Sampling optimized code for type feedback

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F20%3A00347259" target="_blank" >RIV/68407700:21240/20:00347259 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1145/3426422.3426984" target="_blank" >https://doi.org/10.1145/3426422.3426984</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3426422.3426984" target="_blank" >10.1145/3426422.3426984</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Sampling optimized code for type feedback

Popis výsledku v původním jazyce

To efficiently execute dynamically typed languages, many language implementations have adopted a two-tier architecture. The first tier aims for low-latency startup times and collects dynamic profiles, such as the dynamic types of variables. The second tier provides high-throughput using an optimizing compiler that specializes code to the recorded type information. If the program behavior changes to the point that not previously seen types occur in specialized code, that specialized code becomes invalid, it is deoptimized, and control is transferred back to the first tier execution engine which will start specializing anew. However, if the program behavior becomes more specific, for instance, if a polymorphic variable becomes monomorphic, nothing changes. Once the program is running optimized code, there are no means to notice that an opportunity for optimization has been missed. We propose to employ a sampling-based profiler to monitor native code without any instrumentation. The absence of instrumentation means that when the profiler is not active, no overhead is incurred. We present an implementation is in the context of the A just-in-time, optimizing compiler for the R language. Based on the sampled profiles, we are able to detect when the native code produced by A is specialized for stale type feedback and recompile it to more type-specific code. We show that sampling adds an overhead of less than 3 2020 ACM.

Název v anglickém jazyce

Sampling optimized code for type feedback

Popis výsledku anglicky

To efficiently execute dynamically typed languages, many language implementations have adopted a two-tier architecture. The first tier aims for low-latency startup times and collects dynamic profiles, such as the dynamic types of variables. The second tier provides high-throughput using an optimizing compiler that specializes code to the recorded type information. If the program behavior changes to the point that not previously seen types occur in specialized code, that specialized code becomes invalid, it is deoptimized, and control is transferred back to the first tier execution engine which will start specializing anew. However, if the program behavior becomes more specific, for instance, if a polymorphic variable becomes monomorphic, nothing changes. Once the program is running optimized code, there are no means to notice that an opportunity for optimization has been missed. We propose to employ a sampling-based profiler to monitor native code without any instrumentation. The absence of instrumentation means that when the profiler is not active, no overhead is incurred. We present an implementation is in the context of the A just-in-time, optimizing compiler for the R language. Based on the sampled profiles, we are able to detect when the native code produced by A is specialized for stale type feedback and recompile it to more type-specific code. We show that sampling adds an overhead of less than 3 2020 ACM.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/EF15_003%2F0000421" target="_blank" >EF15_003/0000421: Big Code: Škálovatelná analýza rozsáhlých bází programů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

DSL_Proceedings of the 16th ACM SIGPLAN International Symposium on Dynamic Languages

ISBN

978-1-4503-8175-8

ISSN

—

e-ISSN

—

Počet stran výsledku

13

Strana od-do

99-111

Název nakladatele

ACM

Místo vydání

New York

Místo konání akce

virtual

Datum konání akce

15. 11. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—