Facing the Linux 8000 Feature Nightmare
نویسندگان
چکیده
System software, especially operating systems, tends to be highly configurable. The relatively small eCos operating system for embedded applications [1] already offers more than 750 features; current versions of Linux provide even more than 8000 configuration options – which technically are implemented and enforced in the code by means of preprocessor macros. To assist the user on his path through this variability, both systems additionally employ a dedicated variability model, together with some interactive configuration tool (such as qconf for Linux), which not only organizes, lists, and explains these options, but also enforces a myriad of inter-feature constraints and sanity checks to guarantee the outcome of a sound configuration. These two views onto the same system (variability model vs. variability implementation) are a nightmare with respect to maintenance and evolution. Without sophisticated tool support, they quickly lead to “dead” code and “zombie” features because of inconsistencies. Our initial analysis of just the referential integrity between both worlds in Linux has uncovered around 400 bugs – and we thereby have just touched the tip of the iceberg!
منابع مشابه
APIs and End Users Cluster and Grid Computing 2006
There exist quite a few different systems for low level tasks in the Grid world – a number of environments for jobs scheduling, file transfers, resource discovering is overwhelming. It complicates applications and makes porting a real nightmare. GridRPC (see 3.2.1) regards [11] programming directly on top of Globus I/O as performing parallel programming using only the Linux API on a cluster. An...
متن کاملOptimal Feature Selection for Data Classification and Clustering: Techniques and Guidelines
In this paper, principles and existing feature selection methods for classifying and clustering data be introduced. To that end, categorizing frameworks for finding selected subsets, namely, search-based and non-search based procedures as well as evaluation criteria and data mining tasks are discussed. In the following, a platform is developed as an intermediate step toward developing an intell...
متن کاملOptimal Feature Selection for Data Classification and Clustering: Techniques and Guidelines
In this paper, principles and existing feature selection methods for classifying and clustering data be introduced. To that end, categorizing frameworks for finding selected subsets, namely, search-based and non-search based procedures as well as evaluation criteria and data mining tasks are discussed. In the following, a platform is developed as an intermediate step toward developing an intell...
متن کاملروابط چندگانه بین راهبردهای کنترل فکر و علائم اختلالات شخصیت با علائم رؤیاهای ترسناک
Objectives Nightmare is one of the sleep problems among adolescents. This research aimed to examine multiple relations among thought control strategies and symptoms of personality disorders among female students in Shahreza City, Iran in 2015-2016. Methods In this correlational study, the statistical population comprised all female students in pre-university schools and the sample was 300 gir...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010