Files
rocm-systems/src/include/nvtx3/nvtxDetail/nvtxExtImplCounters_v1.h
T
Mark Santesson f1308997d0 NCCL 2.28.3-1
Device API (Experimental)
 * Introduces device-side APIs to integrate NCCL communication directly into application kernels.
 * Supports LSA (Load/Store Access) for CUDA P2P communication over NVLink and some PCIe platforms.
 * Supports Multimem for hardware multicast using NVLink SHARP.
 * Adds initial framework for GIN (GPU-Initiated Networking), currently under development.
 * Introduces device communicators created using ncclDevCommCreate.
 * Enables device-side communication operations with synchronization (ncclLsaBarrierSession) and memory accessors (ncclGetLsaPointer, ncclGetLsaMultimemPointer).
 * Experimental APIs - signatures and functionality may evolve in future releases.
 * No ABI compatibility is guaranteed — applications must be recompiled with each new NCCL release.

Symmetric memory improvements
 * Support for aggregating symmetric operations using ncclGroupStart/End APIs.
 * Reimplement symmetric kernels using device API.

New Host APIs
 * Introduce new host collective APIs: ncclAlltoAll, ncclScatter, ncclGather.

CE (Copy Engine) Collectives
 * Reduce SM utilization for alltoall, scatter, gather, and allgather within a single (MN)NVL domain.
 * Free up SM capacity for the application to do computation at the same time.
 * To enable the feature for ncclAllGather, ncclAlltoAll, ncclGather, ncclScatter, register buffers into symmetric windows and use the NCCL_CTA_POLICY_ZERO flag in the communicator config_t.

NCCL Inspector Plugin
 * Introduces an Inspector plugin for always-on performance monitoring.
 * Produces structured JSON output with metadata, execution time, bandwidth, and optional event traces for each NCCL operation.
 * Enables integration with analysis tools such as Performance Exporter to visualize NCCL performance bottlenecks.
 * Lightweight to enable via environment variables NCCL_PROFILER_PLUGIN and NCCL_INSPECTOR_ENABLE.

CMake support (Experiemental)
 * Adds a CMake build system as an alternative to existing Makefiles.
 * Known issues: pkg.build and Device API currently do not work with CMake.
 * The known issues will be addressed in a future release.

Decreased max CTA count from 32 to 16 on Blackwell
 * SM overhead is decreased by 50% with this improvement.
 * This may cause some perf drop on Blackwell because of the reduced SM usage.
 * If the extra SM capacity is not desired, two options are available to restore to previous behavior: 1) Setting NCCL_MIN_CTAS=32 NCCL_MAX_CTAS=32 environment variables; 2) setting communicator config to over-write max CTA count to 32.
 * Based on community feedback, future versions may consider different trade-offs between performance and SM overhead.

Plugins
 * Network
   * App-aware Network plugin. NCCL passes information about communication operations to be executed on the network end point. This allows for better tuning of network end points and their use in the plugins.
   * Improve handling of physical and virtual network devices and load/unload.
   * Network plugin version 11 - add explicit context and communication ID support for per communicator init/finalize.
   * Add Multi-Request Net API. Using this will help NCCL to anticipate multiple send/recv requests and optimize for it. See maxMultiRequestSize field in ncclNetProperties_v11_t.
 * Profiler
   * Add support for API events (group, collective, and p2p) and for tracking kernel launches in the profiler plugin.
   * Add Inspector Profiler Plugin (see section above).
   * Add a hook to Google’s CoMMA profiler on github.
 * Tuner
   * Expose NCCL tuning constants at tuner initialization via ncclTunerConstants_v5_t.
   * Add NVL Domain Information API.
 * Support multiple plugin types from a single shared object.

