KSM: full_scans will stop increasing after stopping ksm

Quoted 6 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",

so group_check will be hanged on "new_num < old_num * 3".

So, i think we should break the test if "run" is not expecting.

Quoted 3 lines

Quoted 2 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop, i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up, because new_num does not be increased.

Regards
Bian

Quoted 96 lines

CAI Qian

Quoted 95 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +-
testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2, int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
-        int pages_to_scan);
+        int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);
 void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +9 -8

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
-        tst_resm(TFAIL, "%s is not %ld.", path, value);
+        tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2, int end2)

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start, int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
-        int sleep_millisecs, int pages_to_scan)
+        int sleep_millisecs, int pages_to_scan,
+        int scans)
 {
   int fd;
   char buf[BUFSIZ];

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
-    while (new_num < old_num * 3) {
+    while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)
           tst_brkm(TBROK|TERRNO, cleanup, "read");

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
-    group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 * num);
+    group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 * num, 3);
 
   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
-    group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 * num);
+    group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 * num, 3);
 
   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
-    group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 * num);
+    group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 * num, 3);
 
   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
-    group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 * num);
+    group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 * num, 3);
 
   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
-    group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+    group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 16 lines

Quoted 15 lines

testcases/kernel/mem/lib/mem.c +1 -1

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
-    group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+    group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)
           tst_resm(TFAIL, "child exit status is %d",

Signature · 1 line

1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 11 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

So, i think we should break the test if "run" is not expecting.

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 115 lines

CAI Qian

Quoted 114 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);

void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +1 -1

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int
 pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)

           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +1 -1

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int
 unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 18 lines

Quoted 11 lines

testcases/kernel/mem/lib/mem.c +0 -0

   tst_resm(TINFO, "wait for all children to stop.");

   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int
 unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)

Quoted 5 lines

tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

----- Original Message -----

Quoted 17 lines

Quoted 10 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There are some patches for ksm recently merged upstream. If the problem still persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is hung, please upload sysrq-t output somewhere.

CAI Qian

Quoted 166 lines

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 115 lines

CAI Qian

Quoted 114 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);

void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +1 -1

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int
 pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)

           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +1 -1

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int
 unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 17 lines

Quoted 11 lines

testcases/kernel/mem/lib/mem.c +0 -0

   tst_resm(TINFO, "wait for all children to stop.");

   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int
 unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)

Quoted 5 lines

tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 22 lines · CAI Qian

----- Original Message -----

Quoted 20 lines

Quoted 4 lines · Qian Cai

Quoted 2 lines · Bian Naimeng

There are some problem in ksm tests.

Quoted 8 lines

Quoted 1 line

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent
the rest of the tests from running, so there is no need to stop the
tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2

Quoted 1 line

/sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 9 lines

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There
are some patches for ksm recently merged upstream. If the problem still
persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is
hung, please upload sysrq-t output somewhere.

Maybe there are some bugs in the RHEL6's kernel, but the purpose of this patch
is not to workround these bugs, i want to fix the test's bug.

Would you explain to me why we do this loop "while (new_num < old_num
* 3)" in
group_check, i think "while (new_num < old_num + 3)" is better.

Some time ago, the following patch insert this loop.
http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3Dg%3DojJu0m%2B556rnekYenRTXtX%2BVBOj%3DrPmnjSw%40mail.gmail.com&forum_name=ltp-list

The changelog of this patch said "we should wait 3~5 increments of the /sys/kernel/mm/ksm/full_scans before checking ksm* testcases's results", but
it do "while (new_num < old_num * 3)" actually.

Regards
Bian

Quoted 181 lines

CAI Qian

Quoted 177 lines

Quoted 5 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this
loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the
process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 131 lines

CAI Qian

Quoted 130 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int
       sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);

void check_ksm_options(int *size, int *num, int *unit); void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +0 -0

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared,

int
pages_sharing,

   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)
           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +0 -0

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num,

int
unit)

       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);

}
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 * num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 * num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +0 -0

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num,

int
unit)

   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");

- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 * num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 * num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +0 -0

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num,

int
unit)

       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);

}
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 * num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 * num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +0 -0

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num,

int
unit)

   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");

- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 * num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 * num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +0 -0

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num,

int
unit)

       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");

- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +0 -0

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num,

int
unit)
if (write(fd, "0", 1) != 1)

       tst_brkm(TBROK|TERRNO, cleanup, "write");

close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);

Quoted 9 lines

Quoted 8 lines

+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0); while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0) if (WEXITSTATUS(status) != 0) tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 19 lines · CAI Qian

----- Original Message -----

Quoted 17 lines

Quoted 10 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 6 lines

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There are some patches for ksm recently merged upstream. If the problem still persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is hung, please upload sysrq-t output somewhere.

