Despite the progress made in developing more advanced compilers for embedded systems, programming of embedded high-performance computing systems based on Digital Signal Processors (DSPs) is still a highly skilled manual task. This is true for single-processor systems, and even more for embedded systems based on multiple DSPs. Compilers often fail to optimise existing DSP codes written in C due to the employed programming style. Parallelisation is hampered by the complex multiple address space memory architecture, which can be found in most commercial multi-DSP configurations. This thesis develops an integrated optimisation and parallelisation strategy that can deal with low-level C codes and produces optimised parallel code for a homogeneous multi-DSP architecture with distributed physical memory and multiple logical address spaces. In a first step, low-level programming idioms are identified and recovered. This enables the application of high-level code and data transformations well-known in the field of scientific computing. Iterative feedback-driven search for “good” transformation sequences is being investigated. A novel approach to parallelisation based on a unified data and loop transformation framework is presented and evaluated. Performance optimisation is achieved through exploitation of data locality on the one hand, and utilisation of DSP-specific architectural features such as Direct Memory Access (DMA) transfers on the other hand. The proposed methodology is evaluated against two benchmark suites (DSPstone & UTDSP) and four different high-performance DSPs, one of which is part of a commercial four processor multi-DSP board also used for evaluation. Experiments confirm the effectiveness of the program recovery techniques as enablers of high-level transformations and automatic parallelisation. Source-to-source transformations of DSP codes yield an average speedup of 2.21 across four different DSP architectures. The parallelisation scheme is – in conjunction with a set of locality optimisations – able to produce linear and even super-linear speedups on a number of relevant DSP kernels and applications.

Το αντικείμενο που πραγματεύεται η παρούσα διδακτορική διατριβή σχετίζεται με την ανάπτυξη βελτιστοποιητικών τεχνικών μεταγλώττισης για επαναπροσδιοριζόμενα ολοκληρωμένα συστήματα γενικού και ειδικού σκοπού. Στόχος είναι η βελτιστοποίηση της εκτέλεσης των εφαρμογών ως προς την ταχύτητα, την επιφάνεια ολοκλήρωσης και την κατανάλωση ισχύος. Αυτό επιτυγχάνεται με την εισαγωγή πρωτότυπων τεχνικών μεταγλώττισης αλλά και από την ανεύρεση βέλτιστων αρχιτεκτονικών. Η αυτοματοποίηση των μεθοδολογιών επιτυγχάνεται με την ανάπτυξη εργαλείων βελτιστοποίησης που υλοποιούν την μεθοδολογία μεταγλώττισης. Τα πειράματα έδειξαν γρήγορο προσδιορισμό βέλτιστων λύσεων και σημαντικές βελτιώσεις στην ταχύτητα, επιφάνεια ολοκλήρωσης και κατανάλωση ισχύος για μια σειρά από εφαρμογές ψηφιακής επεξεργασίας σήματος. / The research material that is presented in this PhD Phesis is related with developement of compilation techniques for reconfigurable systems and application specific integrated circuits. The objective is the optimization of the execution of the applications in terms of speed area and power consumption in these architectures. This is achieved by developing original compiling techniques and efficient architecture instances. Moreover, one of the fundamental objectives of this thesis is the automation of these techniques for fast solution determination. Experiments showed that applications are executed faster while keeping the area and power overhead low. The experiments are based on a set of Digital Signal Processing applications.