Comments (5)
Closing as fixed by updating docs.
from pytorch.
from pytorch.
This is actually a known limitation. We should better document it though.
NCCL's API does not support tags, so there isn't a clear way that we could make use of this, even though our APIs expose it (such that it can be used by backends that do support a tag).
https://docs.nvidia.com/deeplearning/nccl/user-guide/docs/api/p2p.html#c.ncclSend
cc @kwen2501 @H-Huang Please keep me honest here.
from pytorch.
Wanted to note that tagging is not intended for supporting out-of-order send/recv calls (in particular the blocking version). Neither NCCL nor MPI would be able to support the example in this issue.
from pytorch.
In that case wouldn't we at least expect hangs (with MPI)? Wouldn't both processes block on differently tagged send/recv?
from pytorch.
Related Issues (20)
- torch compiler bisect tool
- [AOTI][UX] One has no way of knowing whether they need to load DSO as CPU or CUDA runner HOT 1
- 2D TP+FSDP with device mesh HOT 1
- Error: Exporting the operator 'aten::searchsorted' to ONNX opset version 17
- [RFC] Deprecation support for Amazon Linux 2 support for PyTorch Release 2.5 HOT 1
- log shapes + dtypes in Flight Recorder logs
- `torch.compiler.allow_in_graph` does not create a `call_module` op in fx.Graph in torch 2.3.0
- codegen error on .item() as a Triton kernel arg
- RuntimeError when using Adam(fused=True) with torch.compile
- DCP sees 1/2 of the expected size of each tensor in 3D parallel
- [Pipelining] Support for uneven microbatch sizes
- Distributed Checkpoint doesn't verify shapes are correct
- UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR
- [ONNX] export() with dynamic shapes fails when only part of input dimensions are dynamic HOT 2
- `aten.index_put_` runtime shape mismatch HOT 1
- FSDP hangs when combing MoE architecture
- inductor compile failling for batched tensor on fuse
- OnnxExporterError using dynamo_export
- compile inductor falling operator βat::vec::CPU_CAPABILITY::VectorizedN<long int, 2>β and βintβ
- dataloader crashes after several epochs if the trained model contains triton-based operators HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. πππ
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch.