aseipp

42 ms·

421.
by aseipp 3y ago
Are the approval rates that harsh? It never seemed that way to me. Anecdote: I applied in 2019 when it first came out, and hadn't had a credit card in several years, and my credit score was, to my knowledge, pretty much shit and non-ex
422.
by aseipp 3y ago
Module support will be stabilized in CMake 3.28, actually: https://www.reddit.com/r/cpp/comments/16y9qv2/cmake_c_module... The other points (mainly integration points) are definitely valid though.
423.
by aseipp 3y ago
> In this case, it's a bid for biometrics Biometrics are used to unlock an local, on-device key storage mechanism which contains a private key, and from that you can derive a public key, and that's what Passkeys fundamentally a
424.
by aseipp 3y ago
People constantly make this comparison, but it's stupid, the Linux kernel's DCO is just a CLA by a slightly different name and slightly different signatory procedure; giving it a slightly different acronym doesn't make it som
425.
by aseipp 3y ago
This week I upgraded my kernel on a 2017 workstation to 6.5.5 and when I rebooted and looked at 'dmesg' there were no less than 7 kernel faults with stack traces in my 'dmesg' from amdgpu. Just from booting up. This is a
426.
by aseipp 3y ago
User-space apps have a lot more avenues for timely updates than middleboxes or kernel-space implementations do though, and developers have lots of experience with it. If middleboxes actually received timely updates and bugfixes, there would
427.
by aseipp 3y ago
The RPi4 has had a functional port of TianoCore for a while now, which is likely what they were referring to. You can use e.g. generic aarch64 UEFI Fedora images out of the box with it. https://github.com/pftf/RPi4
428.
by aseipp 3y ago
Finally! The Cortex-A76 is ARMv8.2 at least, so it has a non-LL/SC atomics model (Large System Extensions). You'd normally have to call out to libatomics for outlined implementations that could choose the right instructions at run
429.
by aseipp 3y ago
1. Because you will piss off would-be users. That simple. I have also complained about this endlessly. I also despise Tcl and had to use it to help automate lots of EDA tools in ASIC and FPGA designs, so I'm with you on this. But it is
430.
by aseipp 3y ago
Not really. Phi never had a very large market in reality, and at the time it existed there were very few non-niche workloads it was actually cost effective at. Intel was also on top at the time so they could afford to experiment there; now
431.
by aseipp 3y ago
> whereas password auth can be done in several lines. Secure password auth is complex and because of externally imposed constraints like phishing, generally needs more work to support. You also need recovery flows, etc, since just losing
432.
by aseipp 3y ago
It's a really recent addition to Chrome/Edge to support that, and it only was supported on iOS in v17 which came out last week. Actually, 1Password probably released their support for Passkeys last week precisely because iOS 17 wa
433.
by aseipp 3y ago
I can't read anyone's mind. I assume it's simply because passkeys can be used outside the browser, in theory for arbitrary applications, and like every standard, it tries to satisfy a large set of users, there are tradeoffs,
434.
by aseipp 3y ago
FIDO itself specifies that attestation can be used but it's extremely likely browsers aren't going to ever support it for the same reason Apple has publicly said they won't: because attestation is a misfeature for Passkeys, b
435.
by aseipp 3y ago
The ability to support GC'able DOM references in browsers through integration with the host-side GC is absolutely being worked on, it just took several years to stabilize because it's a tricky feature. It will also allow GC'd
436.
by aseipp 3y ago
I think the reality is that faster startup, smaller deployments, and lower memory usage is considered much more valuable today than it was when GCJ was an actively maintained project. It simply did not offer things people needed at that tim
437.
by aseipp 3y ago
Haskell has always been a language where there are many ways to skin any particular cat, for reasons from ergonomics to expressiveness to theoretical niceness or practicality. "Learn the idioms" has nothing to do with it. Why didn
438.
by aseipp 3y ago
There's a pretty good argument to be made that Linux, the most high-profile GPL project in existence, not trying to enforce its license/copyright terms on reasonable grounds, is far, far worse for the FOSS ecosystem than the alter
439.
by aseipp 3y ago
You're completely right, the argument is a red herring. People did not move away from cooperative multitasking for operating systems because it solved every "run-away usage of resource" problem for all time. They did it speci
440.
by aseipp 3y ago
Amazing, thanks for the correction(s)!
441.
by aseipp 3y ago
Yeah, I think a good simple litmus test for this is "can I directly call mmap(2) on a file, and then launch a kernel on that mmap'd memory, with no extra steps, and it works as I expect it to". With these newer features in CU
442.
by aseipp 3y ago
Note: that isn't the same thing as what the OP describes, at least according to those release notes, but it does fall under the "HMM" umbrella. You still need to specifically allocate your memory with hipMallocManaged before
443.
by aseipp 3y ago
I think that's still the case; though I assume you're talking about Linux-on-Linux vGPUs, the same is true of e.g. WSL2 where unified memory isn't supported. Sucks, because it's a great feature.
444.
by aseipp 3y ago
You could already do that with Unified Memory which has existed for a while and IIRC supported paging and swapping, assuming you `cudaMalloc` and `cudaFree` appropriately for your allocations. This is not a change to "features" bu
445.
by aseipp 3y ago
The advantage is simple: Git submodules suck and are a chore to manage for any dependency that sees remotely high traffic or requires frequent synchronization. As the number of developers, submodules, and synchronization requirements increa
446.
by aseipp 3y ago
Because those people can just keep using older versions of the compiler if they need binaries to run on those platforms (easier than ever with tools like rustup), and because actually maintaining things like build farms and testing a platfo
447.
by aseipp 3y ago
I think it's more that the compile was unconstrained, so no floorplanning or handholding synthesis anything (aside from the usual bullshit to e.g. make sure DSPs map correctly). Which is pretty good; yeah, the architecture is also good
448.
by aseipp 3y ago
The jj daemon thing is something I've had on my mind to maybe hack up, but in practice it's not that huge of a deal I've found. It is worth noting that jj is designed as a CLI and a library. So, for the hypothetical IDE integ
449.
by aseipp 3y ago
Realistically, Jujutsu is a pretty new project and there are many features to add and lots and lots of code that needs to be written. There is a lot of talking that needs to happen, in other words. So, most of the developers and committers
450.
by aseipp 3y ago
Personal anecdote: linking a debug build of buck2 (big Rust codebase) went from 30s to 3s for me. Pretty wonderful.

More ›