Merge commit '3d4813d99196bb349eccd50a925e2addc8f1622c' into develop

projects/rccl/ext-profiler/example/nccl/profiler.h

@@ -11,17 +11,20 @@
 #include <stdlib.h>
 
 #include "common.h"
-#include "err.h"
 
 enum {
-  ncclProfileGroup     = (1 << 0),  // group event type
-  ncclProfileColl      = (1 << 1),  // host collective call event type
-  ncclProfileP2p       = (1 << 2),  // host point-to-point call event type
-  ncclProfileProxyOp   = (1 << 3),  // proxy operation event type
-  ncclProfileProxyStep = (1 << 4),  // proxy step event type
-  ncclProfileProxyCtrl = (1 << 5),  // proxy control event type
-  ncclProfileKernelCh  = (1 << 6),  // kernel channel event type
-  ncclProfileNetPlugin = (1 << 7),  // network plugin-defined, events
+  ncclProfileGroup          = (1 << 0),  // group event type
+  ncclProfileColl           = (1 << 1),  // host collective call event type
+  ncclProfileP2p            = (1 << 2),  // host point-to-point call event type
+  ncclProfileProxyOp        = (1 << 3),  // proxy operation event type
+  ncclProfileProxyStep      = (1 << 4),  // proxy step event type
+  ncclProfileProxyCtrl      = (1 << 5),  // proxy control event type
+  ncclProfileKernelCh       = (1 << 6),  // kernel channel event type
+  ncclProfileNetPlugin      = (1 << 7),  // network plugin-defined, events
+  ncclProfileGroupApi       = (1 << 8),  // Group API events
+  ncclProfileCollApi        = (1 << 9),  // Collective API events
+  ncclProfileP2pApi         = (1 << 10), // Point-to-Point API events
+  ncclProfileKernelLaunch   = (1 << 11), // Kernel launch events
 };
 
 typedef enum {
@@ -56,21 +59,27 @@ typedef enum {
 
   /* Kernel event states */
   ncclProfilerKernelChStop             = 22,
+
+  /* Group API States */
+  ncclProfilerEndGroupApiStart         = 23,
+  ncclProfilerBeginGroupApiEnd         = 24
 } ncclProfilerEventState_t;
 
 typedef ncclProfilerEventState_t ncclProfilerEventState_v1_t;
 typedef ncclProfilerEventState_t ncclProfilerEventState_v2_t;
 typedef ncclProfilerEventState_t ncclProfilerEventState_v3_t;
 typedef ncclProfilerEventState_t ncclProfilerEventState_v4_t;
+typedef ncclProfilerEventState_t ncclProfilerEventState_v5_t;
 
+#include "profiler_v5.h"
 #include "profiler_v4.h"
 #include "profiler_v3.h"
 #include "profiler_v2.h"
 #include "profiler_v1.h"
 #include "profiler_net.h"
 
-typedef ncclProfiler_v4_t ncclProfiler_t;
-typedef ncclProfilerEventDescr_v4_t ncclProfilerEventDescr_t;
-typedef ncclProfilerEventStateArgs_v4_t ncclProfilerEventStateArgs_t;
+typedef ncclProfiler_v5_t ncclProfiler_t;
+typedef ncclProfilerEventDescr_v5_t ncclProfilerEventDescr_t;
+typedef ncclProfilerEventStateArgs_v5_t ncclProfilerEventStateArgs_t;
 
