`torch.compile` can be used either as decorator or to optimize model directly, for example:
```
@torch.compile
def foo(x):
return torch.sin(x) + x.max()
```
or
```
mod = torch.nn.ReLU()
optimized_mod = torch.compile(mod, mode="max-autotune")
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/89607
Approved by: https://github.com/soumith
Preparation for the next PR in this stack: #89559.
I replaced
- `self.assertTrue(torch.equal(...))` with `self.assertEqual(..., rtol=0, atol=0, exact_device=True)`,
- the same for `self.assertFalse(...)` with `self.assertNotEqual(...)`, and
- `assert torch.equal(...)` with `torch.testing.assert_close(..., rtol=0, atol=0)` (note that we don't need to set `check_device=True` here since that is the default).
There were a few instances where the result of `torch.equal` is used directly. In that cases I've replaced with `(... == ...).all().item()` while sometimes also dropping the `.item()` depending on the context.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/89527
Approved by: https://github.com/mruberry
After our failed attempt to remove `assert_allclose` in #87974, we decided to add it to the documentation after all. Although we drop the expected removal date, the function continues to be deprecated in favor of `assert_close`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/89526
Approved by: https://github.com/mruberry
Summary: The recommended way to use QConfigMapping is through
`get_default_qconfig_mapping`. However, the docs still references
usages that use `QConfigMapping().set_global(...)`. This doesn't
actually work well in practice when the model has fixed qparams
ops for example. This commit updates these usages.
Reviewers: vkuzo
Subscribers: vkuzo
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87299
Approved by: https://github.com/jerryzh168
Fixes#43144
This uses the Backend system added by [82682](https://github.com/pytorch/pytorch/pull/82682) to change allocators dynamically during the code execution. This will allow us to use RMM, use CUDA managed memory for some portions of the code that do not fit in GPU memory. Write static memory allocators to reduce fragmentation while training models and improve interoperability with external DL compilers/libraries.
For example, we could have the following allocator in c++
```c++
#include <sys/types.h>
#include <cuda_runtime_api.h>
#include <iostream>
extern "C" {
void* my_malloc(ssize_t size, int device, cudaStream_t stream) {
void *ptr;
std::cout<<"alloc "<< size<<std::endl;
cudaMalloc(&ptr, size);
return ptr;
}
void my_free(void* ptr) {
std::cout<<"free "<<std::endl;
cudaFree(ptr);
}
}
```
Compile it as a shared library
```
nvcc allocator.cc -o alloc.so -shared --compiler-options '-fPIC'
```
And use it from PyTorch as follows
```python
import torch
# Init caching
# b = torch.zeros(10, device='cuda')
new_alloc = torch.cuda.memory.CUDAPluggableAllocator('alloc.so', 'my_malloc', 'my_free')
old = torch.cuda.memory.get_current_allocator()
torch.cuda.memory.change_current_allocator(new_alloc)
b = torch.zeros(10, device='cuda')
# This will error since the current allocator was already instantiated
torch.cuda.memory.change_current_allocator(old)
```
Things to discuss
- How to test this, needs compiling external code ...
Pull Request resolved: https://github.com/pytorch/pytorch/pull/86786
Approved by: https://github.com/albanD
# Summary
Creates a callable native function that can determine which implementation of scaled dot product will get called. This allows to bump re-order the runtime dispatch of SDP to enable autograd.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/89029
Approved by: https://github.com/cpuhrsch
Currently all of the distributed errors are thrown from the `TORCH_CHECK` macro which throws a generic `RuntimeError`. This change introduced a new error type `DistBackendError` which derives from `RuntimeError` to signify there was an error with the backend communication library. This allows for better error handling and analysis at higher levels in the stack. Motivation: https://docs.google.com/document/d/1j6VPOkC6znscliFuiDWMuMV1_fH4Abgdq7TCHMcXai4/edit#heading=h.a9rc38misyx8
Changes:
- introduce new error type
- Update `C10D_NCCL_CHECK`
Sample script to demonstrate new error type
```python
# python -m torch.distributed.run --nproc_per_node=2 <script>.py
import torch
import torch.distributed as dist
if __name__ == "__main__":
dist.init_process_group("nccl")
dist.broadcast(torch.tensor([1, 2, 3]).cuda(), 0)
```
Differential Revision: [D40998803](https://our.internmc.facebook.com/intern/diff/D40998803)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88134
Approved by: https://github.com/rohan-varma
Summary:
Improved roundup_power2_divisions knob so it allows better control of rouding in the PyTorch CUDA Caching Allocator.
