作者｜KIDGINBROOK
更新｜潘丽晨

NCCL是英伟达开源的GPU通信库，支持集合通信和点对点通信。

看下官方给的一个demo：

#include <stdio.h>
#include "cuda_runtime.h"
#include "nccl.h"
#include "mpi.h"
#include <unistd.h>
#include <stdint.h>#define MPICHECK(cmd) do {                          \int e = cmd;                                      \if( e != MPI_SUCCESS ) {                          \printf("Failed: MPI error %s:%d '%d'\n",        \__FILE__,__LINE__, e);   \exit(EXIT_FAILURE);                             \}                                                 \
} while(0)#define CUDACHECK(cmd) do {                         \cudaError_t e = cmd;                              \if( e != cudaSuccess ) {                          \printf("Failed: Cuda error %s:%d '%s'\n",             \__FILE__,__LINE__,cudaGetErrorString(e));   \exit(EXIT_FAILURE);                             \}                                                 \
} while(0)#define NCCLCHECK(cmd) do {                         \ncclResult_t r = cmd;                             \if (r!= ncclSuccess) {                            \printf("Failed, NCCL error %s:%d '%s'\n",             \__FILE__,__LINE__,ncclGetErrorString(r));   \exit(EXIT_FAILURE);                             \}                                                 \
} while(0)static uint64_t getHostHash(const char* string) {// Based on DJB2a, result = result * 33 ^ charuint64_t result = 5381;for (int c = 0; string[c] != '\0'; c++){result = ((result << 5) + result) ^ string[c];}return result;
}static void getHostName(char* hostname, int maxlen) {gethostname(hostname, maxlen);for (int i=0; i< maxlen; i++) {if (hostname[i] == '.') {hostname[i] = '\0';return;}}
}int main(int argc, char* argv[])
{int size = 32*1024*1024;int myRank, nRanks, localRank = 0;//initializing MPIMPICHECK(MPI_Init(&argc, &argv));MPICHECK(MPI_Comm_rank(MPI_COMM_WORLD, &myRank));MPICHECK(MPI_Comm_size(MPI_COMM_WORLD, &nRanks));//calculating localRank which is used in selecting a GPUuint64_t hostHashs[nRanks];char hostname[1024];getHostName(hostname, 1024);hostHashs[myRank] = getHostHash(hostname);MPICHECK(MPI_Allgather(MPI_IN_PLACE, 0, MPI_DATATYPE_NULL, hostHashs, sizeof(uint64_t), MPI_BYTE, MPI_COMM_WORLD));for (int p=0; p<nRanks; p++) {if (p == myRank) break;if (hostHashs[p] == hostHashs[myRank]) localRank++;}//each process is using two GPUsint nDev = 2;float** sendbuff = (float**)malloc(nDev * sizeof(float*));float** recvbuff = (float**)malloc(nDev * sizeof(float*));cudaStream_t* s = (cudaStream_t*)malloc(sizeof(cudaStream_t)*nDev);//picking GPUs based on localRankfor (int i = 0; i < nDev; ++i) {CUDACHECK(cudaSetDevice(localRank*nDev + i));CUDACHECK(cudaMalloc(sendbuff + i, size * sizeof(float)));CUDACHECK(cudaMalloc(recvbuff + i, size * sizeof(float)));CUDACHECK(cudaMemset(sendbuff[i], 1, size * sizeof(float)));CUDACHECK(cudaMemset(recvbuff[i], 0, size * sizeof(float)));CUDACHECK(cudaStreamCreate(s+i));}ncclUniqueId id;ncclComm_t comms[nDev];//generating NCCL unique ID at one process and broadcasting it to allif (myRank == 0) ncclGetUniqueId(&id);MPICHECK(MPI_Bcast((void *)&id, sizeof(id), MPI_BYTE, 0, MPI_COMM_WORLD));//initializing NCCL, group API is required around ncclCommInitRank as it is//called across multiple GPUs in each thread/processNCCLCHECK(ncclGroupStart());for (int i=0; i<nDev; i++) {CUDACHECK(cudaSetDevice(localRank*nDev + i));NCCLCHECK(ncclCommInitRank(comms+i, nRanks*nDev, id, myRank*nDev + i));}NCCLCHECK(ncclGroupEnd());//calling NCCL communication API. Group API is required when using//multiple devices per thread/processNCCLCHECK(ncclGroupStart());for (int i=0; i<nDev; i++)NCCLCHECK(ncclAllReduce((const void*)sendbuff[i], (void*)recvbuff[i], size, ncclFloat, ncclSum,comms[i], s[i]));NCCLCHECK(ncclGroupEnd());//synchronizing on CUDA stream to complete NCCL communicationfor (int i=0; i<nDev; i++)CUDACHECK(cudaStreamSynchronize(s[i]));//freeing device memoryfor (int i=0; i<nDev; i++) {CUDACHECK(cudaFree(sendbuff[i]));CUDACHECK(cudaFree(recvbuff[i]));}//finalizing NCCLfor (int i=0; i<nDev; i++) {ncclCommDestroy(comms[i]);}//finalizing MPIMPICHECK(MPI_Finalize());printf("[MPI Rank %d] Success \n", myRank);return 0;
}