Maybe there are some bugs in the RHEL6's kernel, but the purpose of this patch is not to workround these bugs, i want to fix the test's bug.

Would you explain to me why we do this loop "while (new_num < old_num * 3)" in group_check, i think "while (new_num < old_num + 3)" is better.

Some time ago, the following patch insert this loop.
http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3Dg%3DojJu0m%2B556rnekYenRTXtX%2BVBOj%3DrPmnjSw%40mail.gmail.com&forum_name=ltp-list

The changelog of this patch said "we should wait 3~5 increments of the /sys/kernel/mm/ksm/full_scans before checking ksm* testcases's results", but it do "while (new_num < old_num * 3)" actually.

Regards
Bian

Quoted 170 lines

CAI Qian

Quoted 166 lines

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 115 lines

CAI Qian

Quoted 114 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);

void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +1 -1

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int
 pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)

           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +1 -1

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int
 unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 17 lines

Quoted 11 lines

testcases/kernel/mem/lib/mem.c +0 -0

   tst_resm(TINFO, "wait for all children to stop.");

   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int
 unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)

Quoted 5 lines

tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 35 lines · Bian Naimeng

Quoted 18 lines · CAI Qian

----- Original Message -----

Quoted 17 lines

Quoted 10 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 6 lines

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There are some patches for ksm recently merged upstream. If the problem still persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is hung, please upload sysrq-t output somewhere.

Maybe there are some bugs in the RHEL6's kernel, but the purpose of this patch is not to workround these bugs, i want to fix the test's bug.

Would you explain to me why we do this loop "while (new_num < old_num * 3)" in group_check, i think "while (new_num < old_num + 3)" is better.

Some time ago, the following patch insert this loop.
http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3Dg%3DojJu0m%2B556rnekYenRTXtX%2BVBOj%3DrPmnjSw%40mail.gmail.com&forum_name=ltp-list

Quoted 3 lines

The changelog of this patch said "we should wait 3~5 increments of the /sys/kernel/mm/ksm/full_scans before checking ksm* testcases's results", but it do "while (new_num < old_num * 3)" actually.

I made a mistake. The code is what I wanted to do, but the changelog was wrong. When testing the new ksm patch, the developer told us we must wait 3~5 times increments of the number before checking testing results. So I coded to wait til new_num >= old_num * 3 before checking the results.

About 'echo 2 > /sys/kernel/mm/ksm/run' problem, I have tested it with ksm01.
If I run the 'echo 2 > /sys/kernel/mm/ksm/run' before issue ksm test, the content of /sys/kernel/mm/ksm/run will be changed to 1 and the test can finished successfully.

Only if I echo the 2 between the testing process,
ksm01 will hang up. On that time, new_num will be zero, so your plus 3 method won't work either. So what should we do in this circumstance?

Thanks.

Han Pingtian

Quoted 181 lines

Regards
Bian

Quoted 167 lines

CAI Qian

Quoted 164 lines

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 115 lines

CAI Qian

Quoted 114 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);

void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +1 -1

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int
 pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)

           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +1 -1

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int
 unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 16 lines

Quoted 11 lines

testcases/kernel/mem/lib/mem.c +0 -0

   tst_resm(TINFO, "wait for all children to stop.");

   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int
 unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)

Quoted 5 lines

tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 39 lines · Han Pingtian

Quoted 38 lines · Bian Naimeng

Quoted 18 lines · CAI Qian

----- Original Message -----

Quoted 17 lines

Quoted 10 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 6 lines

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There are some patches for ksm recently merged upstream. If the problem still persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is hung, please upload sysrq-t output somewhere.

Maybe there are some bugs in the RHEL6's kernel, but the purpose of this patch is not to workround these bugs, i want to fix the test's bug.

Would you explain to me why we do this loop "while (new_num < old_num * 3)" in group_check, i think "while (new_num < old_num + 3)" is better.

Some time ago, the following patch insert this loop.
http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3Dg%3DojJu0m%2B556rnekYenRTXtX%2BVBOj%3DrPmnjSw%40mail.gmail.com&forum_name=ltp-list

The changelog of this patch said "we should wait 3~5 increments of the /sys/kernel/mm/ksm/full_scans before checking ksm* testcases's results", but it do "while (new_num < old_num * 3)" actually.

Quoted 6 lines

I made a mistake. The code is what I wanted to do, but the changelog was wrong. When testing the new ksm patch, the developer told us we must wait 3~5 times increments of the number before checking testing results. So I coded to wait til new_num >= old_num * 3 before checking the results.

The bigger old_new is, the longer time test takes, it's strange.

Quoted 6 lines

About 'echo 2 > /sys/kernel/mm/ksm/run' problem, I have tested it with ksm01.
If I run the 'echo 2 > /sys/kernel/mm/ksm/run' before issue ksm test, the content of /sys/kernel/mm/ksm/run will be changed to 1 and the test can finished successfully.

