This is an automated email from the ASF dual-hosted git repository.
junrushao pushed a commit to branch unity
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/unity by this push:
new afb2e421f0 [Unity][Disco] separate computation and communication into
2 stream (#15742)
afb2e421f0 is described below
commit afb2e421f0d0345f2e6e120ec17c665e82e6bd8b
Author: Hongyi Jin <[email protected]>
AuthorDate: Thu Sep 14 21:37:11 2023 -0700
[Unity][Disco] separate computation and communication into 2 stream (#15742)
put computation on default stream and put communication on a new stream.
---
src/runtime/disco/nccl/nccl.cc | 41 +++++++++++++++++++++++++++++------------
tests/python/disco/test_nccl.py | 11 +++--------
2 files changed, 32 insertions(+), 20 deletions(-)
diff --git a/src/runtime/disco/nccl/nccl.cc b/src/runtime/disco/nccl/nccl.cc
index e404e3c2bb..88eaf03c44 100644
--- a/src/runtime/disco/nccl/nccl.cc
+++ b/src/runtime/disco/nccl/nccl.cc
@@ -39,12 +39,13 @@ namespace nccl {
struct NCCLThreadLocalContext {
DiscoWorker* worker;
int device_id;
- cudaStream_t stream;
+ cudaStream_t comm_stream;
+ cudaStream_t compute_stream = nullptr;
ncclComm_t comm;
void Clear() {
NCCL_CALL(ncclCommDestroy(comm));
- CUDA_CALL(cudaStreamDestroy(stream));
+ CUDA_CALL(cudaStreamDestroy(comm_stream));
}
static NCCLThreadLocalContext* Get() {
@@ -74,9 +75,8 @@ void InitCCLPerWorker(ShapeTuple device_ids, std::string
unique_id_bytes) {
// Step up local context of NCCL
int device_id = device_ids[worker->worker_id];
CUDA_CALL(cudaSetDevice(device_id));
- CUDA_CALL(cudaStreamCreate(&ctx->stream));
+ CUDA_CALL(cudaStreamCreate(&ctx->comm_stream));
Device device{DLDeviceType::kDLCUDA, device_id};
- DeviceAPI::Get(device)->SetStream(device, ctx->stream);
worker->default_device = device;
worker->ccl = "nccl";
ctx->worker = worker;
@@ -91,9 +91,12 @@ void AllReduce(NDArray send, ReduceKind reduce_kind, NDArray
recv) {
NCCLThreadLocalContext* ctx = NCCLThreadLocalContext::Get();
ShapeTuple shape = send.Shape();
int64_t numel = shape->Product();
+ Device device = ctx->worker->default_device;
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->compute_stream,
ctx->comm_stream);
NCCL_CALL(ncclAllReduce(send->data, recv->data, numel,
/*datatype=*/AsNCCLDataType(DataType(send->dtype)),
- /*op=*/AsNCCLRedOp(reduce_kind), ctx->comm,
ctx->stream));
+ /*op=*/AsNCCLRedOp(reduce_kind), ctx->comm,
ctx->comm_stream));
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->comm_stream,
ctx->compute_stream);
}
void BroadcastFromWorker0(NDArray send, NDArray recv) {
@@ -101,9 +104,12 @@ void BroadcastFromWorker0(NDArray send, NDArray recv) {
ICHECK(send.Shape()->Product() == recv.Shape()->Product());
ShapeTuple shape = send.Shape();
int64_t numel = shape->Product();
+ Device device = ctx->worker->default_device;
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->compute_stream,
ctx->comm_stream);
NCCL_CALL(ncclBroadcast(send->data, recv->data, numel,
/*datatype=*/AsNCCLDataType(DataType(send->dtype)),
- /*root=*/0, ctx->comm, ctx->stream));
+ /*root=*/0, ctx->comm, ctx->comm_stream));
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->comm_stream,
ctx->compute_stream);
}
void ScatterFromWorker0(Optional<NDArray> send, NDArray recv) {
@@ -111,6 +117,8 @@ void ScatterFromWorker0(Optional<NDArray> send, NDArray
recv) {
NCCLThreadLocalContext* ctx = NCCLThreadLocalContext::Get();
int worker_id = ctx->worker->worker_id;
int num_workers = ctx->worker->num_workers;
+ Device device = ctx->worker->default_device;
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->compute_stream,
ctx->comm_stream);
if (worker_id == 0) {
CHECK(send.defined()) << "ValueError: buffer `send` must be provided when
worker_id == 0.";
NDArray buffer = send.value();
@@ -129,7 +137,8 @@ void ScatterFromWorker0(Optional<NDArray> send, NDArray
recv) {
NCCL_CALL(ncclGroupStart());
uint8_t* data = static_cast<uint8_t*>(buffer->data);
for (int i = 0; i < num_workers; ++i) {
- NCCL_CALL(ncclSend(data, numel_per_shard, AsNCCLDataType(dtype), i,
