This file lists some rules and limitations on the implementation of this library for pto-isa developers.【免费下载链接】pto-isaParallel Tile Operation (PTO) is a virtual instruction set architecture designed by Ascend CANN, focusing on tile-level operations. This repository offers high-performance, cross-platform tile operations across Ascend platforms.项目地址: https://gitcode.com/cann/pto-isaNot following the rules can result in any of the following:Cant compile (including source-code level compile errors and crash in compiler)Functionally incorrect (e.g., precision issues)Bad performance1 - Remember thatpto::(Conv)Tile::data()returns vector type instead of pointer type in auto modeThe return type of.data()member function isTileDType, which is defined differently in manual vs auto mode. In manual mode this is simply a pointer, while in auto mode its a vector type. See the details ininclude/pto/common/memory.hpp.You should always keep this in mind to avoid using the returned value of.data()function directly as a pointer type outside tile functions.2 - Avoid default initializer for a struct/class memberIts a very common practice to default-initialize data members in a struct or class in C, for instance:struct ConvTile { public: ... int shape[ConvTileDetail::MAX_CONVTILE_DIM] {1}; };This turns out to cause problems for the SROA pass in the compiler (SROA cant eliminate theAllocaInstof the struct plus all the load and store instructions associated with it). At least in auto mode, please DONT default initialize the members:#ifdef __PTO_AUTO__ // In auto mode, do not have default initialization in the class definition itself for its members int shape[ConvTileDetail::MAX_CONVTILE_DIM]; #else int shape[ConvTileDetail::MAX_CONVTILE_DIM] {1}; #endifEven though we are programming in C, we encourage to use POD (Plain Old Data) Aggregate programming to describe structs and classes that is compatible with the C-programming language.3 - Explicit synchronization is still needed inside tile functions and their calleesTL;DR:Useset_flag,wait_flagorpipe_barrierexplicitly in tile functions and all of their callees.UsePtoSetWaitFlagorTSYNCanywhere else.Reason: The auto-sync will NOT traverse inside tile functions; as a matter of fact, the whole auto mode compiler works on the tile function level, meaning that everything inside tile function is a complete black box to auto-mode.For this reason, if any synchronization is needed inside tile function, the library developers should still add synchronizations manually. Thats why usingPtoSetWaitFlagandTSYNCwont work in auto mode because its no-op. Most of the cases this interface is used by kernel developers.4 - Avoid usingTASSIGNfor implementationCurrently implementations of some pto instructions directly useTASSIGN_IMPL. This may be a problem for auto mode because its no-op.If you useTASSIGNjust to alias 2 tiles, you should useTRESHAPEorTSUBVIEWto achieve the same goal depending on your needs. Anything else wont work for auto mode.For instance, if you callTASSIGNto allocate memory based on some kind of algorithm, this will never work for auto-mode because the compiler cant possibly recognize the specific algorithm logic and do the same allocation as you want to do in manual mode.After all, the whole memory allocation in auto mode is based on each individual tiles liveness analysis, without knowing any other context. This is why the current implementation ofTPUSHandTPOPwont work for auto mode.5 - Some general rules for*_IMPLfunctionsSome consistency must be ensured for*_IMPLand tile function interface:The function signature must havePTO_INTERNALmacroIts implementation should directly call tile functions inside, dont call any non-tile functions unless theyre inlined.Always call.data()function to pass into tile functions, or return-by-reference for all return values of.data(). For example:TExp(dstTile.data(), srcTile.data()); // correct auto dst dstTile.data(); // wrong: return by value auto src srcTile.data(); // correct: return by reference TExp(dst, src);6 - Some general rules for tile functionsEnsure to usetypename ...::TileDTypeinstead oftypename ...::DType *for tile types on tile function parametersEnsure thesetypename ...::TileDTypeparameters are pass-by-value, not by pointer or referenceEnsure__in__or__out__attributes are properly attached to thesetypename ...::TileDTypeparametersAlways call__cce_get_tile_ptron thesetypename ...::TileDTypearguments to get a tiles underlying buffer pointerThe return type should always bevoid. Otherwise the compilers assumption about TF interface is broken and its an undefined behavior. Please make all return values as pass-by-value arguments, even just for a single scalar.7 - Avoid having runtime control flow before tile functionsHaving runtime control flows imposes great challenges for auto-sync to work properly. We encourage developers to either try to remove these runtime conditions if theyre not necessary or move them inside tile functions if possible.Some examples includeTROWEXPANDDIV_IMPLandTMULS_IMPL.【免费下载链接】pto-isaParallel Tile Operation (PTO) is a virtual instruction set architecture designed by Ascend CANN, focusing on tile-level operations. This repository offers high-performance, cross-platform tile operations across Ascend platforms.项目地址: https://gitcode.com/cann/pto-isa创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考