Files
rocm-systems/src/context_host.cpp
T
Anatolii Rozanov d0c8380650 Add host API for *_on_stream operations (#340)
* Add functional test for barrier_all_on_stream

* Add rocshmem_barrier_all_on_stream support for GDA and RO backends

Implements rocshmem_barrier_all_on_stream operation for
GPU Direct Access and Reverse Offload backends.

Previously, rocshmem_barrier_all_on_stream was only supported for IPC backend.

* Add functional test for rocshmem_broadcastmem_on_stream

* Add host-side rocshmem_broadcastmem_on_stream API

Implement stream-based broadcast collective operation

- Add rocshmem_broadcastmem_on_stream host API and kernel implementation
- Add functional test TeamBroadcastmemOnStreamTester with multi-stream
  support and correctness verification
- Use per-workgroup contexts to avoid contention across parallel streams

API:
rocshmem_broadcastmem_on_stream(team, dest, source, nelems, pe_root, stream)

* Add functional test for rocshmem_getmem_on_stream

* Add host-side rocshmem_getmem_on_stream API

Implement stream-based point-to-point RMA get operation

- Add rocshmem_getmem_on_stream host API and kernel implementation
- Support for asynchronous getmem operations on HIP streams
- Add backend support for GDA, RO, and IPC contexts
- Use work-group collective getmem for efficient memory transfer

API:
rocshmem_getmem_on_stream(dest, source, nelems, pe, stream)

(AI Assist)

* Add host-side rocshmem_putmem_on_stream API

- Add rocshmem_putmem_on_stream for asynchronous remote writes
- Support for concurrent RMA operations on HIP streams
- Add backend support for GDA, RO, and IPC contexts
- Use work-group device collective operation

API:
rocshmem_putmem_on_stream(dest, source, bytes, pe, stream)

(AI Assist)

* Add functional test for rocshmem_putmem_on_stream

* Add host-side rocshmem_putmem_signal_on_stream API

Enables asynchronous putmem operations with signaling on HIP streams.

The implementation includes:
- Kernel wrapper rocshmem_putmem_signal_kernel
- Host interface putmem_signal_on_stream method
- Context layer support across all backends (IPC, GDA, RO)
- Public API

Function signature:
void rocshmem_putmem_signal_on_stream(void *dest, const void *source,
                                      size_t bytes, uint64_t *sig_addr,
                                      uint64_t signal, int sig_op,
                                      int pe, hipStream_t stream);

* Add functional test for rocshmem_putmem_signal_on_stream

* Add host-side rocshmem_signal_wait_until_on_stream API

Enables asynchronous signal wait operations on HIP streams.

The implementation includes:
- Kernel wrapper rocshmem_signal_wait_until_kernel
- Host interface signal_wait_until_on_stream method
- Context layer support across all backends (IPC, GDA, RO)
- Native uint64_t support in wait_until API (generated from P2P_SYNC.py)

Function signature:
void rocshmem_signal_wait_until_on_stream(uint64_t *sig_addr, int cmp,
                                          uint64_t cmp_value,
                                          hipStream_t stream);

(AI Assist)

* Add functional test for rocshmem_signal_wait_until_on_stream

* Add documentation for stream API functions

This commit adds API documentation for the following host-side
stream functions:

- rocshmem_barrier_all_on_stream (collective routines)
- rocshmem_broadcastmem_on_stream (collective routines)
- rocshmem_getmem_on_stream (RMA operations)
- rocshmem_putmem_on_stream (RMA operations)
- rocshmem_putmem_signal_on_stream (signaling operations)
- rocshmem_signal_wait_until_on_stream (point-to-point sync)

The documentation includes function signatures, parameter descriptions,
and detailed explanations of asynchronous behavior and stream handling.

(AI Assist)

* Rename "bytes" -> "nelems"

* Add "_TEST_" to the variables used in tests

* Remove incorrect hipStreamDefault usage

hipStreamDefault is not a default stream. This is a flag.

