Update Barrier_All and Sync_All APIs (#72)

* Fix deadlock in `rocshmem_ctx_wg_barrier_all` API in IPC conduit by adding per-context pSync buffers and context IDs
  - Added separate pSync buffers for each device context
  - Resolved deadlock when invoking barrier API (`rocshmem_ctx_wg_barrier_all`) concurrently from multiple contexts

* Update barrier_all functional tests for multi-context support

* Add thread, wavefront, and workgroup-level barrier_all APIs in IPC and RO conduits
  - Implemented barrier_all APIs at thread, wavefront, and workgroup granularity
  - Added support in both IPC and RO conduits
  - Updated functional tests to cover all `barrier_all` APIs

* Add thread, wavefront, and workgroup-level sync_all APIs in IPC and RO conduits
  - Implemented sync_all APIs for thread, wavefront, and workgroup scopes
  - Added support into both IPC and RO conduits
  - Added functional tests to cover all `sync_all` APIs

src/rocshmem_gpu.cpp

@@ -570,12 +570,24 @@ __device__ int rocshmem_test(T *ivars, int cmp, T val) {
   return ctx_internal->test(ivars, cmp, val);
 }
 
-__device__ void rocshmem_ctx_wg_barrier_all(rocshmem_ctx_t ctx) {
+__device__ void rocshmem_ctx_barrier_all(rocshmem_ctx_t ctx) {
   GPU_DPRINTF("Function: rocshmem_ctx_barrier_all\n");
 
   get_internal_ctx(ctx)->barrier_all();
 }
 
+__device__ void rocshmem_ctx_wave_barrier_all(rocshmem_ctx_t ctx) {
+  GPU_DPRINTF("Function: rocshmem_ctx_wave_barrier_all\n");
+
+  get_internal_ctx(ctx)->barrier_all_wave();
+}
+
+__device__ void rocshmem_ctx_wg_barrier_all(rocshmem_ctx_t ctx) {
+  GPU_DPRINTF("Function: rocshmem_ctx_wg_barrier_all\n");
+
+  get_internal_ctx(ctx)->barrier_all_wg();
+}
+
 __device__ void rocshmem_wg_barrier_all() {
   rocshmem_ctx_wg_barrier_all(ROCSHMEM_CTX_DEFAULT);
 }
@@ -586,12 +598,32 @@ __device__ void rocshmem_barrier(rocshmem_team_t team) {
   get_internal_ctx(ROCSHMEM_CTX_DEFAULT)->barrier(team);
 }
 
-__device__ void rocshmem_ctx_wg_sync_all(rocshmem_ctx_t ctx) {
+__device__ void rocshmem_ctx_sync_all(rocshmem_ctx_t ctx) {
   GPU_DPRINTF("Function: rocshmem_ctx_sync_all\n");
 
   get_internal_ctx(ctx)->sync_all();
 }
 
+__device__ void rocshmem_sync_all() {
+  rocshmem_ctx_sync_all(ROCSHMEM_CTX_DEFAULT);
+}
+
+__device__ void rocshmem_ctx_wave_sync_all(rocshmem_ctx_t ctx) {
+  GPU_DPRINTF("Function: rocshmem_ctx_wave_sync_all\n");
+
+  get_internal_ctx(ctx)->sync_all_wave();
+}
+
+__device__ void rocshmem_wave_sync_all() {
+  rocshmem_ctx_wave_sync_all(ROCSHMEM_CTX_DEFAULT);
+}
+
+__device__ void rocshmem_ctx_wg_sync_all(rocshmem_ctx_t ctx) {
+  GPU_DPRINTF("Function: rocshmem_ctx_wg_sync_all\n");
+
+  get_internal_ctx(ctx)->sync_all_wg();
+}
+
 __device__ void rocshmem_wg_sync_all() {
   rocshmem_ctx_wg_sync_all(ROCSHMEM_CTX_DEFAULT);
 }
@@ -600,7 +632,7 @@ __device__ void rocshmem_ctx_wg_team_sync(rocshmem_ctx_t ctx,
                                            rocshmem_team_t team) {
   GPU_DPRINTF("Function: rocshmem_ctx_sync_all\n");
 
-  get_internal_ctx(ctx)->sync(team);
+  get_internal_ctx(ctx)->sync_wg(team);
 }
 
 __device__ void rocshmem_wg_team_sync(rocshmem_team_t team) {