OpenHMPP (HMPP[1] for Hybrid Multicore Parallel Programming) - programming standard for heterogeneous computing. Based on a set of compiler directives, standard is a programming model designed to handle hardware accelerators without the complexity associated with GPU programming. This approach based on directives has been implemented because they enable a loose relationship between an application code and the use of a hardware accelerator (HWA).
The OpenHMPP directive-based programming model offers a syntax to offload computations on hardware accelerators and to optimize data movement to/from the hardware memory.
The model is based on works initialized by CAPS (Compiler and Architecture for Embedded and Superscalar Processors), a common project from INRIA, CNRS, the University of Rennes 1 and the INSA of Rennes.
OpenHMPP concept
OpenHMPP is based on the concept of codelets, functions that can be remotely executed on HWAs.
It does not contain static or volatile variable declarations nor refer to any global variables except if these have been declared by a HMPP directive “resident”
It does not contain any function calls with an invisible body (that cannot be inlined). This includes the use of libraries and system functions such as malloc, printf, ...
Every function call must refer to a static pure function (no function pointers).
It does not return any value (void function in C or a subroutine in Fortran).
The number of arguments should be fixed (i.e. it can not be a variadic function as in stdarg.h in C).
It does not contain callsite directives (i.e. RPC to another codelet) or other HMPP directives.
These properties ensure that a codelet RPC can be remotely executed by a HWA. This RPC and its associated data transfers can be asynchronous.
Codelet RPCs
HMPP provides synchronous and asynchronous RPC. Implementation of asynchronous operation is hardware dependent.
Synchronous versus asynchronous RPC
HMPP Memory Model
HMPP considers two address spaces: the host processor one and the HWA memory.
HMPP memory Model
Directives concept
The OpenHMPP directives may be seen as “meta-information” added in the application source code. They are safe meta-information i.e. they do not change the original code behavior. They address the remote execution (RPC) of a function as well as the transfers of data to/from the HWA memory.
The table below introduces the OpenHMPP directives. OpenHMPP directives address different needs: some of them are dedicated to declarations and others are dedicated to the management of the execution.
Control flow instructions
Directives for data management
Declarations
codelet group
resident map mapbyname
Operational Directives
callsite synchronize region
allocate release advancedload delegatedstore
Concept of set of directives
One of the fundamental points of the HMPP approach is the concept of directives and their associated labels which makes it possible to expose a coherent structure on a whole set of directives disseminated in an application.
There are two kinds of labels:
One associated to a codelet. In general, the directives carrying this kind of labels are limited to the management of only one codelet (called stand-alone codelet in the remainder of the document to distinguish it from the group of codelets).
One associated to a group of codelets. These labels are noted as follow: “<LabelOfGroup>“, where “LabelOfGroup” is a name specified by the user. In general, the directives which have a label of this type relate to the whole group. The concept of group is reserved to a class of problems which requires a specific management of the data throughout the application to obtain performance.
OpenHMPP Directives Syntax
In order to simplify the notations, regular expressions will be used to describe the syntax of the HMPP directives.
The color convention below is used for the description of syntax directives:
Reserved HMPP keywords are in green;
Elements of grammar which can be declined in HMPP keywords are in red;
<grp_label>: is a unique identifier naming a group of codelets. In cases where no groups are defined in the application, this label can simply miss. Legal label name must follow this grammar: [a-zA-Z_][a-zA-Z0-9_]*. Note that the “< >” characters belong to the syntax and are mandatory for this kind of label.
codelet_label: is a unique identifier naming a codelet. Legal label name must follow this grammar: [a-zA-Z_][a-zA-Z0-9_]*
directive: is the name of the directive;
directive_parameters: designates some parameters associated to the directive. These parameters may be of different kinds and specify either some arguments given to the directive either a mode of execution (asynchronous versus synchronous for example);
[&]: is a character used to continue the directive on the next line (same for C and FORTRAN).
