ring-buffer: Bring back context level recursive checks

Commit 1a149d7d ("ring-buffer: Rewrite trace_recursive_(un)lock() to be
simpler") replaced the context level recursion checks with a simple counter.
This would prevent the ring buffer code from recursively calling itself more
than the max number of contexts that exist (Normal, softirq, irq, nmi). But
this change caused a lockup in a specific case, which was during suspend and
resume using a global clock. Adding a stack dump to see where this occurred,
the issue was in the trace global clock itself:


The function graph tracer traced queued_spin_lock_slowpath that was called
by trace_clock_global. This pointed out that the trace_clock_global() is not
reentrant, as it takes a spin lock. It depended on the ring buffer recursive
lock from letting that happen.

By removing the context detection and adding just a max number of allowable
recursions, it allowed the trace_clock_global() to be entered again and try
to retake the spinlock it already held, causing a deadlock.

Fixes: 1a149d7d ("ring-buffer: Rewrite trace_recursive_(un)lock() to be simpler")
Reported-by: default avatarDavid Weinehall <david.weinehall@gmail.com>
Signed-off-by: default avatarSteven Rostedt (VMware) <rostedt@goodmis.org>
parent 4397f045
......@@ -2534,29 +2534,59 @@ rb_wakeups(struct ring_buffer *buffer, struct ring_buffer_per_cpu *cpu_buffer)
* The lock and unlock are done within a preempt disable section.
* The current_context per_cpu variable can only be modified
* by the current task between lock and unlock. But it can
* be modified more than once via an interrupt. There are four
* different contexts that we need to consider.
* be modified more than once via an interrupt. To pass this
* information from the lock to the unlock without having to
* access the 'in_interrupt()' functions again (which do show
* a bit of overhead in something as critical as function tracing,
* we use a bitmask trick.
* Normal context.
* SoftIRQ context
* IRQ context
* NMI context
* bit 0 = NMI context
* bit 1 = IRQ context
* bit 2 = SoftIRQ context
* bit 3 = normal context.
* If for some reason the ring buffer starts to recurse, we
* only allow that to happen at most 4 times (one for each
* context). If it happens 5 times, then we consider this a
* recusive loop and do not let it go further.
* This works because this is the order of contexts that can
* preempt other contexts. A SoftIRQ never preempts an IRQ
* context.
* When the context is determined, the corresponding bit is
* checked and set (if it was set, then a recursion of that context
* happened).
* On unlock, we need to clear this bit. To do so, just subtract
* 1 from the current_context and AND it to itself.
* (binary)
* 101 - 1 = 100
* 101 & 100 = 100 (clearing bit zero)
* 1010 - 1 = 1001
* 1010 & 1001 = 1000 (clearing bit 1)
* The least significant bit can be cleared this way, and it
* just so happens that it is the same bit corresponding to
* the current context.
static __always_inline int
trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer)
if (cpu_buffer->current_context >= 4)
unsigned int val = cpu_buffer->current_context;
unsigned long pc = preempt_count();
int bit;
bit = pc & NMI_MASK ? RB_CTX_NMI :
if (unlikely(val & (1 << bit)))
return 1;
/* Interrupts must see this update */
val |= (1 << bit);
cpu_buffer->current_context = val;
return 0;
......@@ -2564,9 +2594,7 @@ trace_recursive_lock(struct ring_buffer_per_cpu *cpu_buffer)
static __always_inline void
trace_recursive_unlock(struct ring_buffer_per_cpu *cpu_buffer)
/* Don't let the dec leak out */
cpu_buffer->current_context &= cpu_buffer->current_context - 1;
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment