I recently published a long post on the Intel Community Blog, talking about how my colleague Evgeny solved a nicely complicated bug using Simics-on-Simics. The bug involved UEFI, an operating system, SMM, SMI, and virtualization. Just another day in the office (or more like a year, given how long it took to get this one resolved).
The SystemC Evolution Fika on April 7 had threading/parallelism as its theme. There were four speakers who presented various angles on how to parallelize SystemC models. The presentations and following discussion provided a variety of perspectives on threading as it can be applied in virtual platforms and other computer architecture simulations. It was pretty clear that the presenters and audience had quite different ideas about just what the target domain looks like and the best way to introduce parallelism to SystemC. Here is my take on what was said.Continue reading “SystemC Evolution Fika: Parallel SystemC”
This blog post was originally posted at Intel back in 2018, but it has since been retired from the Intel blog system. As it is of general interest (in my opinion), here is a reposting (with a few small updates here and there).
Temporal decoupling is a key technology in virtual platforms, and can speed up the execution of a system by several orders of magnitude. In my own experiments, I have seen it provide a speedup of more than 1000x. Here, I will dig a little deeper into temporal decoupling and its semantic effects.Continue reading “Some Notes on Temporal Decoupling (Reposted)”
The July 2020 edition of the Communications of the ACM (CACM) had a front-page theme of “Domains-Specific Hardware Accelerators”, or DSAs. It contained two articles about the subject, one about an academic genomics accelerator, and one about the Google TPU. Hardware accelerators dedicated to particular types of computation are basically everywhere today, and an accepted part of the evolution of computers. The CACM articles have some good tidbits and points about how accelerators are designed and used today. At the same time, I also found a youtube talk about the first hardware accelerator, the IBM Stretch HARVEST, showing both contrasts with today as well as a remarkable continuity in concept.Continue reading “CACM on DSAs”
A new short blog post on my Intel Developer Zone blog talks about the improved threading simulation core we have added in Simics version 6… and about how a colleague of mine climbed to the top of the highest mountain in Europe and showed a flag with our new Simics icon! Read the story at https://software.intel.com/en-us/blogs/2019/09/10/simics-6-at-the-mountain-top.
There are some things in computing that seem “obviously” true and that “clearly” make it “impossible” to do some things. One example of this is the idea that you cannot go backwards in time from the current state of a program or computer system and recover previous state by just reversing the semantics of the instructions in the program. In particular, that you cannot take a core dump from a failed system and reverse-execute back from it – how could you? In order to do reverse debugging and reverse execution, you “have to” record the state at the first point in time that you want to be able to go back to, and then record all changes to the state. Turns out I was wrong, as shown by a recent Usenix OSDI paper.Continue reading “Microsoft REPT: You CAN Reverse from a Core Dump!”
I had many interesting conversations at the HiPEAC 2017 conference in Stockholm back in January 2017. One topic that came up several times was the GEM5 research simulator, and some cool tricks implemented in it in order to speed up the execution of computer architecture experiments. Later, I located some research papers explaining the “full speed ahead” technology in more detail. The mix of fast simulation using virtualization and clever tricks with cache warming is worth a blog post.
The SiCS Multicore Day took place last week, for the tenth year in a row! It is still a very good event to learn about multicore and computer architecture, and meet with a broad selection of industry and academic people interested in multicore in various ways. While multicore is not bright shiny new thing it once was, it is still an exciting area of research – even if much of the innovation is moving away from the traditional field of making a bunch of processor cores work together, towards system-level optimizations. For the past few years, SiCS has had to good taste to publish all the lectures online, so you can go to their Youtube playlist and see all the talks for free, right now!
Once upon a time, when multicore processors were novelties, multicore was motivated by the simple fact that it was impossible to keep raising the clock frequency of processors. More “clocks” simply would result in an overheated mess. Instead, by adding more cores, much more performance could be obtained without having to go to extreme frequencies and power budgets. The first multicore processors pretty much kept clock frequencies of the single-core processors preceding them, and that has remained the mainstream fact until today. Desktop and laptop processors tend to stay at 4 cores or less. But when you go beyond 4 cores, clock frequencies tend to start to go down in order to keep power consumption per package under control. A nice example of this can be found in Intel’s Xeon lineup.
Continue reading “Clocks or Cores? Choose One”
I love bug and debug stories in general. Bugs are a fun and interesting part of software engineering, programming, and systems development. Stories that involve running Simics on Simics to find bugs are a particular category that is fascinating, as it shows how to apply serious software technology to solve problems related to said serious software technology. On the Intel Software and Services blog, I just posted a story about just that: debugging a Linux kernel bug provoked by Simics, by running Simics on a small network of machines inside of Simics. See https://blogs.intel.com/evangelists/2016/05/30/finding-kernel-1-2-3-bug-running-wind-river-simics-simics/ for the full story.
IEEE Micro published an article called “Architectural Simulators Considered Harmful”, by Nowatski et al, in the November-December 2015 issue. It is a harsh critique of how computer architecture research is performed today, and its uninformed overreliance on architectural simulators. I have to say I mostly agree with what they say. The article follows in a good tradition of articles from the University of Wisconsin-Madison of critiquing how computer architecture research is performed, and I definitely applaud this type of critique.
A new record, replay, and reverse debugger has appeared, and I just had to take a look at what they do and how they do it. “rr” has been developed by the Firefox developers at Mozilla Corporation, initially for the purpose of debugging Firefox itself. Starting at a debugger from the angle of attacking a particular program does let you get things going quickly, but the resulting tool is clearly generally useful, at least for Linux user-land programs on x86. Since I have tried to keep up with the developments in this field, a write-up seems to be called for.
I have read some recent IBM articles about the POWER8 processor and its hardware debug and trace facilities. They are very impressive, and quite interesting to compare to what is usually found in the embedded world. Instead of being designed to help with software debug, it seems the hardware mechanisms in the Power8 are rather focused on silicon bringup and performance analysis and verification in IBM’s own labs. As well as supporting virtual machines and JIT-based systems!
In a blog post at Wind River, I describe how the Wind River Helix Lab Cloud system can be used to communicate hardware design to software developers. The idea is that you upload a virtual platform to the cloud-based system, and then share it to the software developers. In this way, there is no need to install or build a virtual platform locally, and the sender has perfect control over access and updates. It is a realization of the hardware communication principles I presented in an earlier blog post on use cases for Lab Cloud.
But the past part is that the targets I talk about in the blog post and use in the video are available for anyone! Just register on Lab Cloud, and you can try your own threaded software and check how it scales on a simulated 8-core ARM!
On November 3, 2015, I will give a presentation at the Embedded Conference Scandinavia about simulating IoT systems. The conference program can be found at http://www.svenskelektronik.se/ECS/ECS15/Program.html, with my session detailed at http://www.svenskelektronik.se/ECS/ECS15/Program/IoT%20Development.html.
My topic is how to realistically simulate very large IoT networks for software testing and system development. This is a fun field where I have spent significant time recently. Only a couple of weeks ago, I tried my hand simulating a 1000-node network. Which worked! I had 1000 ARM-based nodes running VxWorks running at the same time, inside a single Simics process, and at speeds close to real time! It did use some 55GB of RAM, which I think is a personal record for largest use of system resources from a single process. Still, it only took a dozen processors to do it.
While I was on vacation, Wind River published a blog post I wrote about the new multicore accelerator feature of Simics 5. The post has some details on what we did, and some of the things we learnt about simulation performance.
When mobile phones first appeared, they were powered by very simple cores like the venerable ARM7 and later the ARM9. Low clock frequencies, zero microarchitectural sophistication, sufficient for the job. In recent years, as smartphones have come into their own as the most important computing device for most people, the processor performance of mobile phones have increased tremendously. Today, cutting-edge phones and tablets contain four or eight cores, running at clock frequencies well above 2 gigahertz. The performance race for most of the market (more about that in a moment) was mostly about pushing higher clock frequencies and more cores, even while microarchitecture was left comparatively simple. Mobile meant “fairly simple”, and IPC was nowhere near what you would get with a typical Intel processor for a laptop or desktop.
Today, that seems to be changing, as the Nvidia Denver core and Apple’s Cyclone core both go the route of a few fat cores rather than many thin cores.
At the ISCA 2014 conference (the biggest event in computer architecture research), a group of researchers from Microsoft Research presented a paper on their Catapult system. The full title of the paper is “A Reconfigurable Fabric for Accelerating Large-Scale Datacenter Services“, and it is about using FPGAs to accelerate search engine queries at datacenter scale. It has 23 authors, which is probably the most I have ever seen on an interesting paper. There are many things to be learnt from and discussed about this paper, and here are my thoughts on it.
