(Porting most of #161008)
Hooking SymmetricMemory Allocator to MemPool so that user can create symmetric tensors with regular `torch.zeros`, `torch.arange` etc factories. Also so that our ops can have functional variants that create `out` tensors on symmetric memory.
To end users, this PR supports a python UI as follows:
```
allocator = symm_mem.get_mempool_allocator(device)
mempool = torch.cuda.MemPool(allocator)
with torch.cuda.use_mem_pool(mempool):
tensor = torch.arange(numel, dtype=dtype, device=device)
```
Added tests for both use cases above.
Differential Revision: [](https://our.internmc.facebook.com/intern/diff/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161471
Approved by: https://github.com/ngimel
ghstack dependencies: #161470
(Porting most of #161008)
Hooking SymmetricMemory Allocator to MemPool so that user can create symmetric tensors with regular `torch.zeros`, `torch.arange` etc factories. Also so that our ops can have functional variants that create `out` tensors on symmetric memory.
To end users, this PR supports a python UI as follows:
```
allocator = symm_mem.get_mempool_allocator(device)
mempool = torch.cuda.MemPool(allocator)
with torch.cuda.use_mem_pool(mempool):
tensor = torch.arange(numel, dtype=dtype, device=device)
```
Added tests for both use cases above.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161471
Approved by: https://github.com/ngimel
ghstack dependencies: #161470
(Porting most of #161008)
Hooking SymmetricMemory Allocator to MemPool so that user can create symmetric tensors with regular `torch.zeros`, `torch.arange` etc factories. Also so that our ops can have functional variants that create `out` tensors on symmetric memory.
To end users, this PR supports a python UI as follows:
```
allocator = symm_mem.get_mempool_allocator(device)
mempool = torch.cuda.MemPool(allocator)
with torch.cuda.use_mem_pool(mempool):
tensor = torch.arange(numel, dtype=dtype, device=device)
```
Added tests for both use cases above.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161471
Approved by: https://github.com/ngimel
ghstack dependencies: #161470
This pull request adds the following ops for sparse matrices using Eigen library:
```python
add(a_csr, b_csr)
add(a_csc, b_csc)
addmm(c_csr, a_csr, b_csr)
addmm(c_csr, a_csr, b_csc)
addmm(c_csr, a_csc, b_csc)
addmm(c_csr, a_csc, b_csr)
addmm(c_csc, a_csr, b_csr)
addmm(c_csc, a_csr, b_csc)
addmm(c_csc, a_csc, b_csc)
addmm(c_csc, a_csc, b_csr)
```
Currently, the operations for sparse matrices on CPU are available through MKL only. The non-existence of MKL on `aarch64` causes the unavailability of these ops on any machines with ARM based CPUs, including Apple Silicon, AWS Graviton and NVIDIA Grace. This PR addresses this issue by using Eigen as a backend for the above ops.
This is a re-factored version of my previous PR #101814. The main difference with the old one, this does not enable Eigen by default.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/155357
Approved by: https://github.com/pearu, https://github.com/eqy
Co-authored-by: Eli Uriegas <eliuriegas@meta.com>
This pull request adds the following ops for sparse matrices using Eigen library:
```python
add(a_csr, b_csr)
add(a_csc, b_csc)
addmm(c_csr, a_csr, b_csr)
addmm(c_csr, a_csr, b_csc)
addmm(c_csr, a_csc, b_csc)
addmm(c_csr, a_csc, b_csr)
addmm(c_csc, a_csr, b_csr)
addmm(c_csc, a_csr, b_csc)
addmm(c_csc, a_csc, b_csc)
addmm(c_csc, a_csc, b_csr)
```
Currently, the operations for sparse matrices on CPU are available through MKL only. The non-existence of MKL on `aarch64` causes the unavailability of these ops on any machines with ARM based CPUs, including Apple Silicon, AWS Graviton and NVIDIA Grace. This PR addresses this issue by using Eigen as a backend for the above ops.