If stream == nullptr, then just pass it to kernel. It will launch the kernel on the default stream
2025-12-09 08:55:46 -06:00

175 строки
5.8 KiB
C++

/******************************************************************************
* Copyright (c) Advanced Micro Devices, Inc. All rights reserved.
*
* SPDX-License-Identifier: MIT
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*****************************************************************************/
#include "rocshmem/rocshmem_config.h" // NOLINT(build/include_subdir)
#include "backend_bc.hpp"
#include "context_incl.hpp"
namespace rocshmem {
__host__ Context::Context(Backend* handle)
: num_pes(handle->getNumPEs()),
my_pe(handle->getMyPE()),
btype(handle->type) {
}
__host__ Context::~Context() {
}
/******************************************************************************
********************** CONTEXT DISPATCH IMPLEMENTATIONS **********************
*****************************************************************************/
__host__ void Context::putmem(void* dest, const void* source, size_t nelems,
int pe) {
if (nelems == 0) {
return;
}
ctxHostStats.incStat(NUM_HOST_PUT);
HOST_DISPATCH(putmem(dest, source, nelems, pe));
}
__host__ void Context::getmem(void* dest, const void* source, size_t nelems,
int pe) {
if (nelems == 0) {
return;
}
ctxHostStats.incStat(NUM_HOST_GET);
HOST_DISPATCH(getmem(dest, source, nelems, pe));
}
__host__ void Context::putmem_nbi(void* dest, const void* source, size_t nelems,
int pe) {
if (nelems == 0) {
return;
}
ctxHostStats.incStat(NUM_HOST_PUT_NBI);
HOST_DISPATCH(putmem_nbi(dest, source, nelems, pe));
}
__host__ void Context::getmem_nbi(void* dest, const void* source, size_t nelems,
int pe) {
if (nelems == 0) {
return;
}
ctxHostStats.incStat(NUM_HOST_GET_NBI);
HOST_DISPATCH(getmem_nbi(dest, source, nelems, pe));
}
__host__ void Context::fence() {
ctxHostStats.incStat(NUM_HOST_FENCE);
HOST_DISPATCH(fence());
}
__host__ void Context::quiet() {
ctxHostStats.incStat(NUM_HOST_QUIET);
HOST_DISPATCH(quiet());
}
__host__ void* Context::shmem_ptr(const void* dest, int pe) {
ctxHostStats.incStat(NUM_HOST_SHMEM_PTR);
HOST_DISPATCH_RET_PTR(shmem_ptr(dest, pe));
}
__host__ void Context::sync_all() {
ctxHostStats.incStat(NUM_HOST_SYNC_ALL);
HOST_DISPATCH(sync_all());
}
__host__ void Context::barrier_all() {
ctxHostStats.incStat(NUM_HOST_BARRIER_ALL);
HOST_DISPATCH(barrier_all());
}
__host__ void Context::barrier_all_on_stream(hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_BARRIER_ALL);
HOST_DISPATCH(barrier_all_on_stream(stream));
}
__host__ void Context::alltoallmem_on_stream(rocshmem_team_t team, void *dest,
const void *source, size_t size,
hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_ALLTOALL);
HOST_DISPATCH(alltoallmem_on_stream(team, dest, source, size, stream));
}
__host__ void Context::broadcastmem_on_stream(rocshmem_team_t team, void *dest,
const void *source, size_t nelems,
int pe_root, hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_BROADCAST);
HOST_DISPATCH(
broadcastmem_on_stream(team, dest, source, nelems, pe_root, stream));
}
__host__ void Context::getmem_on_stream(void *dest, const void *source,
size_t nelems, int pe,
hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_GET);
HOST_DISPATCH(getmem_on_stream(dest, source, nelems, pe, stream));
}
__host__ void Context::putmem_on_stream(void *dest, const void *source,
size_t nelems, int pe,
hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_PUT);
HOST_DISPATCH(putmem_on_stream(dest, source, nelems, pe, stream));
}
__host__ void Context::putmem_signal_on_stream(void *dest, const void *source,
size_t nelems,
uint64_t *sig_addr,
uint64_t signal, int sig_op,
int pe, hipStream_t stream) {
ctxHostStats.incStat(NUM_HOST_PUT);
HOST_DISPATCH(putmem_signal_on_stream(dest, source, nelems, sig_addr, signal,
sig_op, pe, stream));
}
__host__ void Context::signal_wait_until_on_stream(uint64_t *sig_addr, int cmp,
uint64_t cmp_value,
hipStream_t stream) {
HOST_DISPATCH(signal_wait_until_on_stream(sig_addr, cmp, cmp_value, stream));
}
} // namespace rocshmem