ltp-cvs — LTP CVS Commits

[Ltp-cvs] utils/analysis/lcov lcovrc,1.11,1.12

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv19705

Modified Files:
	lcovrc 
Log Message:
lcov: improve list output

Improve list output by separating directory and file names. Also provide
an option to show full path names.


Index: lcovrc
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/lcovrc,v
retrieving revision 1.11
retrieving revision 1.12
diff -C2 -d -r1.11 -r1.12
*** lcovrc	22 Jan 2010 16:17:36 -0000	1.11
--- lcovrc	19 Jul 2010 14:27:08 -0000	1.12
***************
*** 116,117 ****
--- 116,121 ----
  # Location for temporary directories
  lcov_tmp_dir = /tmp
+ 
+ # Show full paths during list operation if non-zero (same as --list-full-path
+ # option of lcov)
+ lcov_list_full_path = 0

[Ltp-cvs] utils/analysis/lcov/bin genhtml,1.47,1.48

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv23179

Modified Files:
	genhtml 
Log Message:
genhtml: fix large numbers being shown as negative in html output

genhtml uses a "%d" format string for printing execution counts. For
counts exceeding integer range, the output becomes negative. Fix this
by using the "%.0f" format string instead.

Reported by kky...@ya....


Index: genhtml
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/genhtml,v
retrieving revision 1.47
retrieving revision 1.48
diff -C2 -d -r1.47 -r1.48
*** genhtml	29 Jan 2010 11:07:25 -0000	1.47
--- genhtml	19 Jul 2010 12:12:43 -0000	1.48
***************
*** 3845,3849 ****
  	my $exp;
  
! 	$result = sprintf("%*d", $width, $count);
  	while (length($result) > $width) {
  		last if ($count < 10);
--- 3845,3849 ----
  	my $exp;
  
! 	$result = sprintf("%*.0f", $width, $count);
  	while (length($result) > $width) {
  		last if ($count < 10);

[Ltp-cvs] utils/analysis/lcov/bin geninfo,1.75,1.76

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv14886

Modified Files:
	geninfo 
Log Message:
geninfo: ensure that exclusion markers apply to --initial

Fix a problem where exclusion markers are ignored when gathering
initial coverage data.

Problem was reported by ahm...@me....


Index: geninfo
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/geninfo,v
retrieving revision 1.75
retrieving revision 1.76
diff -C2 -d -r1.75 -r1.76
*** geninfo	1 Jun 2010 13:48:29 -0000	1.75
--- geninfo	7 Jun 2010 12:22:18 -0000	1.76
***************
*** 103,106 ****
--- 103,108 ----
  sub read_config($);
  sub apply_config($);
+ sub get_exclusion_data($);
+ sub apply_exclusion_data($$);
  sub process_graphfile($$);
  sub filter_fn_name($);
***************
*** 112,115 ****
--- 114,118 ----
  sub graph_skip(*$;$);
  sub sort_uniq(@);
+ sub sort_uniq_lex(@);
  sub graph_cleanup($);
  sub graph_find_base($);
***************
*** 1924,1927 ****
--- 1927,2081 ----
  
  
+ #
+ # get_exclusion_data(filename)
+ #
+ # Scan specified source code file for exclusion markers and return
+ #   linenumber -> 1
+ # for all lines which should be excluded.
+ #
+ 
+ sub get_exclusion_data($)
+ {
+ 	my ($filename) = @_;
+ 	my %list;
+ 	my $flag = 0;
+ 	local *HANDLE;
+ 
+ 	if (!open(HANDLE, "<$filename")) {
+ 		warn("WARNING: could not open $filename\n");
+ 		return undef;
+ 	}
+ 	while (<HANDLE>) {
+ 		if (/$EXCL_STOP/) {
+ 			$flag = 0;
+ 		} elsif (/$EXCL_START/) {
+ 			$flag = 1;
+ 		}
+ 		if (/$EXCL_LINE/ || $flag) {
+ 			$list{$.} = 1;
+ 		}
+ 	}
+ 	close(HANDLE);
+ 
+ 	if ($flag) {
+ 		warn("WARNING: unterminated exclusion section in $filename\n");
+ 	}
+ 
+ 	return \%list;
+ }
+ 
+ 
+ #
+ # apply_exclusion_data(instr, graph)
+ #
+ # Remove lines from instr and graph data structures which are marked
+ # for exclusion in the source code file.
+ #
+ # Return adjusted (instr, graph).
+ #
+ # graph         : file name -> function data
+ # function data : function name -> line data
+ # line data     : [ line1, line2, ... ]
+ #
+ # instr     : filename -> line data
+ # line data : [ line1, line2, ... ]
+ #
+ 
+ sub apply_exclusion_data($$)
+ {
+ 	my ($instr, $graph) = @_;
+ 	my $filename;
+ 	my %excl_data;
+ 	my $excl_read_failed = 0;
+ 
+ 	# Collect exclusion marker data
+ 	foreach $filename (sort_uniq_lex(keys(%{$graph}), keys(%{$instr}))) {
+ 		my $excl = get_exclusion_data($filename);
+ 
+ 		# Skip and note if file could not be read
+ 		if (!defined($excl)) {
+ 			$excl_read_failed = 1;
+ 			next;
+ 		}
+ 
+ 		# Add to collection if there are markers
+ 		$excl_data{$filename} = $excl if (keys(%{$excl}) > 0);
+ 	}
+ 
+ 	# Warn if not all source files could be read
+ 	if ($excl_read_failed) {
+ 		warn("WARNING: some exclusion markers may be ignored\n");
+ 	}
+ 
+ 	# Skip if no markers were found
+ 	return ($instr, $graph) if (keys(%excl_data) == 0);
+ 
+ 	# Apply exclusion marker data to graph
+ 	foreach $filename (keys(%excl_data)) {
+ 		my $function_data = $graph->{$filename};
+ 		my $excl = $excl_data{$filename};
+ 		my $function;
+ 
+ 		next if (!defined($function_data));
+ 
+ 		foreach $function (keys(%{$function_data})) {
+ 			my $line_data = $function_data->{$function};
+ 			my $line;
+ 			my @new_data;
+ 
+ 			# To be consistent with exclusion parser in non-initial
+ 			# case we need to remove a function if the first line
+ 			# was excluded
+ 			if ($excl->{$line_data->[0]}) {
+ 				delete($function_data->{$function});
+ 				next;
+ 			}
+ 			# Copy only lines which are not excluded
+ 			foreach $line (@{$line_data}) {
+ 				push(@new_data, $line) if (!$excl->{$line});
+ 			}
+ 
+ 			# Store modified list
+ 			if (scalar(@new_data) > 0) {
+ 				$function_data->{$function} = \@new_data;
+ 			} else {
+ 				# All of this function was excluded
+ 				delete($function_data->{$function});
+ 			}
+ 		}
+ 
+ 		# Check if all functions of this file were excluded
+ 		if (keys(%{$function_data}) == 0) {
+ 			delete($graph->{$filename});
+ 		}
+ 	}
+ 
+ 	# Apply exclusion marker data to instr
+ 	foreach $filename (keys(%excl_data)) {
+ 		my $line_data = $instr->{$filename};
+ 		my $excl = $excl_data{$filename};
+ 		my $line;
+ 		my @new_data;
+ 
+ 		next if (!defined($line_data));
+ 
+ 		# Copy only lines which are not excluded
+ 		foreach $line (@{$line_data}) {
+ 			push(@new_data, $line) if (!$excl->{$line});
+ 		}
+ 
+ 		# Store modified list
+ 		if (scalar(@new_data) > 0) {
+ 			$instr->{$filename} = \@new_data;
+ 		} else {
+ 			# All of this file was excluded
+ 			delete($instr->{$filename});
+ 		}
+ 	}
+ 
+ 	return ($instr, $graph);
+ }
+ 
+ 
  sub process_graphfile($$)
  {
***************
*** 1935,1941 ****
  	my $instr;
  	my $filename;
! 	my $ln_found;
  	my $total_ln_found = 0;
! 	my $fn_found;
  	my $total_fn_found = 0;
  	local *INFO_HANDLE;
--- 2089,2095 ----
  	my $instr;
  	my $filename;
! 	my $ln_found = 0;
  	my $total_ln_found = 0;
! 	my $fn_found = 0;
  	my $total_fn_found = 0;
  	local *INFO_HANDLE;
***************
*** 1983,1986 ****
--- 2137,2145 ----
  	}
  
