2.16.2-1

Add support for CUDA 12.0, drop Kepler (sm_35).
Support for H100 features.
Make socket code more robust and protected. Solves #555.
Improve performance on large CUDA graphs, reducing dependencies.
Reduce inter-socket bandwidth on AMD CPUs to favor better paths.
Various fixes to ncclCommAbort.
Make service thread polling resistant to EINTR.
Compile with profiling API by default.
Extend NVTX instrumentation with call arguments.

src/include/comm.h

@@ -171,9 +171,12 @@ struct ncclComm {
   uint64_t* connectSend;
   uint64_t* connectRecv;
 
+  uint64_t magic; // Magic number for all network communication. Not a security key -- only goal is to detect mismatches.
+
   int rank;    // my rank in the communicator
   int nRanks;  // number of GPUs in communicator
   int cudaDev; // my cuda device index
+  int compCap; // compute capability of the GPU
   int64_t busId;   // my PCI bus ID in int format
   cpu_set_t cpuAffinity; // CPU affinity of the GPU
 
@@ -208,7 +211,7 @@ struct ncclComm {
 
   // Buffer sizes
   int buffSizes[NCCL_NUM_PROTOCOLS];
-  int p2pNetChunkSize;
+  int p2pChunkSize;
 
   // Algorithm/Protocols thresholds
   ssize_t threadThresholds[NCCL_NUM_ALGORITHMS][NCCL_NUM_PROTOCOLS];
@@ -240,7 +243,6 @@ struct ncclComm {
   // Intra-process sync
   struct ncclComm* intraComm0; // leader of intra-process comms (self possible)
   struct ncclComm* intraNext; // next of intra-process comms, intraComm0 is head
-  int intraRefs; // reference count from intra-process comms (zero if not leader else intraRanks)
   int intraRank;
   int intraRanks;
   uint32_t intraBarrierPhase;