FPGAs Have the Wrong Abstraction for Computing

FPGAs Have the Wrong Abstraction for Computing

Like GPUs, FPGAs need a hardware abstraction that embodies this computational pattern:

What’s missing here is an ISA-like abstraction for the software that FPGAs run. The problem with Verilog for computational FPGAs is that it neither does a good job as a low-level hardware abstraction nor as a high-level programming abstraction. The long-term resolution to this problem might only come when the hardware evolves, as GPUs once did:

If computational FPGAs are accelerators for a particular class of algorithmic patterns, there’s no reason to believe that today’s FPGAs are the ideal implementation of that goal.

Source: www.cs.cornell.edu

Building an LTE Access Point with a Raspberry Pi

Building an LTE Access Point with a Raspberry Pi

With this setup, you can now connect another computer through the wired network interface, statically configure the computer with an IP address, i.e. , and SSH into the Pi using your changed credentials. This makes the initial setup quite easy as we just need to add another network device at the end of and configure it to automatically receive IP-addresses:

For the routing we need to enable IP forwarding within the configuration file by setting:

In addition we need some firewall rules to forward and masquerade traffic from our bridge ( ) to the LTE modem. The weak point of this setup seems to be the modem: the Huawei E3372 HiLink network mode might be the best for an easy setup, but I really would like to compare it with a “traditional” modem setup – e.g., with an assigned IP address and no additional NAT being performed by the modem (through the fake network interface).

Source: snikt.net

Compiler basics (5): LLVM system calls

Compiler basics (5): LLVM system calls

Previously in compiler basics:

Compiler basics (1): lisp to assembly

Compiler basics (2): user-defined functions and variables

Compiler basics (3): LLVM

Compiler basics (4): LLVM conditionals and compiling fibonacci

In this post we’ll extend the ulisp compiler’s LLVM backend to support printing integers to stdout. Under the hood LLVM will likely optimize this into a local variable reference instead of first pushing to the stack. Since LLVM IR is typed, the value representing the address of a local variable will be a pointer type.

Source: notes.eatonphil.com

PulseAudio Under the Hood

PulseAudio Under the Hood

It shows three clients (employing three different APIs), one local PulseAudio server, two remote PulseAudio servers (connected via “native” and RTP protocols), one remote RTP receiver, ALSA backend, and a set of modules required to serve this setup. PulseAudio server supports a variety of network protocols to communicate with clients, remote servers, and third-party software. Each tunnel connects a single pair of a local device and remote stream:

Each tunnel acts as a regular PulseAudio client and connects to a remote PulseAudio server via the “native” protocol over TCP.

Source: gavv.github.io

Building an LTE+Wireless Access Point with a Raspberry Pi

Building an LTE+Wireless Access Point with a Raspberry Pi

With this setup, you can now connect another computer through the wired network interface, statically configure the computer with an IP address, i.e. , and SSH into the Pi using your changed credentials. This makes the initial setup quite easy as we just need to add another network device at the end of and configure it to automatically receive IP-addresses:

For the routing we need to enable IP forwarding within the configuration file by setting:

In addition we need some firewall rules to forward and masquerade traffic from our bridge ( ) to the LTE modem. The weak point of this setup seems to be the modem: the Huawei E3372 HiLink network mode might be the best for an easy setup, but I really would like to compare it with a “traditional” modem setup – e.g., with an assigned IP address and no additional NAT being performed by the modem (through the fake network interface).

Source: snikt.net

US air quality is slipping after years of improvement

US air quality is slipping after years of improvement

There were noticeably more polluted air days each year in the president’s first two years in office than any of the four years before, according to new Environmental Protection Agency data analyzed by The Associated Press. The EPA quietly posted new air quality data online last month that shows a recent uptick in polluted days. El Paso, Texas, saw one of the biggest increases in bad air days from the mid-2010s among metro areas with at least 750,000 people.

Source: www.apnews.com

Google clears up AMP’s confusing URLs by hiding google.com

Google clears up AMP’s confusing URLs by hiding google.com

AMP aims to allow websites to load faster by fetching pages from Google’s cache but had a drawback, as each URL for AMP pages began with https://google.com/amp. Last January Google announced a solution that preserved the original URL on AMP pages, which was available in a developer preview in November for Chrome 71, and is now being rolled out more widely. SEE: Tech budgets 2019: A CXO’s guide (ZDNet special report) | Download the report as a PDF (TechRepublic)
The answer to the AMP URL problem are “signed exchanges”, which allow a publisher’s domain to be displayed in the browser address bar, even though the content is loaded from Google’s cache when a user clicks on an AMP link in Google Search results.

Source: www.zdnet.com

The internet is an SEO landfill

The internet is an SEO landfill

I’m being wooed by a couple of ‘SEO Consultants’ who claim their expertise is guaranteed to get me on to the first page of a Search Engine’s results. The logic peddled is once i start paying the ‘search tax’ more people will start clicking my link, and in a few months my site will start to show up organically in the first page of search results. With many different people (and sock puppets) writing and talking about my product, the Search Engine assumes my product is popular and slowly starts bubbling it up in search results.

Source: docs.sendwithses.com

Ask HN: What’s the Legality of Web Scraping?

Ask HN: What’s the Legality of Web Scraping?

Many students ask me whether it’s legally OK to scrape websites without using an API and use the data for their projects for my course. Note that the concern is less about what’s ethical and more about what’s legal. This will also help me in my research because these days some reviewers are raising this concern when they see authors used web scraped data.

Source: news.ycombinator.com

Jim Allison Won a Nobel Prize for His Cancer Immunotherapy Breakthrough (2018)

Jim Allison Won a Nobel Prize for His Cancer Immunotherapy Breakthrough (2018)

Allison’s breakthrough was the discovery of a sort of secret handshake that cancer uses to evade the immune system, and a means to block that handshake—what the Nobel committee hailed as “a landmark in our fight against cancer,” which has “revolutionized cancer treatment, fundamentally changing the way we view how cancer can be managed.” But despite the mounting mockery of the larger scientific community and dwindling research funds, a handful of immunotherapy researchers continued to believe—and continued searching, decade after decade, for the missing piece of the cancer immunity puzzle, a factor that prevented the immune system from recognizing and attacking cancer cells. The spring cleaning is carried out by hungry, blobby Pac-Man cells in our blood—part of a 500 million-year-old personal defense force that Allison’s textbooks called the innate immune system.

Source: www.wired.com