frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 12:54:11 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
c6a8db0b9acbefc66f02e7ff46ad6bbedabd8b4b
pytorch/torch/_inductor/fx_passes/reinplace.py
Laith Sakka c6a8db0b9a Fix issues with generalized_scatter and setitem allocated unbacked symbols. (#164341) 
Three fixes:
1. When doing t[u0] +=1  if u0 is unbacked we could allocate a new unbacked symbol during the the indexing of t[u0] (when we fake trace setitem), namely because meta_select does allocate a new unbacked symbol for the storage offset when we do not know if u0>=0 or u0<0.  but the output size/stride of setitem(), does not depend on that new symbol. it's self consumed in setitem so we shall ignore it.

2. Also when we trace through generalized_scatter the applications of the views could allocate unbacked symints
but those do not effect final output, we also shall ignore them.

3.Before accessing strides in lowering we shall materialize.

Address  https://github.com/pytorch/pytorch/issues/114293 and https://github.com/pytorch/pytorch/issues/131911

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164341
Approved by: https://github.com/bobrenjc93
2025-10-18 03:20:30 +00:00

784 lines
30 KiB
Python
Raw Blame History

# mypy: allow-untyped-defs
import itertools
import logging
import operator
from collections import defaultdict
from collections.abc import Sequence
from contextlib import nullcontext
from dataclasses import dataclass
from typing import Any, Callable, cast
import torch
import torch.fx.node
from torch._C._dynamo.guards import compute_overlapping_tensors
from torch._dispatch.python import enable_python_dispatcher
from torch._dynamo.utils import ReinplaceCounters, ReInplaceTrigger
from torch._guards import detect_fake_mode
from torch._higher_order_ops.triton_kernel_wrap import (
kernel_side_table,
triton_kernel_wrapper_functional,
)
from torch._inductor import config, inductor_prims
from torch._inductor.fx_utils import get_node_storage, is_node_realized
from torch._inductor.lowering import (
inplaceable_foreach_ops as inplaceable_foreach_ops_lowerings,
)
from torch._inductor.virtualized import V
from torch.fx.experimental.symbolic_shapes import GuardOnDataDependentSymNode
from torch.fx.immutable_collections import immutable_dict, immutable_list
from torch.fx.passes.reinplace import _is_view_op
from torch.utils import _pytree as pytree
from torch.utils._ordered_set import OrderedSet
log = logging.getLogger(__name__)
aten = torch.ops.aten
@dataclass(frozen=True)
class InplaceableOp:
inplace_op: Callable[..., Any]
mutated_arg: int
extra_check: Callable[[torch.fx.Node], bool] = lambda node: True
_SCATTER_OP_TO_VIEW = {
torch.ops.aten.diagonal_scatter.default: torch.ops.aten.diagonal.default,
torch.ops.aten.select_scatter.default: torch.ops.aten.select.int,
torch.ops.aten.slice_scatter.default: torch.ops.aten.slice.Tensor,
torch.ops.aten.as_strided_scatter.default: torch.ops.aten.as_strided.default,
}
_VIEW_OP_TO_SCATTER = {v: k for k, v in _SCATTER_OP_TO_VIEW.items()}
def graph_call_function(graph: torch.fx.Graph, fn, *args, **kwargs):
fake_args, fake_kwargs = pytree.tree_map(
lambda node: node.meta["val"] if isinstance(node, torch.fx.Node) else node,
(args, kwargs),
)
with V.fake_mode:
fake_result = fn(*fake_args, **fake_kwargs)
node = graph.call_function(fn, args, kwargs)
node.meta["val"] = fake_result
return node
@dataclass
class ViewOp:
target: torch._ops.OpOverload
args: tuple[Any, ...]