+ 	if (!$no_markers) {
+ 		# Apply exclusion marker data to graph file data
+ 		($instr, $graph) = apply_exclusion_data($instr, $graph);
+ 	}
+ 
  	# Check whether we're writing to a single file
  	if ($output_filename)
***************
*** 2201,2204 ****
--- 2360,2380 ----
  
  #
+ # sort_uniq_lex(list)
+ #
+ # Return list in lexically ascending order and without duplicate entries.
+ #
+ 
+ sub sort_uniq_lex(@)
+ {
+ 	my (@list) = @_;
+ 	my %hash;
+ 
+ 	foreach (@list) {
+ 		$hash{$_} = 1;
+ 	}
+ 	return sort keys(%hash);
+ }
+ 
+ #
  # graph_cleanup(graph)
  #

[Ltp-cvs] utils/analysis/lcov/bin geninfo,1.74,1.75

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv5063

Modified Files:
	geninfo 
Log Message:
lcov: fix problem with relative path names

Fix a problem where coverage data is missing because gcov produces
output files starting with a dot.

Problem reported by wes...@op....


Index: geninfo
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/geninfo,v
retrieving revision 1.74
retrieving revision 1.75
diff -C2 -d -r1.74 -r1.75
*** geninfo	28 Feb 2010 20:57:37 -0000	1.74
--- geninfo	1 Jun 2010 13:48:29 -0000	1.75
***************
*** 724,727 ****
--- 724,753 ----
  
  #
+ # get_filenames(directory, pattern)
+ #
+ # Return a list of filenames found in directory which match the specified
+ # pattern.
+ #
+ # Die on error.
+ #
+ 
+ sub get_filenames($$)
+ {
+ 	my ($dirname, $pattern) = @_;
+ 	my @result;
+ 	my $directory;
+ 	local *DIR;
+ 
+ 	opendir(DIR, $dirname) or
+ 		die("ERROR: cannot read directory $dirname\n");
+ 	while ($directory = readdir(DIR)) {
+ 		push(@result, $directory) if ($directory =~ /$pattern/);
+ 	}
+ 	closedir(DIR);
+ 
+ 	return @result;
+ }
+ 
+ #
  # process_dafile(da_filename, dir)
  #
***************
*** 919,923 ****
  
  	# Collect data from resulting .gcov files and create .info file
! 	@gcov_list = glob("*.gcov");
  
  	# Check for files
--- 945,949 ----
  
  	# Collect data from resulting .gcov files and create .info file
! 	@gcov_list = get_filenames('.', '\.gcov$');
  
  	# Check for files

[Ltp-cvs] Ith pathetic fervency he fell upon his knees and imp

[Hedrix H. <ac...@nt...>]

Ould he say to her with his present imperfect knowledge? How

could she write to him if that knowledge

was correct? "Ef," said Nott, kindly, with a laying on of
large benedictory and paternal hands, "ef yer are willin' to see Rosey
agin, without SPEAKIN' to her, I reckon I ken fix it for yer. I'm goin'
to
take her down to the boat in half an
hour. Ef yer should happen--mind, ef yer should HAPPEN to be down there,
seein' some friends off and sorter promenadin' up and down the wharf
like them high-toned chaps on Montgomery Street--ye might ketch her eye
unconscious like. Or, ye might

do this!" He
rose after a moment's cogitation and

with a face of profound mystery opened
the door and beckoned Renshaw to follow him. Leading the way cautiously,
he brought the young man into an open unpartitioned recess beside her
stateroom. It seemed to be used as a storeroom, and Renshaw's eye was
caught by a trunk the
size and shape of the one

[Ltp-cvs] [Git] ltp updated.

[Rishi k. K R. <ris...@li...>]

The branch, maint, has been updated
       via  2784e9a6e36cacad4192f41a8c9044a85428ed27 (commit)
       via  00c8b3450d151482e9f7424224fc8a64bff27fdd (commit)
       via  a25f31b6f0a774b75ad2bf664f84ff0500086785 (commit)
       via  007dce57f645ad796f8cb60a13de936ac04fd1e8 (commit)
       via  a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f (commit)
       via  737694964966bcf7a807b335d51923ee9c9c27a5 (commit)
       via  ce944320bea9e3d7044a36c64ef2e1719312370d (commit)
       via  71256e18bb4d21d1b82c4031a1983cb72ef215aa (commit)
      from  b8ed5460176838d9b714dddbc17685983b4a8b9b (commit)


- Log -----------------------------------------------------------------
commit 2784e9a6e36cacad4192f41a8c9044a85428ed27
Merge: b8ed5460176838d9b714dddbc17685983b4a8b9b 00c8b3450d151482e9f7424224fc8a64bff27fdd
Author: Rishikesh K Rajak <ris...@li...>
Date:   Sat Mar 13 12:49:29 2010 +0530

    Merge branches 'master' and 'maint' into maint

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

Summary of changes:
 TODO                                               |    6 ++++++
 .../memcg/control/memcg_control_test.sh            |    2 +-
 testcases/kernel/security/filecaps/Makefile        |    2 +-
 testcases/kernel/syscalls/libevent/buffer.c        |    4 ++--
 4 files changed, 10 insertions(+), 4 deletions(-)
 
diff --git a/TODO b/TODO
index 1756168..5e289ca 100644
--- a/TODO
+++ b/TODO
@@ -32,6 +32,9 @@ testcases/ballista:
 3. Uses cshell for some weird arse reason, instead of using Bourne shell
    compat scripts.
 
