Sylvain Jeaugey
178b6b7590
Rework core for NVIDIA Trusted Computing * Compress work structs so that they are shared between channels * Utilize the full amount of kernel argument space permitted (4k) before resorting to work fifo. * Rework the task preprocessing phase. * Use a separate abortDevFlag which is kept in sync with abortFlag using cudaMemcpy operations. * Rename src/include/align.h to src/include/bitops.h Add lazy connection establishment for collective operations * Move buffer allocation and connection establishment to the first collective operation using that algorithm. * Accelerate init time and reduce memory usage. * Avoid allocating NVLS buffers if all calls are registered. * Compute algo/proto in ncclLaunchCollTasksInfo early on. * Connect peers in ncclCollPreconnectFunc if not connected already. * Also move shared buffer creation to the first send/recv call. Accelerate intra-node NVLink detection * Make each rank only detect NVLinks attached to its GPU. * Fuse XMLs to reconstruct the full NVLink topology Add init profiling to report time spend in different init phases. * Report timings of bootstrap, allgather, search, connect, etc. * Add new "PROFILE" category for NCCL_DEBUG_SUBSYS. Add support for PCI p2p on split PCI switches * Detect split PCI switches through a kernel module exposing switch information. * Update the topology XML and graph to add those inter-switch connections. Add cost estimation API * Add a new ncclGroupEndSimulate primitive to return the estimated time a group would take. Net/IB: Add separate traffic class for fifo messages * Add NCCL_IB_FIFO_TC to control the traffic class of fifo messages independently from NCCL_IB_TC. Merges PR #1194 Net/IB: Add support for IB router * Use flid instead of lid if subnets do not match * Warn if flid is 0 Optimizations and fixes for device network offload (unpack) * Double the default number of channels * Cache netDeviceType * Fix save/increment head logic to enable Tree support. Support ncclGroupStart/End for ncclCommAbort/Destroy * Allow Abort/Destroy to be called within a group when managing multiple GPUs with a single process. Improve Tuner API * Provide to the plugin the original cost table so that the plugin can leave unknown or disabled algo/proto combinations untouched. * Remove nvlsSupport and collnetSupport. Do not print version to stdout when using a debug file * Also print version from all processes with INFO debug level. Fixes issue #1271 Fix clang warnings in NVTX headers * Update NVTX headers to the latest version Fixes issue #1270 Disable port fusion in heterogeneous systems * Do not fuse ports if a mix of multi-port and single port are detected. Fix NVLS graphs search for dual NICs. * Fix NVLS graph search when we have more than one NIC per GPU. Fix crash with collnetDirect * Add separate graph search for collnetDirect, testing alltoall paths and working similarly to the NVLS search. Fix hang when nodes have different CPU types * Add the CPU type to the rank peer info. * Align all ranks on the CPU type after the first allgather. * Only use the aligned CPU type for all tuning operations. Fixes issue #1136 Fixes issue #1184 Fix performance of registered send/recv operations * Allow for single full size operations * Add INFO to confirm the registration of send/recv buffers. Move all sync ops to finalize stage * Ensure ncclCommDestroy is non-blocking if ncclCommFinalize has been called. Improve error reporting during SHM segment creation Improve support of various compilers Merges PR #1177 Merges PR #1228 Allow net and tuner plugins to be statically linked * Search for ncclNet or ncclTuner symbols in the main binary. Merges PR #979 Plugin examples includes cleanup * Harmonize err.h and common.h usage. * Add mixed plugin with both net and tuner.
142 خطوط
4.5 KiB
C
142 خطوط
4.5 KiB
C
/*
|
|
* Copyright 2009-2022 NVIDIA Corporation. All rights reserved.
|
|
*
|
|
* Licensed under the Apache License v2.0 with LLVM Exceptions.
|
|
* See https://llvm.org/LICENSE.txt for license information.
|
|
* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
*/
|
|
|
|
#include "nvToolsExt.h"
|
|
|
|
#include "cuda.h"
|
|
|
|
#ifndef NVTOOLSEXT_CUDA_V3
|
|
#define NVTOOLSEXT_CUDA_V3
|
|
|
|
#ifdef __cplusplus
|
|
extern "C" {
|
|
#endif /* __cplusplus */
|
|
|
|
/* ========================================================================= */
|
|
/** \name Functions for CUDA Resource Naming
|
|
*/
|
|
/** \addtogroup RESOURCE_NAMING
|
|
* \section RESOURCE_NAMING_CUDA CUDA Resource Naming
|
|
*
|
|
* This section covers the API functions that allow to annotate CUDA resources
|
|
* with user-provided names.