I have a silly demo program that I have been using for a few years to demonstrate the Simics Analyzer ability to track software programs as they are executing and plot which threads run where and when. This demo involves using that plot window to virtually draw text, in a way akin to how I used to make my old ZX Spectrum “draw” things in the border. But when I brought it up in a new setting it failed to run properly and actually starting hanging on me. Strange, but also quite funny when I realized that I had originally foreseen this very problem and consciously decided not to put in a fix for it… which now came back to bite me in a pretty spectacular way. But at least I did get an interesting bug to write about.
Via the EETimes, I found a very interesting talk by Bristol professor David May, presented at the 4th Annual Bristol Multicore Challenge, in June of 2013. The talk can be found as a Youtube movie here, and the slides are available here. The EETimes focused on the idea to cut down ARM to be really RISC, but I think the more interesting part is Professor May’s observations on multicore computing in general, and the case for and against heterogeneity in (parallel) computers.
Probably thanks to the yearly Mobile World Congress, there have been a slew of recent announcements of mobile application processors recently. Everything is ARM-based, but show quite some variety in the CPU core configurations used. Indeed, I think this variety has something to say on the general state of multicore.
There is a new post at my Wind River blog, telling the story of how some of the Simics developers used Simics itself to debug an intermittent Simics program crash caused by a timing-sensitive race condition.
Running Simics on itself is pretty cool, and shows the power of the simulator and its applicability even to really complex software.
There is a new blog post on my Wind River blog, about the Landslide system from CMU. It is a pretty impressive Master’s Thesis project that used the control that Simics has over interrupts to systematically try different OS kernel thread interleavings to find concurrency bugs. The blog is an interview with Ben Blum, the student who did the work. Ben is now a PhD student, and I bet that he will continue to generate cool stuff in the future.
The 2012 edition of the SiCS Multicore Day was fun, like they have always been in the past. I missed it in 2010 and 2011, but could make it back this year. It was interesting to see that the points where keynote speakers disagreed was similar to previous years, albeit with some new twists. There was also a trend in architecture, moving crypto operations into the core processor ISA, that indicates another angle on the hardware accelerator space.
A few years ago, I built a demo on Simics that used a hacked Freescale MPC8641D target that was forced to scale from 1 to 8 cores. Some interesting experiements could be made using this target, and it was nicely scalable for its time. However, I always wanted to have something just a bit bigger. Say 20 cores, or 100. Just to see what would happen. Finally, I got it.
The Simics QSP target that we quietly launched earlier this Summer is such a scalable target. As discussed in a blog post describing the architecture, it is designed to scale to 128 cores currently. Using this ability, I repeated my old experiments, but trying very large threads counts and target core counts. The results show clearly that the way that I coded my parallel computation program was pretty bad, and I really would like to try to rewrite it using some more modern threading library. All I need is time and a way to cross-compile Wool…
Anyway, the new blog post is here.
When IBM moved their mainframe systems (the S/360 family that is today called System Z) from BiCMOS to mainstream CMOS in 1994, the net result was a severe loss in clock frequency and thus single-processor performance. Still, the move had to be done, since CMOS would scale much better into the future. As a result, IBM introduced additional parallelism to the system in order to maintain performance parity. Parallelism as a patch, essentially.
Once upon a time, all programming was bare metal programming. You coded to the processor core, you took care of memory, and no operating system got in your way. Over time, as computer programmers, users, and designers got more sophisticated and as more clock cycles and memory bytes became available, more and more layers were added between the programmer and the computer. However, I have recently spotted what might seem like a trend away from ever-thicker software stacks, in the interest of performance and, in particular, latency.
I was recently pointed to a 2011 SPLASH presentation by David Ungar, an IBM researcher working on parallel programming for manycore systems. In particular, in a project called Renaissance, run together with the Vrije Universiteit Brussels in Belgium (VUB) and Portland State University in the US. The title of the presentation is “Everything You Know (about Parallel Programming) Is Wrong! A Wild Screed about the Future“, and it has provoked some discussion among people I know about just how wrong is wrong.
I just read a quite interesting article by Christian Pinto et al, “GPGPU-Accelerated Parallel and Fast Simulation of Thousand-core Platforms“, published at the CCGRID 2011 conference. It discusses some work in using a GPGPU to run simulations of massively parallel computers, using the parallelism of the GPU to speed the simulation. Intriguing concept, but the execution is not without its flaws and it is unclear at least from the paper just how well this generalizes, scales, or compares to parallel simulation on a general-purpose multicore machine.