ctx->comm, ctx->stream));
+ NCCL_CALL(
+ ncclSend(data, numel_per_shard, AsNCCLDataType(dtype), i, ctx->comm,
ctx->comm_stream));
data += bytes_per_shard;
}
} else {
@@ -142,8 +151,9 @@ void ScatterFromWorker0(Optional<NDArray> send, NDArray
recv) {
}
int64_t numel = recv.Shape()->Product();
DataType dtype(recv->dtype);
- NCCL_CALL(ncclRecv(recv->data, numel, AsNCCLDataType(dtype), 0, ctx->comm,
ctx->stream));
+ NCCL_CALL(ncclRecv(recv->data, numel, AsNCCLDataType(dtype), 0, ctx->comm,
ctx->comm_stream));
NCCL_CALL(ncclGroupEnd());
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->comm_stream,
ctx->compute_stream);
}
void GatherToWorker0(NDArray send, Optional<NDArray> recv) {
@@ -151,6 +161,8 @@ void GatherToWorker0(NDArray send, Optional<NDArray> recv) {
NCCLThreadLocalContext* ctx = NCCLThreadLocalContext::Get();
int worker_id = ctx->worker->worker_id;
int num_workers = ctx->worker->num_workers;
+ Device device = ctx->worker->default_device;
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->compute_stream,
ctx->comm_stream);
if (worker_id == 0) {
CHECK(recv.defined()) << "ValueError: buffer `recv` must be provided when
worker_id == 0.";
NDArray buffer = recv.value();
@@ -169,7 +181,8 @@ void GatherToWorker0(NDArray send, Optional<NDArray> recv) {
NCCL_CALL(ncclGroupStart());
uint8_t* data = static_cast<uint8_t*>(buffer->data);
for (int i = 0; i < num_workers; ++i) {
- NCCL_CALL(ncclRecv(data, numel_per_shard, AsNCCLDataType(dtype), i,
ctx->comm, ctx->stream));
+ NCCL_CALL(
+ ncclRecv(data, numel_per_shard, AsNCCLDataType(dtype), i, ctx->comm,
ctx->comm_stream));
data += bytes_per_shard;
}
} else {
@@ -182,24 +195,28 @@ void GatherToWorker0(NDArray send, Optional<NDArray>
recv) {
}
int64_t numel = send.Shape()->Product();
DataType dtype(send->dtype);
- NCCL_CALL(ncclSend(send->data, numel, AsNCCLDataType(dtype), 0, ctx->comm,
ctx->stream));
+ NCCL_CALL(ncclSend(send->data, numel, AsNCCLDataType(dtype), 0, ctx->comm,
ctx->comm_stream));
NCCL_CALL(ncclGroupEnd());
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->comm_stream,
ctx->compute_stream);
}
void RecvFromWorker0(NDArray buffer) {
NCCLThreadLocalContext* ctx = NCCLThreadLocalContext::Get();
CHECK_NE(ctx->worker->worker_id, 0)
<< "ValueError: Worker 0 is not allowed to call RecvFromWorker0.";
+ Device device = ctx->worker->default_device;
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->compute_stream,
ctx->comm_stream);
NCCL_CALL(ncclGroupStart());
NCCL_CALL(ncclRecv(buffer->data, buffer.Shape()->Product(),
AsNCCLDataType(buffer.DataType()), 0,
- ctx->comm, ctx->stream));
+ ctx->comm, ctx->comm_stream));
NCCL_CALL(ncclGroupEnd());
+ DeviceAPI::Get(device)->SyncStreamFromTo(device, ctx->comm_stream,
ctx->compute_stream);
}
void SyncWorker() {
NCCLThreadLocalContext* ctx = NCCLThreadLocalContext::Get();
ICHECK(ctx->worker != nullptr);
- CUDA_CALL(cudaStreamSynchronize(ctx->stream));
+ CUDA_CALL(cudaStreamSynchronize(ctx->compute_stream));
}
TVM_REGISTER_GLOBAL("runtime.disco.nccl.init_ccl").set_body_typed(InitCCL);
diff --git a/tests/python/disco/test_nccl.py b/tests/python/disco/test_nccl.py
index f0f949ab80..e86c973fc2 100644
--- a/tests/python/disco/test_nccl.py
+++ b/tests/python/disco/test_nccl.py
@@ -22,6 +22,7 @@ import numpy as np
import pytest
import tvm
+import tvm.testing
from tvm import dlight as dl
from tvm import relax as rx
from tvm.runtime import disco as di
@@ -103,7 +104,7 @@ def test_scatter(session_kind):
@pytest.mark.parametrize("session_kind", _all_session_kinds)
def test_gather(session_kind):
- devices = [1, 2]
+ devices = [0, 1]
sess = session_kind(num_workers=len(devices))
sess.init_ccl("nccl", *devices)
@@ -376,10 +377,4 @@ def test_attention(session_kind): # pylint:
disable=too-many-locals,too-many-st
if __name__ == "__main__":
- test_init(di.ProcessSession)
- test_allreduce(di.ProcessSession)
- test_broadcast_from_worker0(di.ProcessSession)
- test_scatter(di.ProcessSession)
- test_gather(di.ProcessSession)
- test_mlp(di.ProcessSession)
- test_attention(di.ProcessSession)
+ tvm.testing.main()