|
|
*
|
|
* @{
|
|
*/
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/* \cond SHOW_HIDDEN
|
|
* \brief Used to build a non-colliding value for resource types separated class
|
|
* \version \NVTX_VERSION_2
|
|
*/
|
|
#define NVTX_RESOURCE_CLASS_CUDA 4
|
|
/** \endcond */
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/** \brief Resource types for CUDA
|
|
*/
|
|
typedef enum nvtxResourceCUDAType_t
|
|
{
|
|
NVTX_RESOURCE_TYPE_CUDA_DEVICE = NVTX_RESOURCE_MAKE_TYPE(CUDA, 1), /* CUdevice */
|
|
NVTX_RESOURCE_TYPE_CUDA_CONTEXT = NVTX_RESOURCE_MAKE_TYPE(CUDA, 2), /* CUcontext */
|
|
NVTX_RESOURCE_TYPE_CUDA_STREAM = NVTX_RESOURCE_MAKE_TYPE(CUDA, 3), /* CUstream */
|
|
NVTX_RESOURCE_TYPE_CUDA_EVENT = NVTX_RESOURCE_MAKE_TYPE(CUDA, 4), /* CUevent */
|
|
} nvtxResourceCUDAType_t;
|
|
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/** \brief Annotates a CUDA device.
|
|
*
|
|
* Allows the user to associate a CUDA device with a user-provided name.
|
|
*
|
|
* \param device - The handle of the CUDA device to name.
|
|
* \param name - The name of the CUDA device.
|
|
*
|
|
* \version \NVTX_VERSION_1
|
|
* @{ */
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuDeviceA(CUdevice device, const char* name);
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuDeviceW(CUdevice device, const wchar_t* name);
|
|
/** @} */
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/** \brief Annotates a CUDA context.
|
|
*
|
|
* Allows the user to associate a CUDA context with a user-provided name.
|
|
*
|
|
* \param context - The handle of the CUDA context to name.
|
|
* \param name - The name of the CUDA context.
|
|
*
|
|
* \par Example:
|
|
* \code
|
|
* CUresult status = cuCtxCreate( &cuContext, 0, cuDevice );
|
|
* if ( CUDA_SUCCESS != status )
|
|
* goto Error;
|
|
* nvtxNameCuContext(cuContext, "CTX_NAME");
|
|
* \endcode
|
|
*
|
|
* \version \NVTX_VERSION_1
|
|
* @{ */
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuContextA(CUcontext context, const char* name);
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuContextW(CUcontext context, const wchar_t* name);
|
|
/** @} */
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/** \brief Annotates a CUDA stream.
|
|
*
|
|
* Allows the user to associate a CUDA stream with a user-provided name.
|
|
*
|
|
* \param stream - The handle of the CUDA stream to name.
|
|
* \param name - The name of the CUDA stream.
|
|
*
|
|
* \version \NVTX_VERSION_1
|
|
* @{ */
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuStreamA(CUstream stream, const char* name);
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuStreamW(CUstream stream, const wchar_t* name);
|
|
/** @} */
|
|
|
|
/* ------------------------------------------------------------------------- */
|
|
/** \brief Annotates a CUDA event.
|
|
*
|
|
* Allows the user to associate a CUDA event with a user-provided name.
|
|
*
|
|
* \param event - The handle of the CUDA event to name.
|
|
* \param name - The name of the CUDA event.
|
|
*
|
|
* \version \NVTX_VERSION_1
|
|
* @{ */
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuEventA(CUevent event, const char* name);
|
|
NVTX_DECLSPEC void NVTX_API nvtxNameCuEventW(CUevent event, const wchar_t* name);
|
|
/** @} */
|
|
|
|
/** @} */ /* END RESOURCE_NAMING */
|
|
|
|
/* ========================================================================= */
|
|
#ifdef UNICODE
|
|
#define nvtxNameCuDevice nvtxNameCuDeviceW
|
|
#define nvtxNameCuContext nvtxNameCuContextW
|
|
#define nvtxNameCuStream nvtxNameCuStreamW
|
|
#define nvtxNameCuEvent nvtxNameCuEventW
|
|
#else
|
|
#define nvtxNameCuDevice nvtxNameCuDeviceA
|
|
#define nvtxNameCuContext nvtxNameCuContextA
|
|
#define nvtxNameCuStream nvtxNameCuStreamA
|
|
#define nvtxNameCuEvent nvtxNameCuEventA
|
|
#endif
|
|
|
|
#ifdef __cplusplus
|
|
}
|
|
#endif /* __cplusplus */
|
|
|
|
#ifndef NVTX_NO_IMPL
|
|
#define NVTX_IMPL_GUARD_CUDA /* Ensure other headers cannot be included directly */
|
|
#include "nvtxDetail/nvtxImplCuda_v3.h"
|
|
#undef NVTX_IMPL_GUARD_CUDA
|
|
#endif /*NVTX_NO_IMPL*/
|
|
|
|
#endif /* NVTOOLSEXT_CUDA_V3 */
|