[LTP] [PATCH 1/3] m4/ltp-numa.m4: fixed numa config

[LTP] [PATCH 1/3] m4/ltp-numa.m4: fixed numa config

[Zhouping Liu]

numa_alloc_onnode() and numa_move_pages() all need numa lib '-lnuma',
but in ltp-numa.m4, we just checked the two functions using
AC_CHECK_FUNCS, without the patch, it always report such error
log in numa system: "undefined reference to `numa_alloc_onnode`",
which blocked syscalls/move_pages/* cases, the patch fixed it.

Signed-off-by: Zhouping Liu <zliu@redhat.com>
---
 m4/ltp-numa.m4 |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/m4/ltp-numa.m4 b/m4/ltp-numa.m4
index bcc5e6b..d03bc80 100644
--- a/m4/ltp-numa.m4
+++ b/m4/ltp-numa.m4
@@ -26,7 +26,14 @@ AC_DEFUN([LTP_CHECK_SYSCALL_NUMA],
 [dnl
 AC_CHECK_HEADERS([linux/mempolicy.h numa.h numaif.h],[
 	LTP_SYSCALL_NUMA_HEADERS=yes
-	AC_CHECK_FUNCS(numa_alloc_onnode,numa_move_pages)
+	AC_CHECK_LIB(numa,numa_alloc_onnode,[have_numa_alloc_onnode="yes"])
+	if  test "x$have_numa_alloc_onnode" = "xyes"; then
+		AC_DEFINE(HAVE_NUMA_ALLOC_ONNODE,1,[define to 1 if you have 'numa_alloc_onnode' function])
+	fi
+	AC_CHECK_LIB(numa,numa_move_pages,[have_numa_move_pages="yes"])
+	if  test "x$have_numa_move_pages" = "xyes"; then
+		AC_DEFINE(HAVE_NUMA_MOVE_PAGES,1,[define to 1 if you have 'numa_move_pages' function])
+	fi
 ]
 	AC_CHECK_LIB(numa,numa_available,[
 NUMA_CPPFLAGS="-DNUMA_VERSION1_COMPATIBILITY"
-- 
1.7.10


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

[LTP] [PATCH 2/3] syscalls/move_pages: modified HAS_NUMA_H as HAVE_NUMA_H

[Zhouping Liu]

HAS_NUMA_H should be HAVE_NUMA_H, and most functions are defined
in numa.h and numaif.h, so added the include headers into
move_pages_support.h

Signed-off-by: Zhouping Liu <zliu@redhat.com>
---
 testcases/kernel/syscalls/move_pages/move_pages_support.c |   12 ++++++------
 testcases/kernel/syscalls/move_pages/move_pages_support.h |    7 +++++++
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.c b/testcases/kernel/syscalls/move_pages/move_pages_support.c
index 168805e..6bd95d3 100644
--- a/testcases/kernel/syscalls/move_pages/move_pages_support.c
+++ b/testcases/kernel/syscalls/move_pages/move_pages_support.c
@@ -37,7 +37,7 @@ long get_page_size()
 void free_pages(void **pages, unsigned int num)
 {
 
-#if HAS_NUMA_H
+#if HAVE_NUMA_H
 	int i;
 	size_t onepage = get_page_size();
 
@@ -112,7 +112,7 @@ int alloc_pages_linear(void **pages, unsigned int num)
 {
 	int nodes[num];
 
-#if HAS_NUMA_H
+#if HAVE_NUMA_H
 	unsigned int i;
 	unsigned int n;
 
@@ -208,7 +208,7 @@ verify_pages_on_nodes(void **pages, int *status, unsigned int num, int *nodes)
  */
 void verify_pages_linear(void **pages, int *status, unsigned int num)
 {
-#if HAS_NUMA_H
+#if HAVE_NUMA_H
 	unsigned int i;
 	unsigned int n;
 	int nodes[num];
@@ -257,7 +257,7 @@ void verify_pages_on_node(void **pages, int *status, unsigned int num, int node)
  */
 int alloc_shared_pages_on_node(void **pages, unsigned int num, int node)
 {
-#if HAS_NUMA_H
+#if HAVE_NUMA_H
 	char *shared;
 	unsigned int i;
 	int nodes[num];
@@ -380,7 +380,7 @@ void free_sem(sem_t * sem, int num)
  */
 void check_config(unsigned int min_nodes)
 {
-#if HAS_NUMA_H
+#if HAVE_NUMA_H && HAVE_NUMAIF_H
 	if (numa_available() < 0) {
 		tst_resm(TCONF, "NUMA support is not available");
 	} else if (numa_max_node() < (min_nodes - 1)) {
@@ -392,4 +392,4 @@ void check_config(unsigned int min_nodes)
 	tst_resm(TCONF, "NUMA support not provided");
 #endif
 	tst_exit();
-}
\ No newline at end of file
+}
diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.h b/testcases/kernel/syscalls/move_pages/move_pages_support.h
index 4081625..fd78572 100644
--- a/testcases/kernel/syscalls/move_pages/move_pages_support.h
+++ b/testcases/kernel/syscalls/move_pages/move_pages_support.h
@@ -19,6 +19,13 @@
 #ifndef MOVE_PAGES_SUPPORT_H
 #define MOVE_PAGES_SUPPORT_H
 
