Unify environment variable management (#235)

* Add environment variable configuration infrastructure
  - Namespace rocshmem::envvar
  - Track all config env vars in per-category lists
  - Remove duplicates from list of allowed env var types
  - Reject negative inputs for unsigned integer types
  - Accept empty strings for std::string
  - Print error source location using C++20 std::source_location
  - Unit tests
* Port environment variables
  - ROCSHMEM_UNIQUEID_WITH_MPI
  - ROCSHMEM_RO_DISABLE_IPC
  - ROCSHMEM_BOOTSTRAP_TIMEOUT
  - ROCSHMEM_BOOTSTRAP_HOSTID
  - ROCSHMEM_BOOTSTRAP_SOCKET_IFNAME
  - ROCSHMEM_RO_PROGRESS_DELAY
  - ROCSHMEM_BOOTSTRAP_SOCKET_FAMILY
  - ROCSHMEM_MAX_NUM_CONTEXTS
    + Merge the independent per-backend copies into a single variable
      that is used by all three backends (IPC, RO, GDA).
    + Set default to 32 (for GDA); prior default for IPC and RO was 1024.
  - ROCSHMEM_MAX_NUM_HOST_CONTEXTS
  - ROCSHMEM_MAX_WF_BUFFERS
  - ROCSHMEM_SQ_SIZE
  - ROCSHMEM_RO_NET_CPU_QUEUE
    + Renamed from RO_NET_CPU_QUEUE
    + Change env var input type to bool, default to false
    + Invert code logic: setting RO_NET_CPU_QUEUE to anything
      would /disable/ a variable gpu_queue, which defaulted to true.
      Variable is now named config::ro::net_cpu_queue,
      with all prior checks for gpu_queue inverted.
  - ROCSHMEM_USE_IB_HCA
  - ROCSHMEM_HEAP_SIZE
    + Defaults to 1L << 30 i.e. 1 GiB,
      from default heap size in memory/heap_memory.hpp.
  - ROCSHMEM_MAX_NUM_TEAMS
    + Unlike other env vars, this can be referenced from devices.
    + Function currently narrows from size_t to int: uses need to be audited
      for safety and correctness in using size_t directly.
  - ROCSHMEM_GDA_ALTERNATE_QP_PORTS
* New env var ROCSHMEM_DEBUG
  - Debug levels:
    + NONE
    + VERSION
    + WARN
    + INFO
    + TRACE
  - Currently unused - will be added later
  - Mirrors RCCL debug control
* Remove rocshmem::rocshmem_env_config
* Change interface for GetClosestNicToGpu
  to accept const char** instead of char**:
  the pointed-to string does not need to be modified
  - Files were not audited for inclusion of util.hpp only for env vars
---------
Signed-off-by: Omri Mor <Omri.Mor@amd.com>

[ROCm/rocshmem commit: a0fcbf8d35]

projects/rocshmem/src/host/host.cpp

@@ -25,6 +25,7 @@
 #include "host.hpp"
 
 #include "rocshmem/rocshmem_config.h"  // NOLINT(build/include_subdir)
+#include "envvar.hpp"
 #include "host_helpers.hpp"
 #include "memory/window_info.hpp"
 #include "util.hpp"
@@ -69,7 +70,7 @@ __host__ void HostInterface::release_window_context(WindowInfo* window_info) {
 }
 
 int HostInterface::find_avail_pool_entry() {
-  for (int i{0}; i < max_num_ctxs_; i++) {
+  for (size_t i = 0; i < envvar::max_num_host_contexts; i++) {
     if (host_window_context_pool_[i]->is_avail()) {
       return i;
     }
@@ -78,7 +79,7 @@ int HostInterface::find_avail_pool_entry() {
 }
 
 int HostInterface::find_win_info_in_pool(WindowInfo* window_info) {
-  for (int i{0}; i < max_num_ctxs_; i++) {
+  for (size_t i = 0; i < envvar::max_num_host_contexts; i++) {
     if (host_window_context_pool_[i]->is_avail()) {
       continue;
     }
@@ -109,16 +110,11 @@ __host__ HostInterface::HostInterface(HdpPolicy* hdp_policy,
   /*
    * Allocate and initialize pool of windows for contexts
    */
-  char* value{nullptr};
-  if ((value = getenv("ROCSHMEM_MAX_NUM_HOST_CONTEXTS"))) {
-    max_num_ctxs_ = atoi(value);
-  }
-
-  size_t pool_size = max_num_ctxs_ * sizeof(HostContextWindowInfo*);
+  size_t pool_size = envvar::max_num_host_contexts * sizeof(HostContextWindowInfo*);
   host_window_context_pool_ =
       reinterpret_cast<HostContextWindowInfo**>(malloc(pool_size));
 
-  for (int ctx_i = 0; ctx_i < max_num_ctxs_; ctx_i++) {
+  for (size_t ctx_i = 0; ctx_i < envvar::max_num_host_contexts; ctx_i++) {
     host_window_context_pool_[ctx_i] =
         new HostContextWindowInfo(host_comm_world_, heap);
   }
@@ -164,16 +160,11 @@ __host__ HostInterface::HostInterface(HdpPolicy* hdp_policy,
   /*
    * Allocate and initialize pool of windows for contexts
    */
-  char* value{nullptr};
-  if ((value = getenv("ROCSHMEM_MAX_NUM_HOST_CONTEXTS"))) {
-    max_num_ctxs_ = atoi(value);
-  }
-
-  size_t pool_size = max_num_ctxs_ * sizeof(HostContextWindowInfo*);
+  size_t pool_size = envvar::max_num_host_contexts * sizeof(HostContextWindowInfo*);
   host_window_context_pool_ =
       reinterpret_cast<HostContextWindowInfo**>(malloc(pool_size));
 
-  for (int ctx_i = 0; ctx_i < max_num_ctxs_; ctx_i++) {
+  for (size_t ctx_i = 0; ctx_i < envvar::max_num_host_contexts; ctx_i++) {
     host_window_context_pool_[ctx_i] =
         new HostContextWindowInfo(heap);
   }
@@ -194,7 +185,7 @@ __host__ HostInterface::~HostInterface() {
   /* Detroy the pool of contexts */
 
   if (host_window_context_pool_ != nullptr) {
-    for (int ctx_i = 0; ctx_i < max_num_ctxs_; ctx_i++) {
+    for (size_t ctx_i = 0; ctx_i < envvar::max_num_host_contexts; ctx_i++) {
       delete host_window_context_pool_[ctx_i];
     }
     free(host_window_context_pool_);

projects/rocshmem/src/host/host.hpp

@@ -349,11 +349,6 @@ class HostInterface {
   MPI_Win hdp_win;
 #endif  // USE_HDP_FLUSH
 
-  /**
-   * @brief Max number of contexts for the application
-   */
-  int max_num_ctxs_{1};
-
   /**
    * @brief Pool of HostContexWindowInfos
    */