non-hipGraph MSCCL++ tests for allReduce and allGather (#1503)

* working tests for a single message size * move call_RCCL routine StandaloneUtils, create .cpp file for StandaloneUtils so that it can be included in several tests * simplify test invocation * remove unecessary logs and exit from ncclCommRegister * set expected results for allGather * skip test if nranks doesn't match number of gpus, call getAndDistributeNCCLid only from parent process * fix improper size of expected-results vector * Removing unused changes. * Refactored to create a new file for the forked collectives call, as StandaloneUtils is for the Standalone tests. Renamed the functions to be slightly more accurate and follow existing naming conventions. * Apply suggestions from code review Co-authored-by: corey-derochie-amd <161367113+corey-derochie-amd@users.noreply.github.com> --------- Co-authored-by: isaki001 <isakioti@banff-pla-r27-38.pla.dcgpu> Co-authored-by: corey-derochie-amd <161367113+corey-derochie-amd@users.noreply.github.com> Co-authored-by: Corey Derochie <corey.derochie@amd.com> [ROCm/rccl commit: 3398fa78fe]
2025-02-04 09:11:32 -06:00
parent d2b5ba80a7
commit a40d4eb960
7 changed files with 339 additions and 68 deletions
@@ -0,0 +1,76 @@
+/*************************************************************************
+ * Copyright (c) 2022 Advanced Micro Devices, Inc. All rights reserved.
+ *
+ * See LICENSE.txt for license information
+ ************************************************************************/
+#include "CollectiveArgs.hpp"
+#include "StandaloneUtils.hpp"
+#include <iostream>
+#include <regex>
+
+
+namespace RcclUnitTesting
+{
+
+std::string executeCommand(const char* cmd) {
+    std::string result;
+    FILE* pipe = popen(cmd, "r");
+
+    if (!pipe) {
+        std::cerr << "Error executing command: " << cmd << std::endl;
+        return result;
+    }
+
+    char buffer[128];
+    while (!feof(pipe)) {
+        if (fgets(buffer, 128, pipe) != NULL) {
+            result += buffer;
+        }
+    }
+
+    pclose(pipe);
+    return result;
+}
+
+std::vector<std::string> splitString(const std::string& str, char delimiter) {
+    std::vector<std::string> result;
+    std::istringstream iss(str);
+
+    std::string line;
+    while(std::getline(iss, line, delimiter)) {
+        result.push_back(line);
+    }
+
+    return result;
+}
+
+
+ArchInfo parseMetadata(const std::vector<std::string>& list) {
+    ArchInfo archInfo;
+    KernelInfo currKernelInfo;
+    
+    std::regex amdhsaTargetRegex("amdhsa.target:\\s+(?:'?)amdgcn-amd-amdhsa--(\\w+)(?:'?)");
+    std::regex kernelNameRegex("\\.name:\\s+(\\w+)");
+    std::regex privateSegmentSizeRegex("\\.private_segment_fixed_size:\\s+(\\d+)");
+    
+    for (const auto& line : list) {
+        std::smatch match;
+
+        if (std::regex_search(line, match, amdhsaTargetRegex)) {
+            archInfo.archName = match[1];
+        } else if (std::regex_search(line, match, kernelNameRegex)) {
+            currKernelInfo.name = match[1];
+        } else if (std::regex_search(line, match, privateSegmentSizeRegex)) {
+            currKernelInfo.privateSegmentFixedSize = std::stoi(match[1]);
+        }
+        
+        if (!currKernelInfo.name.empty() && currKernelInfo.privateSegmentFixedSize != 0) {
+            archInfo.kernels.push_back(currKernelInfo);
+            currKernelInfo = {}; // Empty kernelInfo
+        }
+    }
+    
+    return archInfo;
+}
+
+}