kwargs: dict[str, Any]
def _inplace_generalized_scatter(
inp: torch.Tensor, src: torch.Tensor, view_ops: list[ViewOp]
) -> torch.Tensor:
tmp = inp
for view in view_ops:
fake_args, fake_kwargs = pytree.tree_map(
lambda node: node.meta["val"] if isinstance(node, torch.fx.Node) else node,
(view.args, view.kwargs),
)
# slice and select can allocate new unbacked symints, but those won't be reflected
# in the output of this function, hence shall be ignored.
fake_mode = detect_fake_mode(fake_args)
with (
fake_mode.shape_env.ignore_fresh_unbacked_symbols()
if fake_mode and fake_mode.shape_env
else nullcontext()
):
tmp = view.target(tmp, *fake_args, **fake_kwargs)
try:
tmp.copy_(src)
except RuntimeError as e:
raise RuntimeError(
f"shape error in scatter op, can not broadcast {src.shape} to {tmp.shape}"
) from e
return inp
def _generalized_scatter(
inp: torch.Tensor, src: torch.Tensor, view_ops: list[ViewOp]
) -> torch.Tensor:
out = inp.clone()
return _inplace_generalized_scatter(out, src, view_ops)
def _decompose_scatter_functional_helper(
graph: torch.fx.Graph,
inp: torch.Tensor,
src: torch.Tensor,
view_ops: list[ViewOp],
) -> torch.fx.Node:
view_op, view_ops_tail = view_ops[0], view_ops[1:]
if view_ops_tail:
view = graph_call_function(
graph, view_op.target, inp, *view_op.args, **view_op.kwargs
)
src = _decompose_scatter_functional_helper(graph, view, src, view_ops[1:]) # type: ignore[assignment]
return graph_call_function(
graph,
_VIEW_OP_TO_SCATTER[view_op.target],
inp,
src,
*view_op.args,
**view_op.kwargs,
)
def _decompose_scatter_functional(
graph: torch.fx.Graph, node: torch.fx.Node
) -> torch.fx.Node:
"""Decompose _generalized_scatter to a sequence of view_scatter operations
e.g. _generalized_scatter(inp, src, [(aten.slice, 0, 0, 10), (aten.slice, 1, 10, -10)])
will become
view = aten.slice(inp, 0, 0, 10)
view_updated = aten.slice_scatter(view, src, 1, 10, -10)
inp_updated = aten.slice_scatter(inp, view_updated, 0, 0, 10)
"""
assert node.target is _generalized_scatter
return _decompose_scatter_functional_helper(graph, *node.args) # type: ignore[arg-type]
def _decompose_scatter_mutating(
graph: torch.fx.Graph, node: torch.fx.Node
) -> torch.fx.Node:
"""Decompose _generalized_scatter using mutations
e.g. _generalized_scatter(inp, src, [(aten.slice, 0, 0, 10), (aten.slice, 1, 10, -10)])
will become
inp_updated = aten.clone(inp)
slice1 = aten.slice(inp_updated, 0, 0, 10)
slice2 = aten.slice(slice1, 1, 10, -10)
slice2.copy_(src)
"""
assert node.target in (_generalized_scatter, _inplace_generalized_scatter)
inp, src, view_ops = node.args
assert not node.kwargs
if node.target is _generalized_scatter:
inp = graph_call_function(graph, aten.clone, inp)
tmp = inp
for view in view_ops: # type: ignore[union-attr]
tmp = graph_call_function(graph, view.target, tmp, *view.args, **view.kwargs) # type: ignore[union-attr]
graph_call_function(graph, aten.copy_.default, tmp, src)
return inp # type: ignore[return-value]
# View ops whose view_scatter op is lowered into mutations anyway,
# so is never a pessimisation to decompose.
_ALWAYS_MUTATING_SCATTER_OPS = OrderedSet(
[
aten.as_strided.default,
aten.diagonal.default,
]
)
def scatter_always_uses_mutation(node: torch.fx.Node) -> bool:
_, _, view_ops = node.args
view_ops = cast(Sequence[torch.fx.node.Argument], view_ops)
return any(
target in _ALWAYS_MUTATING_SCATTER_OPS
for view in view_ops
if isinstance(target := getattr(view, "target", None), torch._ops.OpOverload)
)
def should_reinplace_scatter(node: torch.fx.Node) -> bool:
"""Choose between mutating and functional scatter decompositions
Reinplacing view scatter ops can be pessimising as it blocks fusion with the
input or output tensor computations. However, it is still profitable if the
input and output would have been realized anyway.