+testcases/kernel/controllers/freezer/libcgroup_freezer:
+1. Uses pushd / popd, which isn't present in busybox.
+
 testcases/kernel/fs/dmapi/Makefile:
 1. This test requires JFS support, which most systems don't have. So a set of
    proper autoconf tests need to be added for this as well.
@@ -86,6 +89,9 @@ testcases/network/stress/udp/...:
    that all of the scripts are being linked to the file
    $(abs_top_srcdir)/testcases/bin .
 
+testcases/realtime/run.sh:
+1. Uses pushd / popd, which isn't present in busybox.
+
 tools/Makefile:
 1. netpipe* needs fixing and reorg, because it's referencing /home/mpich, etc.
 
diff --git a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
index 5f52a18..d926041 100644
--- a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
+++ b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
@@ -27,7 +27,7 @@
 ##                                                                            ##
 ################################################################################
 
-if [ "$(grep -w memory /proc/cgroups | cut -f4)" -ne "1" ]; then
+if [ "x$(grep -w memory /proc/cgroups | cut -f4)" != "x1" ]; then
 	echo "WARNING:"
 	echo "Either kernel does not support memory resource controller or feature not enabled"
 	echo "Skipping all memcg_control testcases...."
diff --git a/testcases/kernel/security/filecaps/Makefile b/testcases/kernel/security/filecaps/Makefile
index acad441..d6efeee 100644
--- a/testcases/kernel/security/filecaps/Makefile
+++ b/testcases/kernel/security/filecaps/Makefile
@@ -22,7 +22,7 @@ top_srcdir		?= ../../../..
 
 include $(top_srcdir)/include/mk/testcases.mk
 
-INSTALL_TARGETS		:= filecapstest.sh
+INSTALL_TARGETS		:= *.sh
 
 CPPFLAGS		+= -D_GNU_SOURCE
 DEBUG_CPPFLAGS		+= -DDEBUG=1
diff --git a/testcases/kernel/syscalls/libevent/buffer.c b/testcases/kernel/syscalls/libevent/buffer.c
index 5fae50c..583b812 100644
--- a/testcases/kernel/syscalls/libevent/buffer.c
+++ b/testcases/kernel/syscalls/libevent/buffer.c
@@ -37,8 +37,8 @@
 
 #include <err.h>
 #include <errno.h>
-/* FIXME (garrcoop): this is a bug according to the manpage; need this for
- * vasprintf(3). */
+/* FIXME (garrcoop): this wasn't present in the upstream sources [is
+ * requirement according to the manpage for vasprintf(3)]. */
 #ifndef _GNU_SOURCE
 #define _GNU_SOURCE
 #endif


hooks/post-receive
-- 
ltp

[Ltp-cvs] [Git] ltp updated.

[Rishi k. K R. <ris...@li...>]

The branch, pu, has been updated
       via  d05498f44de5d3e542b2031b458d84e907604728 (commit)
       via  00c8b3450d151482e9f7424224fc8a64bff27fdd (commit)
       via  a25f31b6f0a774b75ad2bf664f84ff0500086785 (commit)
       via  007dce57f645ad796f8cb60a13de936ac04fd1e8 (commit)
       via  a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f (commit)
       via  737694964966bcf7a807b335d51923ee9c9c27a5 (commit)
       via  ce944320bea9e3d7044a36c64ef2e1719312370d (commit)
       via  71256e18bb4d21d1b82c4031a1983cb72ef215aa (commit)
      from  615e1e37de053855b6bd56bae8b00d8a17133f8a (commit)


- Log -----------------------------------------------------------------
commit d05498f44de5d3e542b2031b458d84e907604728
Merge: 615e1e37de053855b6bd56bae8b00d8a17133f8a 00c8b3450d151482e9f7424224fc8a64bff27fdd
Author: Rishikesh K Rajak <ris...@li...>
Date:   Sat Mar 13 12:50:13 2010 +0530

    Merge branches 'master' and 'pu' into pu

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

Summary of changes:
 TODO                                               |    6 ++++++
 .../memcg/control/memcg_control_test.sh            |    2 +-
 testcases/kernel/security/filecaps/Makefile        |    2 +-
 testcases/kernel/syscalls/libevent/buffer.c        |    4 ++--
 4 files changed, 10 insertions(+), 4 deletions(-)
 
diff --git a/TODO b/TODO
index 1756168..5e289ca 100644
--- a/TODO
+++ b/TODO
@@ -32,6 +32,9 @@ testcases/ballista:
 3. Uses cshell for some weird arse reason, instead of using Bourne shell
    compat scripts.
 
+testcases/kernel/controllers/freezer/libcgroup_freezer:
+1. Uses pushd / popd, which isn't present in busybox.
+
 testcases/kernel/fs/dmapi/Makefile:
 1. This test requires JFS support, which most systems don't have. So a set of
    proper autoconf tests need to be added for this as well.
@@ -86,6 +89,9 @@ testcases/network/stress/udp/...:
    that all of the scripts are being linked to the file
    $(abs_top_srcdir)/testcases/bin .
 