在上边的示例中，rank0会执行ncclGetUniqueId获取Id，然后通过mpi广播给其他rank，接下来看下UniqueId是怎么产生的。

ncclResult_t ncclGetUniqueId(ncclUniqueId* out) {NCCLCHECK(ncclInit());NCCLCHECK(PtrCheck(out, "GetUniqueId", "out"));return bootstrapGetUniqueId(out);
}

然后看下ncclInit。

首先执行initEnv，设置环境变量。

然后执行initNet，用来初始化nccl所需要的网络，包括两个，一个是bootstrap网络，另外一个是数据通信网络，bootstrap网络主要用于初始化时交换一些简单的信息，比如每个机器的ip端口，由于数据量很小，而且主要是在初始化阶段执行一次，因此bootstrap使用的是tcp；而通信网络是用于实际数据的传输，因此会优先使用rdma（支持gdr的话会优先使用gdr）。

ncclResult_t initNet() {// Always initialize bootstrap networkNCCLCHECK(bootstrapNetInit());NCCLCHECK(initNetPlugin(&ncclNet, &ncclCollNet));if (ncclNet != NULL) return ncclSuccess;if (initNet(&ncclNetIb) == ncclSuccess) {ncclNet = &ncclNetIb;} else {NCCLCHECK(initNet(&ncclNetSocket));ncclNet = &ncclNetSocket;}return ncclSuccess;
}

bootstrapNetInit就是bootstrap网络的初始化，主要就是通过findInterfaces遍历机器上所有的网卡信息，通过prefixList匹配选择使用哪些网卡，将可用网卡的信息保存下来，将ifa_name保存到全局的bootstrapNetIfNames，ip地址保存到全局bootstrapNetIfAddrs，默认除了docker和lo其他的网卡都可以使用。

例如在测试机器上有三张网卡，分别是xgbe0、xgbe1、xgbe2，那么就会把这三个ifaname和对应的ip地址保存下来，另外nccl提供了环境变量NCCL_SOCKET_IFNAME可以用来指定想用的网卡名，例如通过export NCCL_SOCKET_IFNAME=xgbe0来指定使用xgbe0，其实就是通过prefixList来匹配做到的。

static int findInterfaces(const char* prefixList, char* names, union socketAddress *addrs, int sock_family, int maxIfNameSize, int maxIfs) {struct netIf userIfs[MAX_IFS];bool searchNot = prefixList && prefixList[0] == '^';if (searchNot) prefixList++;bool searchExact = prefixList && prefixList[0] == '=';if (searchExact) prefixList++;int nUserIfs = parseStringList(prefixList, userIfs, MAX_IFS);int found = 0;struct ifaddrs *interfaces, *interface;getifaddrs(&interfaces);for (interface = interfaces; interface && found < maxIfs; interface = interface->ifa_next) {if (interface->ifa_addr == NULL) continue;int family = interface->ifa_addr->sa_family;if (family != AF_INET && family != AF_INET6)continue;if (sock_family != -1 && family != sock_family)continue;if (family == AF_INET6) {struct sockaddr_in6* sa = (struct sockaddr_in6*)(interface->ifa_addr);if (IN6_IS_ADDR_LOOPBACK(&sa->sin6_addr)) continue;}if (!(matchIfList(interface->ifa_name, -1, userIfs, nUserIfs, searchExact) ^ searchNot)) {continue;}bool duplicate = false;for (int i = 0; i < found; i++) {if (strcmp(interface->ifa_name, names+i*maxIfNameSize) == 0) { duplicate = true; break; }}if (!duplicate) {strncpy(names+found*maxIfNameSize, interface->ifa_name, maxIfNameSize);int salen = (family == AF_INET) ? sizeof(sockaddr_in) : sizeof(sockaddr_in6);memcpy(addrs+found, interface->ifa_addr, salen);found++;}}freeifaddrs(interfaces);return found;
}

开始初始化通信网络。

ncclNet_t结构体是一系列的函数指针，比如初始化，发送，接收等；socket，IB等通信方式都实现了自己的ncclNet_t，如ncclNetSocket，ncclNetIb，初始化通信网络的过程就是依次看哪个通信模式可用，然后赋值给全局的ncclNet。

首先执行initNetPlugin，查看是否有libnccl-net.so，测试环境没有这个so，所以直接返回。

然后尝试使用IB网络：

首先执行ncclNetIb的init函数，就是ncclIbInit。

ncclResult_t ncclIbInit(ncclDebugLogger_t logFunction) {static int shownIbHcaEnv = 0;if(wrap_ibv_symbols() != ncclSuccess) { return ncclInternalError; }if (ncclParamIbDisable()) return ncclInternalError;if (ncclNIbDevs == -1) {pthread_mutex_lock(&ncclIbLock);wrap_ibv_fork_init();if (ncclNIbDevs == -1) {ncclNIbDevs = 0;if (findInterfaces(ncclIbIfName, &ncclIbIfAddr, MAX_IF_NAME_SIZE, 1) != 1) {WARN("NET/IB : No IP interface found.");return ncclInternalError;}// Detect IB cardsint nIbDevs;struct ibv_device** devices;// Check if user defined which IB device:port to usechar* userIbEnv = getenv("NCCL_IB_HCA");if (userIbEnv != NULL && shownIbHcaEnv++ == 0) INFO(NCCL_NET|NCCL_ENV, "NCCL_IB_HCA set to %s", userIbEnv);struct netIf userIfs[MAX_IB_DEVS];bool searchNot = userIbEnv && userIbEnv[0] == '^';if (searchNot) userIbEnv++;bool searchExact = userIbEnv && userIbEnv[0] == '=';if (searchExact) userIbEnv++;int nUserIfs = parseStringList(userIbEnv, userIfs, MAX_IB_DEVS);if (ncclSuccess != wrap_ibv_get_device_list(&devices, &nIbDevs)) return ncclInternalError;for (int d=0; d<nIbDevs && ncclNIbDevs<MAX_IB_DEVS; d++) {struct ibv_context * context;if (ncclSuccess != wrap_ibv_open_device(&context, devices[d]) || context == NULL) {WARN("NET/IB : Unable to open device %s", devices[d]->name);continue;}int nPorts = 0;struct ibv_device_attr devAttr;memset(&devAttr, 0, sizeof(devAttr));if (ncclSuccess != wrap_ibv_query_device(context, &devAttr)) {WARN("NET/IB : Unable to query device %s", devices[d]->name);if (ncclSuccess != wrap_ibv_close_device(context)) { return ncclInternalError; }continue;}for (int port = 1; port <= devAttr.phys_port_cnt; port++) {struct ibv_port_attr portAttr;if (ncclSuccess != wrap_ibv_query_port(context, port, &portAttr)) {WARN("NET/IB : Unable to query port %d", port);continue;}if (portAttr.state != IBV_PORT_ACTIVE) continue;if (portAttr.link_layer != IBV_LINK_LAYER_INFINIBAND&& portAttr.link_layer != IBV_LINK_LAYER_ETHERNET) continue;// check against user specified HCAs/portsif (! (matchIfList(devices[d]->name, port, userIfs, nUserIfs, searchExact) ^ searchNot)) {continue;}TRACE(NCCL_INIT|NCCL_NET,"NET/IB: [%d] %s:%d/%s ", d, devices[d]->name, port,portAttr.link_layer == IBV_LINK_LAYER_INFINIBAND ? "IB" : "RoCE");ncclIbDevs[ncclNIbDevs].device = d;ncclIbDevs[ncclNIbDevs].guid = devAttr.sys_image_guid;ncclIbDevs[ncclNIbDevs].port = port;ncclIbDevs[ncclNIbDevs].link = portAttr.link_layer;ncclIbDevs[ncclNIbDevs].speed = ncclIbSpeed(portAttr.active_speed) * ncclIbWidth(portAttr.active_width);ncclIbDevs[ncclNIbDevs].context = context;strncpy(ncclIbDevs[ncclNIbDevs].devName, devices[d]->name, MAXNAMESIZE);NCCLCHECK(ncclIbGetPciPath(ncclIbDevs[ncclNIbDevs].devName, &ncclIbDevs[ncclNIbDevs].pciPath, &ncclIbDevs[ncclNIbDevs].realPort));ncclIbDevs[ncclNIbDevs].maxQp = devAttr.max_qp;ncclNIbDevs++;nPorts++;pthread_create(&ncclIbAsyncThread, NULL, ncclIbAsyncThreadMain, context);}if (nPorts == 0 && ncclSuccess != wrap_ibv_close_device(context)) { return ncclInternalError; }}if (nIbDevs && (ncclSuccess != wrap_ibv_free_device_list(devices))) { return ncclInternalError; };}if (ncclNIbDevs == 0) {INFO(NCCL_INIT|NCCL_NET, "NET/IB : No device found.");} else {char line[1024];line[0] = '\0';for (int d=0; d<ncclNIbDevs; d++) {snprintf(line+strlen(line), 1023-strlen(line), " [%d]%s:%d/%s", d, ncclIbDevs[d].devName,ncclIbDevs[d].port, ncclIbDevs[d].link == IBV_LINK_LAYER_INFINIBAND ? "IB" : "RoCE");}line[1023] = '\0';char addrline[1024];INFO(NCCL_INIT|NCCL_NET, "NET/IB : Using%s ; OOB %s:%s", line, ncclIbIfName, socketToString(&ncclIbIfAddr.sa, addrline));}pthread_mutex_unlock(&ncclIbLock);}return ncclSuccess;
}

首先第三行通过wrap_ibv_symbols加载动态库libibverbs.so，然后获取动态库的各个函数。

然后通过wrap_ibv_fork_init避免fork引起rdma网卡读写出错。

后面会讲到ib网络也会用到socket进行带外网络的传输，所以这里也通过findInterfaces获取一个可用的网卡保存到ncclIbIfAddr。

通过ibv_get_device_list获取所有rdma设备到devices中，遍历devices的每个device，因为每个HCA可能有多个物理port，所以对每个device遍历每一个物理port，获取每个port的信息。

然后将相关信息保存到全局的ncclIbDevs中，比如是哪个device的哪个port，使用的是IB还是ROCE，device的pci路径，maxqp，device的name等，注意这里也有类似bootstrap网络NCCL_SOCKET_IFNAME的环境变量，叫NCCL_IB_HCA，可以指定使用哪个IB HCA。

到这里整个初始化的过程就完成了，一句话总结就是，获取了当前机器上所有可用的IB网卡和普通以太网卡之后保存下来。

然后开始生成UniqueId。

ncclResult_t bootstrapCreateRoot(ncclUniqueId* id, bool idFromEnv) {ncclNetHandle_t* netHandle = (ncclNetHandle_t*) id;void* listenComm;NCCLCHECK(bootstrapNetListen(idFromEnv ? dontCareIf : 0, netHandle, &listenComm));pthread_t thread;pthread_create(&thread, NULL, bootstrapRoot, listenComm);return ncclSuccess;
}

ncclNetHandle_t也是一个字符数组，然后执行bootstrapNetListen。

static ncclResult_t bootstrapNetListen(int dev, ncclNetHandle_t* netHandle, void** listenComm) {union socketAddress* connectAddr = (union socketAddress*) netHandle;static_assert(sizeof(union socketAddress) < NCCL_NET_HANDLE_MAXSIZE, "union socketAddress size is too large");// if dev >= 0, listen based on devif (dev >= 0) {NCCLCHECK(bootstrapNetGetSocketAddr(dev, connectAddr));} else if (dev == findSubnetIf) {...} // Otherwise, handle stores a local addressstruct bootstrapNetComm* comm;NCCLCHECK(bootstrapNetNewComm(&comm));NCCLCHECK(createListenSocket(&comm->fd, connectAddr));*listenComm = comm;return ncclSuccess;
}

依次看下这三个函数，通过bootstrapNetGetSocketAddr获取一个可用的ip地址。

static ncclResult_t bootstrapNetGetSocketAddr(int dev, union socketAddress* addr) {if (dev >= bootstrapNetIfs) return ncclInternalError;memcpy(addr, bootstrapNetIfAddrs+dev, sizeof(*addr));return ncclSuccess;
}

此时dev是0， bootstrapNetIfs是初始化bootstrap网络的时候一共找到了几个可用的网卡，这里就是获取了第0个可用的ip地址。

然后通过bootstrapNetNewComm创建bootstrapNetComm，bootstrapNetComm其实就是fd，bootstrapNetNewComm其实就是new了一个bootstrapNetComm。

struct bootstrapNetComm {int fd;
};

通过createListenSocket启动socker server。

static ncclResult_t createListenSocket(int *fd, union socketAddress *localAddr) {/* IPv4/IPv6 support */int family = localAddr->sa.sa_family;int salen = (family == AF_INET) ? sizeof(sockaddr_in) : sizeof(sockaddr_in6);/* Create socket and bind it to a port */int sockfd = socket(family, SOCK_STREAM, 0);if (sockfd == -1) {WARN("Net : Socket creation failed : %s", strerror(errno));return ncclSystemError;}if (socketToPort(&localAddr->sa)) {// Port is forced by env. Make sure we get the port.int opt = 1;
#if defined(SO_REUSEPORT)SYSCHECK(setsockopt(sockfd, SOL_SOCKET, SO_REUSEADDR | SO_REUSEPORT, &opt, sizeof(opt)), "setsockopt");
#elseSYSCHECK(setsockopt(sockfd, SOL_SOCKET, SO_REUSEADDR, &opt, sizeof(opt)), "setsockopt");
#endif}// localAddr port should be 0 (Any port)SYSCHECK(bind(sockfd, &localAddr->sa, salen), "bind");/* Get the assigned Port */socklen_t size = salen;SYSCHECK(getsockname(sockfd, &localAddr->sa, &size), "getsockname");#ifdef ENABLE_TRACEchar line[1024];TRACE(NCCL_INIT|NCCL_NET,"Listening on socket %s", socketToString(&localAddr->sa, line));
#endif/* Put the socket in listen mode* NB: The backlog will be silently truncated to the value in /proc/sys/net/core/somaxconn*/SYSCHECK(listen(sockfd, 16384), "listen");*fd = sockfd;return ncclSuccess;
}

创建监听fd，ip由localaddr指定，初始端口为0，bind时随机找一个可用端口，并通过getsockname(sockfd, &localAddr->sa, &size)将ip端口写回到localaddr，这里localaddr就是UniqueId。

到这里UniqueId也就产生了，其实就是当前机器的ip和port。

（本文经授权后由OneFlow发布。原文：https://blog.csdn.net/KIDGIN7439/article/details/126712106?spm=1001.2014.3001.5502）

其他人都在看

• One-YOLOv5 v1.2.0发布

• 超越ChatGPT：大模型的智能极限

• 对抗软件系统复杂性③：恰当分层，不多不少

• ChatGPT作者Schulman：我们成功的秘密武器

• 比快更快，开源Stable Diffusion刷新作图速度

• OneEmbedding:单卡训练TB级推荐模型不是梦

• GLM训练加速：性能最高提升3倍，显存节省1/3

欢迎Star、试用OneFlow新版本：