This is a re-factored version of my previous PR #101814. The main difference with the old one, this does not enable Eigen by default.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/155357
Approved by: https://github.com/pearu, https://github.com/eqy
Summary: Undo highlevel BUCKification in favor of something more organized by moving it to the dir itself
Test Plan:
CI
Rollback Plan:
Reviewed By: swolchok
Differential Revision: D76920013
Pull Request resolved: https://github.com/pytorch/pytorch/pull/156503
Approved by: https://github.com/swolchok
We arrive at a point when so many files are related to symmetric memory and files are scattered around in the cpp side. Let's first put all related code (symmetric memory related) into a separate folder. We can do further refactoring later if needed.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/155573
Approved by: https://github.com/fegin, https://github.com/d4l3k
Summary: The goal of this PR and future follow-up PRs is to group a set of header files required by AOTInductor Standalone in a separate directory, ensuring they are implemented in a header-only manner.
Test Plan: CI
Bifferential Revision: D75756619
Pull Request resolved: https://github.com/pytorch/pytorch/pull/154850
Approved by: https://github.com/janeyx99
Make Functorch interpreters serializable most of the time, so that we can save the guards on functorch states.
## Test Cases:
0. torch.compile() without functorch layers present. Guard should fail with any layer being pushed.
1. torch.compile() nested in vmap.
2. torch.compile() nested in grad.
3. torch.compile() nested in jvp + vmap
4. torch.compile() nested functionalize
5. torch.compile() nested in vmap + grad
Differential Revision: [D74008787](https://our.internmc.facebook.com/intern/diff/D74008787/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/152616
Approved by: https://github.com/zou3519
ghstack dependencies: #152615
Adding NVSHMEM as a backend for `SymmetricMemory`, implementation of which is in `NVSHMEMSymmetricMemory.cu`.
Moving some helper functions in `CUDASymmetricMemory.cu` to `CUDASymmetricMemoryUtils.cpp`, so that they can be shared by `NVSHMEMSymmetricMemory`. These functions are mostly side-band exchange helpers (`store_all_gather`, `IpcChannel`, etc).
Adding `TORCH_SYMMEM` to control which implementation to use for CUDA tensors, currently support: `CUDA` (in-house impl), `NVSHMEM`.
The NVSHMEM feature is gated by build-time flag: `USE_NVSHMEM=1`. And `NVSHMEM_HOME` setting is required (TODO).
Ported most code from #146593.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/151261
Approved by: https://github.com/fegin, https://github.com/fduwjj
This is similar to how we handle protobufs and it makes it more convenient for bazel users to handle their version of flatbuffers. (Flatbuffers is very picky about the generated code matching the runtime). Instead of using the checked in generated code, we generate it on the fly.
This is relevant to https://github.com/pytorch/pytorch/issues/112903, because having the version of flatbuffers tied to pytorch will make pytorch difficult to use as an external workspace.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/151364
Approved by: https://github.com/malfet
Fix: #141974
This PR makes `ViewMeta` sequence, present in functional tensors,
serializable with pickle. In order to accomplish that, it makes
`ViewMeta` an abstract class with overridable `forward` and `reverse`
functions. In this context, each operation that once instanciated
`ViewMeta`, should now create a new specialized class that inherits from
`ViewMeta. Therefore, this PR also uses codegen for creating these
specializations.
In summary, these are the changes this PR introduces:
- `ViewMeta` is turned into an abstract class (see
_FunctionalStorageImpl.cpp_). `forward` and `reverse` are pure virtual
functions that need to be implemented. `to_out_index` should be
implemented by operations that might return more than 1 output.
- New `ViewMeta` specializations for `resize_` and `_unsafe_view` are
created (see _FunctionalizeFallbackKernel.h_).
- New templates _ViewMetaClasses.{cpp,h}_ are created. They hold the
declaration and definition of the `ViewMeta` specializations, which
are automatically generated in the ATen codegen (see _gen.py_).
- New `_functionalization` Python sub-module is created (see
_Module.cpp_). It serves as namespace for the `ViewMeta`
specializations and `InverseReturnMode` enum.
- New template _ViewMetaClassesPythonBinding.cpp_ is created. It holds
the automatically generated Python bindings for the `ViewMeta`
specialization, which are generated in the torch codegen (see
_generate_code.py_).
Note that this PR makes use of codegen at 2 different moments:
- ATen codegen (_gen.py_): generates the `ViewMeta` specialized classes.
