From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Wed, 14 Apr 2010 16:56:53 +0000 (-0400)
Subject: Simplify qsbr online condition
X-Git-Tag: v0.4.4~4
X-Git-Url: https://git.lttng.org./?a=commitdiff_plain;h=92cb20a84557a365cabcde63cb19809a7cbb2651;p=userspace-rcu.git

Simplify qsbr online condition

Alan Stern wrote:

>       return v && (v - rcu_gp_ctr > ULONG_MAX / 2);
>
> Wouldn't it be better to write:
>
>       return v != 0 && v != rcu_gp_ctr;
>
> or something equivalent?  The point is, it's foolish to return 0 when
> rcu_gp_ctr has wrapped around to below v.  This can happen only when
> rcu_thread_online() is delayed (as by preemption), in which case we
> know that there's a good chance the algorithm will fail.  My version of
> the test eliminates the possibility of this failure, although it
> doesn't eliminate the possibility that rcu_gp_ctr has wrapped around so
> far that it is equal to v.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---

diff --git a/urcu-qsbr-static.h b/urcu-qsbr-static.h
index 8822e05..125301f 100644
--- a/urcu-qsbr-static.h
+++ b/urcu-qsbr-static.h
@@ -169,7 +169,7 @@ static inline int rcu_gp_ongoing(unsigned long *ctr)
 	unsigned long v;
 
 	v = LOAD_SHARED(*ctr);
-	return v && (v - rcu_gp_ctr > ULONG_MAX / 2);
+	return v && (v != rcu_gp_ctr);
 }
 #endif  /* !(BITS_PER_LONG < 64) */