refactored threading wait group handling

- added `async_for` and `go_for` allows passing multi arguments directly, by converting macro to func, and update examples
- add `grouping` field to wait group type pointer to hold wait result pointer
- fix various atomic race conditions,  example `benchmark.c` now execute and complete as expected,

README.md

@@ -110,14 +110,14 @@ a terminated/finish status.
 
 The initialization ends when `wait_for()` is called, as such current coroutine will pause, and
 execution will begin for the group of coroutines, and wait for all to finished. */
-C_API wait_group_t *wait_group(void);
+C_API wait_group_t wait_group(void);
 
 /* Pauses current coroutine, and begin execution for given coroutine wait group object, will
 wait for all to finished. Returns hast table of results, accessible by coroutine id. */
-C_API wait_result_t wait_for(wait_group_t *);
+C_API wait_result_t wait_for(wait_group_t);
 
 /* Returns results of the given completed coroutine id, value in union value_t storage format. */
-C_API value_t wait_result(wait_result_t *, int);
+C_API value_t wait_result(wait_result_t, int);
 
 /* Creates an unbuffered channel, similar to golang channels. */
 C_API channel_t *channel(void);
@@ -619,12 +619,12 @@ void_t worker(void_t arg)
 int co_main(int argc, char **argv)
 {
     int cid[5];
-    wait_group_t *wg = wait_group();
+    wait_group_t wg = wait_group();
     for (int i = 1; i <= 5; i++)
     {
        cid[i-1] = go(worker, &i);
     }
-    wait_result_t *wgr = wait_for(wg);
+    wait_result_t wgr = wait_for(wg);
 
     printf("\nWorker # %d returned: %d\n",
            cid[2],

docs/index.md

@@ -110,14 +110,14 @@ a terminated/finish status.
 
 The initialization ends when `wait_for()` is called, as such current coroutine will pause, and
 execution will begin for the group of coroutines, and wait for all to finished. */
-C_API wait_group_t *wait_group(void);
+C_API wait_group_t wait_group(void);
 
 /* Pauses current coroutine, and begin execution for given coroutine wait group object, will
 wait for all to finished. Returns hast table of results, accessible by coroutine id. */
-C_API wait_result_t wait_for(wait_group_t *);
+C_API wait_result_t wait_for(wait_group_t);
 
 /* Returns results of the given completed coroutine id, value in union value_t storage format. */
-C_API value_t wait_result(wait_result_t *, int);
+C_API value_t wait_result(wait_result_t, int);
 
 /* Creates an unbuffered channel, similar to golang channels. */
 C_API channel_t *channel(void);
@@ -634,11 +634,11 @@ void_t worker(void_t arg) {
 int co_main(int argc, char **argv)
 {
     int cid[5];
-    wait_group_t *wg = wait_group();
+    wait_group_t wg = wait_group();
     for (int i = 1; i <= 5; i++) {
        cid[i-1] = go(worker, &i);
     }
-    wait_result_t *wgr = wait_for(wg);
+    wait_result_t wgr = wait_for(wg);
 
     printf("\nWorker # %d returned: %d\n",
            cid[2],

examples/benchmark.c

@@ -47,7 +47,7 @@ int co_main(int argc, char **argv) {
         numRoutines = (u32)atoi(argv[1]);
 
     co_interrupt_off();
-    wait_group_t *wg = wait_group_by(numRoutines);
+    wait_group_t wg = wait_group_by(numRoutines);
     for (i = 0; i < numRoutines; i++) {
         go(func, NULL);
     }

examples/go_multi_args.c

@@ -1,7 +1,7 @@
 
 #include "coroutine.h"
 
-void *worker(void *arg) {
+void *worker(args_t *arg) {
     int i, count = args_get(arg, 0).integer;
     char *text = args_get(arg, 1).char_ptr;
 
@@ -13,9 +13,9 @@ void *worker(void *arg) {
 }
 
 int co_main(int argc, char **argv) {
-    go(worker, args_for("is", 4, "a"));
-    go(worker, args_for("is", 2, "b"));
-    go(worker, args_for("is", 3, "c"));
+    go_for(worker, "is", 4, "a");
+    go_for(worker, "is", 2, "b");
+    go_for(worker, "is", 3, "c");
 
     sleep_for(100);
 

examples/go_wait_group.c

@@ -23,11 +23,11 @@ void *worker(void *arg) {
 int co_main(int argc, char **argv) {
     int cid[50], i;
 
-    wait_group_t *wg = wait_group_by(50);
+    wait_group_t wg = wait_group_by(50);
     for (i = 0; i < 50; i++) {
         cid[i] = go(worker, args_for("i", i));
     }
-    wait_result_t *wgr = wait_for(wg);
+    wait_result_t wgr = wait_for(wg);
 
     printf("\nWorker # %d returned: %d\n", cid[2], wait_result(wgr, cid[2]).integer);
     printf("\nWorker # %d returned: %s\n", cid[1], wait_result(wgr, cid[1]).char_ptr);