the8472

34 ms·

631.
by the8472 3y ago
In some alternative history there would have been a push to make http 1.1 pipelining work, trim fat from bloated websites (loading cookie consent banners from a 3rd party domain is a travesty on several levels) and maybe use websockets for
632.
by the8472 3y ago
So we needed HTTP2 to deliver ads, trackers and bloated frontend frameworks faster. And now it delivers attacks faster too.
633.
by the8472 3y ago
Yes, the pressure information is used by userspace OOM killers but afaik also internally to detect whether progress has been made on reclaiming memory and it's supposed to be better than the previous progress heuristic.
634.
by the8472 3y ago
I've had important background processes die on windows when the offender didn't die and the OOM situation persisted for some time - I assume because it was using fallible allocations while the other processes weren't.
635.
by the8472 3y ago
Which can lead to dozens of unrelated applications dying on windows when they assume infallible allocators while linux keeps going (sluggishly) until it has to kill just the biggest one.
636.
by the8472 3y ago
Over the last few years there has been ongoing work to improve this. Including improved pressure detection, multi-generational LRU, large huge page swap and a bunch of other things. Some aren't enabled by default, some need userspace d
637.
by the8472 3y ago
But it does. https://files.catbox.moe/l9je82.png orange is the part used by caches
638.
by the8472 3y ago
> but 23andme could have prevented this by requiring 2FA for all accounts They could have prevented that by not keeping the data longer than needed to send it to the user.
639.
by the8472 3y ago
Isn't that just BT.2100?
640.
by the8472 3y ago
> Have you been completely asleep at the wheel over the past 20 years as ISPs around the world have started injecting ads in people's unencrypted HTTP traffic? ISPs are in my local jurisdiction, under regulators I may have voted for
641.
by the8472 3y ago
Oh, I already have implemented something[0] like that. It doesn't need lambdas or anything "cloud scale" like that. You "just" need a few dozen to a hundred IP addresses assigned to one machine and run a multi-homed
642.
by the8472 3y ago
I don't quite understand what you're proposing here. Generally you only control and operate ~1 node per IPv4 address or per IPv6 /64. All other nodes are operated by someone else, so they don't cooperate on anything beyo
643.
by the8472 3y ago
From a brief look at each it seems like they're scraping things like torrent websites, usenet or maybe RSS feeds. Not the DHT.
644.
by the8472 3y ago
These things need uptime of hours and days to do it properly and also to stay up to date. There are millions of nodes and torrents and to be non-abusive you have to issue requests at a somewhat sedate pace. And activity kind of moves with t
645.
by the8472 3y ago
It seems like every single of these things always cut corners and don't implement proper, spec-compliant nodes that provide the same services as they use. You know, the "peer" in p2p. BEP51 was designed to make it easier to n
646.
by the8472 3y ago
This is probably something you can setup with netfilter, not a sysfs knob.
647.
by the8472 3y ago
I like the clean slate we get from boot-from-scratch. Some pieces of hardware have subtle state corruption issues that can accumulate over uptime (kinda like Windows) that get flushed out by a reinitialization. Instead as much as possible o
648.
by the8472 3y ago
> No matter what, loading up 2+ GB of ram will always take noticeable time. Barely so. NVMe sequential throughput is measured in gigabytes per second. So you can get this under 300ms. And you can optimize the order in which things are lo
649.
by the8472 3y ago
Similar claims have been made about MRAM, FeRAM and similar devices for many years, hailed as replacement and unification of both storage and DRAM. MRAM isn't completely vaporware, but it's not available at the prices or densities
650.
by the8472 3y ago
That only applies to RLHF'd mealymouthed corporate AI. Unfiltered AI can be as antisocial or manipulative as the worst corners on the internet. Has early Sydney telling people to kill themselves and gaslighting them about the correctne
651.
by the8472 3y ago
Different things for different purposes. Measuring how real software under real production workloads in variable enviornments behaves is useful but inherently high-variance. It doesn't let you track <1% changes commit-by-commit. Fie
652.
by the8472 3y ago
> As I understand it, hardware counters would remain consistent in the face of the normal noisy CI runner. With cloud CI runners you'd still have issues with hardware differences, e.g. different CPUs counting slightly differently. e
653.
by the8472 3y ago
Age limits work for keeping those noisy kids out, should also help with senile old farts.
654.
by the8472 3y ago
It's not the first time that a company got compromised due to microsoft software. This time it was their cloud offering, the previous N times it was AD, Exchange, Outlook, WSUS (for delayed updates), ... And even if they'd move to
655.
by the8472 3y ago
These are probably outdated. A lot has changed in TCP stacks since 2009.
656.
by the8472 3y ago
There's fq_pie that's flow-agnostic.
657.
by the8472 3y ago
So many caused by retries. I wonder how many of those were due to lack of exponential backoff with jitter.
658.
by the8472 3y ago
Wayland has a pretty deep issue that loss of the compositor socket is treated as fatal and kills the client application. And when the IPC socket buffer is full that's leads to a close (so it is one of those fatal errors). So any latenc
659.
by the8472 3y ago
If I uninstall ssh I still want to have have my authorized hosts. If I uninstall some firefox version firefox I want to keep my profiles. XDG defines a thumbnailing hierarchy followed by multiple libraries, uninstalling any of those shouldn
660.
by the8472 3y ago
A rubber mat should work too

More ›