menaerus

23 ms·

421.
by menaerus 11mo ago
Sorry, this is just ridicilous and shows how people fragile really are. This whole topic and whole MR as well. I am routinely looking into the folly implementation, sometimes into the libstdc++, sometimes into libc++, sometimes into boost o
422.
by menaerus 11mo ago
They don't sell their models to individuals only but also to companies with most likely different business and pricing models so that's an overly simplistic view of their business. YoY their spending increases, we can safely assum
423.
by menaerus 11mo ago
It's not philosophical, half of the internet broke. There is a notion between "I really have no other choice but to crash" and "I might wanna crash at this moment because something is wrong but I won't and I'll
424.
by menaerus 11mo ago
Nonsense. Adding more keywords to a language is avoidable? Or making the compiler stricter in checking the integral promotions is avoidable? Sure, it is, but only if you stick compiling all of your code with C89. You're barking at the
425.
by menaerus 11mo ago
The job market is gone but there's an AI hype ongoing? Hmmm ...
426.
by menaerus 11mo ago
So everybody is cheating in your mind? We can't trust anything? How about taking a more balanced take: there's certainly some progress, and while the benchmark results most likely don't represent the world reality, the progre
427.
by menaerus 11mo ago
In demand? In the age where people are being completely replaced by their AI fellow colleagues, it logically doesn't follow that the Linux admin skills won't be or aren't already. The heck, I hate to say this but most of my c
428.
by menaerus 11mo ago
It was. And no, breaking change is not what you seem to imply it is. When talking about breaking changes, introducing new keywords is not what people usually think of. It's irrelevant.
429.
by menaerus 11mo ago
> C++ makes breaking changes all the time, Please don't spread misinformation. Breaking changes are actually almost inexistent with C++. The last one was with the COW std::string and std::list ~15 years ago with the big and major sw
430.
by menaerus 11mo ago
There is - you simply build your code with -std=c++XY, and if your toolchain doesn't support it (which one btw doesn't support at least c++17?), it will simply error out, no? That should be a pretty strong signal that your code re
431.
by menaerus 1y ago
> With "conventional RAII" you need to know if your pointer is the only pointer left to know whether to call the destructor. That requires some sort of pointer tracking. That's not really conventional in terms of RAII - th
432.
by menaerus 1y ago
650M LoC is certainly not a single codebase that you can "checkout" and "build". Also, the figure is a little bit hard to believe.
433.
by menaerus 1y ago
A bucket of 30 questions is not a statistically significant sample size which we can use to support the hypothesis which goes to say that all AI assistants they tested are 45% of the time wrong. That's not how science works. Neither is
434.
by menaerus 1y ago
Yes, and you answered it yourself.
435.
by menaerus 1y ago
> For the current research, a set of 30 “core” news questions was developed Right. Let's talk about statistics for a bit. Or let's put it differently: they found in their report that 45% of the answers for 30 questions they hav
436.
by menaerus 1y ago
What type of workloads?
437.
by menaerus 1y ago
In retrospective, I think this came out for the better in case of LLVM, and probably for GCC too. After all, both compilers emit ~equally optimized code today.
438.
by menaerus 1y ago
JIRA /s
439.
by menaerus 1y ago
Thanks, I'll revisit your comment. Some interesting things you shared.
440.
by menaerus 1y ago
Interesting thought but is it that the sigmoid curve can represent the developments in SW and/or HW? To reach the saturation point we need to be able to define a system up to the point where there are almost no more unknowns (variables
441.
by menaerus 1y ago
> It is not O(N^4/3), it is O(N * M^1/3). There's a good argument to be made that M^(1/3) should be considered a constant Math isn't mathing here. If M^1/3 >> N, like in memory-bound algorithms, then w
442.
by menaerus 1y ago
> Memory access times have not significantly improved in many years. We could say that it actually became worse and not better if we put it more into the context. For example, 90ns latency coupled with a 3GHz core is "better" t
443.
by menaerus 1y ago
I found the old discussion here on HackerNews although I haven't seen it before. I only found it now because I wanted to double-check my hypothesis and see if my understanding is really off. Seems like not: https://news.ycom
444.
by menaerus 1y ago
https://godbolt.org/z/aPcr1bfPe
445.
by menaerus 1y ago
> Not entirely sure what you're getting at here? > I believe the motivation is to avoid needing to pay the cost of atomics when there is no parallelism going on. Obviously yes. What I am wondering is what benefit does it bring in
446.
by menaerus 1y ago
Sharding is if anything a way to reduce the contention.
447.
by menaerus 1y ago
Yes, it can. Why would you be checking the pointer for nullptr after you have dereferenced it? It makes no sense at all, so, compiler indeed can elide the nullptr check before dereferencing the ptr exactly because it is free to _always_ ass
448.
by menaerus 1y ago
Hm, not sure. I can see that shared_ptr::_M_release [0] is implemented in terms of __exchange_and_add_dispatch [1] and which is implemented in terms of __is_single_threaded [2]. __is_single_threaded will use __gthread_active_p iff __GTHREAD
449.
by menaerus 1y ago
Since the compiler will merge/fold what it appears to be a different logic sections of your code into a single one, you can never be sure what the release build codegen looks like unless you read the assembly.
450.
by menaerus 1y ago
It's a platform-agnostic optimization in case of GCC so if your embedded Linux toolchain is based on GCC, and most of them are, it's pretty much the case that it will have this optimization turned on by default. > This option i

More ›