August 15, 2014 – 11:36 am
This writeup concludes our series of in-depth postings about the LPC 2014 microconferences. Thank you for reading so far! Looking forward to an exciting and productive conference in Düsseldorf.
Tracing is heavily used throughout the Linux ecosystem and provides an essential method for extracting information about the underlying code that is running on a system. Although tracing is simple in concept, effective usage and implementation can be quite involved (see this LWN.net article) because of the variety of environments that tracing must handle.
The ultimate goal of tracing is to get as much information as possible with as little overhead as possible. These two requirements are clearly in conflict and different approaches to balance this conflict have been taken by ftrace, perf and LTTng. One of the topics to be discussed at this microconference will be how to better integrate these three differing tracing implementations, for example, using Common Trace Format (CTF) to allow each to read the others’ tracing output. There are also other tools, such as systemtap, which complement tracing, providing more control features, and the kernel may benefit by working more inclusively with them.
Another topic for this microconference will be how tracing could make use of JITted code, given that, surprising though it might seem, the Linux kernel now contains a JIT for the Berkeley Packet Filter (BPF, as explained in this LWN.net article). At the same time, kernel-bloat considerations will generate a discussion of how to reduce the Linux kernel’s current per-tracepoint memory overhead.
This microconference is a follow-on to a Tracing Summit. The Summit will focus more on traditional style presentations, while this microconference will focus on finding and implementing solutions to Linux’s current tracing problems.
(Thanks to Paul McKenney for help with this post)
August 14, 2014 – 8:36 pm
We are pleased to announce that we have reached our designated limit for general registration to LPC 2014. As of now, the general registration for LPC 2014 is closed. This year we have reached our maximum capacity earlier than usual. As previously announced (see blog post), we have deliberately limited the number of registrations in order to allow for meaningful participation in the microconferences and productive interaction among the participants. Thanks to all of you that have registered!
If you still haven’t registered but would like to participate, contact us. We are running a waiting list on a first come first serve basis but with priority given to people who have accepted microconference topics. You could also try to use one of the sponsors tickets if your employer can provide one to you.
We look forward to seeing you in Düsseldorf!
August 12, 2014 – 7:54 pm
The LCP 2014 Planning Committee would like to thank all of you that have registered so far for this year’s conference. We would like to remind everyone that the Linux Plumbers Conference sets a limit on the number of attendees, in the interest of creating an environment conductive to planning, productive interaction and making progress towards the solution of technical problems, with focused meetings and exchanges. Like in past years, we are enforcing a hard limit on the number of registrations for LPC 2014.
We are very rapidly approaching our attendance limit, this year faster than in any past editions of the conference. We expect that the conference general registration will be sold out soon, possibly even within a few days. If you have a vested interest in participating in the discussions, please register now, to guarantee that you will obtain a ticket for the conference. To register please follow the instructions on our Attend Page.
As usual if you have any questions feel free to please contact us.
August 12, 2014 – 3:23 pm
As our series of microconferences explained continues, here is Live Kernel Patching.
There has been a great deal of interest in live kernel patching (see this LWN.net article) over the past few months, with several different approaches proposed, including CRIU+kexec, kGraft, and kpatch, all in addition to ksplice. This microconference will host discussions on required infrastructure (including tracing, checkpoint/restart, kexec, and live patching), along with expositions and comparisons of the various approaches. The purpose, believe it or not, is to work towards a common implementation that everyone can live with. It should be a spirited discussion!
For more details, please see the 2014 LPC wiki.
Please join us at this microconference!
(Thanks to Paul McKenney for help with the post)
August 11, 2014 – 5:05 pm
The LPC 2014 Planning Committee and the Microconference Leaders are pleased to announce that it is now possible to submit discussion topics for individual microconferences. In addition, attendees can also propose ideas for BOF (“Birds of a Feather”) discussion sessions.
Before you submit a topic for discussion within a microconference, make sure you look at the list of accepted microconferences for this year, in the LPC 2014 microconferences page, to determine which microconference is more appropriate for your proposal and who its leader is. Microconference leads have an important role as they are responsible for accepting and scheduling discussion topics within a microconference. Please carefully read the instructions on how to submit and manage microconference discussion topic proposals here: How to Submit Microconferences Discussions Topics.
