Sven-Olof Nyström

A soft-typing system for Erlang

This paper presents a soft-typing system for the programming language Erlang. The system is based on two concepts; a (forward) data flow analysis that determines upper approximations of the possible values of expressions and other constructs, and a specification language that allows the programmer to specify the interface of a module. We examine the programming language Erlang and point to various aspects of the language that make it hard to type. We present experimental result of applying the soft-typing system to some previously written programs.

Postoperative ventilatory and circulatory effects of extended rewarming during cardiopulmonary bypass

Postoperative effects of extended rewarming (ECR) after hypothermic cardiopulmonary bypass (CPB) were studied. All (n = 28) patients were rewarmed to a nasopharyngeal temperature exceeding 38° C before terminating CPB. In 12 patients (control group) the rectal temperature (Tre) was 33.8 ± 1.7° C (mean ± sd) at termination of CPB. In sixteen patients (ECR group) rewarming during CPB was continued to a Tre of 36.8 ±0.5°C. Postoperative body temperatures, heat content, shivering, oxygen uptake, CO2 production and haemodynamic variables were measured. ECR reduced the heat gain required to complete core rewarming to 665 ± 260 kJ, compared with 1037 ± 374 kJ in the control group (p < 0.01). The incidence of shivering was reduced (p < 0.05) as well as shivering intensity and duration. In seven non-shivering ECR group patients this coincided with significantly reduced metabolic and ventilatory demands but these improvements were not valid for the group as a whole. The required ventilation temporarily during postoperative rewarming in both groups increased to 250 per cent of the basal need. Extending CPB rewarming (to at least 36°C Tre) was inefficient when used as the sole measure to reduce the untoward effects of residual hypothermia during recovery after cardiac surgery with hypothermic CPB.RésuméEtude des effets postopératoire du réchauffement prolongé (ECR) après une opération faite avec circulation extracorporelle (CEC) et hypothermie. On a réchauffé tous les malades (n = 28) jusqu’à ce que la température du nasopharynx ait dépassé 38° C avant ta fin de la période en CEC. Douze malades (groupe de contrôle) ont eu une température rectale (Tre) de 33.8 ± 1.7° C (moyenne ± sd) à la fin de la période en CEC. Durant le CEC on a réchauffé seize malades (groupe dénommé ECR) jusqu’à ce que leur Tre soit montée à 36.8 ± 0.5° C. Après ľopération on a mesuré les températures en diverses places du corps, la quantité du chaluer, ľhyperactivité musculaire, le degré ďoxygène, la production de CO2 et les différentes phases de ľhaemodynamique. Le procédé par ECR a réduit le besoin de chaleur nécessaire pour le réchauffement central à 665 ± 260 kJ. A comparer aux 1037 ± 374 kJ dans le groupe de contrôle (p < 0.01). La manifestation de ľhyperactivité musculaire a été réduite (p < 0.05) aussi bien que son intensité et sa durée. Chez sept malades du groupe ECR ľabsence ďhyperactivité musculaire a coincïdé avec une réduction significative au point de vue du métabolisme et de la ventilation mais ce genre ďamélioration n’est pas valable pour le groupe pris dans son ensemble. Après ľopération, le débit ventilatoire dans la période de réchauffement a augmenté dans les deux groupes de 250 pour cent au-dessus du niveau de base. Prolonger le réchauffement en CEC jusqu’à atteindre au moins 36° C rectal n’apas été une méthode efficace si on ľemploie seule pour réduire après ľopération les effets indésirables de ľhypothermie pratiquée en CEC.

Retargetable graph-coloring register allocation for irregular architectures

Global register allocation is one of the most important optimizations in a compiler. Since the early 80’s, register allocation by graph coloring has been the dominant approach. The traditional formulation of graph-coloring register allocation implicitly assumes a single bank of non-overlapping general-purpose registers and does not handle irregular architectural features like overlapping register pairs, special purpose registers, and multiple register banks. We present a generalization of graph-coloring register allocation that can handle all such irregularities. The algorithm is parameterized on a formal target description, allowing fully automatic retargeting. We report on experiments conducted with a prototype implementation in a framework based on a commercial compiler.

Profile-guided optimization across process boundaries

We describe a profile-driven compiler optimization technique for inter-process optimization, which dynamically inlines the effects of sending messages. Profiling is used to find optimization opportunities, and to dynamically trigger recompilation and optimization at run-time. We apply the optimization technique on the concurrent programming language ERLANG, letting recompilation take place in a separate ERLANG process, and taking advantage of the facilities provided by ERLANG to dynamically replace code at run-time. We have implemented a prototype inter-process profiler and optimizer, that can handle small programs. Measurements on synthetic benchmarks show encouraging speedups of up to 1.8 times.

An evaluation of the perioperative efficacy of selective β1-blockade in coronary surgery: Studies with a Late Preoperative Dose of Metoprolol

The plasma concentration of metoprolol was measured hourly following an oral dose on two consecutive days--the day before, and the day of, coronary surgery. No significant difference was found between the two sampling days, though there was a tendency to lower concentrations during and after extracorporeal circulation. After a median dose of 50 mg the peak concentration (reached on average after 1.5 h) was 545 +/- 70 nmol/l on the first day and 388 +/- 57 on the day of surgery. The respective elimination rates from plasma, expressed as half-life, were 3.4 +/- 0.21 and 3.5 +/- 0.19 hours (NS). On the day of surgery the heart rate rose during the second half of the observation period, peaking above 130 beats/min. Heart rate was inversely correlated to metoprolol concentration in plasma with coefficient -0.68 before induction of anesthesia and -0.77 two hours after termination of extracorporeal circulation. The perioperative efficacy of beta-blockade following a late preoperative oral dose of the agent thus appeared to be reduced and inadequate.

There is no fully abstract fixpoint semantics for non-deterministic languages with infinite computations

Abstract It is well known that for many non-deterministic programming languages there is no continuous fully abstract fixpoint semantics. This is usually attributed to “problems with continuity”, that is, the assumption that the semantic functions should be continuous supposedly plays a role in the difficulties of giving a fully abstract fixpoint semantics. We show that for a large class of non-deterministic programming languages there is no fully abstract least fixpoint semantics even if one considers arbitrary functions (not necessarily continuous) over arbitrary partial orders (not necessarily complete).

Optimizing Code Size through Procedural Abstraction

Memory size is an important economic factor in the development of embedded systems. It is therefore desirable to find compiler optimization techniques that reduce the size of the generated code. One such technique is procedural abstraction, where repeated occurrences of equivalent code fragments are factored out into new subroutines. Previously, procedural abstraction has been applied at the machine code level in optimizing linkers or binary rewriting tools. We investigate the prospects of applying procedural abstraction at the intermediate code level of a whole-program optimizing C compiler. Placing the optimization before register allocation makes it target indepentdent, and will allow us to take full advantage of powerful code selection, register allocation and code scheduling techniques.

A fully abstract semantics for concurrent constraint programming

A compositional and fully abstract semantics for concurrent constraint programming is developed. It is the first fully abstract semantics which takes into account both non-determinism, infinite computations, and fairness. We present a simple concurrent constraint programming language, whose semantics is given by a set of reduction rules augmented with fairness requirements. In the fully abstract semantics we consider two aspects of a trace, viz. the function computed by the trace (the functionality) and the set of input and output data (the limit of the trace). We then derive the fully abstract semantics from the set of traces using a closure operation. We give two proofs of full abstraction; the first relies on the use of a syntactically infinite context. The second proof requires only a finite context, but assumes as input a representation of the function to be computed by the context. Finally, we examine the algebraic properties of the programming language with respect to the fully abstract semantics. It turns out that the non-deterministic selection operation can be defined using operations derived from parallel composition and the usual set-theoretic operations on sets of traces.