user/sven/linux.git/kernel/marker.c, branch v2.6.31.2

markers/tracpoints: fix non-modular build

2008-11-16T08:52:03Z

fix: kernel/marker.c: In function 'marker_module_notify': kernel/marker.c:905: error: 'MODULE_STATE_COMING' undeclared (first use in this function) [...] Signed-off-by: Ingo Molnar

markers: auto enable tracepoints (new API : trace_mark_tp())

2008-11-16T08:01:29Z

Impact: new API Add a new API trace_mark_tp(), which declares a marker within a tracepoint probe. When the marker is activated, the tracepoint is automatically enabled. No branch test is used at the marker site, because it would be a duplicate of the branch already present in the tracepoint. Signed-off-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: use module notifier

2008-11-16T08:01:28Z

Impact: cleanup Use module notifiers instead of adding a hook in module.c. Signed-off-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: use rcu_*_sched_notrace and notrace

2008-11-16T08:01:27Z

Make marker critical code use notrace to make sure they can be used as an ftrace callback. Signed-off-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: fix unregister

2008-11-16T08:01:26Z

Impact: fix marker registers/unregister race get_marker() can return a NULL entry because the mutex is released in the middle of those functions. Make sure we check to see if it has been concurrently removed. Signed-off-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: break the redundant loop in kernel/marker.c

2008-10-27T12:02:23Z

Impact: cleanup, no functionality changed Because e->name is unique in list, we don't need to continue the iteration after matched. Signed-off-by: Zhao Lei Acked-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: let marker_table be close to its comments

2008-10-27T12:02:23Z

marker_table is defined far from its comments, this fix make cleanup for it. Signed-off-by: Lai Jiangshan Acked-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: remove exported symbol marker_probe_cb_noarg()

2008-10-27T12:02:23Z

marker_probe_cb_noarg() should not be seen by outer code. this patch remove it. Signed-off-by: Lai Jiangshan Acked-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: simplify marker_set_format()

2008-10-27T12:02:22Z

current marker_set_format() is complex this patch simplify it, and decrease the overhead of marker_update_probes(). Signed-off-by: Lai Jiangshan Acked-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar

markers: bit-field is not thread-safe nor smp-safe

2008-10-14T08:38:45Z

bit-field is not thread-safe nor smp-safe. struct marker_entry.rcu_pending is not protected by any lock in rcu-callback free_old_closure(). so we must turn it into a safe type. detail: I suppose rcu_pending and ptype are store in struct marker_entry.tmp1 free_old_closure() side: change ptype side: | load struct marker_entry.tmp1 --------------------------------|-------------------------------- | change ptype bit in tmp1 load struct marker_entry.tmp1 | change rcu_pending bit in tmp1 | store tmp1 | --------------------------------|-------------------------------- | store tmp1 now this result equals that free_old_closure() do not change rcu_pending bit, bug! This bug will cause redundant rcu_barrier_sched() called. not too harmful. ----- corresponding: free_old_closure() side: change ptype side: load struct marker_entry.tmp1 | --------------------------------|-------------------------------- | load struct marker_entry.tmp1 change rcu_pending bit in tmp1 | | change ptype bit in tmp1 | store tmp1 --------------------------------|-------------------------------- store tmp1 | now this result equals that change ptype side do not change ptype bit, bug! this bug cause marker_probe_cb() access to invalid memory. oops! see also: http://en.wikipedia.org/wiki/Bit_field Signed-off-by: Lai Jiangshan Acked-by: Mathieu Desnoyers Signed-off-by: Ingo Molnar