"""
inp, _src, _view_ops = node.args
# Mutating scatter ops unconditionally realize input and output
if scatter_always_uses_mutation(node):
return True
if is_node_realized(inp) and is_node_realized(node): # type: ignore[arg-type]
return True
# If the output is copied back into the input, this forces both to be
# realized as the output is a user of the input
if inp.op in ("placeholder", "get_attr") and any( # type: ignore[union-attr]
user.target is aten.copy_.default and user.args[0] is inp for user in node.users
):
return True
# Otherwise, assume fusions will make functional variants profitable
return False
def decompose_generalized_scatter(graph: torch.fx.Graph) -> None:
"""Replace _generalized_scatter with normal aten ops"""
for node in itertools.chain(
graph.find_nodes(op="call_function", target=_generalized_scatter),
graph.find_nodes(op="call_function", target=_inplace_generalized_scatter),
):
use_mutation = (
node.target is _inplace_generalized_scatter
or scatter_always_uses_mutation(node)
)
with graph.inserting_before(node):
if use_mutation:
new_node = _decompose_scatter_mutating(graph, node)
else:
new_node = _decompose_scatter_functional(graph, node)
node.replace_all_uses_with(new_node)
graph.erase_node(node)
def canonicalize_view_scatter_ops(graph: torch.fx.Graph) -> None:
"""
This canonicalizes view scatter ops into a generalized form, defined as:
def scatter(inp, src, views):
tmp = inp.clone()
for view in views:
tmp = view(tmp)
tmp.copy_(src)
We also fuse consecutive view scatter ops of the form
a = scatter(view2(self), src, [view1])
b = scatter(self, a, [view2])
which can be rewritten as
b = scatter(self, src, [view2, view1])
a = view2(b)
This is both more efficient as we only do a single scatter, and also
easier to reinplace since there is only one use of `self`
"""
node_to_view_base: dict[torch.fx.Node, torch.fx.Node] = {}
node_to_view_op: dict[torch.fx.Node, list[ViewOp]] = defaultdict(list)
def handle_views(node: torch.fx.Node):
inp = node.args[0]
node_to_view_base[node] = node_to_view_base.get(inp, inp) # type: ignore[arg-type, assignment]
node_to_view_op[node] = [
*node_to_view_op[inp], # type: ignore[index]
ViewOp(
node.target, # type: ignore[arg-type]
args=node.args[1:],
kwargs=node.kwargs,
),
]
def handle_view_scatter(node: torch.fx.Node):
assert len(node.args) >= 2
inp, src = node.args[:2]
assert isinstance(node.target, torch._ops.OpOverload)
scatter_view_op = ViewOp(
_SCATTER_OP_TO_VIEW[node.target],
args=node.args[2:],
kwargs=node.kwargs,
)
def can_fuse():
if src.target is not _generalized_scatter: # type: ignore[union-attr]
return False
src_inp, _src_src, _src_scatter_view_op = src.args # type: ignore[union-attr]
inp_base = node_to_view_base.get(inp, inp) # type: ignore[arg-type]
src_base = node_to_view_base.get(src_inp, src_inp) # type: ignore[arg-type]
return inp_base is src_base and node_to_view_op[src_inp] == [ # type: ignore[index]
*node_to_view_op[inp], # type: ignore[index]
scatter_view_op,
]
if not can_fuse():
with graph.inserting_before(node):
new_node = graph_call_function(
graph,
_generalized_scatter,
inp,
src,
[scatter_view_op],
)
node.replace_all_uses_with(new_node)
graph.erase_node(node)
return
_src_inp, src_src, src_scatter_view_op = src.args # type: ignore[union-attr]
with graph.inserting_before(src): # type: ignore[arg-type]
new_node = graph_call_function(
graph,
_generalized_scatter,
inp,
src_src,
[scatter_view_op, *src_scatter_view_op], # type: ignore[misc]
)
node.replace_all_uses_with(new_node)
graph.erase_node(node)
if src.users: # type: ignore[union-attr]
new_src = graph_call_function(
graph,
_SCATTER_OP_TO_VIEW[node.target],
new_node,
*node.args[2:],
**node.kwargs,
)
handle_views(new_src)
src.replace_all_uses_with(new_src) # type: ignore[union-attr]
graph.erase_node(src) # type: ignore[arg-type]
for node in graph.nodes:
if _is_view_op(node.target):
handle_views(node)
elif node.target in _SCATTER_OP_TO_VIEW:
handle_view_scatter(node)
inplaceable_ops: dict[Callable[..., Any], InplaceableOp] = {
aten.index_put.default: InplaceableOp(aten.index_put_.default, 0),
aten._unsafe_index_put.default: InplaceableOp(inductor_prims._unsafe_index_put_, 0),
_generalized_scatter: InplaceableOp(
_inplace_generalized_scatter,
0,
extra_check=should_reinplace_scatter,
),
}
try:
c10d_functional = torch.ops._c10d_functional
inplaceable_collective_ops: dict[Callable[..., Any], InplaceableOp] = {
c10d_functional.all_reduce.default: InplaceableOp(
c10d_functional.all_reduce_.default, 0
),
c10d_functional.all_reduce_coalesced.default: InplaceableOp(
c10d_functional.all_reduce_coalesced_.default, 0
),
}
inplaceable_ops.update(inplaceable_collective_ops)
except AttributeError:
# _c10d_functional ops are only available when torch
# is built with USE_DISTRIBUTED=1.
pass
inplaceable_foreach_ops: dict[torch._ops.OpOverload, InplaceableOp] = {}
for outplace_op, inplace_op in inplaceable_foreach_ops_lowerings.items():
inplaceable_foreach_ops[outplace_op] = InplaceableOp(inplace_op, 0)
inplaceable_triton_ops = OrderedSet([triton_kernel_wrapper_functional])
# Operators that don't depend on the tensor data
META_ONLY_OPS = OrderedSet(
[
aten.sym_size.int,
aten.sym_stride.int,
aten.sym_numel.default,
aten.sym_storage_offset.default,
]
)
def reinplace_inplaceable_ops_core(graph: torch.fx.Graph) -> None:
"""
Reinplaces in-placeable operations.
If there are no uses of a view of the mutated arg after the current node,
it is possible to inplace the op.
This above algorithm could be justified by observing side effects. While
we traverse the graph in forwards direction, only latter nodes could view
side effects of the current node. If the current node is not used later as
well as no view of this node is used later in the graph, then it is safe to
inplace as there would be no way to observe the side effects.
This condition is slightly different for graph inputs where they can only
be inplaced if the above condition is true and there's a copy_ in the
epilogue that signals that the caller wants to observe the mutation.
Unlike JIT Inductor, AOTInductor currently unlifts weights and buffers from
input args, so instead of checking mutation on placeholder, AOTInductor
checks mutation on get_attr. This is subject to change in future.
"""
copy_args_to_copy_nodes = {}
# maps argument to the first copy_ node that mutates it.
copy_nodes = {}
mutated_inputs = OrderedSet[Any]()
storage_to_nodes = defaultdict(list)
node_order: dict[Any, int] = {}
for i, node in enumerate(reversed(graph.nodes)):
node_order[node] = len(graph.nodes) - i - 1
storage_to_nodes[get_node_storage(node)].append(node)
if node.target == aten.copy_.default and node.args[0].op in (
"placeholder",
"get_attr",
):
dst = node.args[0]
src = node.args[1]
# If the target is a getitem and it indexes a possible clone,
# then skip over it
if src.target == operator.getitem and (
(
src.args[0].target == triton_kernel_wrapper_functional
and src.args[0].kwargs["kwargs"][src.args[1]] == node.args[0]
)
or (src.args[0].target in inplaceable_foreach_ops)
or (src.args[0].target == torch.ops.higher_order.auto_functionalized)
):
src = src.args[0]
copy_args_to_copy_nodes[(dst, src)] = node
copy_nodes[dst] = node
mutated_inputs.add(node.args[0])
def any_use_of_views_after_node(node, shared_view_nodes, *, copy_node, mutated_arg):
node_loc = node_order[node]
copy_node_loc = node_order[copy_node] if copy_node is not None else None
def is_meta_only_user(node):
if _is_view_op(node.target):
return all(is_meta_only_user(u) for u in node.users)
return node.target in META_ONLY_OPS
for view in shared_view_nodes:
for user in view.users:
user_loc = node_order[user]
# Skip all users before node
if user_loc <= node_loc:
continue
# Ignore uses after the copy_ epilogue node, where the input
# has already been mutated anyway
if copy_node_loc is not None and copy_node_loc <= user_loc:
continue
# Reinplacing does not change shape metadata
if is_meta_only_user(user):
