Linux Plumbers Conf 2011

An explanation of RCU appealing to life-experience intuitions.

ARM platform and cross-distribution standardization

The session will detail a design and progress towards getting a unified way to manage common features like snapshots, growth, shrinking for file systems and files.

Byte queue limits are a mechanism to limit NIC HW queues by number of bytes as opposed to number of packets, thus reducing buffer bloat problem caused by these queues.

Checkpoint/restart in the userspace (slides)

Greatly simplifying, securing, improving our traditional filesystem layout without causing major breakage

Codemap is a semantic source code visualizer helping to navigate, visualize and search a large codebase using ideas similar to google maps. (slides)

The new streaming interface proposal provides a framework in sound subsystem to support offloaded decoders and encoders

Container in a file (loopback-like block device)

Discussion of Containers/Cgroups issues

Kernel cgroups are not sufficient for our containment needs. (slides)

lightning fast booting on x86 / x64 platforms

Add dcache accounting/limiting to the kernel

Designing for the disconnect - Enabling the Edge - some lessons in hardware, software and network design, learned in Central America, the hard way.

The verity target provides transparent integrity checking of block devices using a cryptographic digest.

A proposal for an EFI like boot mechanism with a dracut initramfs and the linux kernel using kexec to boot other linux kernels.

Status update and discussion on future directions for the control of DSP features on

A proposal for dynamic firewall management with network zone support.

Efficient utilization by overcommitment.

This presentation will introduce multi-plane HW solution and provide one software solution to make good use of multi-plane and enable translucency.

Container enter functionality

In this presentation, we discuss the set of design and implementation details that play into developing a Linux-based, open-source, large scale, infrastructure level distributed system that implements a private cloud computing platform.

Extending CPU accounting stats

Finding and burying Configuration Defects in Linux with the undertaker (slides)

This talk will present the features Linux is missing to comply with the current AVRCP spec, and a proposal of how to achieve fully compliance.

Building an efficient and maintenable API for allowing userspace to program GPU is one of the most important piece of the graphic stacks (slides)

Using hierarchical cgroups to manage shared resources.

How can we optimise the power management of an entire system? (slides)

A new version of Iogrind for I/O analysis, touching kernel internals this time.

Both Xen and KVM do virtualization. What can we integrate to leverage each other advantages (and reduce the maintainer headache)

Linux Virtualization Stack simulation using libvirt APIs

Remapping legacy interrupts and MSIs onto event channels in Xen HVM guests

Discussion of Linux idle power management vs other OSes (slides)

Presentation of the new LTTng 2.0 kernel and userspace tracer.

An overview of the Linux kernel scalability bottlenecks relating to signals in multithreaded apps.

Proposed improvements to cgroups memory controller

Folproof filesystem freeze API for user space daemons and guest agents. (slides)

OpenGL (and friends) in the Future: A notional view of what may happen in the land of graphics APIs. (slides)

Linker, compiler, runtime linker and the kernel can be improved to start applications significantly faster

A discussion of implementation and performance issues for persistent data storage in KVM-based clouds

An in-container view of /proc fs

The challenges and the benefits of running Qemu in a Linux based stubdom

Remus is a HA system for Xen VMs that preserves complete Guest OS runtime state (including network connections) upon failover.

Scalability problems in/caused by fork (slides)

Advancements in NIC hardware and functionality, as well as things like increasing core counts on systems present new opportunities for taking networking performance to next level.

Embedded Linux routers are everywhere. And many are buggy.

How can we reintroduce the zero-copy behaviour that was removed as part of the process of upstreaming Xen's netback driver (slides)

Session leader for the Storage and Filesystem MC

Tools to aid in the development of Linux code

PM Development in The Linux Kernel Since August 2010 (slides)

Identifying and eliminating the effects of bufferbloat

Scaling Linux, both up and down

Session lead for Virtualization micro conf

Status update on Beating the (buffer) bloat.

systemd was first announced a year ago and has since become an integral part of various Linux distributions, including Fedora 15.

Data exchange during TCP handshake

Improving reliability of Linux device drivers against hardware failures and hardware specification bugs (slides)

This presentation will propose interfaces to expose DCB attributes to MPIO, virtualization, and other applications consistently regardless of underlying device type. (slides)

Can the "trace" utility merge perf and ftrace. (slides)

Tracing

Tracing the Guest from the Host, a discussion (slides)

The Cloud Track focuses on the underlying design methodologies, development tools, and emerging best practices around Linux and Open Source in Public and Private Clouds

The Unified Buffer Management microconference will use a free-form agenda, with time blocked out as shown below

A check-point of the Linaro-driven approach to addressing the memory management issues encountered by the ARM ecosystem. (slides)

Presentation of the Userspace RCU library features. (slides)

VFIO allows exporting PCI devices to userspace drivers. Discussion of using this interface for assigning physical devices to virtual machines and obstacles remaining for integration. (slides)

How does Virtio compare with traditional Xen PV drivers? (slides)

Alternative LXC user-space implementation

An overview of what is involved in making video calls and the issues typically encountered on mobile platforms. (slides)

Yabusame is yet another live migration mechanism for Qemu/KVM, which implements the migration technique known as "postcopy" or "lazy" migration. (slides)