huajsj commented on a change in pull request #8702: URL: https://github.com/apache/tvm/pull/8702#discussion_r698845894
########## File path: python/tvm/contrib/pipeline_executor.py ########## @@ -0,0 +1,395 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +"""Pipeline executor that executes a series of modules in a pipeline fashion.""" +import json +import tvm._ffi +from tvm import relay +from tvm.contrib import graph_executor + + +def pipeline_executor_enabled(): + """check if pipeline executor is enabled. + + Return + ------- + enable: bool + Return pipeline executor is enabled or not. + """ + return tvm._ffi.get_global_func("tvm.pipeline_executor.create", + allow_missing=True) is not None + + +def build(pipe_configs): + """build module list that can use for pipeline execution. + + Parameters + ---------- + mod_n_configs: Dict[IRModule, Dict[str, Any]] + build configuration informaton, structure like following. + {IRModule: {"target":target, + "target_host":target_host, + "params":params, + "mod_name"mod_name, + "build":build}} + + Returns + ------- + ret: List[IRModule] + list of IRModule + string_config: Dict[int, Dict[str, any]] + pipeline configuration + """ + mods = {} + mod_n_configs = pipe_configs.get_config() + config_len = len(mod_n_configs) + string_config = [{} for _ in range(config_len)] + #for _, (ir_mod, mod_config) in enumerate(mod_n_configs.items()): + for ir_mod, mod_config in mod_n_configs.items(): + mconf = mod_config["pipeline"].copy() + mod_indx = mconf["mod_indx"] - 1 + # Get mod device config + dev = mod_config["dev"] + target = mod_config["target"] + build_func = relay.build + # if there is a self defined build function then use it. + if "build" in mod_config and mod_config["build"]: + build_func = mod_config["build"] + + # build IRModule + mod = build_func( + ir_mod, + target, + params=mod_config["params"], + target_host=mod_config["target_host"], + mod_name=mod_config["mod_name"], + ) + + mconf["dev"] = "{},{}".format(dev.device_type, dev.device_id) + # Create pipeline configuration + string_config[mod_indx] = mconf + # associate mod with device + mods[mod] = {"dev": dev} + + # return PipeModuleConfig + return PipeModuleConfig(mods, string_config) + + +def create(pipe_mod_config): + """Create a pipeline runtime executor. + + Parameters + ---------- + + pipe_mod_config : PipeModuleConfig + class to storage IRModule list and pipeline configuration. + + Returns + ------- + submodule : PipelineModule + Runtime pipeline module. + """ + + return PipelineModule(pipe_mod_config) + +class PipelineModule(object): + """Wrapper runtime module. This is a thin wrapper of the underlying TVM module. + + Parameters + ---------- + pipeline_mods : List[GraphModule] + The internal tvm module that holds the actual graph functions. + pipeline_config : Dict[IRModule, Dict[str, Any]] + modules and modules dependency configuration informaiton. + """ + + def __init__(self, pipe_mod_config): + self.pipeline_mods_ = pipe_mod_config.pipeline_mods_ + self.mod_config_ = pipe_mod_config.mods_config_ + mods, config = self.graph_executor_create(self.pipeline_mods_, self.mod_config_) + assert pipeline_executor_enabled(), \ + "Pipeline executor is not enabled. Please \ + re-build TVM with USE_PIPELINE_EXECUTOR=ON" + pipelinecreate = tvm._ffi.get_global_func("tvm.pipeline_executor.create", + allow_missing=False) + assert pipelinecreate + module = pipelinecreate(mods, config) + + self.module_ = module + + def graph_executor_create(self, pipeline_mods, mod_config): + """Create graph_executor list and return string format config. + + Parameters + ---------- + + pipeline_mods : List[IRModule] + list of IRModule + + mod_config : Dict[int, Dict[str, Any]] + modules and modules dependency configuration informaiton. + + Returns + ------- + mods : List[GraphModule] + Runtime graph module. + + mod_config : str + mods configuration + """ + + mods = [] + for pipeline_mod in pipeline_mods: + mod = graph_executor.GraphModule( + pipeline_mod["default"](pipeline_mods[pipeline_mod]["dev"]) + ) + mods.append(mod.module) + + return mods, json.dumps(mod_config) + + +class