continue
# If our graph looks like:
# foo(mutated_arg)
# mutated_arg.copy_(other)
# then it's safe for us to reinplace foo because mutated_arg
# will get overwritten anyways.
if (
user.target is torch.ops.aten.copy_.default
and mutated_arg is user.args[0]
):
continue
return True
return False
def can_inplace(node, mutated_arg):
# ls should be a list of tensors that all shares the same storage.
def _overlap(ls) -> bool:
try:
return len(compute_overlapping_tensors(ls)) != 0
except GuardOnDataDependentSymNode:
# If we fail with data dependent error we assume they all overlap.
return True
if isinstance(mutated_arg, (list, tuple)):
# TODO Using _overlap here causes a several issues.
unique_storages = OrderedSet(get_node_storage(arg) for arg in mutated_arg)
if len(unique_storages) != len(mutated_arg):
# At least two Tensors in mutated_arg alias each other, so we can't reinplace it.
# We can probably do better (that is, reinplace one of them and clone the other)
# but that requires more work and mutable List[Tensor] are not that common.
return False
return all(can_inplace(node, arg) for arg in mutated_arg)
if get_node_storage(mutated_arg) is None:
return False
shared_view_nodes = storage_to_nodes[get_node_storage(mutated_arg)]
# Only keep tensor that might overlap with mutated_arg.
shared_view_nodes = [
v
for v in shared_view_nodes
if _overlap([mutated_arg.meta["val"], v.meta["val"]])
]
if mutated_arg.op in ("placeholder", "get_attr"):
# Get the first copy_ node that mutates the mutated_arg.
copy_node = copy_nodes.get(mutated_arg)
if copy_node is None:
# There is no copy_ back to the candidate mutated_arg (which is a graph input).
# Therefore the semantics of the program are that it does not mutate
# mutated_arg, so we cannot re-inplace it.
return False
if any_use_of_views_after_node(
node, shared_view_nodes, copy_node=copy_node, mutated_arg=mutated_arg
):
return False
return True
elif any(view.op in ("placeholder", "get_attr") for view in shared_view_nodes):
# This should never happen in auto_functionalize_v2 non-inference mode,
# since all mutated_arg are bases.
# If mutated arg is view of any of the inputs of the graph,
# do not allow for inplacing.
# This would require more sophisticated algorithm to handle
return False
else:
return not any_use_of_views_after_node(
node, shared_view_nodes, copy_node=None, mutated_arg=mutated_arg
)
def log_inplace_results(
node_name,
old_tensors_to_clone,
tensors_to_clone,
missed_args,
missed_nodes,
trigger,
):
