re: rcu/nocb: Add bypass callback queueing, bug report

re: rcu/nocb: Add bypass callback queueing, bug report

[Colin Ian King]

Hi,

Static analysis on linux-next today found an issue in the following commit:

commit 1afc4b18724f8f7b7a21fdf66cd43cc4a932812d
Author: Paul E. McKenney <paulmck@linux.ibm.com>
Date:   Tue Jul 2 16:03:33 2019 -0700

    rcu/nocb: Add bypass callback queueing


The coverity report is as follows:

1783        // If we have advanced to a new jiffy, reset counts to allow
1784        // moving back from ->nocb_bypass to ->cblist.
1785        if (j == rdp->nocb_nobypass_last) {
1786                c = rdp->nocb_nobypass_count + 1;
1787        } else {
1788                WRITE_ONCE(rdp->nocb_nobypass_last, j);
1789                c = rdp->nocb_nobypass_count -
nocb_nobypass_lim_per_jiffy;
1790                if (c > nocb_nobypass_lim_per_jiffy)
1791                        c = nocb_nobypass_lim_per_jiffy;

CID 85141 (#1 of 1): Unsigned compared against 0
unsigned_compare: This less-than-zero comparison of an unsigned value is
never true. c < 0UL.

1792                else if (c < 0)
1793                        c = 0;

Variable c is an unsigned long so the c < 0 check is never true. I'm not
sure what the ramifications are if c is made a signed long instead, so
I'm not fixing this and reporting this issue.

Regards,

Colin

Re: rcu/nocb: Add bypass callback queueing, bug report

[Paul E. McKenney]

On Tue, Aug 13, 2019 at 01:34:02PM +0100, Colin Ian King wrote:
> Hi,
> 
> Static analysis on linux-next today found an issue in the following commit:
> 
> commit 1afc4b18724f8f7b7a21fdf66cd43cc4a932812d
> Author: Paul E. McKenney <paulmck@linux.ibm.com>
> Date:   Tue Jul 2 16:03:33 2019 -0700
> 
>     rcu/nocb: Add bypass callback queueing
> 
> 
> The coverity report is as follows:
> 
> 1783        // If we have advanced to a new jiffy, reset counts to allow
> 1784        // moving back from ->nocb_bypass to ->cblist.
> 1785        if (j == rdp->nocb_nobypass_last) {
> 1786                c = rdp->nocb_nobypass_count + 1;
> 1787        } else {
> 1788                WRITE_ONCE(rdp->nocb_nobypass_last, j);
> 1789                c = rdp->nocb_nobypass_count -
> nocb_nobypass_lim_per_jiffy;
> 1790                if (c > nocb_nobypass_lim_per_jiffy)
> 1791                        c = nocb_nobypass_lim_per_jiffy;
> 
> CID 85141 (#1 of 1): Unsigned compared against 0
> unsigned_compare: This less-than-zero comparison of an unsigned value is
> never true. c < 0UL.
> 
> 1792                else if (c < 0)
> 1793                        c = 0;
> 
> Variable c is an unsigned long so the c < 0 check is never true. I'm not
> sure what the ramifications are if c is made a signed long instead, so
> I'm not fixing this and reporting this issue.

Good catch!!!

How about the alleged fix shown below?

							Thanx, Paul

------------------------------------------------------------------------

diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 91cefa3bf943..2defc7fe74c3 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -1787,10 +1787,11 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp,
 	} else {
 		WRITE_ONCE(rdp->nocb_nobypass_last, j);
 		c = rdp->nocb_nobypass_count - nocb_nobypass_lim_per_jiffy;
-		if (c > nocb_nobypass_lim_per_jiffy)
-			c = nocb_nobypass_lim_per_jiffy;
-		else if (c < 0)
+		if (ULONG_CMP_LT(rdp->nocb_nobypass_count,
+				 nocb_nobypass_lim_per_jiffy))
 			c = 0;
+		else if (c > nocb_nobypass_lim_per_jiffy)
+			c = nocb_nobypass_lim_per_jiffy;
 	}
 	WRITE_ONCE(rdp->nocb_nobypass_count, c);