+#include "config.h"
+#if HAVE_NUMA_H
+#include <numa.h>
+#endif
+#if HAVE_NUMAIF_H
+#include <numaif.h>
+#endif
 #include <semaphore.h>
 
 long get_page_size();
-- 
1.7.10


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

[LTP] [PATCH 3/3] syscalls/move_pages: removed tst_exit in check_config()

[Zhouping Liu]

removed tst_exit in check_config(), as it's always involked
at any time, instead, using tst_brkm to displace tst_resm,
since TCONF should be called by tst_brkm.

Signed-off-by: Zhouping Liu <zliu@redhat.com>
---
 testcases/kernel/syscalls/move_pages/move_pages_support.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.c b/testcases/kernel/syscalls/move_pages/move_pages_support.c
index 6bd95d3..320486f 100644
--- a/testcases/kernel/syscalls/move_pages/move_pages_support.c
+++ b/testcases/kernel/syscalls/move_pages/move_pages_support.c
@@ -382,14 +382,13 @@ void check_config(unsigned int min_nodes)
 {
 #if HAVE_NUMA_H && HAVE_NUMAIF_H
 	if (numa_available() < 0) {
-		tst_resm(TCONF, "NUMA support is not available");
+		tst_brkm(TCONF, NULL, "NUMA support is not available");
 	} else if (numa_max_node() < (min_nodes - 1)) {
-		tst_resm(TCONF, "atleast 2 NUMA nodes are required");
+		tst_brkm(TCONF, NULL, "atleast 2 NUMA nodes are required");
 	} else if (tst_kvercmp(2, 6, 18) < 0) {
-		tst_resm(TCONF, "2.6.18 or greater kernel required");
+		tst_brkm(TCONF, NULL, "2.6.18 or greater kernel required");
 	}
 #else
-	tst_resm(TCONF, "NUMA support not provided");
+	tst_brkm(TCONF, NULL, "NUMA support not provided");
 #endif
-	tst_exit();
 }
-- 
1.7.10


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

Re: [LTP] [PATCH 2/3] syscalls/move_pages: modified HAS_NUMA_H as HAVE_NUMA_H

[Garrett Cooper]

On Apr 28, 2012, at 12:48 AM, Zhouping Liu wrote:

> HAS_NUMA_H should be HAVE_NUMA_H, and most functions are defined
> in numa.h and numaif.h, so added the include headers into
> move_pages_support.h
> 
> Signed-off-by: Zhouping Liu <zliu@redhat.com>

Reviewed-by: Garrett Cooper <yanegomi@gmail.com


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

Re: [LTP] [PATCH 2/3] syscalls/move_pages: modified HAS_NUMA_H as HAVE_NUMA_H

[Caspar Zhang]

On 04/28/2012 03:48 PM, Zhouping Liu wrote:
> HAS_NUMA_H should be HAVE_NUMA_H, and most functions are defined
> in numa.h and numaif.h, so added the include headers into
> move_pages_support.h
> 
> Signed-off-by: Zhouping Liu <zliu@redhat.com>

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

> ---
>  testcases/kernel/syscalls/move_pages/move_pages_support.c |   12 ++++++------
>  testcases/kernel/syscalls/move_pages/move_pages_support.h |    7 +++++++
>  2 files changed, 13 insertions(+), 6 deletions(-)
> 
> diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.c b/testcases/kernel/syscalls/move_pages/move_pages_support.c
> index 168805e..6bd95d3 100644
> --- a/testcases/kernel/syscalls/move_pages/move_pages_support.c
> +++ b/testcases/kernel/syscalls/move_pages/move_pages_support.c
> @@ -37,7 +37,7 @@ long get_page_size()
>  void free_pages(void **pages, unsigned int num)
>  {
>  
> -#if HAS_NUMA_H
> +#if HAVE_NUMA_H
>  	int i;
>  	size_t onepage = get_page_size();
>  
> @@ -112,7 +112,7 @@ int alloc_pages_linear(void **pages, unsigned int num)
>  {
>  	int nodes[num];
>  
> -#if HAS_NUMA_H
> +#if HAVE_NUMA_H
>  	unsigned int i;
>  	unsigned int n;
>  
> @@ -208,7 +208,7 @@ verify_pages_on_nodes(void **pages, int *status, unsigned int num, int *nodes)
>   */
>  void verify_pages_linear(void **pages, int *status, unsigned int num)
>  {
> -#if HAS_NUMA_H
> +#if HAVE_NUMA_H
>  	unsigned int i;
>  	unsigned int n;
>  	int nodes[num];
> @@ -257,7 +257,7 @@ void verify_pages_on_node(void **pages, int *status, unsigned int num, int node)
>   */
>  int alloc_shared_pages_on_node(void **pages, unsigned int num, int node)
>  {
> -#if HAS_NUMA_H
> +#if HAVE_NUMA_H
>  	char *shared;
>  	unsigned int i;
>  	int nodes[num];
> @@ -380,7 +380,7 @@ void free_sem(sem_t * sem, int num)
>   */
>  void check_config(unsigned int min_nodes)
>  {
> -#if HAS_NUMA_H
> +#if HAVE_NUMA_H && HAVE_NUMAIF_H
>  	if (numa_available() < 0) {
>  		tst_resm(TCONF, "NUMA support is not available");
>  	} else if (numa_max_node() < (min_nodes - 1)) {
> @@ -392,4 +392,4 @@ void check_config(unsigned int min_nodes)
>  	tst_resm(TCONF, "NUMA support not provided");
>  #endif
>  	tst_exit();
> -}
> \ No newline at end of file
> +}
> diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.h b/testcases/kernel/syscalls/move_pages/move_pages_support.h
> index 4081625..fd78572 100644
> --- a/testcases/kernel/syscalls/move_pages/move_pages_support.h
> +++ b/testcases/kernel/syscalls/move_pages/move_pages_support.h
> @@ -19,6 +19,13 @@
>  #ifndef MOVE_PAGES_SUPPORT_H
>  #define MOVE_PAGES_SUPPORT_H
>  
> +#include "config.h"
> +#if HAVE_NUMA_H
> +#include <numa.h>
> +#endif
> +#if HAVE_NUMAIF_H
> +#include <numaif.h>
> +#endif
>  #include <semaphore.h>
>  
>  long get_page_size();


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

Re: [LTP] [PATCH 3/3] syscalls/move_pages: removed tst_exit in check_config()

[Caspar Zhang]

On 04/28/2012 03:48 PM, Zhouping Liu wrote:
> removed tst_exit in check_config(), as it's always involked
> at any time, instead, using tst_brkm to displace tst_resm,
> since TCONF should be called by tst_brkm.
> 
> Signed-off-by: Zhouping Liu <zliu@redhat.com>

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

> ---
>  testcases/kernel/syscalls/move_pages/move_pages_support.c |    9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/testcases/kernel/syscalls/move_pages/move_pages_support.c b/testcases/kernel/syscalls/move_pages/move_pages_support.c
> index 6bd95d3..320486f 100644
> --- a/testcases/kernel/syscalls/move_pages/move_pages_support.c
> +++ b/testcases/kernel/syscalls/move_pages/move_pages_support.c
> @@ -382,14 +382,13 @@ void check_config(unsigned int min_nodes)
>  {
>  #if HAVE_NUMA_H && HAVE_NUMAIF_H
>  	if (numa_available() < 0) {
> -		tst_resm(TCONF, "NUMA support is not available");
> +		tst_brkm(TCONF, NULL, "NUMA support is not available");
>  	} else if (numa_max_node() < (min_nodes - 1)) {
> -		tst_resm(TCONF, "atleast 2 NUMA nodes are required");
> +		tst_brkm(TCONF, NULL, "atleast 2 NUMA nodes are required");
>  	} else if (tst_kvercmp(2, 6, 18) < 0) {
> -		tst_resm(TCONF, "2.6.18 or greater kernel required");
> +		tst_brkm(TCONF, NULL, "2.6.18 or greater kernel required");
>  	}
>  #else
> -	tst_resm(TCONF, "NUMA support not provided");
> +	tst_brkm(TCONF, NULL, "NUMA support not provided");
>  #endif
> -	tst_exit();
>  }


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

Re: [LTP] [PATCH 1/3] m4/ltp-numa.m4: fixed numa config

[Caspar Zhang]

On 04/28/2012 03:48 PM, Zhouping Liu wrote:
> numa_alloc_onnode() and numa_move_pages() all need numa lib '-lnuma',
> but in ltp-numa.m4, we just checked the two functions using
> AC_CHECK_FUNCS, without the patch, it always report such error
> log in numa system: "undefined reference to `numa_alloc_onnode`",
> which blocked syscalls/move_pages/* cases, the patch fixed it.
> 
> Signed-off-by: Zhouping Liu <zliu@redhat.com>

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

> ---
>  m4/ltp-numa.m4 |    9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/m4/ltp-numa.m4 b/m4/ltp-numa.m4
> index bcc5e6b..d03bc80 100644
> --- a/m4/ltp-numa.m4
> +++ b/m4/ltp-numa.m4
> @@ -26,7 +26,14 @@ AC_DEFUN([LTP_CHECK_SYSCALL_NUMA],
>  [dnl
>  AC_CHECK_HEADERS([linux/mempolicy.h numa.h numaif.h],[
>  	LTP_SYSCALL_NUMA_HEADERS=yes
> -	AC_CHECK_FUNCS(numa_alloc_onnode,numa_move_pages)
> +	AC_CHECK_LIB(numa,numa_alloc_onnode,[have_numa_alloc_onnode="yes"])
> +	if  test "x$have_numa_alloc_onnode" = "xyes"; then
> +		AC_DEFINE(HAVE_NUMA_ALLOC_ONNODE,1,[define to 1 if you have 'numa_alloc_onnode' function])
> +	fi
> +	AC_CHECK_LIB(numa,numa_move_pages,[have_numa_move_pages="yes"])
> +	if  test "x$have_numa_move_pages" = "xyes"; then
> +		AC_DEFINE(HAVE_NUMA_MOVE_PAGES,1,[define to 1 if you have 'numa_move_pages' function])
> +	fi
>  ]
>  	AC_CHECK_LIB(numa,numa_available,[
>  NUMA_CPPFLAGS="-DNUMA_VERSION1_COMPATIBILITY"


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

Re: [LTP] [PATCH 1/3] m4/ltp-numa.m4: fixed numa config

[Wanlong Gao]

On 04/28/2012 03:48 PM, Zhouping Liu wrote:

> numa_alloc_onnode() and numa_move_pages() all need numa lib '-lnuma',
> but in ltp-numa.m4, we just checked the two functions using
> AC_CHECK_FUNCS, without the patch, it always report such error
> log in numa system: "undefined reference to `numa_alloc_onnode`",
> which blocked syscalls/move_pages/* cases, the patch fixed it.



pushed the series, thanks

Wanlong Gao

> 
> Signed-off-by: Zhouping Liu <zliu@redhat.com>
> ---
>  m4/ltp-numa.m4 |    9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/m4/ltp-numa.m4 b/m4/ltp-numa.m4
> index bcc5e6b..d03bc80 100644
> --- a/m4/ltp-numa.m4
> +++ b/m4/ltp-numa.m4
> @@ -26,7 +26,14 @@ AC_DEFUN([LTP_CHECK_SYSCALL_NUMA],
>  [dnl
>  AC_CHECK_HEADERS([linux/mempolicy.h numa.h numaif.h],[
>  	LTP_SYSCALL_NUMA_HEADERS=yes
> -	AC_CHECK_FUNCS(numa_alloc_onnode,numa_move_pages)
> +	AC_CHECK_LIB(numa,numa_alloc_onnode,[have_numa_alloc_onnode="yes"])
> +	if  test "x$have_numa_alloc_onnode" = "xyes"; then
> +		AC_DEFINE(HAVE_NUMA_ALLOC_ONNODE,1,[define to 1 if you have 'numa_alloc_onnode' function])
> +	fi
> +	AC_CHECK_LIB(numa,numa_move_pages,[have_numa_move_pages="yes"])
> +	if  test "x$have_numa_move_pages" = "xyes"; then
> +		AC_DEFINE(HAVE_NUMA_MOVE_PAGES,1,[define to 1 if you have 'numa_move_pages' function])
> +	fi
>  ]
>  	AC_CHECK_LIB(numa,numa_available,[
>  NUMA_CPPFLAGS="-DNUMA_VERSION1_COMPATIBILITY"



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