Git Log


Commit Date  
[a07c78] by subrata_modak subrata_modak

Testcase to test cpu consolidation feature of power management: Testcase Description: When sched_mc_power_saving is set to 1 and when number of threads running in the system is less then or equal to number of cores in the package, all the threads should be running in cores belonging to one package. This automated testcase triggers ebizzy with number of threads equal to number of cores in a package. Identifies cpu's utilized and verifies if all the cpu's utilized belong to same package.
[PATCH 01/05]: This patch will integrate ebizzy into LTP in utils/benchmark path. Ebizzy is used to validate cpu consolidation when sched_mc_power_saving is set to 1. Signed-Off-By: Poornima Nayak <Poornima.Nayak@in.ibm.com>.

2009-02-28 18:10:15 Tree
[576c3b] by subrata_modak subrata_modak

I look at testcases/kernel/include/ and there is no microblaze.in. I am not cvs expert but I hope that is possible to create only simple file. If not please create it with one space. This should be fix before next release. Michal Simek <michal.simek@petalogix.com>.

2009-02-27 10:26:34 Tree
[855766] by subrata_modak subrata_modak

Containers: Remove capability dependency: Review of these test source files suggests that absolutley no capability APIs are in use. I suspect these #ifdefs were added because the original tests had unnecessarily included sys/capability.h. Signed-off-by: Matt Helsley <matthltc@us.ibm.com>.

2009-02-26 13:13:38 Tree
[f58488] by subrata_modak subrata_modak

Update unzip command test: build and run successfully without unzip: Suprisingly, not every Linux distribution comes with unzip installed. Thankfully unzip does not appear to be needed for any other tests. This patch cleans up the unzip test so that if zip or unzip does not exist the test will not be built or run. This patch changes the way the build and run portions of the script are spread across the Makefile and unzip_tests.sh script. Previously part of the build was performed during the test -- building the sample zip archive. Now that sample archive is conditionally built using Make instead of shell script. Also, as near as I can tell, the comments "describing" the test are totally incorrect -- they appear to be a copy-paste artifact. So I updated them. Signed-off-by: Matt Helsley <matthltc@us.ibm.com>.

2009-02-26 13:10:54 Tree
[55929b] by subrata_modak subrata_modak

Fix spelling and grammar nits. Signed-off-by: Matt Helsley <matthltc@us.ibm.com>.

2009-02-26 13:09:05 Tree
[f20da5] by subrata_modak subrata_modak

Rename description files. While working on adding a new testcase I noticed that not all of the description files are named consistently. I guess that 00_ was meant to place it first in any directory listing so presumably that's the correct description name. Signed-off-by: Matt Helsley <matthltc@us.ibm.com>.

2009-02-26 13:07:48 Tree
[bdbaec] by subrata_modak subrata_modak

Clean Trailing Tab: Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 12:14:51 Tree
[4bb656] by subrata_modak subrata_modak

Clear Trailing Whitespace. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 12:02:09 Tree
[a92630] by subrata_modak subrata_modak

Fix return(NULL) -> return NULL. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:48:09 Tree
[134e89] by subrata_modak subrata_modak

fix return(1) -> return 1. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:46:54 Tree
[43337a] by subrata_modak subrata_modak

Fix return(0) to return 0. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:43:51 Tree
[3b42e0] by subrata_modak subrata_modak

uclinux doens't use remap_file_page. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:37:28 Tree
[cd5a78] by subrata_modak subrata_modak

fix uclinux build. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:36:24 Tree
[10de65] by subrata_modak subrata_modak

Use make install for uclinux too. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:34:49 Tree
[6c5710] by subrata_modak subrata_modak

uclinux: It is necessary to compile include. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:32:30 Tree
[240ac9] by subrata_modak subrata_modak

Add microblaze support for linux_syscall_number.h. Signed-off-by: Michal Simek <monstr@monstr.eu>.

2009-02-26 11:22:41 Tree
[5d8dc7] by subrata_modak subrata_modak

Add Freezer Controller Testcases: These testcases cover the basic functionality of the kernel cgroup freezer controller described in the kernel's Documentation/cgroups/freezer-subsystem.txt file. Briefly:
The freezer subsystem in the cgroup filesystem defines a file named freezer.state. Writing "FROZEN" to the state file will freeze all tasks in the cgroup or return with errno EBUSY if the group can't be frozen immediately (use /bin/echo if you wish to check exit codes). After a successful write of "FROZEN" writing "THAWED" will unfreeze the tasks in the cgroup. Reading will return the current state. Note that freezer.state doesn't exist in the root cgroup, which means the root cgroup is non-freezable.
* Sample usage:
$ mkdir /containers
$ mount -t cgroup -ofreezer freezer /containers
$ mkdir /containers/0
$ echo $some_pid > /containers/0/tasks
$ cat /containers/0/tasks/freezer.state
THAWED
$ echo FROZEN > /containers/0/tasks/freezer.state
<IO Error (busy)>
$ cat /containers/0/tasks/freezer.state
FREEZING
$ sleep 0.2
$ echo FROZEN > /containers/0/tasks/freezer.state
$ cat /containers/0/tasks/freezer.state
FROZEN
$ echo FROZEN > /containers/0/tasks/freezer.state
$ cat /containers/0/tasks/freezer.state
THAWED
The bulk of the code is in libcgroup_freezer. Early tests cover the initialization used for the bulk of the tests. Each testcase is a brief shell script composed of your typical three phases:
Initialization
sourcing libcgroup_freezer
mounting the cgroup filesystem
making a sample cgroup
making a sample process [optional]
Test body
Actions:
freeze, thaw, move to (new) cgroups
try to fork new processes
signal sequences
Assertions interleaved with the actions:
cgroup/task existence
cgroup/task state
cgroup membership
Cleanup
The patch runs these tests when the other cgroup controller tests run but is otherwise independent of the infrastructure offered by those controller tests. Also, these tests may be carefully run by hand or via the run.sh script provided. Signed-off-by: Matt Helsley <matthltc@us.ibm.com>.