PipelineConfig(object): + """Pipeline Configuration Class, in this class there are 2 internal class, + first is Module which use to represent Module, second is Interface which use + to represent Module input/output and Pipeline Module input/output, by setting + dependency relation between Interfaces this class can build the module + connection relation. + + The class Hierarchical as following. + PipelineConfig ---> ModuleWrapper ---> Interface(input/output) + """ + + class ModuleWrapper: + """The class use use to represent Module and storage module index and + Interface information. + """ + + class Interface: + """The class that use to storage module connection information. + There are 2 types Interface Input:1 Output:2 + Parameters + ---------- + + owner : ModuleWrapper + The class that own this interface, in such class there are + Module information like index, module name + + itype : integer + Interface type, 1 is input interface, 2 is output interface + + name : str/integer + Interface name, for input that is string for example "data0" + for output that is integer for example 0. + """ + + def __init__(self, owner, itype, name): + self.owner_ = owner + self.itype_ = itype + self.name_ = str(name) + self.dependent_ = [] + + def get_name(self): + mname = "" + if self.owner_: + mname = self.owner_.name_ + + return mname, self.name_ + + def get_owner_indx(self): + return self.owner_.indx_ + + def get_dependent_str(self): + name = "" + for dependent in self.dependent_: + mname, dname = dependent.get_name() + name = name + (mname + ":output(" + dname if self.itype_ == 2 else "") + name = name + (")" if self.itype_ == 2 else mname + ":" + dname) + return name + + def add_dependent(self, dependent): + """ + # check if the dependency setting correct. + # correct connection are following + # 1. global input to module input + # 2. module output to next module input + # 3. module output to global output + """ + owner_indx = self.get_owner_indx() + dep_owner_indx = dependent.get_owner_indx() + assert owner_indx != dep_owner_indx, f"can not set self as dependent." + assert not ( + owner_indx > dep_owner_indx + and not (dependent.itype_ == 2 and dep_owner_indx == 0) + ), f"dependent only can be next module interface or global output." + assert not ( + owner_indx == 0 and dependent.itype_ != 1 + ), f"global input only can set dependent with module input." + + self.dependent_.append(dependent) + + def __init__(self, indx=0): + self.indx_ = indx + self.name_ = "mod" + str(indx) if indx else "" + self.interfaces_ = {1: {}, 2: {}} + self.target_host_ = None + self.mod_name_ = "default" + self.build_func_ = None + self.params_ = None + self.target_ = None + self.dev_ = None + + + def get_interface(self, itype, name): + if name not in self.interfaces_[itype]: + self.interfaces_[itype][name] = self.Interface(self, itype, name) + + return self.interfaces_[itype][name] + + def input(self, name): + return self.get_interface(1, name) + + def output(self, index): + return self.get_interface(2, index) + + def set_target_host(self, host): + self.target_host_ = host + + def set_mod_name(self, name): + self.mod_name_ = name + + def set_build_func(self, build_func): + self.build_func_ = build_func + + def set_params(self, params): + self.params_ = params + + def set_target(self, target): + self.target_ = target + + def set_dev(self, dev): + self.dev_ = dev + + def __init__(self, mods): + self.pipe_module_name_ = "pipeline_module" + self.mod_wrapper = {m: self.ModuleWrapper(i + 1) for m, i in zip(mods, range(len(mods)))} + self.mod_wrapper[self.pipe_module_name_] = self.ModuleWrapper(0) Review comment: self.ModuleWrapper(0) is not for first sub module, this object is the global module that represent pipelinemodule and use for exposing global input and output interface/interface binding, the idea here is to invent a virtual module beside of all sub module to storage the connection relations between pipeline executor input/output and internal sub module for example ``` input: data_0 : mod0: data_0 data_1: mod1: data_0 output: ouput(0) : mod0:ouput(0) ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
