diff --git a/kernel-src/include/kernel/usercopy.h b/kernel-src/include/kernel/usercopy.h
index 290bae4c..aeb25ab4 100644
--- a/kernel-src/include/kernel/usercopy.h
+++ b/kernel-src/include/kernel/usercopy.h
@@ -17,13 +17,20 @@ static inline int _usercopy_strlen_wrapper(void *str, size_t *size) {
 	return 0;
 }
 
+static inline int _usercopy_memset_wrapper(void *dst, uint8_t byte, size_t size) {
+	memset(dst, byte, size);
+	return 0;
+}
+
 #define USERCOPY_POSSIBLY_FROM_USER(kernel, user, size) (IS_USER_ADDRESS(user) ? usercopy_fromuser(kernel, user, size) : _usercopy_memcpy_wrapper(kernel, user, size))
 #define USERCOPY_POSSIBLY_TO_USER(user, kernel, size) (IS_USER_ADDRESS(user) ? usercopy_touser(user, kernel, size) : _usercopy_memcpy_wrapper(user, kernel, size))
 #define USERCOPY_POSSIBLY_STRLEN_FROM_USER(user, sizep) (IS_USER_ADDRESS(user) ? usercopy_strlen(user, sizep) : _usercopy_strlen_wrapper(user, sizep))
+#define USERCOPY_POSSIBLY_MEMSET_TO_USER(user, byte, size) (IS_USER_ADDRESS(user) ? usercopy_memset(user, byte, size) : _usercopy_memset_wrapper(user, byte, size))
 
 int usercopy_touser(void *user, void *kernel, size_t size);
 int usercopy_fromuser(void *kernel, void *user, size_t size);
 int usercopy_fromuseratomic32(uint32_t *user32, uint32_t *value);
 int usercopy_strlen(const char *, size_t *size);
+int usercopy_memset(void *user, uint8_t byte, size_t size);
 
 #endif
diff --git a/kernel-src/mm/usercopy.c b/kernel-src/mm/usercopy.c
index 13e2d540..e65a9cf7 100644
--- a/kernel-src/mm/usercopy.c
+++ b/kernel-src/mm/usercopy.c
@@ -18,6 +18,12 @@ typedef struct {
 	uint32_t *address;
 } atomic32desc_t;
 
+typedef struct {
+	void *dst;
+	uint8_t byte;
+	size_t size;
+} memsetdesc_t;
+
 static void memcpyinternal(context_t *ctx, void *arg) {
 	memcpydesc_t *desc = arg;
 
@@ -60,6 +66,20 @@ static void atomic32internal(context_t *ctx, void *arg) {
 	current_thread()->usercopyctx = NULL;
 }
 
+static void memset_internal(context_t *ctx, void *arg) {
+	memsetdesc_t *desc = arg;
+
+	// this sets up the thread so that in the next invalid page fault, it will jump to
+	// this context with CTX_RET set to EFAULT
+	current_thread()->usercopyctx = ctx;
+
+	memset(desc->dst, desc->byte, desc->size);
+
+	// no errors!
+	CTX_RET(ctx) = 0;
+	current_thread()->usercopyctx = NULL;
+}
+
 int usercopy_touser(void *user, void *kernel, size_t size) {
 	memcpydesc_t desc = {
 		.dst = user,
@@ -109,3 +129,16 @@ int usercopy_fromuseratomic32(uint32_t *user32, uint32_t *value) {
 
 	return arch_context_saveandcall(atomic32internal, NULL, &desc);
 }
+
+int usercopy_memset(void *user, uint8_t byte, size_t size) {
+	memsetdesc_t desc = {
+		.dst = user,
+		.byte = byte,
+		.size = size
+	};
+
+	if (IS_USER_ADDRESS(user) == false)
+		return EFAULT;
+
+	return arch_context_saveandcall(memset_internal, NULL, &desc);
+}