TL/CUDA: add support for mem map

TOOLS: add global and local memory type

TL/CUDA: add push algorithm

TL/CUDA: add alltoallv memh

Author: wfaderhold21

src/components/tl/cuda/Makefile.am

@@ -19,15 +19,17 @@ allgatherv =                       \
 	allgatherv/allgatherv_ring.c   \
 	allgatherv/allgatherv_linear.c
 
-alltoall =                 \
-	alltoall/alltoall.h    \
-	alltoall/alltoall.c    \
-	alltoall/alltoall_ce.c
-
-alltoallv =                  \
-	alltoallv/alltoallv.h    \
-	alltoallv/alltoallv.c    \
-	alltoallv/alltoallv_ce.c
+alltoall =                          \
+	alltoall/alltoall.h             \
+	alltoall/alltoall.c             \
+	alltoall/alltoall_ce.c          \
+	alltoall/alltoall_push.c
+
+alltoallv =                      \
+	alltoallv/alltoallv.h        \
+	alltoallv/alltoallv.c        \
+	alltoallv/alltoallv_ce.c     \
+	alltoallv/alltoallv_push.c
 
 bcast =                  \
 	bcast/bcast.h        \

src/components/tl/cuda/alltoall/alltoall.c

@@ -1,12 +1,25 @@
 /**
- * Copyright (c) 2021-2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+ * Copyright (c) 2021-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
  *
  * See file LICENSE for terms.
  */
 
 #include "alltoall.h"
 #include "components/mc/ucc_mc.h"
 
+ucc_base_coll_alg_info_t
+    ucc_tl_cuda_alltoall_algs[UCC_TL_CUDA_ALLTOALL_ALG_LAST + 1] = {
+        [UCC_TL_CUDA_ALLTOALL_ALG_CE] =
+            {.id   = UCC_TL_CUDA_ALLTOALL_ALG_CE,
+             .name = "ce",
+             .desc = "copy-engine alltoall algorithm"},
+        [UCC_TL_CUDA_ALLTOALL_ALG_PUSH] =
+            {.id   = UCC_TL_CUDA_ALLTOALL_ALG_PUSH,
+             .name = "push",
+             .desc = "push-based alltoall using pre-registered dst handles"},
+        [UCC_TL_CUDA_ALLTOALL_ALG_LAST] = {
+            .id = 0, .name = NULL, .desc = NULL}};
+
 ucc_status_t ucc_tl_cuda_alltoall_ce_init(ucc_tl_cuda_task_t *task);
 
 ucc_status_t ucc_tl_cuda_alltoall_ce_start(ucc_coll_task_t *task);

src/components/tl/cuda/alltoall/alltoall.h

@@ -1,5 +1,5 @@
 /**
- * Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+ * Copyright (c) 2021-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
  *
  * See file LICENSE for terms.
  */
@@ -9,9 +9,33 @@
 
 #include "tl_cuda.h"
 #include "tl_cuda_coll.h"
+#include "components/base/ucc_base_iface.h"
+
+enum {
+    UCC_TL_CUDA_ALLTOALL_ALG_CE   = 0,
+    UCC_TL_CUDA_ALLTOALL_ALG_PUSH = 1,
+    UCC_TL_CUDA_ALLTOALL_ALG_LAST
+};
+
+extern ucc_base_coll_alg_info_t
+    ucc_tl_cuda_alltoall_algs[UCC_TL_CUDA_ALLTOALL_ALG_LAST + 1];
+
+static inline int ucc_tl_cuda_alltoall_alg_from_str(const char *str)
+{
+    int i;
+    for (i = 0; i < UCC_TL_CUDA_ALLTOALL_ALG_LAST; i++) {
+        if (0 == strcasecmp(str, ucc_tl_cuda_alltoall_algs[i].name))
+            return i;
+    }
+    return -1;
+}
 
 ucc_status_t ucc_tl_cuda_alltoall_init(ucc_base_coll_args_t *coll_args,
                                        ucc_base_team_t      *tl_team,
                                        ucc_coll_task_t     **task_p);
 
+ucc_status_t ucc_tl_cuda_alltoall_push_init(ucc_base_coll_args_t *coll_args,
+                                            ucc_base_team_t      *tl_team,
+                                            ucc_coll_task_t     **task_p);
+
 #endif