2009-02-26 07:19:28 Tree
[26e07d] by subrata_modak subrata_modak

I'm a user of LTP. I use LTP for testing SH based Linux kernel for our embedded systems. I'm using a kernel 2.6.23. I've downloaded latest LTP release: ltp-full-20090131.tgz. I've found the following issues: 2) Under $LTP_ROOT/lib/ there are the tlibio.[ch] files which include some code based on Async IO functionalities. The Async IO is well supported if LTP was built with glibc but it is not supported by uclibc as, currently, the uclibc doesn't provide AsyncIO extensions. I think the code for AsyncIO, have to be placed under "#if !defined (__UCLIBC__)". I've made a patch to fix it. Signed-off-by: Francesco Rundo <francesco.rundo@st.com>.

2009-02-26 06:33:39 Tree
[3ffe07] by subrata_modak subrata_modak

I'm a user of LTP. I use LTP for testing SH based Linux kernel for our embedded systems. I'm using a kernel 2.6.23. I've downloaded latest LTP release: ltp-full-20090131.tgz. I've found the following issues: 1) I found some instructions on some LTP Makefiles reporting direct calls to "gcc" or "g++" instead of using the related env. variables $(CC) and/or $(CXX). The usage of gcc/g++ hardcoded, didn't allow to cross-build the related LTP tests. I've done a patch which is attached to this email ( ltp-full-20090131-allow-cross-build-fix.patch). Signed-off-by: Francesco Rundo <francesco.rundo@st.com>.

2009-02-26 06:29:42 Tree
[bb29d9] by subrata_modak subrata_modak

Correct possible errors in the runtest files: This patch changes some runtest lines, st. the same tests are referred to as same test IDs. Also, ot fixes some (presumably by accident) broken lines in runtest files. Signed-off-by: Jiri Palecek <jpalecek@web.de>.

2009-02-24 06:10:16 Tree
[fb6c1a] by subrata_modak subrata_modak

Fix broken line in runtest/scsi_debug.part1 file: This is another broken line in a runtest file. Signed-off-by: Jiri Palecek <jpalecek@web.de>.

2009-02-24 06:07:38 Tree
[c43525] by subrata_modak subrata_modak

Don't declare lseek in the fork10 test to prevent clashes with system definition: This is a patch for bug #2128681 suggested by Fathi Boudra. Signed-off-by: Jiri Palecek <jpalecek@web.de>.

2009-02-24 06:05:33 Tree
[efd3f9] by subrata_modak subrata_modak

I had a look on testcases/kernel/device-drivers/usb/tusb and deleted what is no longer supported by Linux 2.6.28, for example struct usb_operations, http://lwn.net/Articles/196429/. The result is that nearly nothing left which would test any pieces of the USB code. I attached the patch, but I don't know if it is worth to keep the "tusb" directory at all. Signed-Off-By: Németh Márton <nm127@freemail.hu>.

2009-02-24 05:31:50 Tree
[965028] by subrata_modak subrata_modak

RFC: Fixes for Shared Memory test case shmat: Shmat1.c(testcases/kernel/mem/mtest06) is a test case which tries to create 3 threads during its execution. One thread allocates shared memory, second writes to the shared memory and the third reads from the shared memory. All the 3 threads are synchronized using a global variable. In case of signal (sigsegv) sighandler will be called. The current test case implementation is complete only for x86 arc and is not valid for other archs. We have noticed various issues while executing this test case. Test case issues can be summarized as:
1. signals are masked once the signal handler is called
2. comparison signal_context->edi == map address is dubious leading to test case failures under x86 architecture,
Solution:
Issue 1: This is due to calling siglongjmp() with in the signal handler. Once the signal handler is called all the signals will be masked. It wont be set back to the original value unless sigsetjmp() is called with a non zero second parameter. This was not happening earlier leading to segmentation faults while executing the tests.
Issue 2: In the x86 architecture source and destination index with in the ES or DS segments are stored in esi and edi registers. While the shared memory address is being written to edi will have the map_address, returned by shmget, while when the data is read from map_address: will be contained in esi register. So it is inappropriate to just compare map_address to edi register.
Even after fixing these 2 issues I still see the test case failing some time with messages like: process exited with errors -1 etc. I wanted to know whether we should keep this test case in LTP suite or if there is a better way to fix the issues? Whether comparing signal_context->edi (or esi) == map_address is it the right thing to do? Signed-Off-By: Sharyathi Nagesh <sharyath@in.ibm.com>.

2009-02-23 11:48:28 Tree
[4edf14] by subrata_modak subrata_modak

v4l-test 0.8 for LTP: Today I released v4l-test 0.8. Please find attached the patch for LTP. Changes: Test cases added for VIDIOC_G_CROP, VIDIOC_G_CTRL and VIDIOC_S_CTRL. Signed-off-by: Márton Németh <nm127@freemail.hu>.

2009-02-23 11:22:27 Tree
Older >

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks