diff --git a/lib/cpumask.c b/lib/cpumask.c
index 35fe142..beca624 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -33,10 +33,11 @@ EXPORT_SYMBOL(cpumask_next);
 int cpumask_next_and(int n, const struct cpumask *src1p,
 		     const struct cpumask *src2p)
 {
-	while ((n = cpumask_next(n, src1p)) < nr_cpu_ids)
-		if (cpumask_test_cpu(n, src2p))
-			break;
-	return n;
+	/* -1 is a legal arg here. */
+	if (n != -1)
+		cpumask_check(n);
+	return find_next_and_bit(cpumask_bits(src1p), cpumask_bits(src2p),
+		nr_cpumask_bits, n + 1);
 }
 EXPORT_SYMBOL(cpumask_next_and);
 
diff --git a/lib/find_bit.c b/lib/find_bit.c
index 6ed74f7..ee3df93 100644
--- a/lib/find_bit.c
+++ b/lib/find_bit.c
@@ -21,22 +21,29 @@
 #include <linux/export.h>
 #include <linux/kernel.h>
 
-#if !defined(find_next_bit) || !defined(find_next_zero_bit)
+#if !defined(find_next_bit) || !defined(find_next_zero_bit) || \
+		!defined(find_next_and_bit)
 
 /*
- * This is a common helper function for find_next_bit and
- * find_next_zero_bit.  The difference is the "invert" argument, which
- * is XORed with each fetched word before searching it for one bits.
+ * This is a common helper function for find_next_bit, find_next_zero_bit, and
+ * find_next_and_bit. The differences are:
+ *  - The "invert" argument, which is XORed with each fetched word before
+ *    searching it for one bits.
+ *  - The optional "addr2", which is anded with "addr1" if present.
  */
-static unsigned long _find_next_bit(const unsigned long *addr,
-		unsigned long nbits, unsigned long start, unsigned long invert)
+static inline unsigned long _find_next_bit(const unsigned long *addr1,
+		const unsigned long *addr2, unsigned long nbits,
+		unsigned long start, unsigned long invert)
 {
 	unsigned long tmp;
 
 	if (unlikely(start >= nbits))
 		return nbits;
 
-	tmp = addr[start / BITS_PER_LONG] ^ invert;
+	tmp = addr1[start / BITS_PER_LONG];
+	if (addr2)
+		tmp &= addr2[start / BITS_PER_LONG];
+	tmp ^= invert;
 
 	/* Handle 1st word. */
 	tmp &= BITMAP_FIRST_WORD_MASK(start);
@@ -47,7 +54,10 @@ static unsigned long _find_next_bit(const unsigned long *addr,
 		if (start >= nbits)
 			return nbits;
 
-		tmp = addr[start / BITS_PER_LONG] ^ invert;
+		tmp = addr1[start / BITS_PER_LONG];
+		if (addr2)
+			tmp &= addr2[start / BITS_PER_LONG];
+		tmp ^= invert;
 	}
 
 	return min(start + __ffs(tmp), nbits);
@@ -61,7 +71,7 @@ static unsigned long _find_next_bit(const unsigned long *addr,
 unsigned long find_next_bit(const unsigned long *addr, unsigned long size,
 			    unsigned long offset)
 {
-	return _find_next_bit(addr, size, offset, 0UL);
+	return _find_next_bit(addr, NULL, size, offset, 0UL);
 }
 EXPORT_SYMBOL(find_next_bit);
 #endif
@@ -70,11 +80,21 @@ EXPORT_SYMBOL(find_next_bit);
 unsigned long find_next_zero_bit(const unsigned long *addr, unsigned long size,
 				 unsigned long offset)
 {
-	return _find_next_bit(addr, size, offset, ~0UL);
+	return _find_next_bit(addr, NULL, size, offset, ~0UL);
 }
 EXPORT_SYMBOL(find_next_zero_bit);
 #endif
 
+#if !defined(find_next_and_bit)
+unsigned long find_next_and_bit(const unsigned long *addr1,
+		const unsigned long *addr2, unsigned long size,
+		unsigned long offset)
+{
+	return _find_next_bit(addr1, addr2, size, offset, 0UL);
+}
+EXPORT_SYMBOL(find_next_and_bit);
+#endif
+
 #ifndef find_first_bit
 /*
  * Find the first set bit in a memory region.
@@ -146,15 +166,19 @@ static inline unsigned long ext2_swab(const unsigned long y)
 }
 
 #if !defined(find_next_bit_le) || !defined(find_next_zero_bit_le)
-static unsigned long _find_next_bit_le(const unsigned long *addr,
-		unsigned long nbits, unsigned long start, unsigned long invert)
+static inline unsigned long _find_next_bit_le(const unsigned long *addr1,
+		const unsigned long *addr2, unsigned long nbits,
+		unsigned long start, unsigned long invert)
 {
 	unsigned long tmp;
 
 	if (unlikely(start >= nbits))
 		return nbits;
 