- Torch codegen (_generate_code.py_): generated the Python bindings for
them.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/143712
Approved by: https://github.com/bdhirsh
Add networkx as a dependency for test_bazel
Example failure: https://github.com/pytorch/pytorch/actions/runs/12551752021/job/34996706301
```
INFO: From Testing //:test_bazel:
==================== Test output for //:test_bazel:
Traceback (most recent call last):
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/test/_test_bazel.py", line 33, in <module>
test_simple_compile_eager()
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/test/_test_bazel.py", line 27, in test_simple_compile_eager
opt_foo1 = torch.compile(foo, backend="eager")
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/__init__.py", line 2533, in compile
backend = _TorchCompileWrapper(backend, mode, options, dynamic)
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/__init__.py", line 2342, in __init__
self.compiler_fn = lookup_backend(backend)
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_dynamo/backends/registry.py", line 66, in lookup_backend
_lazy_import()
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_dynamo/backends/registry.py", line 102, in _lazy_import
import_submodule(backends)
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_dynamo/utils.py", line 2797, in import_submodule
importlib.import_module(f"{mod.__name__}.{filename[:-3]}")
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/execroot/pytorch/external/python3_10_x86_64-unknown-linux-gnu/lib/python3.10/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1050, in _gcd_import
File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 688, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 883, in exec_module
File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_dynamo/backends/common.py", line 12, in <module>
from torch._functorch.aot_autograd import (
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_functorch/aot_autograd.py", line 147, in <module>
from .partitioners import default_partition
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_functorch/partitioners.py", line 31, in <module>
from ._activation_checkpointing.graph_info_provider import GraphInfoProvider
File "/var/lib/jenkins/.cache/bazel/_bazel_jenkins/fdf6d09bf4b4f04a71e2a7dfceb40620/sandbox/processwrapper-sandbox/6504/execroot/pytorch/bazel-out/k8-fastbuild/bin/test_bazel.runfiles/pytorch/torch/_functorch/_activation_checkpointing/graph_info_provider.py", line 3, in <module>
import networkx as nx
ModuleNotFoundError: No module named 'networkx'
```
No periodic runs on this PR or its main branch commit, but I'm pretty sure its started on https://togithub.com/pytorch/pytorch/pull/143539
Pull Request resolved: https://github.com/pytorch/pytorch/pull/143995
Approved by: https://github.com/huydhn
Description:
1. Quantize Linear Layer Weights to 4-bits:
Quantize the weights of the Linear layer to 4 bits, using symmetric quantization.
Pack two 4-bit weights into one uint8 container.
Choose a quantization scheme (channel-wise or group-wise), with the group size being a multiple of 32.
2. Prepare Quantized Weights, Scales, and Optional Bias:
After quantizing, obtain the quantized_weights, scales, and groupsize.
If the original Linear layer has a bias, prepare it as well.
3. Pack the Weights Efficiently:
Use torch.ops.aten._dyn_quant_pack_4bit_weight to optimally pack the weights, scales, and optional bias.
```python
packed_weights = torch.ops.aten._dyn_quant_pack_4bit_weight(weight, scales_and_zeros, bias, groupsize, in_features, out_features)
```
Input parameters should include:
in_features and out_features (the same as the Linear layer’s corresponding parameters).
4. Perform Dynamic Quantized Matrix Multiplication:
Use torch.ops.aten._dyn_quant_matmul_4bit to perform matrix multiplication with quantized weights.
```python
output = torch.ops.aten._dyn_quant_matmul_4bit(input, packed_weights, groupsize, in_features, out_features)
```
Inputs required include:
The input tensor, packed_weights , groupsize, and the in_features and out_features.
API Usage: https://github.com/pytorch/pytorch/issues/143289
Model Perf :
7B Transformer model:
Prefill : 340 t/s
Decode : 40 t/s
2B Transformer model
Prefill : 747 t/s
Decode : 80 t/s
Tests:
python test/test_linalg.py -k test__dyn_quant_pack_4bit_weight
Ran 1 test in 0.016s
OK
python test/test_linalg.py -k test__dyn_quant_matmul_4bit
Ran 8 tests in 0.077s
OK
python test/test_linalg.py -k test_compile_dyn_quant_matmul_4bit
Ran 8 tests in 11.454s
Change-Id: Ia1672bad5e6ec94e64d8bb1971395d60f4b3a452
Fixes #ISSUE_NUMBER
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134124
Approved by: https://github.com/digantdesai, https://github.com/malfet
Description:
1. Quantize Linear Layer Weights to 4-bits:
Quantize the weights of the Linear layer to 4 bits, using symmetric quantization.
