aseipp

46 ms·

481.
by aseipp 3y ago
Actually, that's a really a good point. It's also especially weird considering we get lots of debugging support through the OS stack on every system. They have APIs, libraries, tools, etc. But actual crash control, a kind of thi
482.
by aseipp 3y ago
Neat, I didn't know Sentry also had a good fork, I haven't tried it! But in contrast, here's an in-process fault library that I whipped up (from forking Phusion Passenger) about 10 years ago that I still reach for sometimes,
483.
by aseipp 3y ago
Yes, I managed to find it: "eXtreme Modeling in Practice" https://arxiv.org/pdf/2006.00915.pdf Unfortunately I got the product wrong; it was not Atlas, it was Realm Sync. All of the test-case generation stuff
484.
by aseipp 3y ago
Ease of integration, because having literally anything is typically better than nothing. Honestly Crashpad isn't fun to integrate unless you use a fork like backtrace's (which adds CMake support), which I think doesn't help
485.
by aseipp 3y ago
Yeah, that's basically an actually good implementation of the pile of crap that I threw together several years ago while writing a few patches. :) And yes, I linked to the spec files because there actually isn't that much test cod
486.
by aseipp 3y ago
Transaction size and duration is limited to keep the latency and throughput of the system manageable under load, from my understanding. It makes sense to some degree even with no background in the design; if you are serving X/rps with
487.
by aseipp 3y ago
Yes, but there really aren't too many good solutions to that that aren't either extremely language or domain specific. And if you're careful you can get a lot of direct mileage out of it. For example, MongoDB (yes, that one!)
488.
by aseipp 3y ago
Snowflake uses FoundationDB for warehouse metadata in the control plane, IIRC. It is not in the data plane path for log ingestion or other warehousing tech. That said the control plane is, uh, pretty important!
489.
by aseipp 3y ago
FWIW, I believe read transactions are unlimited in duration now that the Redwood engine has been available. But I haven't tested Redwood myself. Write transactions are still definitely limited to 5 seconds, though.
490.
by aseipp 3y ago
No, the simulation harness and tests are open source and you can run them. It would be impossible for anyone to contribute anyway without it, for example, Snowflake, which heavily depends on it. It's built into the server binary direct
491.
by aseipp 3y ago
For passkeys in particular, this theory would be more credible if it wasn't for the fact pretty much every password manager has planned support for them, and almost all of those same managers have been integrated with your browser&#x27
492.
by aseipp 3y ago
In the next version of iOS you'll be able to use a third party app to handle the Passkey flow, like how a 3rd party app can handle the password flow today. So you'll be able to remove your passkey from iCloud and use the one insid
493.
by aseipp 3y ago
It really depends on the card. I have an old RDNA workstation card in my server and the driver was a real crapshoot. I eventually started delaying updates to newer kernel releases (which would fix other bugs!) because there would be regress
494.
by aseipp 3y ago
That's fair, and I agree because honestly almost no companies IMO actually know how to build robust FOSS projects outside their own needs, I wouldn't put that totally on one place. It pretty much actually comes down to the enginee
495.
by aseipp 3y ago
Buck2 is IMO much better than even Bazel is from a design POV, because it actually cleanly separates all user rules from the build engine, and has a coherent modern design around a sound theoretical basis. Neil, one of the leads and author
496.
by aseipp 3y ago
> I've never had any problems that couldn't be solved. Stockholm syndrome. I've used Git for 15 years, early GitHub user, etc. Yes, you can solve many of these things, but until recently even things like "I am changin
497.
by aseipp 3y ago
I'm one of the earliest GitHub users, have run FOSS projects, etc. Sapling is absolutely excellent. If anything, the 'sl web' UI alone, which can do rebases/commits, is worth giving it a shot. 'sl web' makes Gi
498.
by aseipp 3y ago
Not for these chips. The Vivado licensing is kinda complex these days because it's basically segregated by the chip series rather than actual software features. But it can basically be summed up as "Free for low/mid-range dev
499.
by aseipp 3y ago
Programming time isn't exactly instantaneous or fast on these; SRAM cells need to all be programmed to the proper value for every LUT (2^k SRAM cells needed for a single k-LUT; k = 6 in this case), and that typically is either done onl
500.
by aseipp 3y ago
Yeah, you often see it come up in lots of discussions among indies or small teams, since people end up just buying One Server that runs all their builds; I didn't think about AMD's V-Cache though, the 7950X3D is packing 144MB of L
501.
by aseipp 3y ago
This thing probably costs more than your car. So, no, it is not a successor in any meaningful sense. There are a lot of alternatives right now that would be an acceptable choice; it's mostly that the DE10-Nano has all the mindshare and
502.
by aseipp 3y ago
UltraScale+ you mean? Well, the real question is P&R time at some utilization, right? We were easily doing 14hr P&R times on fairly utilized US+ chips like, 5+ years ago. We probably could have improved this in a number of ways, tho
503.
by aseipp 3y ago
The two things, Perseus and what Daan calls "FIP" here are a bit apples/oranges. But they are very related. FBIP/Perseus is an algorithm for efficient reference counting with reuse. "FIP" is actually a calculus
504.
by aseipp 3y ago
The fact they did it that way is actually a perfect example of why Google is considered so far ahead of competitors technologically and operationally by their engineers. When you have a powerful building block like Spanner that engineers ca
505.
by aseipp 3y ago
eBPF verification was always a laugh from the very beginning design stages, if you ask me, because as this paper demonstrates, it was never going to be enough. Anyone with a modicum of security or PLT experience could have told you this whe
506.
by aseipp 3y ago
Is it actually that bad on Tiger Lake? Or just for really high-width vectors? On my old Ice Lake laptop, single-core AVX-512 workloads do not decrease frequency at all even with wider registers, and multi-core workloads will result in clo
507.
by aseipp 3y ago
Backblaze B2 and Backblaze's Personal Backup product are completely different and have completely different pricing strategies and customers. B2 is priced at $5/TB/mo in storage, so this is 2x that. But the service also has t
508.
by aseipp 3y ago
Heikki Linnakangas is one of the top Postgres contributors of all time, he isn't just "someone." The fact he's working for a startup on a fork (that already exists, which you can run right now on your local machine) does
509.
by aseipp 3y ago
Interesting, I didn't look far back enough. And I didn't realize it cloned from Mercurial first instead of using the tarball either; that'll hurt a lot more, too!
510.
by aseipp 3y ago
The repository in question, based on the email from Mike Blacker, sent to gmp-devel[1]: https://github.com/BtbN/FFmpeg-Builds There's a bunch of build variants here, for various platforms, in different ways. Class

More ›