Joky

23 ms·

211.
by Joky 10y ago
A big big problem is having enough hardware to have a CI throughput one or two order of magnitude what it is now. Unfortunately that's not an easy thing considering how "heavy" it is to build and test the full toolchain.
212.
by Joky 10y ago
Yes! Such work is all done on a volunteering basis, when we find time for it on top of the "real work" (bug fixes, features, ...). All the infrastructure work is not always rewarding, you just get the upset people yelling at you :
213.
by Joky 10y ago
Swift also has a lot of pre-commit CI: - here is the Jenkins bots page: https://ci.swift.org/view/Pull%20Request/ - example of PR test reporting: https://github.com/apple/swift/pull/
214.
by Joky 10y ago
That's a very simplified view, if the A/C is right in front of your bedroom and you can't sleep anymore, it's not $70/month that'll make for it because you basically "can't" live here anymore. I&
215.
by Joky 10y ago
That's fine, as long as you don't say your benchmark is just "regular browsing". Enabling developer options put you outside of what the vendor really tested and optimized for. Also I find quite dubious your claim about `
216.
by Joky 10y ago
The point was that STL is unusable without exception because of OOM. If you're not using exception anywhere in your codebase and you don't plan to recover from OOM, is this still an issue? (I don't have much experience with
217.
by Joky 10y ago
I believe I read that it is rather that clang is (was?) not feature complete (until recently at least) for debugging in MSVC. Support for PDB/Codeview is very recent in clang, and I'm not sure it is complete yet.
218.
by Joky 10y ago
Are there many applications that recover from OOM?
219.
by Joky 10y ago
Yep, worth giving another look: I couldn't stand C++ before C++11, and now I can't stand C anymore :)
220.
by Joky 10y ago
In my field (Software), I find more profitable to me to watch one 40 min video from a conference rather than trying to understand a paper where the author had to compress all the information in a silly 10 pages limit from a random conferenc
221.
by Joky 10y ago
Something smells with the structure of the project if a small change in a single file triggers a long rebuild. Including headers that define template should be "free", it is costly when you start instantiating all of them. There m
222.
by Joky 10y ago
> Modules are the answer, but it seems nobody cares enough to get them in the standard or, more likely, it's just too hard a problem. Aren't modules especially limited with the mentioned problem of templates? If I have a vector
223.
by Joky 10y ago
> The converse is allowing US companies to break the law in other jurisdictions. Imagine if Valve just kept the money and didn't deliver the games at all? Would this be allowed in any US jurisdiction?
224.
by Joky 10y ago
While possibly true for some specific sports, I highly doubt this is a generality. So +1 to [citation needed].
225.
by Joky 10y ago
> I really wanted to say swift is headache and not a good language for Server development. Followed by a long description about how the ecosystem is young, and not mature enough for server production dev. I don't see how it justify
226.
ThinLTO: Towards Always-Enabled Link-Time Optimizations in Clang/LLVM(youtube.com)
5 points by Joky 10y ago|0 comments
227.
by Joky 10y ago
I think the point was that it was an answer to the jail threat...
228.
by Joky 10y ago
> in California where the Divorce laws are heavily in favor of the woman I thought they were in favor of the less rich of the two? I.e. marry a richer women that you are and you'd be the favored one.
229.
by Joky 10y ago
You're right, but what you're describing seems to be on the order of "algorithmic changes" or "data structure changes" to be more suited to your use-case. It seems to me that it is a higher level than what I wa
230.
by Joky 10y ago
I don't get it, this is presenting "design review" as an alternative to code review, while they fundamentally go together. The problem described in this post with code reviews is when there has been a lack of design review
231.
by Joky 10y ago
Fair. As a compiler engineer (full-time on clang/LLVM), I wouldn't make such ridiculous claim (even though the compiler is capable of nasty tricks that "normal" humans wouldn't be able to pull). Someone that pretend
232.
by Joky 10y ago
Because of noise in general, "best case" seems always like the best metric to me. Over a large number of run, you're likely to hit the "perfect" measurement with on a microbenchmark. Otherwise, for an "adaptive
233.
by Joky 10y ago
This seems quite ridiculous to me, I have seldom seen "modern compilers are always faster than you" but rather "they are good enough that it is not worth it". It provides a very over-confident "conclusion" base
234.
by Joky 10y ago
It depends what is meant by "unify" and what is the "stack". What I just read that a language is meant to "unify the full stack", I think I'll be able to rewrite with it any piece of the stack used, includ
235.
by Joky 10y ago
If I own a house but I have a new job far away and I rent a house there and have a tenant in the one I own, that's a zero-sum game (assuming the rent I pay is equivalent to one I get from my tenant). The owner of the house I'm occ
236.
by Joky 10y ago
That supposes that the backup is local, I have a time capsule (remote) and I also plug once every other day an external hard-drive for my time machine backups.
237.
by Joky 10y ago
> C++ is a non-starter. Uh?
238.
by Joky 10y ago
It's OK to point something as "Completely wrong" and as a "Urban myth", but a source is appreciated.
239.
by Joky 10y ago
I hope I'm gonna stay Junior longer! I don't want to start using goto...
240.
by Joky 10y ago
Xray is currently being implemented in LLVM: it will allow exactly this. The code is instrumented at compile-time by leaving at the beginning of the function just the minimum amount of extra space (nop) so that at runtime a selected list

More ›