Re: [LTP] readahead02: change the test pass to 2M

2014-05-08 Thread Jan Stancek
- Original Message - > From: "Madan" > To: ltp-list@lists.sourceforge.net > Sent: Friday, 9 May, 2014 7:45:49 AM > Subject: [LTP] readahead02: change the test pass to 2M > > Max readahead size has been limited since commit: >commit 6d2be915e589b58cb11418cbe1f22ff90732b6ac >Au

[LTP] [PATCH v2] lib/tst_sig.c: output signal name when got unexpected signal

2014-05-08 Thread Xiaoguang Wang
When testcase is killed by unexpected signal, usually it just prints signal's value. fcntl30 1 TBROK : unexpected signal 2 received (pid = 6714). fcntl30 2 TBROK : Remaining cases broken Here we also print signal's name to output more informative message. fcnt

[LTP] readahead02: change the test pass to 2M

2014-05-08 Thread Madan
Max readahead size has been limited since commit: commit 6d2be915e589b58cb11418cbe1f22ff90732b6ac Author: Raghavendra K T Date: Thu Apr 3 14:48:23 2014 -0700 mm/readahead.c: fix readahead failure for memoryless NUMA nodes and limit readahead pages Updating the

[LTP] [PATCH v3] direct_io/dma_thread_diotest7.c: cleanup and add it to run default

2014-05-08 Thread Xiaoguang Wang
This test will create 100 files(every is 12MB), so the total needed free space is 1200MB at least. If the current test temporary directory do not satisfy this requirement, wo need to specify a big block device by running runltp with -z option. If satisfied, we also need to check whether filesyste

[LTP] [PATCH] setpriority/setpriority01.c: cleanup

2014-05-08 Thread Zeng Linggang
* Delete some useless commtents. * Move the test body from main() to setpriority_verify(). * Expand tests([-20,19]). * Add tst_require_root for lowering priorities testes. * Some cleanup. Signed-off-by: Zeng Linggang --- .../kernel/syscalls/setpriority/setpriority01.c| 166 +++---

[LTP] [PATCH v5 2/2] openat/openat02.c: add a new case to test flags

2014-05-08 Thread Xing Gu
From: "gux.f...@cn.fujitsu.com" create a new case to test flags for openat(2): O_APPEND O_CLOEXEC O_LARGEFILE O_NOATIME O_NOFOLLOW O_TRUNC Signed-off-by: Xing Gu --- configure.ac | 1 + m4/ltp-openat.m4 | 25 ++ runtest/s

Re: [LTP] [PATCH v4 1/2] lib/tst_path_has_mnt_flags.c: create a function tst_path_has_mnt_flags()

2014-05-08 Thread gux.fnst
Thanks for your suggestion. I will send a new version v5. Regards, Xing Gu 于 05/07/2014 01:26 AM, chru...@suse.cz 写道: > Hi! >> +/* lib/tst_path_has_mnt_flags.c >> + * >> + * Check whether a path is on a filesystem that is mounted with >> + * specified flags. >> + * @path: path to file >> + * @fla

[LTP] [PATCH v5 1/2] lib/tst_path_has_mnt_flags.c: create a function tst_path_has_mnt_flags()

2014-05-08 Thread Xing Gu
From: "gux.f...@cn.fujitsu.com" Create a function tst_path_has_mnt_flags() for checking whether a path is on a system that is mounted with specified flags. Signed-off-by: Xing Gu --- include/test.h | 12 lib/tst_path_has_mnt_flags.c | 71 +

[LTP] [PATCH 4/4] rt_sigaction*, rt_sigprocmask01, rt_sigsuspend01: SPARC support

2014-05-08 Thread Stanislav Kholmanskikh
Added SPARC support for tests: * rt_sigaction01, rt_sigaction02, rt_sigaction03 * rt_sigprocmask01 * rt_sigsuspend01 Author: Jose E. Marchesi Co-author: Stanislav Kholmanskikh Signed-off-by: Stanislav Kholmanskikh --- include/ltp_signal.h | 39 ++

[LTP] [PATCH 1/4] linux syscall numbers: fixed the checkings for SPARC

2014-05-08 Thread Stanislav Kholmanskikh
For sparc64 platforms cpp does not define '__sparc64__', but defines '__sparc__' and '__arch64__'. For sparc32 platforms cpp just defines '__sparc__'. This may cause problems because sparc64 and sparc32 have not equal sets of system calls. Modified the ckeckings for SPARC in regen.sh and execute

[LTP] [PATCH 3/4] syscalls: corrected the rest of __sparc64__ defines

2014-05-08 Thread Stanislav Kholmanskikh
'__sparc64__' is not defined by cpp. We should use '__sparc__' and '__arch64__' instead. Author: Jose E. Marchesi Author: Stanislav Kholmanskikh Signed-off-by: Stanislav Kholmanskikh --- testcases/kernel/syscalls/recvmsg/recvmsg01.c |3 ++- testcases/kernel/syscalls/sendmsg/sendmsg01.c |

[LTP] SPARC fixes

2014-05-08 Thread Stanislav Kholmanskikh
Hi! Here are some fixes to execute LTP on SPARC. Thanks! -- Is your legacy SCM system holding you back? Join Perforce May 7 to find out: • 3 signs your SCM is hindering your productivity • Requirements for releasing sof

[LTP] [PATCH 2/4] linux syscall numbers: removed unimplemented syscalls for sparc64

2014-05-08 Thread Stanislav Kholmanskikh
In accordance with arch/sparc/include/uapi/asm/unistd.h syscalls with names containing '32' are not defined for sparc64. Signed-off-by: Stanislav Kholmanskikh --- testcases/kernel/include/linux_syscall_numbers.h | 57 -- testcases/kernel/include/sparc64.in |