Pack two 4-bit weights into one uint8 container.
Choose a quantization scheme (channel-wise or group-wise), with the group size being a multiple of 32.
2. Prepare Quantized Weights, Scales, and Optional Bias:
After quantizing, obtain the quantized_weights, scales, and groupsize.
If the original Linear layer has a bias, prepare it as well.
3. Pack the Weights Efficiently:
Use torch.ops.aten._dyn_quant_pack_4bit_weight to optimally pack the weights, scales, and optional bias.
```python
packed_weights = torch.ops.aten._dyn_quant_pack_4bit_weight(weight, scales_and_zeros, bias, groupsize, in_features, out_features)
```
Input parameters should include:
in_features and out_features (the same as the Linear layer’s corresponding parameters).
4. Perform Dynamic Quantized Matrix Multiplication:
Use torch.ops.aten._dyn_quant_matmul_4bit to perform matrix multiplication with quantized weights.
```python
output = torch.ops.aten._dyn_quant_matmul_4bit(input, packed_weights, groupsize, in_features, out_features)
```
Inputs required include:
The input tensor, packed_weights , groupsize, and the in_features and out_features.
API Usage: https://github.com/pytorch/pytorch/issues/143289
Model Perf :
7B Transformer model:
Prefill : 340 t/s
Decode : 40 t/s
2B Transformer model
Prefill : 747 t/s
Decode : 80 t/s
Tests:
python test/test_linalg.py -k test__dyn_quant_pack_4bit_weight
Ran 1 test in 0.016s
OK
python test/test_linalg.py -k test__dyn_quant_matmul_4bit
Ran 8 tests in 0.077s
OK
python test/test_linalg.py -k test_compile_dyn_quant_matmul_4bit
Ran 8 tests in 11.454s
Change-Id: Ia1672bad5e6ec94e64d8bb1971395d60f4b3a452
Fixes #ISSUE_NUMBER
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134124
Approved by: https://github.com/digantdesai, https://github.com/malfet
Description:
1. Quantize Linear Layer Weights to 4-bits:
Quantize the weights of the Linear layer to 4 bits, using symmetric quantization.
Pack two 4-bit weights into one uint8 container.
Choose a quantization scheme (channel-wise or group-wise), with the group size being a multiple of 32.
2. Prepare Quantized Weights, Scales, and Optional Bias:
After quantizing, obtain the quantized_weights, scales, and groupsize.
If the original Linear layer has a bias, prepare it as well.
3. Pack the Weights Efficiently:
Use torch.ops.aten._dyn_quant_pack_4bit_weight to optimally pack the weights, scales, and optional bias.
```python
packed_weights = torch.ops.aten._dyn_quant_pack_4bit_weight(weight, scales_and_zeros, bias, groupsize, in_features, out_features)
```
Input parameters should include:
in_features and out_features (the same as the Linear layer’s corresponding parameters).
4. Perform Dynamic Quantized Matrix Multiplication:
Use torch.ops.aten._dyn_quant_matmul_4bit to perform matrix multiplication with quantized weights.
```python
output = torch.ops.aten._dyn_quant_matmul_4bit(input, packed_weights, groupsize, in_features, out_features)
```
Inputs required include:
The input tensor, packed_weights , groupsize, and the in_features and out_features.
API Usage: https://github.com/pytorch/pytorch/issues/143289
Model Perf :
7B Transformer model:
Prefill : 340 t/s
Decode : 40 t/s
2B Transformer model
Prefill : 747 t/s
Decode : 80 t/s
Tests:
python test/test_linalg.py -k test__dyn_quant_pack_4bit_weight
Ran 1 test in 0.016s
OK
python test/test_linalg.py -k test__dyn_quant_matmul_4bit
Ran 8 tests in 0.077s
OK
python test/test_linalg.py -k test_compile_dyn_quant_matmul_4bit
Ran 8 tests in 11.454s
Change-Id: Ia1672bad5e6ec94e64d8bb1971395d60f4b3a452
Fixes #ISSUE_NUMBER
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134124
Approved by: https://github.com/digantdesai, https://github.com/malfet