I think we shoud not care this.

Quoted 5 lines

Only if I echo the 2 between the testing process,
ksm01 will hang up. On that time, new_num will be zero, so your plus 3 method won't work either. So what should we do in this circumstance?

Please look at my patch, after stopping ksmd in the testing(echo 2 > /sys/kernel/mm/ksm/run or echo 0 > /sys/kernel/mm/ksm/run), group_check will skip waiting at the loop "new_num >= old_num * 3".

Regards
Bian

Quoted 185 lines

Thanks.

Han Pingtian

Quoted 181 lines

Regards
Bian

Quoted 167 lines

CAI Qian

Quoted 164 lines

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian

Quoted 115 lines

CAI Qian

Quoted 114 lines

Signed-off-by Bian Naimeng <[email protected]>

---
testcases/kernel/mem/include/mem.h | 2 +- testcases/kernel/mem/lib/mem.c | 19 ++++++++++---------

2 files changed, 11 insertions(+), 10 deletions(-)

testcases/kernel/mem/include/mem.h +1 -1

index 778d403..b640a63 100644
@@ -42,7 +42,7 @@ void check(char *path, long int value);
 void verify(char value, int proc, int start, int end, int start2,
 int end2);
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared, int sleep_millisecs,
- int pages_to_scan);
+ int pages_to_scan, int scans);
 void create_same_memory(int size, int num, int unit);
 void check_ksm_options(int *size, int *num, int *unit);

void write_cpusets(void);

testcases/kernel/mem/lib/mem.c +1 -1

index 12e61e9..db1a7dd 100644
@@ -284,7 +284,7 @@ void check(char *path, long int value)
 
   tst_resm(TINFO, "%s is %ld.", path, atol(buf));
   if (atol(buf) != value)
- tst_resm(TFAIL, "%s is not %ld.", path, value);
+ tst_brkm(TFAIL, tst_exit, "%s is not %ld.", path, value);
 }
 
 void verify(char value, int proc, int start, int end, int start2,

int end2)

testcases/kernel/mem/lib/mem.c +2 -1

@@ -312,7 +312,8 @@ void verify(char value, int proc, int start,
 int end, int start2, int end2)
 
 void group_check(int run, int pages_shared, int pages_sharing,
       int pages_volatile, int pages_unshared,
- int sleep_millisecs, int pages_to_scan)
+ int sleep_millisecs, int pages_to_scan,
+ int scans)
 {
   int fd;

   char buf[BUFSIZ];

testcases/kernel/mem/lib/mem.c +1 -1

@@ -332,7 +333,7 @@ void group_check(int run, int pages_shared, int
 pages_sharing,
   old_num = new_num = atoi(buf);
   if (lseek(fd, 0, SEEK_SET) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "lseek");
- while (new_num < old_num * 3) {
+ while (new_num < old_num + scans) {
       sleep(1);
       if (read(fd, buf, BUFSIZ) < 0)

           tst_brkm(TBROK|TERRNO, cleanup, "read");

testcases/kernel/mem/lib/mem.c +1 -1

@@ -587,7 +588,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 2, size * num * 256 - 2, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -599,7 +600,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 3, size * num * 256 - 3, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -613,7 +614,7 @@ void create_same_memory(int size, int num, int
 unit)
       if (kill(child[k], SIGCONT) == -1)
           tst_brkm(TBROK|TERRNO, cleanup, "kill child[%d]", k);
   }
- group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 1, 0, 0, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for all children to stop.");
   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -627,7 +628,7 @@ void create_same_memory(int size, int num, int
 unit)
   tst_resm(TINFO, "resume child 1.");
   if (kill(child[1], SIGCONT) == -1)
       tst_brkm(TBROK|TERRNO, cleanup, "kill");
- group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num);
+ group_check(1, 1, size * num * 256 - 2, 0, 1, 0, size * 256 *
 num, 3);

   tst_resm(TINFO, "wait for child 1 to stop.");
   if (waitpid(child[1], &status, WUNTRACED) == -1)

testcases/kernel/mem/lib/mem.c +1 -1

@@ -647,7 +648,7 @@ void create_same_memory(int size, int num, int
 unit)
       tst_brkm(TBROK|TERRNO, cleanup, "open");
   if (write(fd, "2", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
- group_check(2, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(2, 0, 0, 0, 0, 0, size * 256 * num, 0);
 

Quoted 16 lines

Quoted 11 lines

testcases/kernel/mem/lib/mem.c +0 -0

   tst_resm(TINFO, "wait for all children to stop.");

   for (k = 0; k < num; k++) {

testcases/kernel/mem/lib/mem.c +1 -1

@@ -668,7 +669,7 @@ void create_same_memory(int size, int num, int
 unit)
   if (write(fd, "0", 1) != 1)
       tst_brkm(TBROK|TERRNO, cleanup, "write");
   close(fd);
- group_check(0, 0, 0, 0, 0, 0, size * 256 * num);
+ group_check(0, 0, 0, 0, 0, 0, size * 256 * num, 0);
   while (waitpid(-1, &status, WUNTRACED | WCONTINUED) > 0)
       if (WEXITSTATUS(status) != 0)

Quoted 5 lines

tst_resm(TFAIL, "child exit status is %d",
--
1.7.1

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Quoted 42 lines · Bian Naimeng

Quoted 39 lines · Han Pingtian

Quoted 38 lines · Bian Naimeng

Quoted 18 lines · CAI Qian

----- Original Message -----

Quoted 17 lines

Quoted 10 lines · Qian Cai

Quoted 3 lines · Bian Naimeng

There are some problem in ksm tests.

1. We should break the test when checking is failure.

No, this is not the intention. The design here is to run all tests to check for all stats to give a full picture even if the a single failure has been observed. This type of the failures do not prevent the rest of the tests from running, so there is no need to stop the tests now which also give more insight to track down root causes.

Various reason can make checking failure, someone can make the test hangup.
I did this test on RHEL5, i found ksmd stopped before doing "echo 2 > /sys/kernel/mm/ksm/run",
so group_check will be hanged on "new_num < old_num * 3".

Quoted 6 lines

Quoted 1 line

So, i think we should break the test if "run" is not expecting.

What happened if you run the tests for a recent upstream kernel? There are some patches for ksm recently merged upstream. If the problem still persistent, please paste the EXACT OUTPUT from the ksm01 test. If it is hung, please upload sysrq-t output somewhere.

Maybe there are some bugs in the RHEL6's kernel, but the purpose of this patch is not to workround these bugs, i want to fix the test's bug.

Would you explain to me why we do this loop "while (new_num < old_num * 3)" in group_check, i think "while (new_num < old_num + 3)" is better.

Some time ago, the following patch insert this loop.
http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTi%3Dg%3DojJu0m%2B556rnekYenRTXtX%2BVBOj%3DrPmnjSw%40mail.gmail.com&forum_name=ltp-list

The changelog of this patch said "we should wait 3~5 increments of the /sys/kernel/mm/ksm/full_scans before checking ksm* testcases's results", but it do "while (new_num < old_num * 3)" actually.

Quoted 8 lines

Quoted 6 lines

I made a mistake. The code is what I wanted to do, but the changelog was wrong. When testing the new ksm patch, the developer told us we must wait 3~5 times increments of the number before checking testing results. So I coded to wait til new_num >= old_num * 3 before checking the results.

The bigger old_new is, the longer time test takes, it's strange.

I agree that it's strange. But it seems that we have to do this way.

Quoted 8 lines

Quoted 6 lines

About 'echo 2 > /sys/kernel/mm/ksm/run' problem, I have tested it with ksm01.
If I run the 'echo 2 > /sys/kernel/mm/ksm/run' before issue ksm test, the content of /sys/kernel/mm/ksm/run will be changed to 1 and the test can finished successfully.

I think we shoud not care this.

Yep, I think so.

Quoted 6 lines

Quoted 5 lines

Only if I echo the 2 between the testing process,
ksm01 will hang up. On that time, new_num will be zero, so your plus 3 method won't work either. So what should we do in this circumstance?

Quoted 3 lines

Please look at my patch, after stopping ksmd in the testing(echo 2 > /sys/kernel/mm/ksm/run or echo 0 > /sys/kernel/mm/ksm/run), group_check will skip waiting at the loop "new_num >= old_num * 3".

Run 'echo 2 > /sys/kernel/mm/ksm/run' before calling group_check() won't cause the while loops infinitely, because old_num and new_num would be all zero before the while, so new_num == old_num * 3, the while won't be ran.

Quoted 32 lines

Regards
Bian

Quoted 29 lines

Thanks.

Han Pingtian

Quoted 26 lines

Regards
Bian

Quoted 23 lines

CAI Qian

Quoted 21 lines

Quoted 4 lines

Quoted 3 lines

2. The condition "new_num < old_num * 3" seems uncomfortable, i think
it should be "new_num < old_num + 3"

I don't understand. What error did you see from the testing output?

Sometimes, the old_num is a big number, so it takes long time in this loop,
i don't understand the purpose.
Would you explain to me why we expect this condition "new_num < old_num * 3".

Quoted 5 lines

Quoted 3 lines

3. After stopping ksm(echo 2 > /sys/kernel/mm/ksm/run), the ksmd will stop scaning pages, so looping in "new_num < old_num * 3" is wrong.

Ditto.

After stopping ksm, looping in "new_num < old_num * 3" will make the process hang up,
because new_num does not be increased.

Regards
Bian