New Parameterization and ncclConfig changes:
 * Add new option NCCL_MNNVL_CLIQUE_ID=-2 which will use rack serial number to partition the MNNVL clique. This will limit NVLink domains to GPUs within a single rack.
 * Add NCCL_NETDEVS_POLICY to control how NET devices are assigned to GPUs. The default (AUTO) is the policy used in previous versions.
 * Add NCCL_SINGLE_PROC_MEM_REG_ENABLE control variable to enable NVLS UB registration in the “one process, multiple ranks” case as opt in.
 * Move nChannelsPerNetPeer into ncclConfig. NCCL_NCHANNELS_PER_NET_PEER can override the value in ncclConfig.
 * Enable PxN over C2C by default
   * PxN over C2C will improve performance for Grace-Blackwell platforms by allowing NCCL to leverage the NIC attached to a peer GPU over NVLINK, C2C, and PCIe.
   * This behavior can be overridden by setting NCCL_PXN_C2C=0.

Other Improvements:
 * Allow FP8 support for non-reductive operations on pre sm90 devices. (See https://github.com/pytorch/pytorch/pull/151594#discussion_r2135777776)
 * Fix NVLS+CollNet and temporarily disables COLLNET_CHAIN for >8 GPUs.
 * Only consider running interfaces for socket traffic. NCCL will not attempt to use interfaces that do not have the IFF_RUNNING bit. (https://github.com/NVIDIA/nccl/issues/1798)
 * Modernize mutex management. Convert to std::mutex and std::lock_guard.
 * Remove sm35 and sm50 GENCODE targets which have long been deprecated and were causing issues with the latest NCCL release builds.
 * Improved NVLS/NVLSTree tuning prediction to improve algorithm and protocol selection.
 * NVLSTree Tuning Fixes. Update tuning data for H100, GB200-NV72.
 * Respond better to RoCE link flaps. Instead of reporting an “unknown event” it will now report “GID table changed”.
 * Move libvirt bridge interface to the end of possible interfaces so that they are considered last. These interfaces are usually virtual bridges to relay traffic to containers running on the host and cannot be used for traffic to a remote node and are therefore unsuitable.
2025-09-02 13:53:34 -07:00