This new version allows setting the number of divisions per power of two interval starting from 1MB and ending at 64GB and above. An example use case is when rouding is desirable for small allocations but there are also very large allocations which are persistent, thus would not benefit from rounding and take up extra space.
Test Plan: Tested locally
Differential Revision: D40103909
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87290
Approved by: https://github.com/zdevito
- This PR defines a new `api.py` meant to hold the public API for FSDP (minus `FullyShardedDataParallel` itself). This is needed because several of the `_<...>_utils.py` files rely on the public API, and we cannot import from `torch.distributed.fsdp.fully_sharded_data_parallel` without a circular import. Calling the file `api.py` follows the convention used by `ShardedTensor`.
- This PR cleans up the wording in the `BackwardPrefetch`, `ShardingStrategy`, `MixedPrecision`, and `CPUOffload` docstrings.
- This PR adds the aforementioned classes to `fsdp.rst` to have them rendered in public docs.
- To abide by the public bindings contract (`test_public_bindings.py`), the aforementioned classes are removed from `fully_sharded_data_parallel.py`'s `__all__`. This is technically BC breaking if someone uses `from torch.distributed.fsdp.fully_sharded_data_parallel import *`; however, that does not happen in any of our own external or internal code.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87917
Approved by: https://github.com/mrshenli
# Summary
Add in a torch.backends.cuda flag and update context manager to pic between the three implementations of the scaled_dot_product_attention.
cc @cpuhrsch @jbschlosser @bhosmer @mikaylagawarecki
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87946
Approved by: https://github.com/cpuhrsch
This refactor was prompted by challenges handling mixed int/float
operations in C++. A previous version of this patch
added overloads for each permutation of int/float and was unwieldy
https://github.com/pytorch/pytorch/pull/87722/ This PR takes a different
approach.
The general outline of the patch is to combine the C++ types SymIntNode
and SymFloatNode into a single type, SymNode. This is type erased; we
no longer know statically at C++ if we have an int/float and have to test
it with the is_int()/is_float() virtual methods. This has a number of
knock on effects.
- We no longer have C++ classes to bind to Python. Instead, we take an
entirely new approach to our Python API, where we have a SymInt/SymFloat
class defined entirely in Python, which hold a SymNode (which corresponds
to the C++ SymNode). However, SymNode is not pybind11-bound; instead,
it lives as-is in Python, and is wrapped into C++ SymNode using PythonSymNode
when it goes into C++. This implies a userland rename.
In principle, it is also possible for the canonical implementation of SymNode
to be written in C++, and then bound to Python with pybind11 (we have
this code, although it is commented out.) However, I did not implement
this as we currently have no C++ implementations of SymNode.
Because we do return SymInt/SymFloat from C++ bindings, the C++ binding
code needs to know how to find these classes. Currently, this is done
just by manually importing torch and getting the attributes.
- Because SymInt/SymFloat are easy Python wrappers, __sym_dispatch__ now
takes SymInt/SymFloat, rather than SymNode, bringing it in line with how
__torch_dispatch__ works.
Some miscellaneous improvements:
- SymInt now has a constructor that takes SymNode. Note that this
constructor is ambiguous if you pass in a subclass of SymNode,
so an explicit downcast is necessary. This means toSymFloat/toSymInt
are no more. This is a mild optimization as it means rvalue reference
works automatically.
- We uniformly use the caster for c10::SymInt/SymFloat, rather than
going the long way via the SymIntNode/SymFloatNode.
- Removed some unnecessary toSymInt/toSymFloat calls in normalize_*
functions, pretty sure this doesn't do anything.
- guard_int is now a free function, since to guard on an int you cannot
assume the method exists. A function can handle both int and SymInt
inputs.
- We clean up the magic method definition code for SymInt/SymFloat/SymNode.
ONLY the user classes (SymInt/SymFloat) get magic methods; SymNode gets
plain methods; this is to help avoid confusion between the two types.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
cc @jansel @mlazos @soumith @voznesenskym @yanboliang @penguinwu @anijain2305
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87817
Approved by: https://github.com/albanD, https://github.com/anjali411