-	tmp = addr[start / BITS_PER_LONG] ^ invert;
+	tmp = addr1[start / BITS_PER_LONG];
+	if (addr2)
+		tmp &= addr2[start / BITS_PER_LONG];
+	tmp ^= invert;
 
 	/* Handle 1st word. */
 	tmp &= ext2_swab(BITMAP_FIRST_WORD_MASK(start));
@@ -165,7 +189,10 @@ static unsigned long _find_next_bit_le(const unsigned long *addr,
 		if (start >= nbits)
 			return nbits;
 
-		tmp = addr[start / BITS_PER_LONG] ^ invert;
+		tmp = addr1[start / BITS_PER_LONG];
+		if (addr2)
+			tmp &= addr2[start / BITS_PER_LONG];
+		tmp ^= invert;
 	}
 
 	return min(start + __ffs(ext2_swab(tmp)), nbits);
@@ -176,7 +203,7 @@ static unsigned long _find_next_bit_le(const unsigned long *addr,
 unsigned long find_next_zero_bit_le(const void *addr, unsigned
 		long size, unsigned long offset)
 {
-	return _find_next_bit_le(addr, size, offset, ~0UL);
+	return _find_next_bit_le(addr, NULL, size, offset, ~0UL);
 }
 EXPORT_SYMBOL(find_next_zero_bit_le);
 #endif
@@ -185,7 +212,7 @@ EXPORT_SYMBOL(find_next_zero_bit_le);
 unsigned long find_next_bit_le(const void *addr, unsigned
 		long size, unsigned long offset)
 {
-	return _find_next_bit_le(addr, size, offset, 0UL);
+	return _find_next_bit_le(addr, NULL, size, offset, 0UL);
 }
 EXPORT_SYMBOL(find_next_bit_le);
 #endif
diff --git a/lib/find_bit_benchmark.c b/lib/find_bit_benchmark.c
index 67b1923..5985a25 100644
--- a/lib/find_bit_benchmark.c
+++ b/lib/find_bit_benchmark.c
@@ -35,6 +35,7 @@
 #define SPARSE		500
 
 static DECLARE_BITMAP(bitmap, BITMAP_LEN) __initdata;
+static DECLARE_BITMAP(bitmap2, BITMAP_LEN) __initdata;
 
 /*
  * This is Schlemiel the Painter's algorithm. It should be called after
@@ -103,6 +104,22 @@ static int __init test_find_last_bit(const void *bitmap, unsigned long len)
 	return 0;
 }
 
+static int __init test_find_next_and_bit(const void *bitmap,
+		const void *bitmap2, unsigned long len)
+{
+	unsigned long i, cnt;
+	cycles_t cycles;
+
+	cycles = get_cycles();
+	for (cnt = i = 0; i < BITMAP_LEN; cnt++)
+		i = find_next_and_bit(bitmap, bitmap2, BITMAP_LEN, i+1);
+	cycles = get_cycles() - cycles;
+	pr_err("find_next_and_bit:\t\t%llu cycles, %ld iterations\n",
+		(u64)cycles, cnt);
+
+	return 0;
+}
+
 static int __init find_bit_test(void)
 {
 	unsigned long nbits = BITMAP_LEN / SPARSE;
@@ -110,23 +127,29 @@ static int __init find_bit_test(void)
 	pr_err("\nStart testing find_bit() with random-filled bitmap\n");
 
 	get_random_bytes(bitmap, sizeof(bitmap));
+	get_random_bytes(bitmap2, sizeof(bitmap2));
 
 	test_find_next_bit(bitmap, BITMAP_LEN);
 	test_find_next_zero_bit(bitmap, BITMAP_LEN);
 	test_find_last_bit(bitmap, BITMAP_LEN);
 	test_find_first_bit(bitmap, BITMAP_LEN);
+	test_find_next_and_bit(bitmap, bitmap2, BITMAP_LEN);
 
 	pr_err("\nStart testing find_bit() with sparse bitmap\n");
 
 	bitmap_zero(bitmap, BITMAP_LEN);
+	bitmap_zero(bitmap2, BITMAP_LEN);
 
-	while (nbits--)
+	while (nbits--) {
 		__set_bit(prandom_u32() % BITMAP_LEN, bitmap);
+		__set_bit(prandom_u32() % BITMAP_LEN, bitmap2);
+	}
 
 	test_find_next_bit(bitmap, BITMAP_LEN);
 	test_find_next_zero_bit(bitmap, BITMAP_LEN);
 	test_find_last_bit(bitmap, BITMAP_LEN);
 	test_find_first_bit(bitmap, BITMAP_LEN);
+	test_find_next_and_bit(bitmap, bitmap2, BITMAP_LEN);
 
 	/*
 	 * Everything is OK. Return error just to let user run benchmark