ctz

7 ms·

31.
by ctz 2y ago
Also unlimited ponies and puppies for everyone!
32.
by ctz 2y ago
> TLS libraries by default don't have this behavior. No, but the most popular one gives you just a callback and people end up using that to build their own insecure, weird strategies. That's how we end up with things like &quot
33.
by ctz 3y ago
From the people who bought you WebP CVE-2023-41064/CVE-2023-4863...
34.
by ctz 3y ago
AIUI the RP2040 has -- as a deliberate design choice -- no security features: no secure boot, and no debug protection. That means it is incapable of offering any firmware integrity, or keeping any data secret at runtime.
35.
by ctz 3y ago
"yes, please run this loop minus 10 times" - statements dreamed up by the utterly Deranged
36.
by ctz 3y ago
I'm not the OP, but agree with their assessment and stayed 3.5 years despite that. For me, the reason was the vesting schedule: 4 years with a 1 year cliff.
37.
by ctz 3y ago
The IETF SSH working group was disbanded in 2006. You will be waiting a long time if you expect anything else to come from there by magic.
38.
by ctz 3y ago
You just need to repeat the work for each linux distribution package manager, and then again for Windows, and then again Mac and iOS. Simples!
39.
by ctz 3y ago
Smells like leverage for the next time they renew their ARM license.
40.
by ctz 3y ago
One possible riff: cook, chill and reheat all the potatoes. That is known to increase resistant starch.
41.
by ctz 3y ago
I think the GP has read but not comprehended the "best public cryptanalysis" part of that page.
42.
Hickory DNS(bluejekyll.github.io)
74 points by ctz 3y ago|14 comments
43.
by ctz 3y ago
I think your intuition on the cost of not-taken bounds check branches is off, especially in the context of devices that spend most of their resources browsing the web, running radios, powering screen backlights. I bet any holistic measureme
44.
by ctz 3y ago
Not a single test, either here or in adjacent commits. !?!
45.
by ctz 3y ago
> Not sure what the answer is for existing memory-unsafe code. It's an engineering approach that involves writing "A buffer overflow issue was addressed with improved memory handling" an awful lot. Hopefully one day they w
46.
by ctz 3y ago
I am @ctz and I'm forever being pinged by zig people lol
47.
by ctz 3y ago
If you find yourself wondering this, the answers are in an undergraduate economics course. Please learn the basics ;)
48.
by ctz 4y ago
I got the same poem, so off to a good start on the vapourware front.
49.
by ctz 4y ago
Obviously it is quite seasonal. In February we generated 350kWh or 12.5kWh per day average, importing 13.2kWh in total and exporting 153kWh. In January it was 8.5kWh per day average.
50.
by ctz 4y ago
That's an awful lot of vitriolic words to express: comparing kW and kWp is a type error. > This is caused by the unfortunate fact which is that the sun doesn't shine at all for at least half of the year And factually incorrect.
51.
by ctz 4y ago
> impossible Exaggeration. I can walk to a shop 20 minutes away and buy them off the shelf for RRP.
52.
by ctz 5y ago
> non-segmented architectures don't have an architecturally defined way to know the stack base ARMv8-M does have stack limit registers, which do precisely this. Being a much simpler mechanism than the MPU, they are quicker to switch
53.
by ctz 5y ago
OT: has 5ghz WiFi been announced for an espressif part?
54.
by ctz 5y ago
During the development of rustls, we also found a similar bug in Apple's TLS stack. We reported that to Apple and they fixed it. So it is feasible -- at scale -- for TLS clients to validate that DNS names in the SNI extension really a
55.
by ctz 5y ago
SNI is specified to only take DNS hostnames. Not bits of URL. DNS hostnames only. OpenSSL and gnutls are faulty in this respect.
56.
by ctz 5y ago
It's circle.goals.leader, which really underlines the point.
57.
by ctz 5y ago
It also supports 6GHz, but that doesn't mean a specific radio supports all bands. This is a 2.4GHz-only part.
58.
by ctz 6y ago
> The proof-of-work is completely independent of the number of transactions. A block could have 1 or 1000 transactions, but the energy consumption would be the same. There's a block limit, though, right? It doesn't scale up in
59.
by ctz 6y ago
You think they would've learned their lesson? https://en.wikipedia.org/wiki/2008_Universal_Studios_fire
60.
by ctz 6y ago
Does this mean there's a free and open-source version of the Nordic softdevice?

More ›