diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2013-06-07 18:46:51 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-06-07 18:46:51 -0700 |
commit | 14d0ee051753f30e23033bdb153110f00b03c8ce (patch) | |
tree | 88a19ee630628e8904b96b586521d3f3755c14b8 | |
parent | ea7f665612fcc73da6b7698f468cd5fc03a30d47 (diff) | |
parent | f17a5194859a82afe4164e938b92035b86c55794 (diff) |
Merge tag 'trace-fixes-v3.10-rc3-v3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing fixes from Steven Rostedt:
"This contains 4 fixes.
The first two fix the case where full RCU debugging is enabled,
enabling function tracing causes a live lock of the system. This is
due to the added debug checks in rcu_dereference_raw() that is used by
the function tracer. These checks are also traced by the function
tracer as well as cause enough overhead to the function tracer to slow
down the system enough that the time to finish an interrupt can take
longer than when the next interrupt is triggered, causing a live lock
from the timer interrupt.
Talking this over with Paul McKenney, we came up with a fix that adds
a new rcu_dereference_raw_notrace() that does not perform these added
checks, and let the function tracer use that.
The third commit fixes a failed compile when branch tracing is
enabled, due to the conversion of the trace_test_buffer() selftest
that the branch trace wasn't converted for.
The forth patch fixes a bug caught by the RCU lockdep code where a
rcu_read_lock() is performed when rcu is disabled (either going to or
from idle, or user space). This happened on the irqsoff tracer as it
calls task_uid(). The fix here was to use current_uid() when possible
that doesn't use rcu locking. Which luckily, is always used when
irqsoff calls this code."
* tag 'trace-fixes-v3.10-rc3-v3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Use current_uid() for critical time tracing
tracing: Fix bad parameter passed in branch selftest
ftrace: Use the rcu _notrace variants for rcu_dereference_raw() and friends
rcu: Add _notrace variation of rcu_dereference_raw() and hlist_for_each_entry_rcu()
-rw-r--r-- | include/linux/rculist.h | 20 | ||||
-rw-r--r-- | include/linux/rcupdate.h | 9 | ||||
-rw-r--r-- | kernel/trace/ftrace.c | 18 | ||||
-rw-r--r-- | kernel/trace/trace.c | 10 | ||||
-rw-r--r-- | kernel/trace/trace_selftest.c | 2 |
5 files changed, 48 insertions, 11 deletions
diff --git a/include/linux/rculist.h b/include/linux/rculist.h index 8089e35d47ac..f4b1001a4676 100644 --- a/include/linux/rculist.h +++ b/include/linux/rculist.h @@ -461,6 +461,26 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev, &(pos)->member)), typeof(*(pos)), member)) /** + * hlist_for_each_entry_rcu_notrace - iterate over rcu list of given type (for tracing) + * @pos: the type * to use as a loop cursor. + * @head: the head for your list. + * @member: the name of the hlist_node within the struct. + * + * This list-traversal primitive may safely run concurrently with + * the _rcu list-mutation primitives such as hlist_add_head_rcu() + * as long as the traversal is guarded by rcu_read_lock(). + * + * This is the same as hlist_for_each_entry_rcu() except that it does + * not do any RCU debugging or tracing. + */ +#define hlist_for_each_entry_rcu_notrace(pos, head, member) \ + for (pos = hlist_entry_safe (rcu_dereference_raw_notrace(hlist_first_rcu(head)),\ + typeof(*(pos)), member); \ + pos; \ + pos = hlist_entry_safe(rcu_dereference_raw_notrace(hlist_next_rcu(\ + &(pos)->member)), typeof(*(pos)), member)) + +/** * hlist_for_each_entry_rcu_bh - iterate over rcu list of given type * @pos: the type * to use as a loop cursor. * @head: the head for your list. diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 4ccd68e49b00..ddcc7826d907 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -640,6 +640,15 @@ static inline void rcu_preempt_sleep_check(void) #define rcu_dereference_raw(p) rcu_dereference_check(p, 1) /*@@@ needed? @@@*/ +/* + * The tracing infrastructure traces RCU (we want