A BoF is an informal gathering held to discuss a particular topic. Generally, these are used for special interest discussions, topics that aren’t large enough to warrant a full microconference, or to solicit specific input on technical problems. Any LPC attendee may propose a BoF session by following the instructions available on this page: How to Submit a BOF Proposal. BoFs will be scheduled by the LPC Planning Team as the conference draws closer.
If you have questions, please email the planning team.
Let’s take a closer look at the IOMMU and VFIO Microconference.
Increased interest in virtualization and heterogeneous computing has also raised interest in IOMMUs and VFIO, as these can greatly improve the effectiveness and efficiency of compute workloads and virtualization with heavy I/O requirements. This microconference will look at error handling, substituting generic code for heavily replicated architecture-specific code, integrating IOMMUs into the Linux memory management, interactions with device tree, power management, and much more. Please see the LPC 2014 wiki page for more details.
We hope to see you there!
August 6, 2014 – 12:33 pm
Here is another installment in our LPC 2014 MC in-depth series: the Development Tools Microconference.
A very famous 1986 paper entitled “No Silver Bullet. Essence and Accident in Software Engineering” by Fred Brooks (of “Mythical Man Month” fame) states that although there have been modest advances in the field of software engineering, none of the heavily hyped tools or methodologies has really managed to revolutionize software development.
That said, an accumulation of small improvements made over the past decades has made a huge difference in the practice of software development. These improvements are the subject of this microconference (see the LPC 2014 wiki), and include patch generators such as coccinelle, source-code management systems such as git, static analyzers such as sparse (along with gcc and LLVM), emulators such as qemu, and much more. Come to this microconference to see what additional improvements might be in the offing.
Please join us for a great discussion.
(thanks to Paul McKenney for help with this entry)
August 5, 2014 – 12:35 pm
Another Microconference explained: Energy aware-scheduling and CPU power management.
Energy efficiency has received considerable attention, for example, the microconference at last year’s Plumbers (see LPC 2013 schedule, notes and videos) and the workshop at last year’s Kernel Summit (see LWN’s article). However, there is still quite a bit left to be desired in Linux’s energy-aware scheduling and CPU power management, hence this year’s microconference (see the LPC 2014 wiki).
This microconference will look at progress on tools and benchmarks for energy testing and evaluation, integration of existing power-management frameworks with the Linux kernel’s scheduler, and energy-model guided scheduler decisions. It will also look at userspace tools as well as energy efficiency on alternative platforms.
We hope to see you there!
(Thanks to Paul McKenney for help with this entry)
Let’s look at another of our LPC 2104 Microconferences: Android and Mobile.
As has been the case for the past 15 years, 2014 was to be the year of the Linux laptop. But thanks to Android, the past several years really have been the years of the Linux smartphone, as well as the years of an amazing variety of Linux embedded devices, ranging from washing machines to oscilloscopes. Despite Android’s long-standing success, there are quite a few topics of discussion, including use of the LLVM compiler, Android on 64-bit devices, dma-fence and android sync, and of course power management of battery-powered devices, including ARM’s big.LITTLE architecture.
We also have a bonus topic put forward by Greg Kroah-Hartman, motivated by the fact that there is still more than ten thousand lines of code in drivers/staging/android. This naturally raises the question “When is all of this Android stuff going to move out of staging?” Greg didn’t specify whether it would be moving into some non-staging part of mainline or whether it would be moving out of mainline entirely, so it could be an interesting and spirited discussion!
As usual for more details, please see the LPC 2014 Wiki.
Looking forward to see you all in Düsseldorf!
Thanks to Paul McKenney for help with this post.
Welcome to a new installment in our series of microconferences writeups. This time we’ll talk about LLVM.
LLVM and clang (see the LLVM project website) have been increasingly seen as an alternative to the GNU Compiler Collection (gcc), for example, Clang is now distributed as part of the Android Native Development Kit (Android NDK). However, there are some important parts of the Android code base that clang/LLVM cannot build, for example, the Linux kernel (see the LLVMLinux project). Obstacles to use of clang/LLVM to build the kernel include the use of variable arrays in struct (VLAIS), differing __builtin_constant_p() between clang and gcc, handling of init and exit sub-segments, and so on
Other topics include callgraph generation, AArch64 support, x86-specific issues, and LLVM/clang issues with specific open-source packages. For more details please see the MC description on the LPC 2014 wiki
We hope to see you there!
(Thanks to Paul McKenney for help with the writing of this post)