Directive parameters
The parameters associated to a directive may be of different types. Below are the directive parameters defined in OpenHMPP:
version = major.minor[.micro]: specifies the version of the HMPP directives to be considered by the preprocessor.
args[arg_items].size={dimsize[,dimsize]*}: specifies the size of a non scalar parameter (an array).
args[arg_items].io=[in|out|inout]: indicates that the specified function arguments are either input, output or both. By default, unqualified arguments are inputs.
cond = "expr": specifies an execution condition as a boolean C or Fortran expression that needs to be true in order to start the execution of the group or codelets.
target=target_name[:target_name]*: specifies which targets to try to use in the given order.
asynchronous: specifies that the codelet execution is not blocking (default is synchronous).
args[<arg_items>].advancedload=true: indicates that the specified parameters are preloaded. Only in or inout parameters can be preloaded.
args[arg_items].noupdate=true: this property specifies that the data is already available on the HWA and so that no transfer is needed. When this property is set, no transfer is done on the considered argument
args[<arg_items>].addr="<expr>": <expr> is an expression that gives the address of the data to upload.
args[<arg_items>].const=true: indicates that the argument is to be uploaded only once.
OpenHMPP directives
Directives for declaring and executing a codelet
A codelet directive declares a computation to be remotely executed on a hardware accelerator. For the codelet directive:
The codelet label is mandatory and must be unique in the application
The group label is not required if no group is defined.
The codelet directive is inserted just before the function declaration.
More than one codelet directive can be added to a function in order to specify different uses or different execution contexts. However, there can be only one codelet directive for a given call site label.
The callsite directive specifies how the use a codelet at a given point in the program.
/* declaration of the codelet */#pragma hmpp simple1 codelet, args[outv].io=inout, target=CUDAstaticvoidmatvec(intsn,intsm,floatinv[sm],floatinm[sn][sm],float*outv){inti,j;for(i=0;i<sm;i++){floattemp=outv[i];for(j=0;j<sn;j++){temp+=inv[j]*inm[i][j];}outv[i]=temp;}intmain(intargc,char**argv){intn;......../* codelet use */#pragma hmpp simple1 callsite, args[outv].size={n}matvec(n,m,myinc,inm,myoutv);........}
In some cases, a specific management of the data throughout the application is required (CPU/GPU data movements optimization, shared variables...).
The group directive allows the declaration of a group of codelets. The parameters defined in this directive are applied to all codelets belonging to the group. The syntax of the directive is:
#pragma hmpp <grp_label> group [, version = <major>.<minor>[.<micro>]?]? [, target =target_name[:target_name]*]]? [, cond = “expr”]?
Data transfers directives to optimize communication overhead
When using a HWA, the main bottleneck is often the data transfers between the HWA and the main processor. To limit the communication overhead, data transfers can be overlapped with successive executions of the same codelet by using the asynchronous property of the HWA.
allocate directive
The allocate directive locks the HWA and allocates the needed amount of memory.
The synchronize directive specifies to wait until the completion of an asynchronous callsite execution. For the synchronize directive, the codelet label is always mandatory and the group label is required if the codelet belongs to a group.
In the following example, the device initialization, memory allocation and upload of the input data are done only once outside the loop and not in each iteration of the loop.
The synchronize directive allows to wait for the asynchronous execution of the codelet to complete before launching another iteration. Finally the delegatedstore directive outside the loop uploads the sgemm result.
Those directives map together all the arguments sharing the given name for all the group.
The types and dimensions of all mapped arguments must be identical.
The map directive maps several arguments on the device.
#pragma hmpp <grp_label> map, args[arg_items]
This directive is quite similar as the map directive except that the arguments to be mapped are directly specified by their name. The mapbyname directive is equivalent to multiple map directives.
The resident directive declares some variables as global within a group. Those variables can then be directly accessed from any codelet belonging to the group. This directive applies to the declaration statement just following it in the source code.
The notation ::var_name with the prefix ::, indicates an application's variable declared as resident.
Acceleration of regions
A region is a merge of the codelet/callsite directives. The goal is to avoid code restructuration to build the codelet. Therefore, all the attributes available for codelet or callsite directives can be used on regions directives.
In computing, inline expansion, or inlining, is a manual or compiler optimization that replaces a function call site with the body of the called function. Inline expansion is similar to macro expansion, but occurs during compilation, without changing the source code, while macro expansion occurs prior to compilation, and results in different text that is then processed by the compiler.
The C preprocessor is the macro preprocessor for several computer programming languages, such as C, Objective-C, C++, and a variety of Fortran languages. The preprocessor provides inclusion of header files, macro expansions, conditional compilation, and line control.
The syntax of the C programming language is the set of rules governing writing of software in C. It is designed to allow for programs that are extremely terse, have a close relationship with the resulting object code, and yet provide relatively high-level data abstraction. C was the first widely successful high-level language for portable operating-system development.
In computer programming, a function object is a construct allowing an object to be invoked or called as if it were an ordinary function, usually with the same syntax. In some languages, particularly C++, function objects are often called functors.
In computer programming, a directive or pragma is a language construct that specifies how a compiler should process its input. Depending on the programming language, directives may or may not be part of the grammar of the language and may vary from compiler to compiler. They can be processed by a preprocessor to specify compiler behavior, or function as a form of in-band parameterization.
The syntax of Java is the set of rules defining how a Java program is written and interpreted.
In computer programming, an entry point is the place in a program where the execution of a program begins, and where the program has access to command line arguments.
In mathematics and in computer programming, a variadic function is a function of indefinite arity, i.e., one which accepts a variable number of arguments. Support for variadic functions differs widely among programming languages.
The computer programming languages C and Pascal have similar times of origin, influences, and purposes. Both were used to design their own compilers early in their lifetimes. The original Pascal definition appeared in 1969 and a first compiler in 1970. The first version of C appeared in 1972.
In computing, IIf is a function in several editions of the Visual Basic programming language and ColdFusion Markup Language (CFML), and on spreadsheets that returns the second or third parameter based on the evaluation of the first parameter. It is an example of a conditional expression, which is similar to a conditional statement.
sizeof is a unary operator in the programming languages C and C++. It generates the storage size of an expression or a data type, measured in the number of char-sized units. Consequently, the construct sizeof (char) is guaranteed to be 1. The actual number of bits of type char is specified by the preprocessor macro CHAR_BIT, defined in the standard include file limits.h. On most modern computing platforms this is eight bits. The result of sizeof has an unsigned integer type that is usually denoted by size_t.
C++11 is a version of the ISO/IEC 14882 standard for the C++ programming language. C++11 replaced the prior version of the C++ standard, called C++03, and was later replaced by C++14. The name follows the tradition of naming language versions by the publication year of the specification, though it was formerly named C++0x because it was expected to be published before 2010.
In computer programming, an anonymous function is a function definition that is not bound to an identifier. Anonymous functions are often arguments being passed to higher-order functions or used for constructing the result of a higher-order function that needs to return a function. If the function is only used once, or a limited number of times, an anonymous function may be syntactically lighter than using a named function. Anonymous functions are ubiquitous in functional programming languages and other languages with first-class functions, where they fulfil the same role for the function type as literals do for other data types.
The ECL programming language and system were an extensible high-level programming language and development environment developed at Harvard University in the 1970s. The name 'ECL' stood for 'Extensible Computer Language' or 'EClectic Language'. Some publications used the name 'ECL' for the system as a whole and EL/1 for the language.
This article describes the syntax of the C# programming language. The features described are compatible with .NET Framework and Mono.
This article compares a large number of programming languages by tabulating their data types, their expression, statement, and declaration syntax, and some common operating-system interfaces.
The syntax and semantics of PHP, a programming language, form a set of rules that define how a PHP program can be written and interpreted.
In computer programming, variadic templates are templates that take a variable number of arguments.
The structure of the Perl programming language encompasses both the syntactical rules of the language and the general ways in which programs are organized. Perl's design philosophy is expressed in the commonly cited motto "there's more than one way to do it". As a multi-paradigm, dynamically typed language, Perl allows a great degree of flexibility in program design. Perl also encourages modularization; this has been attributed to the component-based design structure of its Unix roots, and is responsible for the size of the CPAN archive, a community-maintained repository of more than 100,000 modules.
Nim is a general-purpose, multi-paradigm, statically typed, compiled high-level systems programming language, designed and developed by a team around Andreas Rumpf. Nim is designed to be "efficient, expressive, and elegant", supporting metaprogramming, functional, message passing, procedural, and object-oriented programming styles by providing several features such as compile time code generation, algebraic data types, a foreign function interface (FFI) with C, C++, Objective-C, and JavaScript, and supporting compiling to those same languages as intermediate representations.
This page is based on this Wikipedia article Text is available under the CC BY-SA 4.0 license; additional terms may apply. Images, videos and audio are available under their respective licenses.