 #endif // end include guard

projects/rccl/ext-profiler/example/nccl/profiler_v5.h

@@ -0,0 +1,152 @@
+/*************************************************************************
+ * Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
+ *
+ * See LICENSE.txt for license information
+ ************************************************************************/
+
+#ifndef PROFILER_V5_H_
+#define PROFILER_V5_H_
+#include <stdbool.h>
+
+typedef struct {
+  uint64_t type;                // event type descriptor: ncclProfileGroupApi, ...
+  void* parentObj;              // pointer to the profiler parent object
+  int rank;                     // originating rank
+  union {
+    struct {
+      int graphCaptured;
+      int groupDepth;
+    } groupApi;
+
+    struct {
+      const char* func;
+      size_t count;
+      const char* datatype;
+      int root;
+      void* stream;
+      bool graphCaptured;
+    } collApi;
+
+    struct {
+      const char* func;
+      size_t count;
+      const char* datatype;
+      void* stream;
+      bool graphCaptured;
+    } p2pApi;
+
+    struct {
+      void* stream;
+    } kernelLaunch;
+
+    struct {
+      uint64_t seqNumber;
+      const char* func;
+      void const* sendBuff;
+      void* recvBuff;
+      size_t count;
+      int root;
+      const char* datatype;
+      uint8_t nChannels;
+      uint8_t nWarps;
+      const char* algo;
+      const char* proto;
+      void* parentGroup; // for backward compatibility with v4
+    } coll;
+
+    struct {
+      const char* func;
+      void* buff;
+      const char* datatype;
+      size_t count;
+      int peer;
+      uint8_t nChannels;
+      void* parentGroup; // for backward compatibility with v4
+    } p2p;
+
+    struct {
+      pid_t pid;                // pid of the originating process
+      uint8_t channelId;        // channel id for this proxy operation
+      int peer;                 // remote rank for send/recv
+      int nSteps;               // number of steps for this proxy operation
+      int chunkSize;            // amount of data transferred by this proxy operation
+      int isSend;
+    } proxyOp;
+
+    struct {
+      int step;
+    } proxyStep;
+
+    struct {
+      uint8_t channelId;
+      uint64_t pTimer;          // start timestamp from GPU globaltimer
+    } kernelCh;
+
+    struct {
+      int64_t id;
+      void* data;
+    } netPlugin;
+  };
+} ncclProfilerEventDescr_v5_t;
+
+typedef union {
+  struct {
+    size_t transSize;
+  } proxyStep;
+
+  struct {
+    int appendedProxyOps;
+  } proxyCtrl;
+
+  struct {
+    void* data;
+  } netPlugin;
+
+  struct {
+    uint64_t pTimer;
+  } kernelCh;
+} ncclProfilerEventStateArgs_v5_t;
+
+typedef struct {
+  const char* name;
+
+  // init - initialize the profiler plugin
+  // Input
+  //  - context        : opaque profiler context object for separating profiler behavior across comms
+  //  - commId         : communicator id
+  //  - commName       : user assigned communicator name
+  //  - nNodes         : number of nodes in communicator
+  //  - nranks         : number of ranks in communicator
+  //  - rank           : rank identifier in communicator
+  //  - logfn          : logger function
+  // Output
+  //  - eActivationMask: bitmask of active events set by the plugin
+  ncclResult_t (*init)(void** context, uint64_t commId, int* eActivationMask, const char* commName, int nNodes, int nranks, int rank, ncclDebugLogger_t logfn);
+
+  // startEvent - initialize and start a new event for the supplied event descriptor inside the eventset
+  // Input
+  //  - context: opaque profiler context object
+  //  - eDescr : pointer to ncclProfilerEventDescr_t object
+  // Output
+  //  - eHandle: return event handle for supplied event descriptor object
+  ncclResult_t (*startEvent)(void* context, void** eHandle, ncclProfilerEventDescr_v5_t* eDescr);
+
+  // stopEvent - stop/finalize an event inside and event set
+  // Input
+  //  - eHandle: handle to event object
+  ncclResult_t (*stopEvent)(void* eHandle);
+
+  // recordEventState - record event state transitions and event attribute updates
+  // Input
+  //  - eHandle   : handle to event object created through startEvent
+  //  - eStateArgs: optional argument used to capture event attribute updates associated with the state transition
+  //  - eState    : event state transition
+  ncclResult_t (*recordEventState)(void* eHandle, ncclProfilerEventState_v5_t eState, ncclProfilerEventStateArgs_v5_t* eStateArgs);
+
+  // finalize - finalize the profiler plugin
+  // Input
+  //  - context: opaque profiler context object
+  ncclResult_t (*finalize)(void* context);
+} ncclProfiler_v5_t;
+
+#endif