pytorch/torch/csrc/jit/node_hashing.cpp
Edward Yang 517c7c9861 Canonicalize all includes in PyTorch. (#14849)
Summary:
Anywhere we used #include "foo.h", we now say #include <foo.h>
Paths are adjusted to be rooted out of aten/src, torch/lib, or
the root level directory.

I modified CMakeLists.txt by hand to remove TH and THC from
the include paths.

I used the following script to do the canonicalization:

```
  import subprocess
  import re
  import os.path

  files = subprocess.check_output(['git', 'ls-files']).decode('utf-8').rstrip().split('\n')
  for fn in files:
      if not any(fn.endswith(suff) for suff in ['.cu', '.cpp', '.in', '.h', '.hpp', '.cu', '.cuh', '.cc']):
          continue
      if not any(fn.startswith(pref) for pref in ["aten/", "torch/"]):
          continue
      with open(fn, 'r') as f:
          c = f.read()
      def fmt(p):
          return "#include <{}>".format(p)
      def repl(m):
          p = m.group(1)
          if p in ["dlfcn.h", "unistd.h", "nvrtc.h", "cuda.h", "cuda_runtime.h", "cstdint", "cudnn.h", "Python.h", "cusparse.h", "cuda_runtime_api.h", "cuda_fp16.h", "cublas_v2.h", "stdint.h", "curand_kernel.h"]:
              return fmt(p)
          if any(p.startswith(pref) for pref in ["torch/csrc", "c10/", "ATen/", "caffe2/", "TH/", "THC/", "Eigen/", "gtest/", "zdl/", "gloo/", "onnx/", "miopen/"]):
              return fmt(p)
          for root in ["aten/src", "torch/lib", ""]:
              for bad_root in [os.path.dirname(fn), "aten/src/TH", "aten/src/THC", "torch/csrc"]:
                  new_p = os.path.relpath(os.path.join(bad_root, p), root)
                  if not new_p.startswith("../") and (os.path.exists(os.path.join(root, new_p)) or os.path.exists(os.path.join(root, new_p + ".in"))):
                      return fmt(new_p)
          print("ERROR: ", fn, p)
          return m.group(0)
      new_c = re.sub(r'#include "([^"]+)"', repl, c)
      if new_c != c:
          print(fn)
          with open(fn, 'w') as f:
              f.write(new_c)
```

Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14849

Reviewed By: dzhulgakov

Differential Revision: D13363445

Pulled By: ezyang

fbshipit-source-id: 52361f878a672785f9306c9e9ab2513128092b68
2018-12-08 19:38:30 -08:00

114 lines
3.4 KiB
C++

#include <torch/csrc/jit/ir.h>
#include <algorithm>
#include <unordered_map>
#include <torch/csrc/jit/assertions.h>
#include <torch/csrc/jit/interned_strings.h>
#include <torch/csrc/jit/passes/common_subexpression_elimination.h>
#include <torch/csrc/jit/node_hashing.h>
#include <torch/csrc/utils/functional.h>
#include <torch/csrc/utils/hash.h>
namespace torch { namespace jit {
namespace {
bool tensorEqual(const at::Tensor& lhs, const at::Tensor& rhs) {
return &lhs.type() == &rhs.type() && lhs.equal(rhs);
}
bool tensorListEqual(const std::vector<at::Tensor>& lhs, const std::vector<at::Tensor>& rhs) {
if (lhs.size() != rhs.size()) return false;
return std::equal(lhs.begin(), lhs.end(), rhs.begin(), tensorEqual);
}
// Check whether two nodes have the same attributes in CSE.
// This function may be too conservative for general use.
// Do NOT support g/gs attributes.
bool attributesEqualCSE(const Node* lhs, const Node* rhs) {
JIT_ASSERT(lhs != nullptr);
JIT_ASSERT(rhs != nullptr);
// One has attributes, the other does not.
if (lhs->hasAttributes() != rhs->hasAttributes()) return false;
// Neither has attributes.
if (!lhs->hasAttributes() && !rhs->hasAttributes()) return true;
auto lnames = lhs->attributeNames();
auto rnames = rhs->attributeNames();
std::sort(lnames.begin(), lnames.end());
std::sort(rnames.begin(), rnames.end());
if (lnames != rnames) return false;
for (auto name : lnames) {
if (lhs->kindOf(name) != rhs->kindOf(name)) return false;
#define COMPARE_ATTRIBUTEVALUE(type) \
case AttributeKind::type: \
{ if (lhs->type(name) != rhs->type(name)) return false; } break;
switch(lhs->kindOf(name)) {
COMPARE_ATTRIBUTEVALUE(f)
COMPARE_ATTRIBUTEVALUE(fs)
COMPARE_ATTRIBUTEVALUE(i)
COMPARE_ATTRIBUTEVALUE(is)
COMPARE_ATTRIBUTEVALUE(s)
COMPARE_ATTRIBUTEVALUE(ss)
case AttributeKind::t: {
if (!tensorEqual(lhs->t(name), rhs->t(name))) return false;
break;
}
case AttributeKind::ts: {
if (!tensorListEqual(lhs->ts(name), rhs->ts(name))) return false;
break;
}
case AttributeKind::g:
case AttributeKind::gs:
return false;
}
#undef COMPARE_ATTRIBUTEVALUE
}
return true;
}
} // anonymous namespace
size_t HashNode::operator()(const Node* k) const {
JIT_ASSERT(k != nullptr);
return get_hash(k->kind(),
fmap(k->outputs(), [](const Value *v) { return v->type()->kind(); }),
fmap(k->inputs(), [](const Value *v) { return v->unique(); }));
};
bool EqualNode::operator()(const Node* lhs, const Node* rhs) const {
if (lhs == nullptr && rhs == nullptr) return true;
if (lhs == nullptr || rhs == nullptr) return false;
if (lhs->kind() != rhs->kind()) return false;
// Check whether the output types are the same.
auto lhs_outputs = lhs->outputs();
auto rhs_outputs = rhs->outputs();
if (lhs_outputs.size() != rhs_outputs.size()) return false;
for (size_t i = 0; i < lhs_outputs.size(); ++i) {
if (*lhs_outputs[i]->type() != *rhs_outputs[i]->type())
return false;
}
// Check whether the inputs are the same.
auto lhs_inputs = lhs->inputs();
auto rhs_inputs = rhs->inputs();
if (lhs_inputs.size() != rhs_inputs.size()) return false;
if (!std::equal(lhs_inputs.begin(), lhs_inputs.end(), rhs_inputs.begin())) return false;
if (!attributesEqualCSE(lhs, rhs)) return false;
return true;
};
}}