byroot

21 ms·

211.
by byroot 3y ago
> there's not compiled machine code that does a jump It's more of an implementation detail, but the Ruby VM actually does a jump if all the cases are "keyable"(typically integers, strings, symbols, etc). It does a has
212.
by byroot 4y ago
Yes, we’ve been doing this at Shopify for quite a while now.
213.
by byroot 4y ago
If it isn't a matter of ego, how to you explain that the fix was to add a special weight to his tweets specifically? https://www.bloomberg.com/news/articles/2023-02-15/musk-forc... I hope Bloomberg is re
214.
by byroot 4y ago
Is Forbes good enough for you? https://www.forbes.com/sites/mattnovak/2023/02/09/elon-musk-...
215.
by byroot 4y ago
https://www.theverge.com/2023/2/9/23593099/elon-musk-twitter...
216.
by byroot 4y ago
I was answering to the person saying to open a PR (on Rails I presume). That comment isn't in the Rails dockerfile, it was added by the OP. https://github.com/rails/rails/blob/4f3af4a67f227ed7998fed57...
217.
by byroot 4y ago
It's intended. Yes serving static files from Ruby is slower than from nginx or whatever, but you'd need to embed nginx in the image and run both processes etc. The assumption here is that there is a caching proxy in front of the c
218.
by byroot 4y ago
> as it requires ~3x more memory. Not sure were you've seen this, but that's absolutely not normal. The YJIT memory overhead is certainly sizeable, but is generally at least an order of magnitude smaller than that.
219.
by byroot 4y ago
There might be a number of reasons why YJIT isn't effective in your application. The best way to know it to enable YJIT runtime statistics [0] to see whether YJIT is doing a lot of side exits and if so why. But it requires a bit of YJI
220.
by byroot 4y ago
> The difference in used RSS is also quite interesting in this comparison. YJIT 3.1 was eagerly allocating it's executable memory region regardless of whether it was needed or not. With the default settings that meant an extra 256M
221.
by byroot 4y ago
If you look at the breakdown per service, EC2 (assuming mostly Ruby) seem to be at best 20% of the bill: https://twitter.com/dhh/status/1613508201953038337 So if you are to make Ruby ~10% faster, you might reduce
222.
by byroot 4y ago
Depends what you are after and what you are constrained by. Some people might be struggling with memory usage and may not see it as worth it, but for others who may have some extra RAM sitting around, it might come as a free perf increase.
223.
by byroot 4y ago
Yes, like all JITs by definition. The actual memory increase depends heavily on the app size so I could quote some figures but it would likely be irrelevant to you.
224.
by byroot 4y ago
It's the same in Ruby. It's even mentioned in TFA.
225.
by byroot 4y ago
Is France justice system unfair too? Because it seems like he's not really eager to face it either.
226.
by byroot 4y ago
You are confusing setters with the assignment evaluation. Setters are just normal methods. ?> def you_are_wrong=(a) ?> a * 2 >> end => :you_are_wrong= >> send(:you_are_wrong=, 24) => 48
227.
by byroot 4y ago
Canada don't care much whether immigrants are "productive members of society". You're free to spend your money in Canada, and as an immigrant without a job, you won't benefit from much welfare and such. The only thi
228.
by byroot 4y ago
Absolutely, I wasn't saying it's a good situation to be in, just that it doesn't legally force them to leave Canada immediately.
229.
by byroot 4y ago
Mostly a guess, but: - Releasing/acquiring the GIL isn't cheap. So you'd need to know in advance whether it's worth it or not. - You need the GIL acquired to instantiate the resulting "match data", captur
230.
by byroot 4y ago
> is likely facing a prospect of having to leave Canada (not familiar with work visa types there) Unless it changed in the last few years, Canadian work permits allow you to continue to reside in the country for the duration of their val
231.
by byroot 4y ago
While I agree with your general point (yes a Ruby web app once basic database access patterns have been optimized is essentially CPU/GVL bound), I'd advise caution when using things like techempower benchmarks of the benchmark gam
232.
by byroot 4y ago
It just needs a small puma config tweak: https://twitter.com/eregontp/status/1588199934796365826
233.
by byroot 4y ago
Rails Core team member working at Shopify here. You seem to be misinterpreting what we’re doing at Shopify. We pretty much never work on the product or anything like that, and we don’t go around the app chasing perf issues either. At best w
234.
by byroot 4y ago
It's much better today. Sure when it was hype a decade ago you'd get lots of flashy tools and library every other days, but a large part of it was really wonky. Now that the dust settled, the tools that remain largely used are m
235.
by byroot 4y ago
> their webhook worker(s) being swamped That's a capacity problem caused by a logic bug. Nothing stack specific. If you throw more work at a system than it is designed to handle, you'll hit a bottleneck. > Your snark doesn&#
236.
by byroot 4y ago
You know you can just click on the post title, that will open the posted link in which you can read the detailed cause of all the outages they had that month. If you do this, you will realize that none are close to what you describe. Also h
237.
by byroot 4y ago
> created an impenetrable fortress along the border Absolutely not. No-one ever expected the Maginot line to hold more than a few weeks. It's goal was two-fold: - Delay the invasion by a few weeks to give time for mobilization.
238.
by byroot 4y ago
Ah I see. As Kevin said, the HTTP response is generally nothing compared to the memory required by the VM. A response might be a couple MB, a medium sized Rails app will need a couple hundred MB to load it’s code etc.
239.
by byroot 4y ago
John Hawthorn from GitHub is a ruby core committer.
240.
by byroot 4y ago
Sorry, I see absolutely no relation between the two.

More ›