# Total size of possibly_missed_reinplacing_opportunities for tensors with static shapes.
missed_bytes = 0
def bytes(node):
t = node.meta.get("val", None)
if (
t is not None
and isinstance(t.element_size(), int)
and isinstance(t.numel(), int)
):
return t.element_size() * t.numel()
else:
return 0
for node in missed_nodes:
if isinstance(node, (list, tuple)):
for n in node:
missed_bytes += bytes(n)
else:
missed_bytes += bytes(node)
log.info(
"For node %s, attempted to reinplace %s. We were unable to reinplace %s; "
"%s (if non-empty) are possible missed reinplacing opportunities that may be bad for "
"memory usage and performance. Total size of missed opportunities with static shapes is"
" : %s bytes.",
node_name,
old_tensors_to_clone,
tensors_to_clone,
missed_args,
missed_bytes,
)
ReinplaceCounters.add_missed_opportunities(trigger, len(missed_args))
ReinplaceCounters.add_missed_bytes(trigger, missed_bytes)
replace_dict: dict[torch.fx.Node, torch.fx.Node] = {}
def reinplace_and_refine_tensors_to_clone(
old_tensors_to_clone, kwargs, node_name, trigger
):
tensors_to_clone: list[str] = []
storage_of_reinplaced_args = OrderedSet[int | None]()
# Those used to count possibly_missed_reinplacing_opportunities
missed_nodes = []
missed_args = []
# TODO this logic can be made more precise using _overlap
def tensor_with_same_storage_already_reinplaced(arg):
if isinstance(arg, (list, tuple)):
return any(
get_node_storage(a) in storage_of_reinplaced_args for a in arg
)
return get_node_storage(mutated_arg) in storage_of_reinplaced_args
for arg in old_tensors_to_clone:
assert arg in kwargs
mutated_arg = kwargs[arg]
# Let's say we have:
# - op(x, y) that mutates both x and y
# - new_x, new_y = functional_op(x, y) is the functional variant
# If we are presented with functional_op(x, x), we must not reinplace
# this into op(x, x), because then it would be writing to the same Tensor.
# Instead, it's OK to reinplace one of them and to clone the other:
# >>> y = x.clone()
# >>> op(x, y)
# This also applies if we have views: functional_op(x, x[0])
# should not reinplace into op(x, x[0]).
should_attempt_reinplace = not tensor_with_same_storage_already_reinplaced(
mutated_arg
)
if should_attempt_reinplace and can_inplace(node, mutated_arg):
# In general, we probably do not need those optimizations.
copy_node = copy_args_to_copy_nodes.get((mutated_arg, node))
if copy_node is not None:
replace_dict[copy_node] = copy_node.args[0]
if trigger != ReInplaceTrigger.AUTO_FUNC_V2:
for user in node.users:
# For auto_functionalize_v2, arg is the index of the base, where base at index i corresponds to
# output atindex size(out)+i.
# This used to compare string with integers before for auto_functionalize_v2. Not sure
# if it was needed for inplaceable_triton_ops?
if user.target == operator.getitem and user.args[1] == arg:
replace_dict[user] = mutated_arg
if isinstance(mutated_arg, (list, tuple)):
for a in mutated_arg:
storage_of_reinplaced_args.add(get_node_storage(a))
else:
storage_of_reinplaced_args.add(get_node_storage(mutated_arg))
else:
if should_attempt_reinplace:
missed_args.append(arg)
missed_nodes.append(mutated_arg)
tensors_to_clone.append(arg)
log_inplace_results(
node_name,
old_tensors_to_clone,
tensors_to_clone,
missed_args,
missed_nodes,
trigger,
)
return tensors_to_clone
for node in graph.nodes:
if (inplaceable_op := inplaceable_ops.get(node.target, None)) is not None:
mutated_arg = node.args[inplaceable_op.mutated_arg]
if can_inplace(node, mutated_arg) and inplaceable_op.extra_check(node):
# TODO(yifu): this doesn't properly remove copy epilogues for
# ops that mutate multiple inputs. Need to revise the copy
# node tracking logic to support the case.
copy_node = copy_args_to_copy_nodes.get((mutated_arg, node))
if copy_node is not None:
replace_dict[copy_node] = copy_node.args[0]
node.target = inplaceable_op.inplace_op
elif node.target == torch.ops.higher_order.auto_functionalized_v2:
_mutable_op = node.args[0]
kwargs = node.kwargs
all_bases = kwargs["_all_bases"]
bases_to_clone = range(len(all_bases))
base_tensors_dct = dict(enumerate(all_bases))
new_bases_to_clone: list[int] = reinplace_and_refine_tensors_to_clone(
bases_to_clone,
base_tensors_dct,
node.target,
ReInplaceTrigger.AUTO_FUNC_V2,
)
# Stash the metadata. There is a pass later on where we decompose
# auto_functionalized into clones + a mutable op; this metadata
# tells the decomp to only clone the following inputs
node.meta["only_clone_these_tensors"] = new_bases_to_clone
elif node.target == torch.ops.higher_order.auto_functionalized:
_mutable_op = node.args[0]
from torch._higher_order_ops.auto_functionalize import get_mutable_args
tensors_to_clone, _ = get_mutable_args(_mutable_op)
