Posts
4417
Following
315
Followers
470
Linux kernel hacker and maintainer etc.

OpenPGP: 3AB05486C7752FE1
@raggi hmm.... I'll just grep async use and write a doc :-) I already have material for the commit message in this thread.
1
0
0
@raggi Right, and lowers the barrier a bit for new contributions. We can probably all admit at least that async can be "a bit" a daunting at first so something to hold on to does make a real difference. Something is at least always better than nothing.
0
0
0

@raggi Hmm…

So one option would be to create Documents/rust/async-guidelines.rs and document pre-existing usage patterns in Linux. I think it would be better try not to over-engineer the document. One example would documenting async itself. It is probably the best idea to keep it compact and punctual.

All you need to do then is just modifying the existing guidelines and adding sometimes completely new guidelines based on the patches.

Now, if I send a Rust patch with some async code, it will now be less likely to be against maintainers expectations. That results less noise in the mailing list, and patches landing more quickly. Or that’s I’d see it.

Way more important benefit would be that async would be “existentially” sealed to the kernel ecosystem early on.

I think this just like basic governance and risk management, not IMHO a big deal :-)

1
0
0
@ljs the idea is so stupid that my best guess is that someone made it on purpose

i should get a fellow position at DARAP from doing [1] if that idea is considered sane

[1] https://idle.slashdot.org/story/18/12/01/2331223/developer-misinterprets-linux-code-of-conduct-suggests-replacing-f-word-with-hug i
0
0
0
@ljs dude, you don't explicitly state that it is not going to work 😂 they really should go to rehabilitation and stop smoking crack tbh
1
0
0
@ljs Since you love AI and blockchains, you might be interested on TRACTOR, which compiles C to Rust using LLM. DARPA created this to turn unsafe C code to memory-safe Rust 🥇 😃

To be honest, I don't know even what to say. Too much is too much, WTF DARPA
1
0
0

PolkaVM is based on RVI20U64, which is a RISC-V profile lacking machine (M) and supervisor (S). The ALU of RVI20U64 has 47 opcodes in total.

I also noted FENCE and FENCE.I are in the profile. Are they useful for a single core CPU package?

Does this architecture have pmpcfg* registers? It would not make any possible sense to me so I’m only sanity checking here.

#polkadot #riscv

0
0
0
@mairacanal Reminds me last Fall ;-) https://github.com/keystone-enclave/keystone/issues/378

Tons of fun with:

- Incompatible CPU implementations
- Incomplete and ambiguous specification
0
0
1
" It’s a paradigm shift for the Linux desktop, crafted in Rust."

OK great it is written in Rust, but what is the paradigm shift? I see windows, icons, a docker and shit.

https://www.both.org/?p=7014
0
0
0

Here's a fascinating look at the first IBM PC 5150, from my friend David who wrote the training documentation.

https://www.both.org/?p=7098

0
3
1
Likely next product launch: AWS Silkroad
0
0
0
LOL, apparently #Wickr is these days an #AWS product. AFAIK, it has been like the choice of modern world drug dealers acting in #Tor. Famous from umh tabloids 🤷 #Amazon
1
0
0

I had missed that AWS discussed how they use to implement network policies, optimize TCP performance, and reduce Lambda function cold starts.

Recording: https://youtu.be/pVJHljuz1F0

0
2
0
@ljs Synths next time sorry ;-) or cats
1
0
1
Edited 10 months ago
@ljs In 21 years of working professionally in software projects from 3D engines to operating systems, I've yet to witness the "we have a shortage of code, please start pushing up the SLOC or hell is breaking loose" situation. I've saved a few projects in the past by deleting tons of useless bad code and rewriting it.

🤝
1
0
1
@ljs Often when someone tells me about how AI making code is so great I usually say that I could switch to Windows fucking Notepad and be as productive as I'm now. My time goes mostly to evaluating code and solving problems ;-) I'm not a SLOC generator as a person. I don't even use auto-complete because it confuses me. I RTFM instead, take things in slow motion and finally get what I'm looking at.
1
0
0
@ljs Thanks and I know :-) I see no value in AI because I have no use for it.

For blockchains I thought that they are equally bad until I realized that the S3 object storage that I love was based on blockchain but only for B2B rewarding from shared space to the network: https://www.storj.io/. It works great and is not slow. I just pay for use with my credit card reasonable bills. That led me to the conclusion that while tokens as a "cash" does not appeal me at all, I still can reason the advantages in B2B collaboration. Shared space is the "proof of stake" in this scheme, not "proving work" by calculating hashes for no good reason.

Parity works in this area providing the most scalable network of networks so that similar companies could do platforms just like Storj. I can think a number of services with unfair sharing of profits in music streaming, food delivery and similar, so I want to help people doing slavery as a service, and thus this is my way to make a world a better place ;-) I go to compiler and virtualization team for smart contracts so I have a chance to make a real difference here.

Being excited about a random cryptocurrency per se is just ultimate stupidity and a hoax. It's like someone would sell you IP addresses with no Internet existing.
1
0
1
Edited 10 months ago

Microsoft breaking a bunch of dual-boot systems by revoking insecure versions of grub during a standard Windows update is, uh, not great and was not supposed to happen, but it's worth mentioning that systems broken by this were running known insecure bootloaders and anyone running a distro that's actually on top of security updates was unaffected

(Edit to add: I wasn't terribly clear here. It's not the user's fault if their distro fails to deal with this, it's the distro's)

9
5
0
@ljs Blockchains felt literally like the more conservative choice than AI. We are living strange times bro ;-) It is founded by Gavin Wood who is co-inventor of Ethereum so I thought also that this can't be the worst. And even tad bit crazy blockchain scene is still flesh and bones humans working together doing stuff that they enjoy. I'd pick any day nutcases over AI ;-)
1
0
0
Show older