diff --git a/winsup/cygwin/include/sys/cpuset.h b/winsup/cygwin/include/sys/cpuset.h
index 4857b879d..2056f6af7 100644
--- a/winsup/cygwin/include/sys/cpuset.h
+++ b/winsup/cygwin/include/sys/cpuset.h
@@ -18,8 +18,8 @@ typedef __SIZE_TYPE__ __cpu_mask;
 #define __NCPUBITS (8 * sizeof (__cpu_mask))  // max size of processor group
 #define __CPU_GROUPMAX (__CPU_SETSIZE / __NCPUBITS)  // maximum group number
 
-#define __CPUELT(cpu)   ((cpu) / __NCPUBITS)
-#define __CPUMASK(cpu)  ((__cpu_mask) 1 << ((cpu) % __NCPUBITS))
+#define __CPUELT(cpu)  ((cpu) / __NCPUBITS)
+#define __CPUMASK(cpu) ((__cpu_mask) 1 << ((cpu) % __NCPUBITS))
 
 typedef struct
 {
@@ -28,6 +28,66 @@ typedef struct
 
 int __sched_getaffinity_sys (pid_t, size_t, cpu_set_t *);
 
+/* These macros alloc or free dynamically-sized cpu sets of size 'num' cpus.
+   Allocations are padded such that full-word operations can be done easily. */
+#define CPU_ALLOC_SIZE(num) ((num+__NCPUBITS-1) / __NCPUBITS) * sizeof (__cpu_mask)
+#define CPU_ALLOC(num)      __builtin_malloc (CPU_ALLOC_SIZE(num))
+#define CPU_FREE(set)       __builtin_free (set)
+
+/* These _S macros operate on dynamically-sized cpu sets of size 'siz' bytes */
+#define CPU_ZERO_S(siz, set)    __builtin_memset (set, 0, siz)
+
+#define CPU_SET_S(cpu,siz,set) \
+	if (cpu < 8 * siz) \
+	  (set)->__bits[__CPUELT(cpu)] |= __CPUMASK(cpu);
+
+#define CPU_CLR_S(cpu,siz,set) \
+	if (cpu < 8 * siz) \
+	  (set)->__bits[__CPUELT(cpu)] &= ~(__CPUMASK(cpu));
+
+#define CPU_ISSET_S(cpu,siz,set) \
+      ({int res = 0; \
+	if (cpu < 8 * siz) \
+	  res = !!((set)->__bits[__CPUELT(cpu)] & __CPUMASK(cpu)); \
+	res;})
+
+#define CPU_COUNT_S(siz, set) \
+      ({int tot = 0;\
+	for (int i = 0; i < siz / sizeof (__cpu_mask); i++) \
+	  tot += __builtin_popcountl ((set)->__bits[i]); \
+	tot;})
+
+#define CPU_AND_S(siz, dst, src1, src2) \
+	for (int i = 0; i < siz / sizeof (__cpu_mask); i++) \
+	  (dst)->__bits[i] = (src1)->__bits[i] & (src2)->__bits[i];
+
+#define CPU_OR_S(siz, dst, src1, src2) \
+	for (int i = 0; i < siz / sizeof (__cpu_mask); i++) \
+	  (dst)->__bits[i] = (src1)->__bits[i] | (src2)->__bits[i];
+
+#define CPU_XOR_S(siz, dst, src1, src2) \
+	for (int i = 0; i < siz / sizeof (__cpu_mask); i++) \
+	  (dst)->__bits[i] = (src1)->__bits[i] ^ (src2)->__bits[i];
+
+#define CPU_EQUAL_S(siz, src1, src2) \
+      ({int res = 1; \
+	for (int i = 0; res && i < siz / sizeof (__cpu_mask); i++) \
+	  res &= (src1)->__bits[i] == (src2)->__bits[i]; \
+	res;})
+
+/* These macros operate on fixed-size cpu sets of size __CPU_SETSIZE cpus */
+#define CPU_ZERO(set)             CPU_ZERO_S(sizeof (cpu_set_t), set)
+
+#define CPU_SET(cpu, set)         CPU_SET_S(cpu, sizeof (cpu_set_t), set)
+#define CPU_CLR(cpu, set)         CPU_CLR_S(cpu, sizeof (cpu_set_t), set)
+#define CPU_ISSET(cpu, set)       CPU_ISSET_S(cpu, sizeof (cpu_set_t), set)
+
+#define CPU_COUNT(set)            CPU_COUNT_S(sizeof (cpu_set_t), set)
+#define CPU_AND(dst, src1, src2)  CPU_AND_S(sizeof (cpu_set_t), dst, src1, src2)
+#define CPU_OR(dst, src1, src2)   CPU_OR_S(sizeof (cpu_set_t), dst, src1, src2)
+#define CPU_XOR(dst, src1, src2)  CPU_XOR_S(sizeof (cpu_set_t), dst, src1, src2)
+#define CPU_EQUAL(src1, src2)     CPU_EQUAL_S(sizeof (cpu_set_t), src1, src2)
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/winsup/cygwin/release/3.1.0 b/winsup/cygwin/release/3.1.0
index d5d3129c8..2c1e8d2a8 100644
--- a/winsup/cygwin/release/3.1.0
+++ b/winsup/cygwin/release/3.1.0
@@ -6,7 +6,7 @@ What's new:
   which uses the nearest color from 16 system colors.
 
 - New APIs: sched_getaffinity, sched_setaffinity, pthread_getaffinity_np,
-  pthread_setaffinity_np.
+  pthread_setaffinity_np, plus CPU_SET macros.
 
 - New APIs: dbm_clearerr, dbm_close, dbm_delete, dbm_dirfno, dbm_error,
   dbm_fetch, dbm_firstkey, dbm_nextkey, dbm_open, dbm_store.
diff --git a/winsup/doc/new-features.xml b/winsup/doc/new-features.xml
index a5dcfb576..5fee581e7 100644
--- a/winsup/doc/new-features.xml
+++ b/winsup/doc/new-features.xml
@@ -29,15 +29,9 @@ If a SA_SIGINFO signal handler changes the ucontext_t pointed to by the
 third parameter, follow it after returning from the handler.
 </para></listitem>
 
-<listitem><para>
-Support for getting and setting process and thread affinities.  New APIs:
-sched_getaffinity, sched_setaffinity, pthread_getaffinity_np,
-pthread_setaffinity_np.
-</para></listitem>
-
 <listitem><para>
 New APIs: sched_getaffinity, sched_setaffinity, pthread_getaffinity_np,
-pthread_setaffinity_np.
+pthread_setaffinity_np, plus CPU_SET macros.
 </para></listitem>
 
 <listitem><para>