Fix queued spinlocks and a bit more

index 443050906018..e3b29eda8074 100644
@@ -448,6 +448,9 @@
  */
 long plpar_hcall_norets(unsigned long opcode, ...);
 
+/* Variant which does not do hcall tracing */
+long plpar_hcall_norets_notrace(unsigned long opcode, ...);
+
 /**
  * plpar_hcall: - Make a pseries hypervisor call
  * @opcode: The hypervisor call to make.

index 5d1726bb28e7..bcb7b5f917be 100644
@@ -28,19 +28,35 @@ static inline u32 yield_count_of(int cpu)
 	return be32_to_cpu(yield_count);
 }
 
+/*
+ * Spinlock code confers and prods, so don't trace the hcalls because the
+ * tracing code takes spinlocks which can cause recursion deadlocks.
+ *
+ * These calls are made while the lock is not held: the lock slowpath yields if
+ * it can not acquire the lock, and unlock slow path might prod if a waiter has
+ * yielded). So this may not be a problem for simple spin locks because the
+ * tracing does not technically recurse on the lock, but we avoid it anyway.
+ *
+ * However the queued spin lock contended path is more strictly ordered: the
+ * H_CONFER hcall is made after the task has queued itself on the lock, so then
+ * recursing on that lock will cause the task to then queue up again behind the
+ * first instance (or worse: queued spinlocks use tricks that assume a context
+ * never waits on more than one spinlock, so such recursion may cause random
+ * corruption in the lock code).
+ */
 static inline void yield_to_preempted(int cpu, u32 yield_count)
 {
-	plpar_hcall_norets(H_CONFER, get_hard_smp_processor_id(cpu), yield_count);
+	plpar_hcall_norets_notrace(H_CONFER, get_hard_smp_processor_id(cpu), yield_count);
 }
 
 static inline void prod_cpu(int cpu)
 {
-	plpar_hcall_norets(H_PROD, get_hard_smp_processor_id(cpu));
+	plpar_hcall_norets_notrace(H_PROD, get_hard_smp_processor_id(cpu));
 }
 
 static inline void yield_to_any(void)
 {
-	plpar_hcall_norets(H_CONFER, -1, 0);
+	plpar_hcall_norets_notrace(H_CONFER, -1, 0);
 }
 #else
 static inline bool is_shared_processor(void)

index 1f3152ad7213..b619568a4d04 100644
@@ -1830,8 +1830,7 @@ void hcall_tracepoint_unregfunc(void)
 
 /*
  * Since the tracing code might execute hcalls we need to guard against
- * recursion. One example of this are spinlocks calling H_YIELD on
- * shared processor partitions.
+ * recursion.
  */
 static DEFINE_PER_CPU(unsigned int, hcall_trace_depth);
 

index b619568a4d04..d79d7410c320 100644
@@ -1835,7 +1835,7 @@ void hcall_tracepoint_unregfunc(void)
 static DEFINE_PER_CPU(unsigned int, hcall_trace_depth);
 
 
-void __trace_hcall_entry(unsigned long opcode, unsigned long *args)
+notrace void __trace_hcall_entry(unsigned long opcode, unsigned long *args)
 {
 	unsigned long flags;
 	unsigned int *depth;

@@ -1863,7 +1863,7 @@ void __trace_hcall_entry(unsigned long opcode, unsigned long *args)
 	local_irq_restore(flags);
 }
 
-void __trace_hcall_exit(long opcode, long retval, unsigned long *retbuf)
+notrace void __trace_hcall_exit(long opcode, long retval, unsigned long *retbuf)
 {
 	unsigned long flags;
 	unsigned int *depth;

index ece84a430701..83e0f701ebc6 100644
@@ -28,7 +28,11 @@ static inline void set_cede_latency_hint(u8 latency_hint)
 
 static inline long cede_processor(void)
 {
-	return plpar_hcall_norets(H_CEDE);
+	/*
+	 * We cannot call tracepoints inside RCU idle regions which
+	 * means we must not trace H_CEDE.
+	 */
+	return plpar_hcall_norets_notrace(H_CEDE);
 }
 
 static inline long extended_cede_processor(unsigned long latency_hint)

index d79d7410c320..ad1cec80019b 100644
@@ -1840,13 +1840,6 @@ notrace void __trace_hcall_entry(unsigned long opcode, unsigned long *args)
 	unsigned long flags;
 	unsigned int *depth;
 
-	/*
-	 * We cannot call tracepoints inside RCU idle regions which
-	 * means we must not trace H_CEDE.
-	 */
-	if (opcode == H_CEDE)
-		return;
-
 	local_irq_save(flags);
 
 	depth = this_cpu_ptr(&hcall_trace_depth);

@@ -1868,9 +1861,6 @@ notrace void __trace_hcall_exit(long opcode, long retval, unsigned long *retbuf)
 	unsigned long flags;
 	unsigned int *depth;
 
-	if (opcode == H_CEDE)
-		return;
-
 	local_irq_save(flags);
 
 	depth = this_cpu_ptr(&hcall_trace_depth);

index ad1cec80019b..dab356e3ff87 100644
@@ -1829,8 +1829,14 @@ void hcall_tracepoint_unregfunc(void)
 #endif
 
 /*
- * Since the tracing code might execute hcalls we need to guard against
- * recursion.
+ * Keep track of hcall tracing depth and prevent recursion. Warn if any is
+ * detected because it may indicate a problem. This will not catch all
+ * problems with tracing code making hcalls, because the tracing might have
+ * been invoked from a non-hcall, so the first hcall could recurse into it
+ * without warning here, but this better than nothing.
+ *
+ * Hcalls with specific problems being traced should use the _notrace
+ * plpar_hcall variants.
  */
 static DEFINE_PER_CPU(unsigned int, hcall_trace_depth);
 

@@ -1844,7 +1850,7 @@ notrace void __trace_hcall_entry(unsigned long opcode, unsigned long *args)
 
 	depth = this_cpu_ptr(&hcall_trace_depth);
 
-	if (*depth)
+	if (WARN_ON_ONCE(*depth))
 		goto out;
 
 	(*depth)++;

@@ -1865,7 +1871,7 @@ notrace void __trace_hcall_exit(long opcode, long retval, unsigned long *retbuf)
 
 	depth = this_cpu_ptr(&hcall_trace_depth);
 
-	if (*depth)
+	if (*depth) /* Don't warn again on the way out */
 		goto out;
 
 	(*depth)++;