENGTH;
+	struct bitmask *allowed = NULL;
+	int i, test_node = -1;
+
+	allowed = numa_allocate_nodemask();
+	// avoid numa_get_mems_allowed(), because of bug in getpol() utility function
+	// in older versions: http://www.spinics.net/lists/linux-numa/msg00849.html
+	TEST(syscall(__NR_get_mempolicy, NULL, allowed->maskp,
+		allowed->size, 0, MPOL_F_MEMS_ALLOWED));
+	if (TEST_RETURN < 0)
+		tst_resm(TFAIL|TERRNO, "get_mempolicy: MPOL_F_MEMS_ALLOWED");
+	for (i = 0; i < allowed->size; i++)
+		if (numa_bitmask_isbitset(allowed, i)) {
+			test_node = i;
+			break;
+		}
 
 #if !defined(LIBNUMA_API_VERSION) || LIBNUMA_API_VERSION < 2
 	nodemask = malloc(sizeof(nodemask_t));
 	nodemask_zero(nodemask);
-	nodemask_set(nodemask, 0);
+	nodemask_set(nodemask, test_node);
 	getnodemask = malloc(sizeof(nodemask_t));
 	nodemask_zero(getnodemask);
 #else
-	numa_bitmask_setbit(nodemask, 0);
+	numa_bitmask_setbit(nodemask, test_node);
 #endif
 	switch (tc->ttype) {
 	case DEFAULT:
diff --git a/testcases/kernel/syscalls/mbind/mbind01.c b/testcases/kernel/syscalls/mbind/mbind01.c
index 4b25db0..c3170ec 100644
--- a/testcases/kernel/syscalls/mbind/mbind01.c
+++ b/testcases/kernel/syscalls/mbind/mbind01.c
@@ -217,15 +217,30 @@ static int do_test(struct test_case *tc)
 	unsigned long maxnode = NUMA_NUM_NODES;
 	unsigned long len = MEM_LENGTH;
 	unsigned long *invalid_nodemask;
+	struct bitmask *allowed = NULL;
+	int i, test_node = -1;
+
+	allowed = numa_allocate_nodemask();
+	// avoid numa_get_mems_allowed(), because of bug in getpol() utility function
+	// in older versions: http://www.spinics.net/lists/linux-numa/msg00849.html
+	TEST(syscall(__NR_get_mempolicy, NULL, allowed->maskp,
+		allowed->size, 0, MPOL_F_MEMS_ALLOWED));
+	if (TEST_RETURN < 0)
+		tst_resm(TFAIL|TERRNO, "get_mempolicy: MPOL_F_MEMS_ALLOWED");
+	for (i = 0; i < allowed->size; i++)
+		if (numa_bitmask_isbitset(allowed, i)) {
+			test_node = i;
+			break;
+		}
 
 #if !defined(LIBNUMA_API_VERSION) || LIBNUMA_API_VERSION < 2
 	nodemask = malloc(sizeof(nodemask_t));
 	nodemask_zero(nodemask);
-	nodemask_set(nodemask, 0);
+	nodemask_set(nodemask, test_node);
 	getnodemask = malloc(sizeof(nodemask_t));
 	nodemask_zero(getnodemask);
 #else
-	numa_bitmask_setbit(nodemask, 0);
+	numa_bitmask_setbit(nodemask, test_node);
 #endif
 	p = mmap(NULL, len, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
 		    0, 0);


[-- Attachment #3: Type: text/plain, Size: 395 bytes --]

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

[-- Attachment #4: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

^ permalink raw reply related	[flat|nested] 2+ messages in thread
end of thread, other threads:[~2012-05-25  0:42 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-05-24 14:44 [LTP] [PATCH v2] mbind01, get_mempolicy01: dont use hardcoded node 0 Jan Stancek
2012-05-25  0:41 ` Wanlong Gao

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox