Timer related syscalls for LTP test

RFC In review Suzuki Poulose 2 participants 2 messages May 27, 2010 – May 29, 2010

lore

Share & tools

lore

Suzuki Poulose

author

2010-05-27 11:19 UTC #0 lore raw

RFC 1/1 Timer related syscalls for LTP test

Hi,

I have found a problem with the timer related syscall tests.

The following syscall tests, uses the direct syscall interface to the kernel for testing the interface, instead of using the GLIBC interface.

testcases/kernel/syscalls
\_ timer_getoverrun/
\_ timer_gettime/

testcases/kernel/timers
\_ timer_create/
\_ timer_delete/
\_ timer_settime/

So they use :
syscall(__NR_timer_XXXX, args )
for directly using the kernel syscall interface than using the

timer_xxxxx(args)

provided by the GLIBC. This is completely fine.

But however there is a mismatch in the "timer_t" type defined by the Linux kernel and the GNU Libc.

Linux kernel defines it as an int for all architectures.

include/linux/types.h :

typedef __kernel_timer_t timer_t;

arch/powerpc/include/asm/posix_types.h :
typedef int __kernel_timer_t;

Now, the glibc provides a wrapper to the timer related system calls. While providing the wrapper, Glibc stores some internal data associated with the timer. So it defines the "timer_t" as a pointer (void*) as below:

/usr/include/time.h :

/* Timer ID returned by `timer_create'. */
typedef __timer_t timer_t;

in /usr/include/bits/types.h :

/* Timer ID returned by `timer_create'. */
__STD_TYPE __TIMER_T_TYPE __timer_t;

in /usr/include/bits/typesizes.h

#define __TIMER_T_TYPE void *

So that leads to the definiton of timer_t as void* by glibc. And this cannot be changed. But the glibc uses the "proper" typed argument for the syscalls it issues by defining a kernel_timer_t.

On 32bit machines the difference is not noticed, since both the timer_t in userspace and the kernel space are of the same size. But on 64bit machines this is different.

Userspace has a width of 64bit and the kernel has 32bit width. Now, when we pass the pointer to timer_t to the kernel, things get complicated and on Big endian machines, ppc64, s390x, this leads to a problem. As the value kept is not available when we try to dereference it as a 32bit and this leads to undesired results like below:

timer_getoverrun01 1 TFAIL : Block 1: test 0 FAILED... errno = 22 :Invalid argument

strace shows :

13017 timer_create(CLOCK_REALTIME, {(nil), SIGALRM, SIGEV_SIGNAL, {...}},{0x26e80}) = 0
13017 fstat(1, {st_dev=makedev(0, 11), st_ino=3, st_mode=S_IFCHR|0620,
st_nlink=1, st_uid=0, st_gid=5, st_blksize=1024, st_blocks=0, st_rdev=makedev(136, 0), st_atime=2010/05/17-15:48:40,st_mtime=2010/05/17-15:48:40, st_ctime=2010/05/17-13:15:48}) = 0 13017 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)= 0x20000028000
13017 write(1, "timer_getoverrun01 0 TINFO "..., 66) = 66
13017 timer_getoverrun(0x26e80) = -1 EINVAL (Invalid argument)

So the conclusion is that, if we are using the System calls directly, we should use the argument which is compatible with the kernel. We simply *cannot* use the glibc definition and the kernel API directly.

I have a patch attached below that fixes this issue.

One more comment regarding the placement of timer_gettime and timer_getoverrun syscall tests. I strongly believe they could be moved to testcases/kernel/timers/ directory where the rest of the timer related syscall tests are placed. If that change is made, we can use the definition added in common_timers.h (in the patch) for those tests too, instead of defining in each of the tests.

Thanks

Suzuki

Signed-off-by Suzuki K P <[email protected]>

Fix timer_t argument for timer related direct syscalls.

The tests timer_getoverrun and timer_gettime should be moved to testcases/kernel/timers directory. If that is done, then we could use the definition in the common_timers.h, like the other tests.

/* Description */
There is a mismatch in the "timer_t" type defined by the Linux kernel and the GNU Libc.

Linux kernel defines it as an int for all architectures.

include/linux/types.h :

typedef __kernel_timer_t timer_t;

arch/powerpc/include/asm/posix_types.h :
typedef int __kernel_timer_t;

/usr/include/time.h :

/* Timer ID returned by `timer_create'. */
typedef __timer_t timer_t;

in /usr/include/bits/types.h :

/* Timer ID returned by `timer_create'. */
__STD_TYPE __TIMER_T_TYPE __timer_t;

in /usr/include/bits/typesizes.h

#define __TIMER_T_TYPE void *

So that leads to the definiton of timer_t as void* by glibc. And this cannot be changed. But the glibc uses the "proper" typed argument for the syscalls it issues by defining a kernel_timer_t.

On 32bit machines the difference is not noticed, since both the timer_t in userspace and the kernel space are of the same size. But on 64bit machines this is different.

testcases/kernel/syscalls/timer_getoverrun/timer_getoverrun01.c +6 -1

Index: ltp/testcases/kernel/syscalls/timer_getoverrun/timer_getoverrun01.c
===================================================================
@@ -49,6 +49,11 @@
 #include "usctest.h"
 #include "linux_syscall_numbers.h"
 
+/* timer_t in kernel(int) is different from  Glibc definition(void*).
+ * Use the kernel definition.
+ */
+typedef int kernel_timer_t;
+
 /* Extern Global Variables */
 extern int Tst_count;           /* counter for tst_xxx routines.         */
 extern char *TESTDIR;           /* temporary dir created by tst_tmpdir() */

@@ -120,7 +125,7 @@
 int main(int ac, char **av) {
         int lc;                 /* loop counter */
         char *msg;              /* message returned from parse_opts */
-	timer_t created_timer_id;
+	kernel_timer_t created_timer_id;
         struct sigevent ev;
 	
         /* parse standard options */

testcases/kernel/syscalls/timer_gettime/timer_gettime01.c +6 -1

Index: ltp/testcases/kernel/syscalls/timer_gettime/timer_gettime01.c
===================================================================
@@ -48,6 +48,11 @@
 #include "usctest.h"
 #include "linux_syscall_numbers.h"
 
+/* timer_t in kernel(int) is different from  Glibc definition(void*).
+ * Use the kernel definition.
+ */
+typedef int kernel_timer_t;
+
 /* Extern Global Variables */
 extern int Tst_count;           /* counter for tst_xxx routines.         */
 extern char *TESTDIR;           /* temporary dir created by tst_tmpdir() */

@@ -122,7 +127,7 @@
         int lc;                 /* loop counter */
         char *msg;              /* message returned from parse_opts */
 
-        timer_t created_timer_id;
+        kernel_timer_t created_timer_id;
         struct sigevent ev;
         struct itimerspec spec;

testcases/kernel/timers/timer_delete/timer_delete02.c +1 -1

Index: ltp/testcases/kernel/timers/timer_delete/timer_delete02.c
===================================================================
@@ -84,7 +84,7 @@
 {
 	int lc;		/* loop counter */
 	char *msg;	/* message returned from parse_opts */
-	timer_t timer_id;
+	kernel_timer_t timer_id;
 
 	/* parse standard options */
 	if ((msg = parse_opts (ac, av, (option_t *) NULL, NULL)) !=

testcases/kernel/timers/timer_delete/timer_delete03.c +1 -1

Index: ltp/testcases/kernel/timers/timer_delete/timer_delete03.c
===================================================================
@@ -73,7 +73,7 @@
 #include "usctest.h"
 #include "common_timers.h"
 
-#define INVALID_ID ((timer_t)-1)
+#define INVALID_ID ((kernel_timer_t)-1)
 
 static void setup();

testcases/kernel/timers/timer_create/timer_create02.c +1 -1

Index: ltp/testcases/kernel/timers/timer_create/timer_create02.c
===================================================================
@@ -86,7 +86,7 @@
 {
 	int lc, i, j;		/* loop counter */
 	char *msg;			/* message returned from parse_opts */
-	timer_t created_timer_id;	/* holds the returned timer_id */
+	kernel_timer_t created_timer_id;	/* holds the returned timer_id */
 	char *message[3] = {
 		"SIGEV_SIGNAL",
 		"NULL",

testcases/kernel/timers/timer_create/timer_create03.c +1 -1

Index: ltp/testcases/kernel/timers/timer_create/timer_create03.c
===================================================================
@@ -100,7 +100,7 @@
 {
 	int lc, i;			/* loop counter */
 	char *msg;			/* message returned from parse_opts */
-	timer_t created_timer_id;	/* holds the returned timer_id */
+	kernel_timer_t created_timer_id;	/* holds the returned timer_id */
 	char *message[] = {
 		"SIGEV_SIGNAL",
 		"NULL",

testcases/kernel/timers/timer_create/timer_create04.c +3 -3

Index: ltp/testcases/kernel/timers/timer_create/timer_create04.c
===================================================================
@@ -111,7 +111,7 @@
 {
 	int lc, i;			/* loop counter */
 	char *msg;			/* message returned from parse_opts */
-	timer_t timer_id, *temp_id;	/* stores the returned timer_id */
+	kernel_timer_t timer_id, *temp_id;	/* stores the returned timer_id */
 	struct sigevent *temp_ev;	/* used for bad address test case */
 
 	clockid_t clocks[6] = {

@@ -158,7 +158,7 @@
 
 			switch (i) {
 			case 2: /* make the timer_id bad address */
-				temp_id = (timer_t *) -1;
+				temp_id = (kernel_timer_t *) -1;
 				break;
 			case 3:
 				/* make the event bad address */

@@ -167,7 +167,7 @@
 			case 4:
 				/* Produce an invalid timer_id address. */
 				if(tst_kvercmp(2, 6, 12) >= 0)
-					temp_id = (timer_t *) -1;
+					temp_id = (kernel_timer_t *) -1;
 				break;
 			case 5:
 				/* Produce an invalid event address. */

testcases/kernel/timers/timer_settime/timer_settime02.c +1 -1

Index: ltp/testcases/kernel/timers/timer_settime/timer_settime02.c
===================================================================
@@ -81,7 +81,7 @@
 extern int Tst_count;		/* Test Case counter for tst_* routines */
 
 static struct itimerspec new_set, old_set, *old_temp;
-static timer_t timer;
+static kernel_timer_t timer;
 static int flag;
 
 int

testcases/kernel/timers/timer_settime/timer_settime03.c +2 -2

Index: ltp/testcases/kernel/timers/timer_settime/timer_settime03.c
===================================================================
@@ -81,7 +81,7 @@
 extern int Tst_count;		/* Test Case counter for tst_* routines */
 
 static struct itimerspec new_set, old_set, *old_temp, *new_temp;
-static timer_t timer, tim;
+static kernel_timer_t timer, tim;
 
 static int exp_enos[] = {EINVAL, EFAULT, 0};

@@ -168,7 +168,7 @@
 		break;
 	case 3:
 		/* make timer_id invalid */
-		tim = (timer_t)-1;
+		tim = (kernel_timer_t)-1;
 		new_set.it_value.tv_nsec = 0;
 		break;
 	case 4:

testcases/kernel/timers/include/common_timers.h +5 -0

Index: ltp/testcases/kernel/timers/include/common_timers.h
===================================================================
@@ -63,4 +63,9 @@
 #include <time.h>
 #include <unistd.h>
 
+/* timer_t in kernel(int) is different from  Glibc definition(void*).
+ * Use the kernel definition for syscall tests
+ */
+typedef int kernel_timer_t;
+
 #endif

------------------------------------------------------------------------------

_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Subrata Modak

2010-05-29 18:23 UTC #1 lore raw

Quoted 91 lines · Suzuki Poulose

Hi,
I have found a problem with the timer related syscall tests.
The following syscall tests, uses the direct syscall interface to the kernel for testing the interface, instead of using the GLIBC interface.
testcases/kernel/syscalls
\_ timer_getoverrun/
\_ timer_gettime/
testcases/kernel/timers
\_ timer_create/
\_ timer_delete/
\_ timer_settime/
So they use :
syscall(__NR_timer_XXXX, args )
for directly using the kernel syscall interface than using the
timer_xxxxx(args)
provided by the GLIBC. This is completely fine.
But however there is a mismatch in the "timer_t" type defined by the Linux kernel and the GNU Libc.
Linux kernel defines it as an int for all architectures.
include/linux/types.h :
typedef __kernel_timer_t timer_t;
arch/powerpc/include/asm/posix_types.h :
typedef int __kernel_timer_t;
Now, the glibc provides a wrapper to the timer related system calls. While providing the wrapper, Glibc stores some internal data associated with the timer. So it defines the "timer_t" as a pointer (void*) as below:
/usr/include/time.h :
/* Timer ID returned by `timer_create'. */
typedef __timer_t timer_t;
in /usr/include/bits/types.h :
/* Timer ID returned by `timer_create'. */
__STD_TYPE __TIMER_T_TYPE __timer_t;
in /usr/include/bits/typesizes.h
#define __TIMER_T_TYPE void *
So that leads to the definiton of timer_t as void* by glibc. And this cannot be changed. But the glibc uses the "proper" typed argument for the syscalls it issues by defining a kernel_timer_t.
On 32bit machines the difference is not noticed, since both the timer_t in userspace and the kernel space are of the same size. But on 64bit machines this is different.
Userspace has a width of 64bit and the kernel has 32bit width. Now, when we pass the pointer to timer_t to the kernel, things get complicated and on Big endian machines, ppc64, s390x, this leads to a problem. As the value kept is not available when we try to dereference it as a 32bit and this leads to undesired results like below:
timer_getoverrun01 1 TFAIL : Block 1: test 0 FAILED... errno = 22
:Invalid argument
strace shows :
13017 timer_create(CLOCK_REALTIME, {(nil), SIGALRM, SIGEV_SIGNAL, {...}},{0x26e80}) = 0
13017 fstat(1, {st_dev=makedev(0, 11), st_ino=3, st_mode=S_IFCHR|0620, st_nlink=1, st_uid=0, st_gid=5, st_blksize=1024, st_blocks=0, st_rdev=makedev(136, 0), st_atime=2010/05/17-15:48:40,st_mtime=2010/05/17-15:48:40, st_ctime=2010/05/17-13:15:48}) = 0
13017 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)= 0x20000028000
13017 write(1, "timer_getoverrun01 0 TINFO "..., 66) = 66
13017 timer_getoverrun(0x26e80) = -1 EINVAL (Invalid argument)

Quoted 5 lines

So the conclusion is that, if we are using the System calls directly, we should use the argument which is compatible with the kernel. We simply *cannot* use the glibc definition and the kernel API directly.
I have a patch attached below that fixes this issue.

Thanks Suzie.

Quoted 7 lines

One more comment regarding the placement of timer_gettime and timer_getoverrun syscall tests. I strongly believe they could be moved to testcases/kernel/timers/ directory where the rest of the timer related syscall tests are placed. If that change is made, we can use the definition added in common_timers.h (in the patch) for those tests too, instead of defining in each of the tests.

Sure, you can move them accordingly in the next patch :-)

Regards--
Subrata

Quoted 13 lines

Thanks
Suzuki

Quoted 6 lines

------------------------------------------------------------------------------
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

------------------------------------------------------------------------------

_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list