add C11 thread creation and related thread functions

based on patch by Jens Gustedt.

the main difficulty here is handling the difference between start
function signatures and thread return types for C11 threads versus
POSIX threads. pointers to void are assumed to be able to represent
faithfully all values of int. the function pointer for the thread
start function is cast to an incorrect type for passing through
pthread_create, but is cast back to its correct type before calling so
that the behavior of the call is well-defined.

changes to the existing threads implementation were kept minimal to
reduce the risk of regressions, and duplication of code that carries
implementation-specific assumptions was avoided for ease and safety of
future maintenance.

src/internal/pthread_impl.h

@@ -128,4 +128,6 @@ void __restore_sigs(void *);
 #define DEFAULT_STACK_SIZE 81920
 #define DEFAULT_GUARD_SIZE PAGE_SIZE
 
+#define __ATTRP_C11_THREAD ((void*)(uintptr_t)-1)
+
 #endif

src/thread/pthread_create.c

@@ -119,7 +119,15 @@ static int start(void *p)
 	if (self->unblock_cancel)
 		__syscall(SYS_rt_sigprocmask, SIG_UNBLOCK,
 			SIGPT_SET, 0, _NSIG/8);
-	pthread_exit(self->start(self->start_arg));
+	__pthread_exit(self->start(self->start_arg));
+	return 0;
+}
+
+static int start_c11(void *p)
+{
+	pthread_t self = p;
+	int (*start)(void*) = (int(*)(void*)) self->start;
+	__pthread_exit((void *)(uintptr_t)start(self->start_arg));
 	return 0;
 }
 
@@ -145,7 +153,7 @@ void *__copy_tls(unsigned char *);
 
 int __pthread_create(pthread_t *restrict res, const pthread_attr_t *restrict attrp, void *(*entry)(void *), void *restrict arg)
 {
-	int ret;
+	int ret, c11 = (attrp == __ATTRP_C11_THREAD);
 	size_t size, guard;
 	struct pthread *self, *new;
 	unsigned char *map = 0, *stack = 0, *tsd = 0, *stack_limit;
@@ -167,7 +175,7 @@ int __pthread_create(pthread_t *restrict res, const pthread_attr_t *restrict att
 		self->tsd = (void **)__pthread_tsd_main;
 		libc.threaded = 1;
 	}
-	if (attrp) attr = *attrp;
+	if (attrp && !c11) attr = *attrp;
 
 	__acquire_ptc();
 
@@ -234,7 +242,7 @@ int __pthread_create(pthread_t *restrict res, const pthread_attr_t *restrict att
 	new->canary = self->canary;
 
 	a_inc(&libc.threads_minus_1);
-	ret = __clone(start, stack, flags, new, &new->tid, TP_ADJ(new), &new->tid);
+	ret = __clone((c11 ? start_c11 : start), stack, flags, new, &new->tid, TP_ADJ(new), &new->tid);
 
 	__release_ptc();
 

src/thread/pthread_detach.c

 #include "pthread_impl.h"
+#include <threads.h>
 
 int __pthread_join(pthread_t, void **);
 
-int __pthread_detach(pthread_t t)
+static int __pthread_detach(pthread_t t)
 {
 	/* Cannot detach a thread that's already exiting */
 	if (a_swap(t->exitlock, 1))
@@ -13,3 +14,4 @@ int __pthread_detach(pthread_t t)
 }
 
 weak_alias(__pthread_detach, pthread_detach);
+weak_alias(__pthread_detach, thrd_detach);

src/thread/pthread_equal.c

 #include <pthread.h>
+#include <threads.h>
+#include "libc.h"
 
-int (pthread_equal)(pthread_t a, pthread_t b)
+static int __pthread_equal(pthread_t a, pthread_t b)
 {
 	return a==b;
 }
+
+weak_alias(__pthread_equal, pthread_equal);
+weak_alias(__pthread_equal, thrd_equal);

src/thread/pthread_self.c

 #include "pthread_impl.h"
+#include <threads.h>
+#include "libc.h"
 
-pthread_t pthread_self()
+static pthread_t __pthread_self_internal()
 {
 	return __pthread_self();
 }
+
+weak_alias(__pthread_self_internal, pthread_self);
+weak_alias(__pthread_self_internal, thrd_current);

src/thread/thrd_create.c

+#include "pthread_impl.h"
+#include <threads.h>
+
+int __pthread_create(pthread_t *restrict, const pthread_attr_t *restrict, void *(*)(void *), void *restrict);
+
+int thrd_create(thrd_t *thr, thrd_start_t func, void *arg)
+{
+	int ret = __pthread_create(thr, __ATTRP_C11_THREAD, (void *(*)(void *))func, arg);
+	switch (ret) {
+	case 0:      return thrd_success;
+	case EAGAIN: return thrd_nomem;
+	default:     return thrd_error;
+	}
+}

src/thread/thrd_exit.c

+#include "pthread_impl.h"
+#include <threads.h>
+
+_Noreturn void __pthread_exit(void *);
+
+_Noreturn void thrd_exit(int result)
+{
+	__pthread_exit((void*)(intptr_t)result);
+}

src/thread/thrd_join.c

+#include <stdint.h>
+#include <threads.h>
+
+int __pthread_join(thrd_t, void**);
+
+int thrd_join(thrd_t t, int *res)
+{
+        void *pthread_res;
+        __pthread_join(t, &pthread_res);
+        if (res) *res = (int)(intptr_t)pthread_res;
+        return thrd_success;
+}

src/thread/thrd_sleep.c

+#include <threads.h>
+#include <errno.h>
+#include "syscall.h"
+
+int thrd_sleep(const struct timespec *req, struct timespec *rem)
+{
+	int ret = __syscall(SYS_nanosleep, req, rem);
+	switch (ret) {
+	case 0:      return 0;
+	case -EINTR: return -1; /* value specified by C11 */
+	default:     return -2;
+	}
+}

src/thread/thrd_yield.c

+#include <threads.h>
+#include "syscall.h"
+
+void thrd_yield()
+{
+	__syscall(SYS_sched_yield);
+}