Skip to main content

‘Tis the season

… to give an end-of-year update of what is happening with the PM language. It is a long time now since I wrote a PM blog. However, behind the scenes there has been a lot of work done on the guts of the compiler with new inference and optimisation passes added and others taken apart, polished, and put back together again. New language features have also been implemented, including methods, improved object lifetime management and, most importantly, a radically simplified approach to coding stencils.

The handling of stencils was one aspect of PM-0.4 with which I was least happy. Since these are a key requirement for nearly all numerical modelling, I have long been looking for a way to make their coding as straightforward as possible. However, this has been one of the greatest challenges in terms of implementation, since the amount of code restructuring need to make a stencil operate efficiently (separating out halo computation, overlapping computation and halo exchange, tiling and the interaction of stencils with sequential loops, merging stencils, etc., etc.) significantly exceeds that of any other language feature and was one of the primary motivators for adding compiler passes. The idea is to enable a simple @ operator to access neighbouring values in the model grid, allowing you to write something like: 

 new_cell = (cell@[-1,0]+cell@[1,0]+cell@[0,-1]+cell@[0,1])/4.0

and then to let the compiler do all of the additional work needed to create optimised distributed stencil code. For this simple example, this is not too difficult. For realistic modelling code, however, this becomes much more complex - stencils will typically be abstracted into their own procedures and then applied to multiple variables at disparate points in the code that make the most mathematical, as opposed to computational, sense. Optimal coding in this context involves inter-procedural inference of halo extents and access patterns and extensive high-level code restructuring, including inlining, outlining and a lot of code motion. Getting all of this to work satisfactorally has taken the best part of a year, but has been a necessary step in making the language as effective as possible in its chosen problem domain.

At the time of writing this, I am planning to release PM-0.5 in the first quarter of 2026, starting with an MPI only version, but then bringing in support to generate hybrid MPI/OpenMP and MPI/OpenAcc target code once these have been tested. I will keep the roadmap on GitHub up to date.

So seasons greetings and hope to have something very interesting for you in the new year!

Comments

Post a Comment

Popular posts from this blog

The PM Type System

In common with other aspects of PM design, the type system is designed to facilitate the use of flexible programming constructs, such as polymorphism and dynamic dispatch, while emphasising the generation of fast, static, code.  In common with many other modern languages, PM also attempts to combine the safety of static typing with the expressivity of dynamically typed languages such as Python . The type of any PM expression may usually be determined by a static examination of the code. Variables take their types from an initialising expression using the ‘:=’ declaration syntax popularised by Go . a:= 100 ! Declare ‘a’ as an integer (int) variable Variables do not change type during the course of their lifetime – polymorphic programming requires the use of special values, described later. Composite values such as structures are generated using specialised expressions rather than through the invocation of type-specific creator functions: b:= struct var_descriptor...
Post a Comment
Read more

Communicating Operators - The Heart of PM

  Communicating operators lie at the heart of the PM parallelisation model. They are designed to provide a compromise between the direct access to global data structures (particularly arrays) offered by approaches such as Partitioned Global Address Space  and the straightforward synchronisation provided by Communicating Sequential Processes . I n common with most data-parallel languages, PM contains a parallel version of the for statement which runs all invocations of its enclosed statement list concurrently: for element1 in array1, element2 in array2 do element2=process_element(element1) endfor Most real models will require some interaction between adjacent array elements. In PM this is achieved by using either a local or a global communicating operator. The global operator @v returns an array whose elements comprise the values of loop-local variable v in each invocation of the enclosing for statement. The neighbourhood operator v@{nbd ...
Post a Comment
Read more