Re: [PATCH] udev enhancements to use kernel event queue

[Patrick Mochel]

> > +	spin_lock(&sequence_lock);
> > +	seq = sequence_num++;
> > +	spin_unlock(&sequence_lock);
> > +
> > +	envp [i++] = scratch;
> > +	scratch += sprintf(scratch, "SEQNUM=%ld", seq) + 1;
> 
> Nice thinking. It is a shame we need a lock for this, but we don't have
> an atomic_inc_and_return().

Those were my sentiments exactly:

16:21  * mochel searches for atomic_inc_and_read() :)

It seems like the following should work. Would someone mind commenting on
it?

Thanks,

	-pat

===== include/asm/atomic.h 1.4 vs edited =====
--- 1.4/include/asm-i386/atomic.h	Mon Feb  4 23:42:13 2002
+++ edited/include/asm/atomic.h	Thu Jun 12 16:32:10 2003
@@ -110,6 +110,23 @@
 		:"m" (v->counter));
 }
 
+
+/**
+ * atomic_inc_and_read - increment atomic variable and return new value
+ * @v: pointer of type atomic_t
+ * 
+ * Atomically increments @v by 1.  Note that the guaranteed
+ * useful range of an atomic_t is only 24 bits.
+ */ 
+static inline int atomic_inc_and_read(atomic_t *v)
+{
+	__asm__ __volatile__(
+		LOCK "incl %0"
+		:"=m" (v->counter)
+		:"m" (v->counter));
+	return v->counter;
+}
+
 /**
  * atomic_dec - decrement atomic variable
  * @v: pointer of type atomic_t