Posts
183
Following
139
Followers
148
Probably some RISC-V stuff, but hopefully other things too ;)
i'm sorry they did WHAT
5
13
1

Framework has stopped selling versions of its base laptop model in the US because of tariffs. Says that if it kept selling them in the US it would lose money. Framework is the only company that makes reliably upgradeable, repairable laptops

https://www.404media.co/framework-stops-selling-some-of-its-laptops-in-the-u-s-due-to-tariffs/

4
7
0

While working on my GCC translation validator, smtgcc, I've encountered several cases where GIMPLE IR semantics are ambiguous or inconsistent with GCC optimizations.

I'll discuss these on the GCC mailing list and link each corresponding thread here. 🧵

2
3
0

David Chisnall (*Now with 50% more sarcasm!*)

One of the interesting commercial differences between RISC-V and Arm is that Arm makes deals that let you ship ISA features first if you contribute a lot to the spec. They've done this with several companies in the past, where companies contribute a lot of design work to the ISA (and, often, patents to the associated patent pool) and, in exchange, they get to ship things that implement it a year or so before anyone else. This has two big advantages:

  • Arm gets the feature in the ISA and everyone can implement it.
  • The company that goes first has first-mover advantage but can promise their customers a second source if it's actually a useful feature (which makes adoption easier).

RISC-V has no analogue of this. Companies that ship useful extensions either:

  • Do a load of work in public. It can take a year or so to get a spec ratified and other people can easily start working on it long before then and get access to it early.
  • Ship the extension as a vendor-specific extension and then work to get it ratified. During ratification, even if nothing changes in the semantics, the encodings of the instructions will move from the vendor space to the standard space and so the thing that you've shipped will not be binary compatible with the standard version.

I'm curious where those different incentives will take the two ecosystems.

0
7
0
Users of w3m and lynx browsers will be asked to identify and fix a bug in the mm subsystem before they are allowed to access git kernel.org.
2
22
46
git.kernel.org and lore.kernel.org now require proof of work. Big thanks to Anubis developers for filling this important need.
10
34
52
Another email sent to the Linux Plumbers mailing list:

"Hi there,

This is Jacob Bryant and I came across the helpful information you shared on the lpc.events website and was wondering if you were open to adding new resources to your page.

Unfortunately, plumbers are at a higher risk of being diagnosed with mesothelioma, a rare cancer caused from exposure to asbestos that takes 20-50 years to develop. It’s extremely important that those who have and are working in plumbing are aware of this type of cancer. That's why we made a guide that covers everything you need to know about mesothelioma as a plumber."
6
6
20

I've been saying this for years: ebikes should have USB-C PD ports for charging!

Finally a manufacturer has actually done it. In this case using 140W power delivery, which is way higher power than any ebike charger I've ever used. And USB-C PD now supports up to 240W, which is more than enough for ebikes.

I hope this becomes the norm:

https://www.theverge.com/news/639681/usb-c-charging-e-bike-ampler-nova-specs-price

4
7
1
I'm sad to say that we're following the lead of many others and putting in proof-of-work proxies into place to protect ourselves against "AI" crawler bots. Yes, I hate this as much as you, but all other options are currently worse (such as locking us into specific vendors).

We'll be rolling it out on lore.kernel.org and git.kernel.org in the next week or so.
9
142
168

@joshbressers @sjvn @TheNewStack I'm using our role as CNA to work within and push against silliness in the program, in part to work against said "need". Together with other open source CNAs (to avoid naming names).

Not saying this will succeed of course...

0
2
0
@joshbressers @sjvn @TheNewStack I'm with @badger Linux is a CNA to help fix the CVE process, and so far we have already achieved some change, more to hopefully come.

And the CRA is going to cause other software projects to come to terms with their reporting process, so becoming a CNA is a good step forward in the whole thing.

And besides, what open source project doesn't want to actually control what other people are saying about your project? Just this week we "took back" a CVE issued by a rogue CNA against Linux when they shouldn't have done so. If we weren't a CNA we would never have been able to do so at all.
1
1
5

Toke Høiland-Jørgensen

I was devastated to learn that Dave Täht passed away. Wrote a bit about it.

https://blog.tohojo.dk/2025/04/remembering-dave-t%C3%A4ht.html
1
4
9

casey at easterhegg (☎️ KCXT)

Edited 17 days ago

turns out Qualcomm dropped the sources for talking to their Embedded USB Debug (EUD) peripheral, and it works on the OnePlus 6!

This means we get JTAG access directly via the USB port, yes seriously! There is also a UART peripheral which we can hook up (so far untested).

Basically you write 1 to a magic register (typically from the Linux driver but i have been testing from U-Boot) and all of a sudden a 7-port USB hub appears on your PC (in addition to whatever USB gadget you had set up) with a single device which is the EUD control interface.

Now that the code to talk to it is public (and functional with this openocd fork https://github.com/linux-msm/openocd) you can get JTAG access to the device for easier debugging of the kernel or U-Boot!

It seems like this works "by accident" on the OnePlus 6, likely the same debug policy misconfiguration that causes the device to go to crashdump instead of just rebooting (so it's unlikely to work on say, the PocoPhone F1, but maybe worth a try!).

There seem to be protections in place so you can't escalate to EL2 or EL3, when in EL2 all registers read as 0

6
24
2

New tool! Meet the iFixit ESD-Safe Hammer.

We hear you: sometimes you need a little more oomph to finish your fix. So, we engineered the world’s first hammer designed for electronics repair. It's built with a non-conductive handle, ESD-safe head, and just enough heft to "persuade" stubborn components.

Perfect for:
🔋 Glued-in batteries
🪛 Stripped screws
🔌Soldered memory
💥 Stress relief

Now unavailable at iFixit.com.

0
14
1

Localization is hard.

Google Titan costs 35 euros in Europe. 35 USD in the US.

They launched in Norway this weekend. What's the price?

35 NOK (3 EUR/USD).

How many can you buy? 10!

11
5
2

I'm really loving how Mastodon has become a refuge for all the grizzled seafarers on the ocean of the internet. They pop up in my feed and their bios all say something like

"I've been online for longer than the internet. I've seen things you people wouldn't believe. 56k modems on fire in the light of Usenet. I watched IRC forks glitter in the dark near the Gateway 3000. All those moments will be lost in slop, like tears in rain. Time to deshittify."

8
31
3

The Meetup is tonight Thursday, March 27th, at the Lucky Labrador Beer Hall on NW Quimby St https://www.meetup.com/portland-linux-kernel-meetup/events/306813644/

0
1
2
"WTF... oh, they saw /crypto/ in the URL."
5
84
231
Show older