vlovich123

4 ms·

1.
by vlovich123 5d ago
This is literally a thread where people are saying “but why laptops no HBM”. It’s a perfectly cromulent response that meets the question where it is. It might be mildly a more defensible question for normal plugged in desktops but that woul
2.
by vlovich123 5d ago
Right, but HBM idle is significantly worse than LPDDR idle or even DDR idle for that matter. That matters a lot precisely because the device is idle most of the time. Your idle power draw dominates.
3.
by vlovich123 5d ago
LPDDR/DDR/GDDR generally still win over HBM when there’s no data being transferred. HBM is primarily better in watts/byte transferred. Consumer electronics spend most of their time idle.
4.
by vlovich123 6d ago
Only if you’re running at 100%. Consumers generally do not.
5.
by vlovich123 6d ago
Endianness yes, but float no. They just force software ieee754. It gives you exact bit accuracy on all build machines but it results in math that would differ between build time and run time depending if the function is evaluated as const o
6.
by vlovich123 6d ago
Neither is using const
7.
by vlovich123 6d ago
build.rs checkmate
8.
by vlovich123 7d ago
1 device out of the replica set I’m assuming so all of them should recover.
9.
by vlovich123 7d ago
I think the Rust approach is incorrect - if the runtime behavior can change based on the machine that the code gets run on, why is it so important the build time behavior has such a restrictive definition? Especially considering build.rs ca
10.
by vlovich123 7d ago
A halt instruction is probably the appropriate representation for an infinite loop - the process gets stuck same as the infinite loop without proceeding. Abort is the instruction you want unreachable() to compile down into. It wasn’t an eit
11.
by vlovich123 7d ago
Traditional stuff in this space are: * floating point differences between the build machine and the target. By far the most common * endiannes - code assumes little median runs on big endian There’s other more subtle issues that can crop up
12.
by vlovich123 7d ago
There’s a few comptime crates out there. Crabtime is iirc the most mature and popular
13.
by vlovich123 7d ago
I would get rid of consistent hashing and ketama for a better system which works save an additional 600TiB. You use the first N bits of your key hash to pick the server partition so it’s a reasonable number (eg 128 servers per partition). T
14.
by vlovich123 8d ago
I think the argument is that the equivalent of an infinite loop would be a halt / abort instruction, not a complete removal of the loop and continue running anything else.
15.
by vlovich123 8d ago
Or you're quoting me which then I don't understand what point you're trying to make.
16.
by vlovich123 8d ago
That's a niche level thing that helps in some scenarios, and generally not as much for C++ which is much more weakly typed than Rust is. Weak typing + static typing is why safety problems in C++ are going to be really difficult to fix
17.
by vlovich123 8d ago
Until Rust proved actually you can get really good or better performance if the language itself is better. I really don’t know how C++ digs itself out of the UB hole it has dug.
18.
by vlovich123 8d ago
not sure what you’re quoting but the last sentence of the post I replied to is: > Just the fact that Firefox got browser tabs before anyone else, and Firebug, just took it to another level compared to all the other browsers available at
19.
by vlovich123 8d ago
There's lots of fake research out there showing fake negative and positive results (admittedly negative is less common currently). What would stop someone from intentionally poisoning this data set to preserve some technical edge or to
20.
by vlovich123 9d ago
That’s a quote you seem to have materialized from thin air - op didn’t make this claim.
21.
by vlovich123 9d ago
Your history is off. Opera was the first mainstream browser to implement tabs and boy was Opera also much faster than Firefox at its peak (both UI, resource usage, and rendering) which overlaps that “peak” Firefox period by a lot. Firefox w
22.
by vlovich123 11d ago
Because anyone building a system needs performance metrics to understand how it’s behaving in aggregate / if they update the model? With the exception of install_id these all seem innocuous. Install_id is maybe needed anyway for purcha
23.
by vlovich123 11d ago
They have development costs to pay for which is substantial. Not sure why you are framing it as greedy and complaining it should be free vs celebrating someone putting this out there.
24.
by vlovich123 11d ago
That’s because the distinction between the Russian mob and the government is nonexistent
25.
by vlovich123 11d ago
This isn’t a show of power. Sabotage works best when your target doesn’t even realize it was sabotage. Similarly it’s best if you misattribute blame because then you’re fighting the wrong party and looking for ghosts. Russia throwing journa
26.
by vlovich123 11d ago
Graal is based on OpenJDK. OpenJ9 while using a separate JVM and JIT leverage the openjdk class path as well as the build environment and various other things. I don’t think there’s a single alternate implementation that doesn’t leverage a
27.
by vlovich123 12d ago
OpenAI is a 501c4 not a 501c3. Also the structure is much more complicated for OpenAI. Nevertheless the point stands - I don’t see what relationship company organizational mission has with their technical responsibilities. Indeed, if the op
28.
by vlovich123 12d ago
It’s already in rustup for what it’s worth. Not sure what “when it lands” means but I’m sure it’s an ever moving goalpost.
29.
by vlovich123 12d ago
https://blog.antoyo.xyz/rustc_codegen_gcc-progress-report-43 They’re plugging away at it. A major milestone was getting unwinding to work. I’d say they’re getting quite close and it’s getting better and better. It’s usable
30.
by vlovich123 13d ago
For what it’s worth the GCC backend for rustc is making progress and rustc more generally supports a lot of embedded stuff already

More ›