microprocessors

3 Articles

Trouble Brewing For RISC-V As Issue Of Technology Transfer Is Questioned

November 7, 2023 by 42 Comments

Within the messy world of international politics, a major consideration by governments concerns which types of kn0w-how and technology can be transferred and sold to other nations, with each type facing restrictions depending on how friendly the political relations are with the target country at that point in time. Amidst all of this, there are signs that a so far relatively minor player in the world of CPU instruction set architectures – RISC-V – may become a victim of this, as a bipartisan group of US politicians is petitioning the White House to restrict transfer of know-how (so-called Intellectual Property, or IP) to RISC-V, as this may benefit adversaries like China.

As a US citizen who is involved in the RISC-V ecosystem, [Andrew ‘bunnie’ Huang] feels rather strongly about this, and has written an open letter to the US President, pleading to not restrict the way that US citizens can deal with the Switzerland-based RISC-V organization. This comes as the California-based RISC-V startup SiFive has announced that it’ll lay off 20% of its workforce. Depending on how a restriction on RISC-V is implemented, this could mean that US citizens would be forbidden from contributing to this ISA and surrounding ecosystem.

China has made it clear that RISC-V is a big part of its strategy to loosen its dependence on the West along with its investments in its MIPS-based Loongson processors, all of which strengthens the case for restricting US participation in RISC-V, even if it forces US companies like SiFive to move countries or cease its operations.

(Thanks to [cbjamo] for the tip)

Direct Memory Access: Data Transfer Without Micro-Management

March 31, 2021 by 16 Comments

In the most simple computer system architecture, all control lies with the CPU (Central Processing Unit). This means not only the execution of commands that affect the CPU’s internal register or cache state, but also the transferring of any bytes from memory to to devices, such as storage and interfaces like serial, USB or Ethernet ports. This approach is called ‘Programmed Input/Output’, or PIO, and was used extensively into the early 1990s for for example PATA storage devices, including ATA-1, ATA-2 and CompactFlash.

Obviously, if the CPU has to handle each memory transfer, this begins to impact system performance significantly. For each memory transfer request, the CPU has to interrupt other work it was doing, set up the transfer and execute it, and restore its previous state before it can continue. As storage and external interfaces began to get faster and faster, this became less acceptable. Instead of PIO taking up a few percent of the CPU’s cycles, a big transfer could take up most cycles, making the system grind to a halt until the transfer completed.

DMA (Direct Memory Access) frees the CPU from these menial tasks. With DMA, peripheral devices do not have to ask the CPU to fetch some data for them, but can do it themselves. Unfortunately, this means multiple systems vying for the same memory pool’s content, which can cause problems. So let’s look at how DMA works, with an eye to figuring out how it can work for us.
Continue reading “Direct Memory Access: Data Transfer Without Micro-Management” →

Hackaday Links Column Banner

Hackaday Links: October 25, 2020

October 25, 2020 by 3 Comments

Siglent has been making pretty big inroads into the mid-range test equipment market, with the manufacturers instruments popping up on benches all over the place. Saulius Lukse, of Kurokesu fame, found himself in possession of a Siglent SPD3303X programmable power supply, which looks like a really nice unit, at least from the hardware side. The software it came with didn’t exactly light his fire, though, so Saulius came up with a Python library to control the power supply. The library lets him control pretty much every aspect of the power supply over its Ethernet port. There are still a few functions that don’t quite work, and he’s only tested it with his specific power supply so far, but chances are pretty good that there’s at least some crossover in the command sets for other Siglent instruments. We’re keen to see others pick this up and run with it.

From the “everyone needs a hobby” department, we found this ultra-detailed miniature of an IBM 1401 mainframe system to be completely enthralling. We may have written this up at an earlier point in its development, but it now appears that the model maker, 6502b, is done with the whole set, so it bears another look. The level of detail is eye-popping — the smallest features of every piece of equipment, from the operator’s console to the line printer, is reproduced . Even the three-ring binders with system documentation are there. And don’t get us started about those tape drives, or the wee chair in period-correct Harvest Gold.

Speaking of diversions, have you ever wondered how many people are in space right now? Or how many humans have had the privilege to hitch a ride upstairs? There’s a database for that: the Astronauts Database over on Supercluster. It lists pretty much everything — human and non-human — that has been intentionally launched into space, starting with Yuri Gagarin in 1961 and up to the newest member of the club, Sergey Kud-Sverchkov, who took off got the ISS just last week from his hometown of Baikonur. Everyone and everything is there, including “some tardigrades” that crashed into the Moon. They even included this guy, which makes us wonder why they didn’t include the infamous manhole cover.

And finally, for the machinists out there, if you’ve ever wondered what chatter looks like, wonder no more. Breaking Taps has done an interesting slow-motion analysis of endmill chatter, and the results are a bit unexpected. The footage is really cool — watching the four-flute endmill peel mild steel off and fling the tiny curlicues aside is very satisfying. The value of the high-speed shots is evident when he induces chatter; the spindle, workpiece, vise, and just about everything starts oscillating, resulting in a poor-quality cut and eventually, when pushed beyond its limits, the dramatic end of the endmill’s life. Interesting stuff — reminds us a bit of Ben Krasnow’s up close and personal look at chip formation in his electron microscope.