Skip to contentSkip to conversation

rcu-tasks miscellaneous fixes

PATCH Superseded Neeraj Upadhyay 2 participants 8 messages Aug 18, 2021 RCU

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover rcu-tasks miscellaneous fixes

Minor typos in comments and fixes for rcu-tasks.

Neeraj Upadhyay (5):
rcu-tasks: Fix s/rcu_add_holdout/trc_add_holdout/ typo in comment
rcu-tasks: Correct firstreport usage in check_all_holdout_tasks_trace
rcu-tasks: Correct check for no_hz_full cpu in show_stalled_task_trace
rcu-tasks: Fix read-side primitives comment for call_rcu_tasks_trace
rcu-tasks: Clarify read side section info for rcu_tasks_rude GP
primitives

 kernel/rcu/tasks.h | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
#1 lore raw

PATCH 1/5 rcu-tasks: Fix s/rcu_add_holdout/trc_add_holdout/ typo in comment

---
 kernel/rcu/tasks.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/rcu/tasks.h +1 -1
index 39cef3c..e62da45 100644
@@ -996,7 +996,7 @@ static void trc_wait_for_one_reader(struct task_struct *t,
 
 	// If this task is not yet on the holdout list, then we are in
 	// an RCU read-side critical section.  Otherwise, the invocation of
-	// rcu_add_holdout() that added it to the list did the necessary
+	// trc_add_holdout() that added it to the list did the necessary
 	// get_task_struct().  Either way, the task cannot be freed out
 	// from under this code.
 
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
#2 lore raw

PATCH 2/5 rcu-tasks: Correct firstreport usage in check_all_holdout_tasks_trace

In reply to Neeraj Upadhyay

In check_all_holdout_tasks_trace(), firstreport is a pointer argument; so, check the dereferenced value, instead of checking the pointer.

---
 kernel/rcu/tasks.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/rcu/tasks.h +1 -1
index e62da45..9db7293 100644
@@ -1134,7 +1134,7 @@ static void check_all_holdout_tasks_trace(struct list_head *hop,
 	cpus_read_unlock();
 
 	if (needreport) {
-		if (firstreport)
+		if (*firstreport)
 			pr_err("INFO: rcu_tasks_trace detected stalls? (Late IPI?)\n");
 		show_stalled_ipi_trace();
 	}
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
#3 lore raw

PATCH 3/5 rcu-tasks: Correct check for no_hz_full cpu in show_stalled_task_trace

In reply to Neeraj Upadhyay

Correct the check for no_hz_full cpu in show_stalled_task_trace(), to include cpu 0.

---
 kernel/rcu/tasks.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/rcu/tasks.h +1 -1
index 9db7293..244e06a 100644
@@ -1091,7 +1091,7 @@ static void show_stalled_task_trace(struct task_struct *t, bool *firstreport)
 		 t->pid,
 		 ".I"[READ_ONCE(t->trc_ipi_to_cpu) > 0],
 		 ".i"[is_idle_task(t)],
-		 ".N"[cpu > 0 && tick_nohz_full_cpu(cpu)],
+		 ".N"[cpu >= 0 && tick_nohz_full_cpu(cpu)],
 		 READ_ONCE(t->trc_reader_nesting),
 		 " N"[!!READ_ONCE(t->trc_reader_special.b.need_qs)],
 		 cpu);
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
#4 lore raw

PATCH 4/5 rcu-tasks: Fix read-side primitives comment for call_rcu_tasks_trace

In reply to Neeraj Upadhyay

call_rcu_tasks_trace() does have read-side primitives - rcu_read_lock_trace() and rcu_read_unlock_trace(). Fix this information in the comments.

---
 kernel/rcu/tasks.h | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)
kernel/rcu/tasks.h +3 -7
index 244e06a..5f9c14c 100644
@@ -1214,13 +1214,9 @@ static void exit_tasks_rcu_finish_trace(struct task_struct *t)
  *
  * The callback function will be invoked some time after a full grace
  * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed. call_rcu_tasks_trace()
- * assumes that the read-side critical sections end at context switch,
- * cond_resched_rcu_qs(), or transition to usermode execution.  As such,
- * there are no read-side primitives analogous to rcu_read_lock() and
- * rcu_read_unlock() because this primitive is intended to determine
- * that all tasks have passed through a safe state, not so much for
- * data-structure synchronization.
+ * read-side critical sections have completed. These read-side
+ * critical sections are delimited by calls to rcu_read_lock_trace()
+ * and rcu_read_unlock_trace().
  *
  * See the description of call_rcu() for more detailed information on
  * memory ordering guarantees.
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
maintainer
#5 lore raw

Re: [PATCH 4/5] rcu-tasks: Fix read-side primitives comment for call_rcu_tasks_trace

Quoted 14 lines · Neeraj Upadhyay

call_rcu_tasks_trace() does have read-side primitives - rcu_read_lock_trace() and rcu_read_unlock_trace(). Fix this information in the comments.