+testcases/realtime/run.sh:
+1. Uses pushd / popd, which isn't present in busybox.
+
 tools/Makefile:
 1. netpipe* needs fixing and reorg, because it's referencing /home/mpich, etc.
 
diff --git a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
index 5f52a18..d926041 100644
--- a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
+++ b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
@@ -27,7 +27,7 @@
 ##                                                                            ##
 ################################################################################
 
-if [ "$(grep -w memory /proc/cgroups | cut -f4)" -ne "1" ]; then
+if [ "x$(grep -w memory /proc/cgroups | cut -f4)" != "x1" ]; then
 	echo "WARNING:"
 	echo "Either kernel does not support memory resource controller or feature not enabled"
 	echo "Skipping all memcg_control testcases...."
diff --git a/testcases/kernel/security/filecaps/Makefile b/testcases/kernel/security/filecaps/Makefile
index acad441..d6efeee 100644
--- a/testcases/kernel/security/filecaps/Makefile
+++ b/testcases/kernel/security/filecaps/Makefile
@@ -22,7 +22,7 @@ top_srcdir		?= ../../../..
 
 include $(top_srcdir)/include/mk/testcases.mk
 
-INSTALL_TARGETS		:= filecapstest.sh
+INSTALL_TARGETS		:= *.sh
 
 CPPFLAGS		+= -D_GNU_SOURCE
 DEBUG_CPPFLAGS		+= -DDEBUG=1
diff --git a/testcases/kernel/syscalls/libevent/buffer.c b/testcases/kernel/syscalls/libevent/buffer.c
index 5fae50c..583b812 100644
--- a/testcases/kernel/syscalls/libevent/buffer.c
+++ b/testcases/kernel/syscalls/libevent/buffer.c
@@ -37,8 +37,8 @@
 
 #include <err.h>
 #include <errno.h>
-/* FIXME (garrcoop): this is a bug according to the manpage; need this for
- * vasprintf(3). */
+/* FIXME (garrcoop): this wasn't present in the upstream sources [is
+ * requirement according to the manpage for vasprintf(3)]. */
 #ifndef _GNU_SOURCE
 #define _GNU_SOURCE
 #endif


hooks/post-receive
-- 
ltp

[Ltp-cvs] [Git] ltp updated.

[Rishi k. K R. <ris...@li...>]

The branch, next, has been updated
       via  b80fdefbd1843484e5a3ccc3d5748fee5b4cd1f7 (commit)
       via  00c8b3450d151482e9f7424224fc8a64bff27fdd (commit)
       via  a25f31b6f0a774b75ad2bf664f84ff0500086785 (commit)
       via  007dce57f645ad796f8cb60a13de936ac04fd1e8 (commit)
       via  a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f (commit)
       via  71256e18bb4d21d1b82c4031a1983cb72ef215aa (commit)
      from  737694964966bcf7a807b335d51923ee9c9c27a5 (commit)


- Log -----------------------------------------------------------------
commit b80fdefbd1843484e5a3ccc3d5748fee5b4cd1f7
Author: Rishikesh K Rajak <ris...@li...>
Date:   Sat Mar 13 12:47:38 2010 +0530

    Merge branches 'master' and 'next' into next

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

Summary of changes:
 TODO                                               |    6 ++++++
 .../memcg/control/memcg_control_test.sh            |    2 +-
 testcases/kernel/syscalls/libevent/buffer.c        |    4 ++--
 3 files changed, 9 insertions(+), 3 deletions(-)
 
diff --git a/TODO b/TODO
index 1756168..5e289ca 100644
--- a/TODO
+++ b/TODO
@@ -32,6 +32,9 @@ testcases/ballista:
 3. Uses cshell for some weird arse reason, instead of using Bourne shell
    compat scripts.
 
+testcases/kernel/controllers/freezer/libcgroup_freezer:
+1. Uses pushd / popd, which isn't present in busybox.
+
 testcases/kernel/fs/dmapi/Makefile:
 1. This test requires JFS support, which most systems don't have. So a set of
    proper autoconf tests need to be added for this as well.
@@ -86,6 +89,9 @@ testcases/network/stress/udp/...:
    that all of the scripts are being linked to the file
    $(abs_top_srcdir)/testcases/bin .
 
+testcases/realtime/run.sh:
+1. Uses pushd / popd, which isn't present in busybox.
+
 tools/Makefile:
 1. netpipe* needs fixing and reorg, because it's referencing /home/mpich, etc.
 
diff --git a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
index 5f52a18..d926041 100644
--- a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
+++ b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
@@ -27,7 +27,7 @@
 ##                                                                            ##
 ################################################################################
 
-if [ "$(grep -w memory /proc/cgroups | cut -f4)" -ne "1" ]; then
+if [ "x$(grep -w memory /proc/cgroups | cut -f4)" != "x1" ]; then
 	echo "WARNING:"
 	echo "Either kernel does not support memory resource controller or feature not enabled"
 	echo "Skipping all memcg_control testcases...."
diff --git a/testcases/kernel/syscalls/libevent/buffer.c b/testcases/kernel/syscalls/libevent/buffer.c
index 5fae50c..583b812 100644
--- a/testcases/kernel/syscalls/libevent/buffer.c
+++ b/testcases/kernel/syscalls/libevent/buffer.c
@@ -37,8 +37,8 @@
 
 #include <err.h>
 #include <errno.h>
-/* FIXME (garrcoop): this is a bug according to the manpage; need this for
- * vasprintf(3). */
+/* FIXME (garrcoop): this wasn't present in the upstream sources [is
+ * requirement according to the manpage for vasprintf(3)]. */
 #ifndef _GNU_SOURCE
 #define _GNU_SOURCE
 #endif


hooks/post-receive
-- 
ltp

[Ltp-cvs] [Git] ltp updated.

[Garrett C. <ris...@li...>]

The branch, master, has been updated
       via  00c8b3450d151482e9f7424224fc8a64bff27fdd (commit)
       via  a25f31b6f0a774b75ad2bf664f84ff0500086785 (commit)
       via  007dce57f645ad796f8cb60a13de936ac04fd1e8 (commit)
       via  71256e18bb4d21d1b82c4031a1983cb72ef215aa (commit)
      from  a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f (commit)


- Log -----------------------------------------------------------------
commit 00c8b3450d151482e9f7424224fc8a64bff27fdd
Merge: a25f31b6f0a774b75ad2bf664f84ff0500086785 a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f
Author: Garrett Cooper <yan...@gm...>
Date:   Wed Mar 10 20:29:15 2010 -0800

    Merge branch 'master' of ssh://yab...@lt.../gitroot/ltp/ltp-dev

commit a25f31b6f0a774b75ad2bf664f84ff0500086785
Author: Garrett Cooper <yan...@gm...>
Date:   Wed Mar 10 20:28:33 2010 -0800

    Update TODO with note about needing to replace pushd and popd logic in a few tests.
    
    Signed-off-by: Garrett Cooper <yan...@gm...>

commit 007dce57f645ad796f8cb60a13de936ac04fd1e8
Author: Garrett Cooper <yan...@gm...>
Date:   Wed Mar 10 20:26:15 2010 -0800

    Hi,
    
    I found following problem about "memcg_control" test.
    ---------
    /opt/ltp/testcases/bin/memcg_control_test.sh: line 30: [: : integer
    expression expected
    ----------
    
    This happens because of following process.
    [ memcg_control_test.sh: line 30 ]
    ---------
    if [ "$(grep -w memory /proc/cgroups | cut -f4)" -ne "1" ]; then
           echo "WARNING:"
           echo "Either kernel does not support memory resource controller or
    feature not ena
    bled"
           echo "Skipping all memcg_control testcases...."
           exit 0
    fi
    ---------
    
    "$(grep -w memory /proc/cgroups | cut -f4)" can't comparable with "1"
    if "$(grep -w memory /proc/cgroups | cut -f4)" = nothing.
    
