Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mgaunard
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
30 ms
·
601.
▲
by
mgaunard
2y ago
multicast is precisely used for low-latency high-throughput message buses.
602.
▲
by
mgaunard
2y ago
Most advanced NICs support flow steering, which makes the NIC write to different buffers depending on the target port. In practice though, you only have a limited amount of these buffers, and it causes complications if multiple processes ne
603.
▲
by
mgaunard
2y ago
ebpf is linux technology, you will never be able to do it from iOS.
604.
▲
by
mgaunard
2y ago
Many organizations don't distinguish the two.
605.
▲
by
mgaunard
2y ago
You can definitely be a hands-on manager who builds things, though that's often what's called a team lead instead.
606.
▲
by
mgaunard
2y ago
Large organizations are very inefficient, so much so that the inefficiency of the cloud is more efficient than them doing the work themselves.
607.
▲
by
mgaunard
2y ago
I have no idea what the rules are, doesn't look like disk fragmentation at all.
608.
▲
by
mgaunard
2y ago
In that picture I'm personally more bothered by the antialiasing than by the misalignment.
609.
▲
by
mgaunard
2y ago
There are several problems with the C++ variant, which could have been easily avoided by just following the original Rust more closely.
610.
▲
by
mgaunard
2y ago
The main applications that I use are a web browser, a terminal, and a text editor. Both my terminal and my text editor support tiling natively, and I assume there are extensions for my web browser to do the same. I'm not sure adding th
611.
▲
by
mgaunard
2y ago
The transformations you're referring to are fully deterministic and guaranteed to be correct. LLMs provide statistically probable answers with no guarantee of correctness. It takes more time to review LLM code than it takes to write it
612.
▲
by
mgaunard
2y ago
Sounds like anti-statist propaganda.
613.
▲
by
mgaunard
2y ago
Depends on the target market. If they need to sell to other men, that's fine, but if they need to sell to other women (which is actually quite likely), then not so much.
614.
▲
by
mgaunard
2y ago
It's already well-known that attractive women also do well at sales. This is the same thing.
615.
▲
by
mgaunard
2y ago
We don't spend 50% of state budgets on primary/secondary education.
616.
▲
by
mgaunard
2y ago
Any code that does divisions and modulos with non-constants that are known to be powers of 2 should do the optimization manually. Even for non-powers-of-two there are also techniques to speed up divisions if the same divisor is used repeate
617.
▲
by
mgaunard
2y ago
School is easy, the point of it is more to learn social skills.
618.
▲
by
mgaunard
2y ago
why does it break userland? if you need to know the page size, you should query sysconf SC_PAGESIZE.
619.
▲
by
mgaunard
2y ago
is there a web framework that managed to significantly iterate on knockout or is it all needless fluff to spout more buzzwords every week?
620.
▲
by
mgaunard
2y ago
London resident not eating junk food either, and indeed I hadn't really noticed the "phenomenon". If anything I've seen the fried chicken with panko much more than the katsu curry sauce.
621.
▲
by
mgaunard
2y ago
It's not just C, but also any object-oriented language. Basically the most popular imperative paradigms are just ill-suited to making the most of SIMD hardware. They're designed with other goals in mind.
622.
▲
by
mgaunard
2y ago
That's not forcing anything, if you're not getting blend you're just targeting an old ISA without that instruction. It's up to you to choose your target adequately based on your deployment requirements.
623.
▲
by
mgaunard
2y ago
normally there is an instruction for this called blend (since SSE4.1). There shouldn't be a need to do any bitwise magic.
624.
▲
by
mgaunard
2y ago
a cmov-based approach is necessarily slower than a branch-based approach that was correctly predicted, since cmov requires computing both branches then selecting the result at the end.
625.
▲
by
mgaunard
2y ago
The main problem of SIMD remains the same: it requires working with structures of arrays rather than arrays of structures, which makes it more difficult to combine and layer objects on top of each other.
626.
▲
by
mgaunard
2y ago
You can just use templates or macros to make it length-agnostic.
627.
▲
by
mgaunard
2y ago
I quit the company I was working for many times. It was usually an opportunity to leave the drama behind, not create more.
628.
▲
by
mgaunard
2y ago
The people that wrote it probably moved on, so ownership and fit must be weak.
629.
▲
by
mgaunard
2y ago
I believe the original proposal for noexcept was that throwing was UB, and the original author believed strongly it should have been that way. In the end what got into the standard is that it is well-defined: it calls std::terminate. If I r
630.
▲
by
mgaunard
2y ago
if you want a fast hash map, just use a library that is built specifically around a particular implementation strategy. std unordered map is slow by design for several reasons. Things of interest are: - open addressing method: linear, quadr
More ›