149 lines
5.3 KiB
C

/*
* Copyright 2023-2024 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
*/
#ifndef NVTX_EXT_IMPL_COUNTERS_GUARD
#error Never include this file directly -- it is automatically included by nvToolsExtCounters.h (except when NVTX_NO_IMPL is defined).
#endif
#define NVTX_EXT_IMPL_GUARD
#include "nvtxExtImpl.h"
#undef NVTX_EXT_IMPL_GUARD
#ifndef NVTX_EXT_IMPL_COUNTERS_V1
#define NVTX_EXT_IMPL_COUNTERS_V1
#ifdef __cplusplus
extern "C" {
#endif /* __cplusplus */
/* Macros to create versioned symbols. */
#define NVTX_EXT_COUNTERS_VERSIONED_IDENTIFIER_L3(NAME, VERSION, COMPATID) \
NAME##_v##VERSION##_bpl##COMPATID
#define NVTX_EXT_COUNTERS_VERSIONED_IDENTIFIER_L2(NAME, VERSION, COMPATID) \
NVTX_EXT_COUNTERS_VERSIONED_IDENTIFIER_L3(NAME, VERSION, COMPATID)
#define NVTX_EXT_COUNTERS_VERSIONED_ID(NAME) \
NVTX_EXT_COUNTERS_VERSIONED_IDENTIFIER_L2(NAME, NVTX_VERSION, NVTX_EXT_COUNTERS_COMPATID)
#ifdef NVTX_DISABLE
#include "nvtxExtHelperMacros.h"
#define NVTX_EXT_COUNTERS_IMPL_FN_V1(ret_val, fn_name, signature, arg_names) \
ret_val fn_name signature { \
NVTX_EXT_HELPER_UNUSED_ARGS arg_names \
return ((ret_val)(intptr_t)-1); \
}
#else /* NVTX_DISABLE */
/*
* Function slots for the counters extension. First entry is the module state,
* initialized to `0` (`NVTX_EXTENSION_FRESH`).
*/
#define NVTX_EXT_COUNTERS_SLOT_COUNT 63
NVTX_LINKONCE_DEFINE_GLOBAL intptr_t
NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersSlots)[NVTX_EXT_COUNTERS_SLOT_COUNT + 1]
= {0};
/* Avoid warnings about missing prototype. */
NVTX_LINKONCE_FWDDECL_FUNCTION void NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersInitOnce)(void);
NVTX_LINKONCE_DEFINE_FUNCTION void NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersInitOnce)()
{
intptr_t* fnSlots = NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersSlots) + 1;
nvtxExtModuleSegment_t segment = {
0, /* unused (only one segment) */
NVTX_EXT_COUNTERS_SLOT_COUNT,
fnSlots
};
nvtxExtModuleInfo_t module = {
NVTX_VERSION, sizeof(nvtxExtModuleInfo_t),
NVTX_EXT_COUNTERS_MODULEID, NVTX_EXT_COUNTERS_COMPATID,
1, &segment, /* number of segments, segments */
NULL, /* no export function needed */
/* bake type sizes and alignment information into program binary */
NULL
};
NVTX_INFO( "%s\n", __FUNCTION__ );
NVTX_VERSIONED_IDENTIFIER(nvtxExtInitOnce)(&module,
NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersSlots));
}
#define NVTX_EXT_COUNTERS_IMPL_FN_V1(ret_type, fn_name, signature, arg_names) \
typedef ret_type (*fn_name##_impl_fntype)signature; \
NVTX_DECLSPEC ret_type NVTX_API fn_name signature { \
intptr_t slot = NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersSlots)[NVTX3EXT_CBID_##fn_name + 1]; \
if (slot != NVTX_EXTENSION_DISABLED) { \
if (slot != NVTX_EXTENSION_FRESH) { \
return (*(fn_name##_impl_fntype)slot) arg_names; \
} else { \
NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersInitOnce)(); \
/* Re-read function slot after extension initialization. */ \
slot = NVTX_EXT_COUNTERS_VERSIONED_ID(nvtxExtCountersSlots)[NVTX3EXT_CBID_##fn_name + 1]; \
if (slot != NVTX_EXTENSION_DISABLED && slot != NVTX_EXTENSION_FRESH) { \
return (*(fn_name##_impl_fntype)slot) arg_names; \
} \
} \
} \
NVTX_EXT_FN_RETURN_INVALID(ret_type) \
}
#endif /*NVTX_DISABLE*/
/* Non-void functions. */
#define NVTX_EXT_FN_RETURN_INVALID(rtype) return ((rtype)(intptr_t)-1);
NVTX_EXT_COUNTERS_IMPL_FN_V1(nvtxCountersHandle_t, nvtxCountersRegister,
(nvtxDomainHandle_t domain, const nvtxCountersAttr_t* attr),
(domain, attr))
#undef NVTX_EXT_FN_RETURN_INVALID
/* END: Non-void functions. */
/* void functions. */
#define NVTX_EXT_FN_RETURN_INVALID(rtype)
#define return
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSampleInt64,
(nvtxDomainHandle_t domain, nvtxCountersHandle_t hCounter, int64_t value),
(domain, hCounter, value))
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSampleFloat64,
(nvtxDomainHandle_t domain, nvtxCountersHandle_t hCounter, double value),
(domain, hCounter, value))
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSample,
(nvtxDomainHandle_t domain, nvtxCountersHandle_t hCounter, void* values, size_t size),
(domain, hCounter, values, size))
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSampleNoValue,
(nvtxDomainHandle_t domain, nvtxCountersHandle_t hCounter, uint8_t reason),
(domain, hCounter, reason))
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSubmitBatch,
(nvtxDomainHandle_t domain, nvtxCountersHandle_t hCounters,
const void* counters, size_t size), (domain, hCounters, counters, size))
NVTX_EXT_COUNTERS_IMPL_FN_V1(void, nvtxCountersSubmitBatchEx,
(nvtxDomainHandle_t domain, const nvtxCountersBatch_t* countersBatch),
(domain, countersBatch))
#undef return
#undef NVTX_EXT_FN_RETURN_INVALID
/* END: void functions. */
/* Keep NVTX_EXT_COUNTERS_IMPL_FN_V1 defined for a future version of this extension. */
#ifdef __cplusplus
} /* extern "C" */
#endif /* __cplusplus */
#endif /* NVTX_EXT_IMPL_COUNTERS_V1 */