From b1cff0ad1062621ae63cb6c5dc4165191fe2e9f1 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 25 May 2011 14:27:43 -0400 Subject: ftrace: Add internal recursive checks Witold reported a reboot caused by the selftests of the dynamic function tracer. He sent me a config and I used ktest to do a config_bisect on it (as my config did not cause the crash). It pointed out that the problem config was CONFIG_PROVE_RCU. What happened was that if multiple callbacks are attached to the function tracer, we iterate a list of callbacks. Because the list is managed by synchronize_sched() and preempt_disable, the access to the pointers uses rcu_dereference_raw(). When PROVE_RCU is enabled, the rcu_dereference_raw() calls some debugging functions, which happen to be traced. The tracing of the debug function would then call rcu_dereference_raw() which would then call the debug function and then... well you get the idea. I first wrote two different patches to solve this bug. 1) add a __rcu_dereference_raw() that would not do any checks. 2) add notrace to the offending debug functions. Both of these patches worked. Talking with Paul McKenney on IRC, he suggested to add recursion detection instead. This seemed to be a better solution, so I decided to implement it. As the task_struct already has a trace_recursion to detect recursion in the ring buffer, and that has a very small number it allows, I decided to use that same variable to add flags that can detect the recursion inside the infrastructure of the function tracer. I plan to change it so that the task struct bit can be checked in mcount, but as that requires changes to all archs, I will hold that off to the next merge window. Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Frederic Weisbecker Cc: Paul E. McKenney Link: http://lkml.kernel.org/r/1306348063.1465.116.camel@gandalf.stny.rr.com Reported-by: Witold Baryluk Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 13 ++++++++++++- kernel/trace/ring_buffer.c | 10 +++++----- kernel/trace/trace.h | 15 +++++++++++++++ 3 files changed, 32 insertions(+), 6 deletions(-) (limited to 'kernel/trace') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 25949b33057c..1ee417fcbfa5 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -109,12 +109,18 @@ ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip); static void ftrace_global_list_func(unsigned long ip, unsigned long parent_ip) { - struct ftrace_ops *op = rcu_dereference_raw(ftrace_global_list); /*see above*/ + struct ftrace_ops *op; + + if (unlikely(trace_recursion_test(TRACE_GLOBAL_BIT))) + return; + trace_recursion_set(TRACE_GLOBAL_BIT); + op = rcu_dereference_raw(ftrace_global_list); /*see above*/ while (op != &ftrace_list_end) { op->func(ip, parent_ip); op = rcu_dereference_raw(op->next); /*see above*/ }; + trace_recursion_clear(TRACE_GLOBAL_BIT); } static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip) @@ -3490,6 +3496,10 @@ ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip) { struct ftrace_ops *op; + if (unlikely(trace_recursion_test(TRACE_INTERNAL_BIT))) + return; + + trace_recursion_set(TRACE_INTERNAL_BIT); /* * Some of the ops may be dynamically allocated, * they must be freed after a synchronize_sched(). @@ -3502,6 +3512,7 @@ ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip) op = rcu_dereference_raw(op->next); }; preempt_enable_notrace(); + trace_recursion_clear(TRACE_INTERNAL_BIT); } static void clear_ftrace_swapper(void) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 0ef7b4b2a1f7..b0c7aa407943 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2216,7 +2216,7 @@ static noinline void trace_recursive_fail(void) printk_once(KERN_WARNING "Tracing recursion: depth[%ld]:" "HC[%lu]:SC[%lu]:NMI[%lu]\n", - current->trace_recursion, + trace_recursion_buffer(), hardirq_count() >> HARDIRQ_SHIFT, softirq_count() >> SOFTIRQ_SHIFT, in_nmi()); @@ -2226,9 +2226,9 @@ static noinline void trace_recursive_fail(void) static inline int trace_recursive_lock(void) { - current->trace_recursion++; + trace_recursion_inc(); - if (likely(current->trace_recursion < TRACE_RECURSIVE_DEPTH)) + if (likely(trace_recursion_buffer() < TRACE_RECURSIVE_DEPTH)) return 0; trace_recursive_fail(); @@ -2238,9 +2238,9 @@ static inline int trace_recursive_lock(void) static inline void trace_recursive_unlock(void) { - WARN_ON_ONCE(!current->trace_recursion); + WARN_ON_ONCE(!trace_recursion_buffer()); - current->trace_recursion--; + trace_recursion_dec(); } #else diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 6b69c4bd306f..229f8591f61d 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -784,4 +784,19 @@ extern const char *__stop___trace_bprintk_fmt[]; FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print)) #include "trace_entries.h" +/* Only current can touch trace_recursion */ +#define trace_recursion_inc() do { (current)->trace_recursion++; } while (0) +#define trace_recursion_dec() do { (current)->trace_recursion--; } while (0) + +/* Ring buffer has the 10 LSB bits to count */ +#define trace_recursion_buffer() ((current)->trace_recursion & 0x3ff) + +/* for function tracing recursion */ +#define TRACE_INTERNAL_BIT (1<<11) +#define TRACE_GLOBAL_BIT (1<<12) + +#define trace_recursion_set(bit) do { (current)->trace_recursion |= (bit); } while (0) +#define trace_recursion_clear(bit) do { (current)->trace_recursion &= ~(bit); } while (0) +#define trace_recursion_test(bit) ((current)->trace_recursion & (bit)) + #endif /* _LINUX_KERNEL_TRACE_H */ -- cgit v1.2.1