# Don't try to reinplace Tensor | None args that are None.
tensors_to_clone = [
t for t in tensors_to_clone if node.kwargs[t] is not None
]
tensors_to_clone = reinplace_and_refine_tensors_to_clone(
tensors_to_clone,
node.kwargs,
_mutable_op._name,
ReInplaceTrigger.AUTO_FUNC_V1,
)
# Stash the metadata. There is a pass later on where we decompose
# auto_functionalized into clones + a mutable op; this metadata
# tells the decomp to only clone the following inputs
node.meta["only_clone_these_tensors"] = tensors_to_clone
elif node.target in inplaceable_triton_ops:
kernel_idx = node.kwargs["kernel_idx"]
kernel = kernel_side_table.get_kernel(kernel_idx)
from triton.runtime.autotuner import Autotuner
from triton.runtime.jit import JITFunction
if isinstance(kernel, JITFunction):
kernel_name = kernel.fn.__name__
elif isinstance(kernel, Autotuner):
if config.is_fbcode():
# Autotuner has different implementations for AMD and NV
if torch.version.hip is None:
kernel_name = kernel.base_fn.__name__
else:
kernel_name = kernel.fn.__name__
else:
kernel_name = kernel.base_fn.__name__
else:
raise AssertionError("Unknown triton kernel type")
# inplaceable_triton_ops take an additional argument called
# tensors_to_clone which contain a list of tensors to clone
# This pass iterates over them and sees which ones are safe
# to eliminate (i.e. no longer need the clones)
tensors_to_clone = reinplace_and_refine_tensors_to_clone(
node.kwargs["tensors_to_clone"],
node.kwargs["kwargs"],
kernel_name,
ReInplaceTrigger.TRITON_OPS,
)
kwargs = dict(node.kwargs)
kwargs["tensors_to_clone"] = tensors_to_clone
node.kwargs = immutable_dict(kwargs)
if "eager_input_vals" in node.meta:
# We changed the kwargs, so we need to update eager_input_vals
# to something sane.
args, kwargs = node.meta["eager_input_vals"]
new_kwargs = {**kwargs}
new_kwargs["tensors_to_clone"] = immutable_list(tensors_to_clone)
new_kwargs = immutable_dict(new_kwargs)
node.meta["eager_input_vals"] = (args, new_kwargs)
elif (
inplaceable_op := inplaceable_foreach_ops.get(node.target, None)
) is not None:
mutated_args = node.args[inplaceable_op.mutated_arg]
if not all((arg, node) in copy_args_to_copy_nodes for arg in mutated_args):
continue
if can_inplace(node, mutated_args):
for arg in mutated_args:
copy_node = copy_args_to_copy_nodes[(arg, node)]
replace_dict[copy_node] = copy_node.args[0]
node.target = inplaceable_op.inplace_op
for node, replacement in replace_dict.items():
while replacement in replace_dict:
replacement = replace_dict[replacement]
replace_dict[node] = replacement
node.replace_all_uses_with(replacement)
graph.erase_node(node)
def reinplace_inplaceable_ops(
fake_tensor_updater: torch._inductor.fx_utils.FakeTensorUpdater,
graph: torch.fx.Graph,
) -> None:
with enable_python_dispatcher():
canonicalize_view_scatter_ops(graph)
# canonicalize_view_scatter_ops adds new operations to the graph.
# We run fake_tensor_updater to update the alias information.
# Correct alias information is required for `reinplace_inplaceable_ops_core`.
fake_tensor_updater.incremental_update()
reinplace_inplaceable_ops_core(graph)
decompose_generalized_scatter(graph)
Reference in New Issue View Git Blame Copy Permalink