---
 kernel/rcu/tasks.h | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)
kernel/rcu/tasks.h +0 -0
index 244e06a..5f9c14c 100644
@@ -1214,13 +1214,9 @@ static void exit_tasks_rcu_finish_trace(struct task_struct *t)
  *

kernel/rcu/tasks.h static void exit_tasks_rcu_finish_trace(struct task_struct *t) · line 1217

Quoted 10 lines
kernel/rcu/tasks.h +1 -7
  * The callback function will be invoked some time after a full grace
  * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed. call_rcu_tasks_trace()
- * assumes that the read-side critical sections end at context switch,
- * cond_resched_rcu_qs(), or transition to usermode execution.  As such,
- * there are no read-side primitives analogous to rcu_read_lock() and
- * rcu_read_unlock() because this primitive is intended to determine
- * that all tasks have passed through a safe state, not so much for
- * data-structure synchronization.
+ * read-side critical sections have completed. These read-side

The above line should start "Tasks Trace read-side critical..." for the benefit of people who stop reading before the next sentence. :-/

Thanx, Paul

Quoted 9 lines
kernel/rcu/tasks.h +2 -0
+ * critical sections are delimited by calls to rcu_read_lock_trace()
+ * and rcu_read_unlock_trace().
  *
  * See the description of call_rcu() for more detailed information on
  * memory ordering guarantees.
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
#6 lore raw

PATCH 5/5 rcu-tasks: Clarify read side section info for rcu_tasks_rude GP primitives

In reply to Neeraj Upadhyay

RCU tasks rude variant does not check whether the current running context on a CPU is usermode. Read side critical section ends on transition to usermode execution, by the virtue of usermode execution being schedulable. Clarify this in comments for call_rcu_tasks_rude() and synchronize_rcu_tasks_rude().

---
 kernel/rcu/tasks.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
kernel/rcu/tasks.h +7 -7
index 5f9c14c..702a4fa 100644
@@ -677,11 +677,11 @@ DEFINE_RCU_TASKS(rcu_tasks_rude, rcu_tasks_rude_wait_gp, call_rcu_tasks_rude,
  * period elapses, in other words after all currently executing RCU
  * read-side critical sections have completed. call_rcu_tasks_rude()
  * assumes that the read-side critical sections end at context switch,
- * cond_resched_rcu_qs(), or transition to usermode execution.  As such,
- * there are no read-side primitives analogous to rcu_read_lock() and
- * rcu_read_unlock() because this primitive is intended to determine
- * that all tasks have passed through a safe state, not so much for
- * data-structure synchronization.
+ * cond_resched_rcu_qs(), or transition to usermode execution (as
+ * usermode execution is schedulable). As such, there are no read-side
+ * primitives analogous to rcu_read_lock() and rcu_read_unlock() because
+ * this primitive is intended to determine that all tasks have passed
+ * through a safe state, not so much for data-structure synchronization.
  *
  * See the description of call_rcu() for more detailed information on
  * memory ordering guarantees.
@@ -699,8 +699,8 @@ EXPORT_SYMBOL_GPL(call_rcu_tasks_rude);
  * grace period has elapsed, in other words after all currently
  * executing rcu-tasks read-side critical sections have elapsed.  These
  * read-side critical sections are delimited by calls to schedule(),
- * cond_resched_tasks_rcu_qs(), userspace execution, and (in theory,
- * anyway) cond_resched().
+ * cond_resched_tasks_rcu_qs(), userspace execution (which is a schedulable
+ * context), and (in theory, anyway) cond_resched().
  *
  * This is a very specialized primitive, intended only for a few uses in
  * tracing and other situations requiring manipulation of function preambles
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation
maintainer
#7 lore raw
In reply to Neeraj Upadhyay
Quoted 1 line · Neeraj Upadhyay

Minor typos in comments and fixes for rcu-tasks.

I queued all but 4/5 for review and testing, thank you!!!

4/5 needs a slight fix that I cannot do in the patch itself.

Thanx, Paul

Quoted 15 lines

Neeraj Upadhyay (5):
rcu-tasks: Fix s/rcu_add_holdout/trc_add_holdout/ typo in comment
rcu-tasks: Correct firstreport usage in check_all_holdout_tasks_trace
rcu-tasks: Correct check for no_hz_full cpu in show_stalled_task_trace
rcu-tasks: Fix read-side primitives comment for call_rcu_tasks_trace
rcu-tasks: Clarify read side section info for rcu_tasks_rude GP
primitives

 kernel/rcu/tasks.h | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, 
hosted by The Linux Foundation