Allocators like that aren't the default for every process because they have higher startup costs. They are targeted to server workloads where startup cost doesn't matter, but it matters a lot if you're doing crud like starting millions of short-lived processes.

mimalloc definitely made claims that could not be reproduced, or at least not by me. That's why I wrote this doc five years ago. "Irreproducible malloc benchmarks" https://www.dropbox.com/scl/fi/evnn6yoornh9p6l7nq1t9/Irrepro...

Just out of curiosity are you getting 1GB huge pages on Xeon or some other platform? I always thought this class of page is the hardest to exploit, considering that the machine only has, if I recall correctly, one TLB slot for those.

Modern x86_64 has supported multiple page sizes for a long time. I'm on commodity Zen 5 hardware (9900X) with 128 GiB of RAM. Linux will still use a base page size of 4kb but also supports both 2 MiB and 1 GiB huge pages. You can pass something like `default_hugepagesz=2M hugepagesz=1G hugepages=16` to your kernel on boot to use 2 MiB pages but reserve 16 1 GiB pages for later use.

The nice thing about mimalloc is that there are a ton of configurable knobs available via env vars. I'm able to hand those 16 1 GiB pages to the program at launch via `MIMALLOC_RESERVE_HUGE_OS_PAGES=16`.

EDIT: after re-reading your comment a few times, I apologize if you already knew this (which it sounds like you did).

Right but on Intel the 1G page size has historically been the odd one. For example Skylake-X has 1536 L2 shared TLB entries for either 4K or 2M pages, but it only has 16 entries that can be used for 1G pages. It wasn't unified until Cascade Lake. But Skylake-like Xeon is still incredibly common in the cloud so it's hard to target the later ones.

So for any process that's using less than 16GB, it's a significant performance boost. And most processes using more RAM, but not splitting accesses across more than 16 zones in rapid succession, will also see a performance boost.

My old Intel CPU only has 4 slots for 1GB pages, and that was enough to get me about a 20% performance boost on Factorio. (I think a couple percent might have been allocator change but the boost from forcing huge pages was very significant)

That strikes me as a common hugepages win. People never believe you, though, when you say you can make their thing 20% faster for free.

Then it should be pretty easy to display that 20% "faster for free", no? But as always the devil is in the details. I experimented a lot with huge pages, and although in theory you should see the performance boost, the workloads I have been using to test this hypothesis did not end up with anything statistically significant/measurable. So, my conclusion was ... it depends.

Of course, it only helps workloads that exhibit high rates of page table walking per instruction. But those are really common.

I'll tell you what AI apparently can't replace and that is information graphics designers who are familiar with the exotic detail known as "contrast".

This is 99.44% slop! You are completely correct. The "exposure" is based entirely on vibes and does not correspond to observable reality. Down here in the real world the very first sector that is being disrupted is manual farm labor. They are out here with machine vision and quadcopters picking fruit. But according to the prompt that produces the treemap, manual labor has an exposure rank of zero.

> Don't project the emptiness of your existence

The irony of such an unnecessarily hostile opening line is ... Absolute cinema

(It's ok if you don't get it, Jeff. This comment is for other people)

Yeah, but unfortunately false nostalgia for subsistence farming is widespread and has traction in the discourse. I guess it's probably because every American who ever suffered from that lifestyle is dead, in other words the same reason that it is now increasingly popular to die from measles.

In reality, the laser-leveled, fully-automated, county-scale factory farm is the only reason anyone on this forum has ever experienced the phenomenon known as "free time".

Farming is harder than people who haven't done it think, and surviving on the production of only your family's property is really, really hard. Source: I grew up in very rural areas, and I've seen what it entails. My grandparents lived through the Great Depression in a farming community on a homestead.

However, I don't think that's the ask, here. You don't have to choose MEGASUPERTROPOLIS or remote solo farm. There's a huge gradient between the two.

It's possible to put a little effort into gardening, share with your community, and massively reduce the overall cost of food while still having free time.

> However, I don't think that's the ask, here. You don't have to choose MEGASUPERTROPOLIS or remote solo farm. There's a huge gradient between the two.

The gradient is where you start to consume a lot of carbon unnecessarily.

It’s an extremely privileged kind of nostalgia. Only the wealthy can romanticize poverty.

The solution to nostalgia for farming is for people to try it.

I used to live in Asheville NC years ago. There were a lot of hippie back to the land types that came there to find themselves. It’s that kind of place. Sometimes they’d decide to become farmers. This usually lasted at best one year. Sometimes it lasted weeks.

There's a book about being the child of back-to-the-land hippies called "Against the Country" that I enjoyed and recommend to all who might be tempted.

Some of the groups around Asheville were "eco villages" that could get rather culty, including the kinds of high control abuses that occur in cults. All groups weren't like this but I recall hearing some crazy stories involving shaming people into giving money to the group (while the higher ups are "more equal" than everyone else) and sexual exploitation.

There was one case where the group leader/organizer ran away with a bunch of cash, and one or two people tried to sue but they'd set up some kind of shell company structure and had conned people into signing things that made it hard to sue or press criminal charges. I think they also left the other members on the hook for a property lease, since the things they'd been conned into signing did that too. It looked a lot like an explicit long con targeting the hippie trust fund kids that would come around. Sex abuse too apparently.

I heard more than one story that mentioned members of the Hells Angels, at least allegedly.

This is another aspect of small insular communities in the good old days that the trads and anarcho-primitivists (left-trads) don't talk about. It still happens in small communities today, but a lot less due to the presence of larger scale nation states with laws and law enforcement. If the sheriff in your tiny rural town rapes you, you could try to bring charges at the state level or at least sue. Worst case you can move, and with a global money system you can sell your home and take the cash with you. Without that larger context, there is no recourse. If your local tribal chief, town sheriff, or cult leader wants to take your stuff, pressure you into polygamy, or rape your kids, there's nothing you can do about it. Back to nature!

Depending on how old you are, I might suggest that you were simply not aware of the details of your grandparents' lives at that time. And it just sounds like you have surrounded yourself with assholes in adulthood.

People made self-guided missiles with 1940s technology, in the 1940s. It can't be too much of a surprise if someone right now can make guided missiles in their garage with 2026 electronics. At this point the "guided" feature is trivial, the "missile" part is doable, and the weapon has probably become the tricky part.

Throwing an aside here that anyone interested in 1940s war technology must check out the old BBC documentary The Secret War (1977) which goes into depth on solving the engineering challenges of the war.

Well worth a watch. I think I watched it on Youtube.

I think the hard part was and will usually continue to be making the whole thing work effectively together with enough performance to actually work in practice. It's a lot of details across a lot of disciplines to get right.