User-Level Device Drivers: Achieved Performance

Leslie, Ben; Chubb, Peter; FitzRoy-Dale, Nicholas; Götz, Stefan; Gray, Charles T.; Macpherson, Luke; Potts, Daniel; Shen, Yang; Elphinstone, Kevin; Heiser, Gernot

doi:10.1007/s11390-005-0654-4

Cited by 93 publications

(56 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Type safety Decaf Drivers [31] net, sound, USB controller, mouse User-level drivers [21] net, disk (ata) 2 teractions [38]. However, network cards that do TCP-offload may have significant protocol state that is only available in the device, and cannot be captured by monitoring the kernel/driver interface.…”

Section: Static Analysis Toolsmentioning

confidence: 99%

Understanding modern device drivers

Kadav

Swift

2012

Proceedings of the Seventeenth International Conference on Architectural Support for Programming Languages and Operating System

View full text Add to dashboard Cite

Device drivers are the single largest contributor to operating-system kernel code with over 5 million lines of code in the Linux kernel, and cause significant complexity, bugs and development costs. Recent years have seen a flurry of research aimed at improving the reliability and simplifying the development of drivers. However, little is known about what constitutes this huge body of code beyond the small set of drivers used for research.In this paper, we study the source code of Linux drivers to understand what drivers actually do, how current research applies to them and what opportunities exist for future research. We determine whether assumptions made by driver research, such as that all drivers belong to a class, are indeed true. We also analyze driver code and abstractions to determine whether drivers can benefit from code re-organization or hardware trends. We develop a set of staticanalysis tools to analyze driver code across various axes. Broadly, our study looks at three aspects of driver code (i) what are the characteristics of driver code functionality and how applicable is driver research to all drivers, (ii) how do drivers interact with the kernel, devices, and buses, and (iii) are there similarities that can be abstracted into libraries to reduce driver size and complexity?We find that many assumptions made by driver research do not apply to all drivers. At least 44% of drivers have code that is not captured by a class definition, 28% of drivers support more than one device per driver, and 15% of drivers do significant computation over data. From the driver interactions study, we find that the USB bus offers an efficient bus interface with significant standardized code and coarse-grained access, ideal for executing drivers in isolation. We also find that drivers for different buses and classes have widely varying levels of device interaction, which indicates that the cost of isolation will vary by class. Finally, from our driver similarity study, we find 8% of all driver code is substantially similar to code elsewhere and may be removed with new abstractions or libraries.

show abstract

Section: Static Analysis Toolsmentioning

confidence: 99%

Understanding modern device drivers

Kadav

Swift

2012

Proceedings of the Seventeenth International Conference on Architectural Support for Programming Languages and Operating System

View full text Add to dashboard Cite

show abstract

“…A recent project ran Linux device drivers in user mode with small changes to the Linux kernel [12]. This work shows that drivers can be isolated in user-mode processes without significant performance degradation.…”

Section: Patching Commodity Systemsmentioning

confidence: 99%

Construction of a Highly Dependable Operating System

Herder

Bos

Gras

et al. 2006

2006 Sixth European Dependable Computing Conference

View full text Add to dashboard Cite

This paper discusses how we designed and implemented the system, which problems we encountered, and how we solved these problems. We also discuss the performance effects of our changes and evaluate how our multiserver design improves operating system dependability over monolithic designs.'Perfection is not achieved when there is nothing left to add, but when there is nothing left to take away.' -Antoine de Saint-Exupéry [2]

show abstract

“…The replacement is done by altering the work of the dynamic loader. In Linux, it is usually done by modifying the LD_PRELOAD variable, as illustrated in Listing 1 [14].…”

Section: Overriding Dynamically Loaded Libraries -Libtrashmentioning

confidence: 99%

“…the open syscall, the library has to define a function with the same exact name as the stubbed function. One example of LD_PRELOAD usage is libtrash 14 . It is a shared library that can be preloaded, and that transparently implements the trash functionality for Linux.…”

Section: Overriding Dynamically Loaded Libraries -Libtrashmentioning

confidence: 99%

Portable Userspace Virtual Filesystem Switch

Faber

Boryczko

2013

csci

View full text Add to dashboard Cite

Multiple different filesystems -including disk-based, network, distributed, abstract -are an integral part of every operating system. They are usually written as kernel modules and abstracted to the user via a virtual filesystem switch.In this paper, we analyze the feasibility of reimplementing the virtual filesystem switch as a userspace daemon and applicability of this approach in real-life usage. Such reimplementation will require a way to virtualize processes behavior related to filesystem operations. The problem is non-trivial, as we assume limited capabilities of the VFS switch implemented in userspace. We present a layered architecture comprising of a monitoring process, the VFS abstraction and real filesystem implementations. All working in userspace. Then, we evaluate this solution in four areas: portability, feasibility, usability, and performance. Our results demonstrate possible gains in the use of a userspacebased approach with monolithic kernels, but also underline problems that are encountered in this approach.

show abstract

User-Level Device Drivers: Achieved Performance

Cited by 93 publications

References 19 publications

Understanding modern device drivers

Understanding modern device drivers

Construction of a Highly Dependable Operating System

Portable Userspace Virtual Filesystem Switch

Contact Info

Product

Resources

About