that), but unfortunately + * some of the RCU checks causes tracing to lock up the system. + * + * The tracing version of rcu_dereference_raw() must not call + * rcu_read_lock_held(). + */ +#define rcu_dereference_raw_notrace(p) __rcu_dereference_check((p), 1, __rcu) + /** * rcu_access_index() - fetch RCU index with no dereferencing * @p: The index to read diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index b549b0f5b977..6c508ff33c62 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -120,22 +120,22 @@ static void ftrace_ops_no_ops(unsigned long ip, unsigned long parent_ip); /* * Traverse the ftrace_global_list, invoking all entries. The reason that we - * can use rcu_dereference_raw() is that elements removed from this list + * can use rcu_dereference_raw_notrace() is that elements removed from this list * are simply leaked, so there is no need to interact with a grace-period - * mechanism. The rcu_dereference_raw() calls are needed to handle + * mechanism. The rcu_dereference_raw_notrace() calls are needed to handle * concurrent insertions into the ftrace_global_list. * * Silly Alpha and silly pointer-speculation compiler optimizations! */ #define do_for_each_ftrace_op(op, list) \ - op = rcu_dereference_raw(list); \ + op = rcu_dereference_raw_notrace(list); \ do /* * Optimized for just a single item in the list (as that is the normal case). */ #define while_for_each_ftrace_op(op) \ - while (likely(op = rcu_dereference_raw((op)->next)) && \ + while (likely(op = rcu_dereference_raw_notrace((op)->next)) && \ unlikely((op) != &ftrace_list_end)) static inline void ftrace_ops_init(struct ftrace_ops *ops) @@ -779,7 +779,7 @@ ftrace_find_profiled_func(struct ftrace_profile_stat *stat, unsigned long ip) if (hlist_empty(hhd)) return NULL; - hlist_for_each_entry_rcu(rec, hhd, node) { + hlist_for_each_entry_rcu_notrace(rec, hhd, node) { if (rec->ip == ip) return rec; } @@ -1165,7 +1165,7 @@ ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) hhd = &hash->buckets[key]; - hlist_for_each_entry_rcu(entry, hhd, hlist) { + hlist_for_each_entry_rcu_notrace(entry, hhd, hlist) { if (entry->ip == ip) return entry; } @@ -1422,8 +1422,8 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip) struct ftrace_hash *notrace_hash; int ret; - filter_hash = rcu_dereference_raw(ops->filter_hash); - notrace_hash = rcu_dereference_raw(ops->notrace_hash); + filter_hash = rcu_dereference_raw_notrace(ops->filter_hash); + notrace_hash = rcu_dereference_raw_notrace(ops->notrace_hash); if ((ftrace_hash_empty(filter_hash) || ftrace_lookup_ip(filter_hash, ip)) && @@ -2920,7 +2920,7 @@ static void function_trace_probe_call(unsigned long ip, unsigned long parent_ip, * on the hash. rcu_read_lock is too dangerous here. */ preempt_disable_notrace(); - hlist_for_each_entry_rcu(entry, hhd, node) { + hlist_for_each_entry_rcu_notrace(entry, hhd, node) { if (entry->ip == ip) entry->ops->func(ip, parent_ip, &entry->data); } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 4d79485b3237..1a41023a1f88 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -843,7 +843,15 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) memcpy(max_data->comm, tsk->comm, TASK_COMM_LEN); max_data->pid = tsk->pid; - max_data->uid = task_uid(tsk); + /* + * If tsk == current, then use current_uid(), as that does not use + * RCU. The irq tracer can be called out of RCU scope. + */ + if (tsk == current) + max_data->uid = current_uid(); + else + max_data->uid = task_uid(tsk); + max_data->nice = tsk->static_prio - 20 - MAX_RT_PRIO; max_data->policy = tsk->policy; max_data->rt_priority = tsk->rt_priority; diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 55e2cf66967b..2901e3b88590 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1159,7 +1159,7 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) /* stop the tracing. */ tracing_stop(); /* check the trace buffer */ - ret = trace_test_buffer(tr, &count); + ret = trace_test_buffer(&tr->trace_buffer, &count); trace->reset(tr); tracing_start(); |