linux/tools/lib/lockdep, branch v4.1

tools/liblockdep: Fix compilation error

2015-05-13T10:12:20Z

Recent changes to kernel/locking/lockdep.c broke the liblockdep build. Fix that. Signed-off-by: Eunbong Song Signed-off-by: Sasha Levin

tools/liblockdep: Fix linker error in case of cross compile

2015-05-13T10:09:09Z

If we try to cross compile liblockdep, even if we set the CROSS_COMPILE variable the linker error can occur because LD is not set with CROSS_COMPILE. This patch adds "LD" can be set automatically with CROSS_COMPILE variable so fixes linker error problem. Signed-off-by: Eunbong Song Signed-off-by: Sasha Levin

Merge tag 'v4.0-rc1' into perf/core, to refresh the tree

2015-02-26T11:24:50Z

Signed-off-by: Ingo Molnar

tools lib lockdep: Use tools build framework

2015-02-12T20:56:01Z

Move the lockdep library building under tools build framework. Signed-off-by: Jiri Olsa Cc: Alexis Berlemont Cc: Borislav Petkov Cc: Corey Ashford Cc: David Ahern Cc: Frederic Weisbecker Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: S. Lockwood-Childs Cc: Sasha Levin Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-i0t25buqyo5jfvzpw2347h1h@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo

tools/liblockdep: don't include host headers

2015-01-30T02:55:19Z

Adding host headers to include path may cause unexpected surprises when cross compiling. Remove /usr/local/include from the default include path. Signed-off-by: Baruch Siach Signed-off-by: Sasha Levin

tools/liblockdep: ignore generated .so file

2015-01-30T02:55:19Z

Signed-off-by: Baruch Siach Signed-off-by: Sasha Levin

tools/liblockdep: Fix debug_check thinko in mutex destroy

2014-12-19T20:50:55Z

In mutex destroy code currently we pass to debug_check_no_locks_freed() [mem_from, mem_end) address region. But debug_check_no_locks_freed() accepts mem_from, mem_*len* i.e. second parameter is region length, not end address. And it was always so, starting from 2006 (fbb9ce95 "lockdep: core"). Fix it, or else on a mutex destroy we wrongly check much-wider-than-mutex region and can find not-yet-released other locks there and wrongly report BUGs on them. Signed-off-by: Kirill Smelkov Signed-off-by: Sasha Levin

tools/liblockdep: Account for bitfield changes in lockdeps lock_acquire

2014-07-07T16:20:16Z

Commit fb9edbe984 shortened held_lock->check from a 2-bit field to a 1-bit field. Make liblockdep compatible with the new definition by passing check=1 to lock_acquire() calls, rather than the old value check=2 (which inadvertently disabled checks by overflowing to 0). Without this fix, several of the test cases in liblockdep run_tests.sh were failing. Signed-off-by: S. Lockwood-Childs Signed-off-by: Sasha Levin

tools/liblockdep: Remove debug print left over from development

2014-07-07T16:14:27Z

Remove a debug print in init_preload() which was left over from development and isn't usefull at all currently. It was also causing false positive test results. Reported-by: S. Lockwood-Childs Signed-off-by: Sasha Levin

tools/liblockdep: Fix comparison of a boolean value with a value of 2

2014-07-07T16:02:21Z

Comparison of a boolean value (!__init_state) with a value of 2 (done) as currently happens in the code is unlikely to succeed and causes repeated initialization of the pthread function pointers. Instead, remove boolean comparison so that we would initialize said function pointers only once. Ref: https://bugzilla.kernel.org/show_bug.cgi?id=76741 Cc: Jean Delvare Reported-by: David Binderman Signed-off-by: Dianfang Zhang Signed-off-by: Sasha Levin