    How about fixing like the other "memcg" tests?
    
    Signed-off-by: Yuki Yao<y_...@ry...>
    Acked-by: Garrett Cooper <yan...@gm...>

commit 71256e18bb4d21d1b82c4031a1983cb72ef215aa
Author: Garrett Cooper <yan...@gm...>
Date:   Mon Mar 8 08:53:05 2010 -0800

    Clean up comment to note the actual problem. It was confusing before.
    
    Signed-off-by: Garrett Cooper <yan...@gm...>

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

Summary of changes:
 TODO                                               |    6 ++++++
 .../memcg/control/memcg_control_test.sh            |    2 +-
 testcases/kernel/syscalls/libevent/buffer.c        |    4 ++--
 3 files changed, 9 insertions(+), 3 deletions(-)
 
diff --git a/TODO b/TODO
index 1756168..5e289ca 100644
--- a/TODO
+++ b/TODO
@@ -32,6 +32,9 @@ testcases/ballista:
 3. Uses cshell for some weird arse reason, instead of using Bourne shell
    compat scripts.
 
+testcases/kernel/controllers/freezer/libcgroup_freezer:
+1. Uses pushd / popd, which isn't present in busybox.
+
 testcases/kernel/fs/dmapi/Makefile:
 1. This test requires JFS support, which most systems don't have. So a set of
    proper autoconf tests need to be added for this as well.
@@ -86,6 +89,9 @@ testcases/network/stress/udp/...:
    that all of the scripts are being linked to the file
    $(abs_top_srcdir)/testcases/bin .
 
+testcases/realtime/run.sh:
+1. Uses pushd / popd, which isn't present in busybox.
+
 tools/Makefile:
 1. netpipe* needs fixing and reorg, because it's referencing /home/mpich, etc.
 
diff --git a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
index 5f52a18..d926041 100644
--- a/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
+++ b/testcases/kernel/controllers/memcg/control/memcg_control_test.sh
@@ -27,7 +27,7 @@
 ##                                                                            ##
 ################################################################################
 
-if [ "$(grep -w memory /proc/cgroups | cut -f4)" -ne "1" ]; then
+if [ "x$(grep -w memory /proc/cgroups | cut -f4)" != "x1" ]; then
 	echo "WARNING:"
 	echo "Either kernel does not support memory resource controller or feature not enabled"
 	echo "Skipping all memcg_control testcases...."
diff --git a/testcases/kernel/syscalls/libevent/buffer.c b/testcases/kernel/syscalls/libevent/buffer.c
index 5fae50c..583b812 100644
--- a/testcases/kernel/syscalls/libevent/buffer.c
+++ b/testcases/kernel/syscalls/libevent/buffer.c
@@ -37,8 +37,8 @@
 
 #include <err.h>
 #include <errno.h>
-/* FIXME (garrcoop): this is a bug according to the manpage; need this for
- * vasprintf(3). */
+/* FIXME (garrcoop): this wasn't present in the upstream sources [is
+ * requirement according to the manpage for vasprintf(3)]. */
 #ifndef _GNU_SOURCE
 #define _GNU_SOURCE
 #endif


hooks/post-receive
-- 
ltp

[Ltp-cvs] [Git] ltp updated.

[Rishi k. K R. <ris...@li...>]

The branch, master, has been updated
       via  a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f (commit)
       via  737694964966bcf7a807b335d51923ee9c9c27a5 (commit)
       via  ce944320bea9e3d7044a36c64ef2e1719312370d (commit)
      from  1a5506c97dcfda0bd16f83ca855bb633c47eea4c (commit)


- Log -----------------------------------------------------------------
commit a3bb7abd2c1fb20dc4d93bef9cda2980f4339a6f
Author: Rishikesh K Rajak <ris...@li...>
Date:   Wed Mar 10 18:34:57 2010 +0530

    Merge branches 'next' and 'master'

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

Summary of changes:
 testcases/kernel/security/filecaps/Makefile |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
 
diff --git a/testcases/kernel/security/filecaps/Makefile b/testcases/kernel/security/filecaps/Makefile
index acad441..d6efeee 100644
--- a/testcases/kernel/security/filecaps/Makefile
+++ b/testcases/kernel/security/filecaps/Makefile
@@ -22,7 +22,7 @@ top_srcdir		?= ../../../..
 
 include $(top_srcdir)/include/mk/testcases.mk
 
-INSTALL_TARGETS		:= filecapstest.sh
+INSTALL_TARGETS		:= *.sh
 
 CPPFLAGS		+= -D_GNU_SOURCE
 DEBUG_CPPFLAGS		+= -DDEBUG=1


hooks/post-receive
-- 
ltp

[Ltp-cvs] [Git] ltp updated.

[Rishi k. K R. <ris...@li...>]

The branch, next, has been updated
       via  737694964966bcf7a807b335d51923ee9c9c27a5 (commit)
      from  ce944320bea9e3d7044a36c64ef2e1719312370d (commit)


- Log -----------------------------------------------------------------
commit 737694964966bcf7a807b335d51923ee9c9c27a5
Author: Rishikesh K Rajak <ris...@li...>
Date:   Wed Mar 10 18:32:35 2010 +0530

    To run filecap test, checkforlibcap.sh file is required to be there in /opt/ltp/testcases/bin
    
    Otherwise you will get this error when you run filecap test.
    
    >>>>
    /opt/ltp/testcases/bin/filecapstest.sh: line 22: checkforlibcap.sh: command not found
    <<<<
    
    Signed-off-by: Rishikesh K Rajak <ris...@li...>

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

Summary of changes:
 testcases/kernel/security/filecaps/Makefile |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
 
diff --git a/testcases/kernel/security/filecaps/Makefile b/testcases/kernel/security/filecaps/Makefile
index acad441..d6efeee 100644
--- a/testcases/kernel/security/filecaps/Makefile
+++ b/testcases/kernel/security/filecaps/Makefile
@@ -22,7 +22,7 @@ top_srcdir		?= ../../../..
 
 include $(top_srcdir)/include/mk/testcases.mk
 
-INSTALL_TARGETS		:= filecapstest.sh
+INSTALL_TARGETS		:= *.sh
 
 CPPFLAGS		+= -D_GNU_SOURCE
 DEBUG_CPPFLAGS		+= -DDEBUG=1


hooks/post-receive
-- 
ltp

[Ltp-cvs] utils/analysis/lcov/bin geninfo,1.73,1.74

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv4520

Modified Files:
	geninfo 
Log Message:
lcov: fix problem with relative paths in build paths

When binaries are built using relative paths, lcov cannot find any coverage
data. Instead, warnings similar to the following are printed:

geninfo: WARNING: cannot find an entry for ^#src#test.c.gcov in .gcno file, skipping file!

The reason for this is that File::Spec::rel2abs does not remove ../ from paths
which results in lcov not being able to match the relative and absolute versions
of the corresponding filenames. Fix this by using the internal function
solve_relative_path instead.


Index: geninfo
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/geninfo,v
retrieving revision 1.73
retrieving revision 1.74
diff -C2 -d -r1.73 -r1.74
*** geninfo	21 Feb 2010 14:57:52 -0000	1.73
--- geninfo	28 Feb 2010 20:57:37 -0000	1.74
***************
*** 53,57 ****
  use File::Basename; 
  use File::Spec::Functions qw /abs2rel catdir file_name_is_absolute splitdir
! 			      splitpath rel2abs/;
  use Getopt::Long;
  use Digest::MD5 qw(md5_base64);
--- 53,57 ----
  use File::Basename; 
  use File::Spec::Functions qw /abs2rel catdir file_name_is_absolute splitdir
! 			      splitpath/;
  use Getopt::Long;
  use Digest::MD5 qw(md5_base64);
***************
*** 2443,2447 ****
  			goto incomplete if (!defined($filename));
  			if ($filename ne "") {
! 				$filename = rel2abs($filename, $base);
  			}
  		} elsif ($value == $minus_two) {
--- 2443,2448 ----
  			goto incomplete if (!defined($filename));
  			if ($filename ne "") {
! 				$filename = solve_relative_path($base,
! 								$filename);
  			}
  		} elsif ($value == $minus_two) {
***************
*** 2564,2568 ****
  				return $filename;
  			}
! 			$filename = rel2abs($string, $base);
  			next;
  		}
--- 2565,2569 ----
  				return $filename;
  			}
! 			$filename = solve_relative_path($base, $string);
  			next;
  		}
***************
*** 2758,2762 ****
  				return $filename;
  			}
! 			$filename = rel2abs($string, $base);
  			next;
  		}
--- 2759,2763 ----
  				return $filename;
  			}
! 			$filename = solve_relative_path($base, $string);
  			next;
  		}
***************
*** 2799,2803 ****
  	$filename = read_gcno_string($handle, $big_endian);
  	return undef if (!defined($filename));
! 	$filename = rel2abs($filename, $base);
  	# Read first line number
  	$lineno = read_gcno_value($handle, $big_endian, "initial line number");
--- 2800,2804 ----
  	$filename = read_gcno_string($handle, $big_endian);
  	return undef if (!defined($filename));
! 	$filename = solve_relative_path($base, $filename);
  	# Read first line number
  	$lineno = read_gcno_value($handle, $big_endian, "initial line number");

[Ltp-cvs] utils/analysis/lcov/bin geninfo,1.72,1.73

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv19108

Modified Files:
	geninfo 
Log Message:
geninfo: write all debugging output to STDERR


Index: geninfo
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/geninfo,v
retrieving revision 1.72
retrieving revision 1.73
diff -C2 -d -r1.72 -r1.73
*** geninfo	21 Feb 2010 14:56:46 -0000	1.72
--- geninfo	21 Feb 2010 14:57:52 -0000	1.73
***************
*** 2928,2932 ****
  
  	return if (!$debug);
! 	print("DEBUG: $msg");
  }
  
--- 2928,2932 ----
  
  	return if (!$debug);
! 	print(STDERR "DEBUG: $msg");
  }
  

[Ltp-cvs] utils/analysis/lcov/bin geninfo,1.71,1.72

[Peter O. <obe...@us...>]

Update of /cvsroot/ltp/utils/analysis/lcov/bin
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv18962

Modified Files:
	geninfo 
Log Message:
geninfo: fix problem with some .gcno files

Some .gcno files contain more data in a line record than
expected. Skip unhandled bytes of a .gcno file record.
This prevents the following unexpected error message:

geninfo: ERROR: file.gcno: reached unexpected end of file


Index: geninfo
===================================================================
RCS file: /cvsroot/ltp/utils/analysis/lcov/bin/geninfo,v
retrieving revision 1.71
retrieving revision 1.72
diff -C2 -d -r1.71 -r1.72
*** geninfo	29 Jan 2010 11:07:25 -0000	1.71
--- geninfo	21 Feb 2010 14:56:46 -0000	1.72
***************
*** 2858,2861 ****
--- 2858,2864 ----
  	graph_skip(*HANDLE, 8, "version and stamp") or goto incomplete;
  	while (!eof(HANDLE)) {
+ 		my $next_pos;
+ 		my $curr_pos;
+ 
  		# Read record tag
  		$tag = read_gcno_value(*HANDLE, $big_endian, "record tag");
***************
*** 2867,2870 ****
--- 2870,2878 ----
  		# Convert length to bytes
  		$length *= 4;
+ 		# Calculate start of next record
+ 		$next_pos = tell(HANDLE);
+ 		goto tell_error if ($next_pos == -1);
+ 		$next_pos += $length;
+ 		# Process record
  		if ($tag == $tag_function) {
  			($filename, $function) = read_gcno_function_record(
***************
*** 2883,2886 ****
--- 2891,2902 ----
  				or goto incomplete;
  		}
+ 		# Ensure that we are at the start of the next record
+ 		$curr_pos = tell(HANDLE);
+ 		goto tell_error if ($curr_pos == -1);
+ 		next if ($curr_pos == $next_pos);
+ 		goto record_error if ($curr_pos > $next_pos);
+ 		graph_skip(*HANDLE, $next_pos - $curr_pos,
+ 			   "unhandled record content")
+ 			or goto incomplete;
  	}
  	close(HANDLE);
***************
*** 2899,2902 ****
--- 2915,2924 ----
  	graph_error($gcno_filename, "found unrecognized gcno file magic");
  	return undef;
+ tell_error:
+ 	graph_error($gcno_filename, "could not determine file position");
+ 	return undef;
+ record_error:
+ 	graph_error($gcno_filename, "found unrecognized record format");
+ 	return undef;
  }
  

[Ltp-cvs] ltp/testcases/kernel/syscalls/mknod mknod02.c,1.7,1.8

[Rishi k. K R. <ris...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/syscalls/mknod
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv28705/testcases/kernel/syscalls/mknod

Modified Files:
	mknod02.c 
Log Message:
>From dd04033034f9e550029178f9d85c95f411ef1009 Mon Sep 17 00:00:00 2001
From: gcooper <gcooper@orangebox.(none)>
Date: Sun, 14 Feb 2010 23:16:22 -0800
Subject: [PATCH 1/1] Correct at least one of these lousy diagnostic messages for not finding
 nobody in /etc/passwd...

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: mknod02.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/syscalls/mknod/mknod02.c,v
retrieving revision 1.7
retrieving revision 1.8
diff -C2 -d -r1.7 -r1.8
*** mknod02.c	23 Mar 2009 13:35:54 -0000	1.7
--- mknod02.c	15 Feb 2010 09:08:44 -0000	1.8
***************
*** 245,249 ****
  	/* Get the uid/gid of ltpuser */
  	if ((user1 = getpwnam(LTPUSER)) == NULL) {
! 		tst_brkm(TBROK, cleanup, "%s not in /etc/passwd", LTPUSER);
  	}
  	user1_uid = user1->pw_uid;
--- 245,250 ----
  	/* Get the uid/gid of ltpuser */
  	if ((user1 = getpwnam(LTPUSER)) == NULL) {
! 		tst_brkm(TBROK | TERRNO, cleanup,
! 			"Couldn't determine if %s was in /etc/passwd", LTPUSER);
  	}
  	user1_uid = user1->pw_uid;

[Ltp-cvs] ltp foo,1.2,1.3

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv6330

Added Files:
	foo 
Log Message:


--- NEW FILE: foo ---

[Ltp-cvs] ltp foo,1.1,NONE

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv5086

Removed Files:
	foo 
Log Message:
Remove test file.



--- foo DELETED ---

[Ltp-cvs] ltp foo,NONE,1.1

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv5066

Added Files:
	foo 
Log Message:
Test.



--- NEW FILE: foo ---

[Ltp-cvs] ltp/testcases/kernel/containers/netns check_netns_enabled.c, 1.2, 1.3

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/containers/netns
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv446/testcases/kernel/containers/netns

Modified Files:
	check_netns_enabled.c 
Log Message:
1. Simplify smoke test further by only checking to see whether or not the syscall works as opposed to checking the kernel version because people can backport support as seen fit and shouldn't be hindered by a stupid kernel version check...
2. Correct compile because I originally checked in the change on a box without a proper toolchain and thus couldn't build before now.

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: check_netns_enabled.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/containers/netns/check_netns_enabled.c,v
retrieving revision 1.2
retrieving revision 1.3
diff -C2 -d -r1.2 -r1.3
*** check_netns_enabled.c	10 Feb 2010 23:12:53 -0000	1.2
--- check_netns_enabled.c	10 Feb 2010 23:24:45 -0000	1.3
***************
*** 23,30 ****
  #include <sched.h>
  #include "config.h"
  #include "linux_syscall_numbers.h"
- #include "libclone/libclone.h"
  #include "test.h"
  
  #ifndef CLONE_NEWNET
  #define CLONE_NEWNET -1
--- 23,34 ----
  #include <sched.h>
  #include "config.h"
+ #include "libclone.h"
  #include "linux_syscall_numbers.h"
  #include "test.h"
  
+ char *TCID = "check_netns_enabled";
+ int TST_COUNT = 1;
+ int TST_TOTAL = 1; 
+ 
  #ifndef CLONE_NEWNET
  #define CLONE_NEWNET -1
***************
*** 38,52 ****
  main()
  {
! 	int rc;
! 
! 	/* Checking if the kernel ver is enough to do NET-NS testing. */
! 	if (tst_kvercmp(2, 6, 24) < 0) {
! 		rc = 1;
!         } else if (syscall(__NR_unshare, CLONE_NEWNET | CLONE_NEWNS) < 0) {
! 		printf ("Error:Unshare syscall failed for network namespace\n");
! 		rc = 3;
! 	} else {
! 		rc = 0;
  	}
! 	return rc;
  }
--- 42,49 ----
  main()
  {
! 	/* Checking if the kernel supports unshare with netns capabilities. */
!         if (syscall(__NR_unshare, CLONE_NEWNET | CLONE_NEWNS) < 0) {
! 		tst_resm (TFAIL | TERRNO, "unshare syscall smoke test failed");
  	}
! 	tst_exit();
  }

[Ltp-cvs] ltp/testcases/kernel/containers/netns check_netns_enabled.c, 1.1, 1.2

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/containers/netns
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv31942/testcases/kernel/containers/netns

Modified Files:
	check_netns_enabled.c 
Log Message:
Not everyone has unshare capability... thanks for the reminder Mitani-san...

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: check_netns_enabled.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/containers/netns/check_netns_enabled.c,v
retrieving revision 1.1
retrieving revision 1.2
diff -C2 -d -r1.1 -r1.2
*** check_netns_enabled.c	19 Sep 2008 12:17:10 -0000	1.1
--- check_netns_enabled.c	10 Feb 2010 23:12:53 -0000	1.2
***************
*** 22,46 ****
  #include <stdio.h>
  #include <sched.h>
! #include "../libclone/libclone.h"
  #include "test.h"
  
! int main()
! {
!     int ret;
!     long flags = 0;
! 
!     flags |= CLONE_NEWNS;
!     flags |= CLONE_NEWNET;
  
  
! 	// Checking if the kernel ver is enough to do NET-NS testing.
! 	if (tst_kvercmp(2,6,24) < 0)
! 		return 1;
  
!         ret = unshare(flags);
! 	if ( ret < 0 ) {
  		printf ("Error:Unshare syscall failed for network namespace\n");
! 		return 3;
  	}
! 	return 0;
  }
--- 22,52 ----
  #include <stdio.h>
  #include <sched.h>
! #include "config.h"
! #include "linux_syscall_numbers.h"
! #include "libclone/libclone.h"
  #include "test.h"
  
! #ifndef CLONE_NEWNET
! #define CLONE_NEWNET -1
! #endif
  
+ #ifndef CLONE_NEWNS
+ #define CLONE_NEWNS -1
+ #endif
  
! int
! main()
! {
! 	int rc;
  
! 	/* Checking if the kernel ver is enough to do NET-NS testing. */
! 	if (tst_kvercmp(2, 6, 24) < 0) {
! 		rc = 1;
!         } else if (syscall(__NR_unshare, CLONE_NEWNET | CLONE_NEWNS) < 0) {
  		printf ("Error:Unshare syscall failed for network namespace\n");
! 		rc = 3;
! 	} else {
! 		rc = 0;
  	}
! 	return rc;
  }

[Ltp-cvs] ltp/testcases/kernel/security/cap_bound exec_without_inh.c, 1.4, 1.5

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/security/cap_bound
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv30635/testcases/kernel/security/cap_bound

Modified Files:
	exec_without_inh.c 
Log Message:
Fixed typo found by Mitani-san.

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: exec_without_inh.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/security/cap_bound/exec_without_inh.c,v
retrieving revision 1.4
retrieving revision 1.5
diff -C2 -d -r1.4 -r1.5
*** exec_without_inh.c	9 Feb 2010 14:49:02 -0000	1.4
--- exec_without_inh.c	10 Feb 2010 23:04:09 -0000	1.5
***************
*** 79,83 ****
  			tst_resm(TBROK, "Failed to drop cap_sys_admin from pI\n");
  			tst_exit();
! 		i}
  	} else if (ret) {
  		tst_brkm(TBROK | TERRNO, tst_exit, "Failed to add \
--- 79,83 ----
  			tst_resm(TBROK, "Failed to drop cap_sys_admin from pI\n");
  			tst_exit();
! 		}
  	} else if (ret) {
  		tst_brkm(TBROK | TERRNO, tst_exit, "Failed to add \

[Ltp-cvs] ltp/runtest syscalls,1.307,1.308

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/runtest
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv20782/runtest

Modified Files:
	syscalls 
Log Message:
Adding runtest entries for rt_sigaction0[23], as noted by Mitani-san, and forgotten by your's truly...

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: syscalls
===================================================================
RCS file: /cvsroot/ltp/ltp/runtest/syscalls,v
retrieving revision 1.307
retrieving revision 1.308
diff -C2 -d -r1.307 -r1.308
*** syscalls	15 Dec 2009 19:32:18 -0000	1.307
--- syscalls	10 Feb 2010 01:28:25 -0000	1.308
***************
*** 802,805 ****
--- 802,807 ----
  
  rt_sigaction01 rt_sigaction01
+ rt_sigaction02 rt_sigaction02
+ rt_sigaction03 rt_sigaction03
  rt_sigprocmask01 rt_sigprocmask01
  rt_sigprocmask02 rt_sigprocmask02

[Ltp-cvs] ltp/testcases/kernel/security/cap_bound exec_without_inh.c, 1.3, 1.4

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/security/cap_bound
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv21663/kernel/security/cap_bound

Modified Files:
	exec_without_inh.c 
Log Message:
An error occured on my i386 box.
(OS: Fedora8, Kernel: 2.6.33-rc6, libcap: libcap-2.16)
-------------
exec_without_inh    1  TFAIL  :  Failed to drop CAP_SYS_ADMIN from bounding set.
exec_without_inh    0  TINFO  :  (ret=-1, errno 38)
-------------
If the macro HAVE_DECL_CAP_BSET_DROP is 0, this error will occurs.
But the program should to be broken when f != CAP_SET, like the similar code 
"if (ret || f != CAP_SET) {" in exec_with_inh.c.

The TBROK message will be outputted if my patch merged.
-------------
exec_without_inh    1  TBROK  :  Failed to add CAP_SYS_ADMIN to pI
-------------

Signed-off-by: Shi Weihua <sh...@cn...>
---



Index: exec_without_inh.c
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/security/cap_bound/exec_without_inh.c,v
retrieving revision 1.3
retrieving revision 1.4
diff -C2 -d -r1.3 -r1.4
*** exec_without_inh.c	9 Oct 2009 17:55:50 -0000	1.3
--- exec_without_inh.c	9 Feb 2010 14:49:02 -0000	1.4
***************
*** 79,83 ****
  			tst_resm(TBROK, "Failed to drop cap_sys_admin from pI\n");
  			tst_exit();
! 		}
  	}
  #if HAVE_DECL_CAP_FREE
--- 79,86 ----
  			tst_resm(TBROK, "Failed to drop cap_sys_admin from pI\n");
  			tst_exit();
! 		i}
! 	} else if (ret) {
! 		tst_brkm(TBROK | TERRNO, tst_exit, "Failed to add \
! 			CAP_SYS_ADMIN to pI");
  	}
  #if HAVE_DECL_CAP_FREE

[Ltp-cvs] ltp/testcases/kernel/containers container_test.sh, 1.9, 1.10

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/containers
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv21088

Modified Files:
	container_test.sh 
Log Message:
1. Remove mqns check app check. It no longer exists in the repo and thus gates these tests unnecessarily from being run..
2. Compile the netns check app.

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: container_test.sh
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/containers/container_test.sh,v
retrieving revision 1.9
retrieving revision 1.10
diff -C2 -d -r1.9 -r1.10
*** container_test.sh	6 Feb 2010 06:39:57 -0000	1.9
--- container_test.sh	9 Feb 2010 14:45:29 -0000	1.10
***************
*** 50,61 ****
  runpidnstest.sh
  
! check_mqns_enabled
! if [ $? -eq 0 ]; then
! 	echo "Running POSIX message queue tests."
! 	runmqnstest.sh
! else
! 	echo "Posix message queues or ipc namespaces not enabled in kernel."
! 	echo "Not running mqns tests."
! fi
  
  check_netns_enabled
--- 50,55 ----
  runpidnstest.sh
  
! echo "Running POSIX message queue tests."
! runmqnstest.sh
  
  check_netns_enabled

[Ltp-cvs] ltp/testcases/kernel/containers/netns Makefile,1.4,1.5

[Garrett C. <yab...@us...>]

Update of /cvsroot/ltp/ltp/testcases/kernel/containers/netns
In directory sfp-cvsdas-2.v30.ch3.sourceforge.com:/tmp/cvs-serv21088/netns

Modified Files:
	Makefile 
Log Message:
1. Remove mqns check app check. It no longer exists in the repo and thus gates these tests unnecessarily from being run..
2. Compile the netns check app.

Signed-off-by: Garrett Cooper <yan...@gm...>



Index: Makefile
===================================================================
RCS file: /cvsroot/ltp/ltp/testcases/kernel/containers/netns/Makefile,v
retrieving revision 1.4
retrieving revision 1.5
diff -C2 -d -r1.4 -r1.5
*** Makefile	9 Oct 2009 17:55:46 -0000	1.4
--- Makefile	9 Feb 2010 14:45:29 -0000	1.5
***************
*** 32,34 ****
--- 32,36 ----
  $(MAKE_TARGETS): %: common.o %.o
  
+ MAKE_TARGETS		+= check_netns_enabled
+ 
  include $(top_srcdir)/include/mk/generic_leaf_target.mk