hsivonen

14 ms·

121.
by hsivonen 8y ago
It'll be interesting to see the numbers again after LLVM thinLTO starts applying across C++ and Rust resulting in cross-language inlining.
122.
by hsivonen 8y ago
You don't need a warranty to get kernel security patches _in practice_ if you run a mainstream distro on x86_64 hardware (which is available passively-cooled with low power consumption).
123.
by hsivonen 8y ago
ARM-based products may be stuck on whatever kernel snapshot the board support package had. With x86_64 one can expect proper kernel security updates.
124.
by hsivonen 8y ago
Passively-cooled Apollo Lake x86_64 computer (e.g. from Compulab) with wifi and two Ethernet ports running a mainstream distro with proper security updates (e.g. Ubuntu). Self-configured dnsmasq, ufw and hostapd.
125.
by hsivonen 8y ago
It may make sense to use non-English words to name stuff that is about local legal concepts in enterprise software.
126.
by hsivonen 8y ago
Bootstrapping is done rarely but ongoing compiler development is done all the time. It seems like a bad idea to optimize for bootstrapping at the cost of not being able to use the best Rust has to offer in compiler development today.
127.
by hsivonen 8y ago
PDF is an unusual format in the sense that it had a rather specific thing it tried to do and then it achieved that goal, so that it could be considered "done", but the product it was most associated with, Acrobat, tried to expand
128.
Encoding_rs: A Web-Compatible Character Encoding Library in Rust(hsivonen.fi)
3 points by hsivonen 8y ago|0 comments
129.
by hsivonen 8y ago
The GTK+ and GStreamer bindings for Rust are already mostly generated code (and seem to work very well): https://github.com/gtk-rs/gir/blob/master/README.md I'm not sure what the exact status of ext
130.
by hsivonen 8y ago
> I don't know if there is in this case In this case, there indeed is no .c compilation unit between the C++ and Rust code that see each other via C linkage.
131.
by hsivonen 8y ago
> who would analyze code that was formerly a part of Firefox? People looking for bugs in Firefox ESR.
132.
by hsivonen 8y ago
> e.g. removed C++ code that has later found to have bugs The article links to a longer article ( https://hsivonen.fi/encoding_rs/ ) about encoding_rs. The longer article mentions a bug that got fixed in Firefox ESR a
133.
by hsivonen 8y ago
> For starters it needs to have classes Rust doesn't have classes. And as discussed in the article, Rust puts the vtable pointer on the reference while C++ puts it on the pointee, so even fully abstract C++ classes ("interfaces
134.
by hsivonen 8y ago
Correct. The Rust API is recreated in C++ using the corresponding C++ facilities with a C API in between.
135.
by hsivonen 8y ago
Not a specific target of having to replace C++ for the sake of replacing C++. The way Rust code gets added includes: * A new feature needs an identifiable library, so the new library can be written in Rust to begin with. (Example: U2F token
136.
How I Wrote a Modern C++ Library in Rust(hsivonen.fi)
406 points by hsivonen 8y ago|116 comments
137.
by hsivonen 8y ago
Personally, I'm very disappointed that the higher frame rate of The Hobbit has been interpreted as bad. For 3D, in my opinion, it was a much better experience than using a lower frame rate for 3D. In my opinion, using a lower frame rat
138.
by hsivonen 8y ago
18.04 prompts you very clearly and the first report they let you see looks innocent. Is there something concretely wrong with 18.94 telemetry beyond objecting to even optional telemetry on principle? (I agree that the Amazon search thing wa
139.
by hsivonen 8y ago
On hardware level, all those use von Neumann architecture. Wasm, however, exposes Harvard architecture (implemented on top on von Neumann machines). Yet, you can compile C to Wasm, which is a data point about how abstract C is: so abstract
140.
by hsivonen 8y ago
Very nice! Do you track opt_level=2 (the Firefox Rust opt level) in addition to the default opt_level=3?
141.
by hsivonen 8y ago
Doesn't asm have the disadvantage that you end up spilling if you exceed 8 xmm registers even when running on x86_64 with 16 xmm registers available? Or do you actually write separate asm for x86 and x86_64?
142.
by hsivonen 8y ago
Especially outside yhe largest countries, it's often hard to find replacement parts for computers that are a older than the current generation. What's the counterfeit situation with chips that can be installed in computers post fa
143.
by hsivonen 8y ago
Is there an "explain like I'm 5" intro to FIDO2 that explains how FIDO2 as the single factor is more secure than the user entering username (not password) and then using a CTAP1 token? (Also, how do you log in to multiple acc
144.
by hsivonen 8y ago
Contrary to (at least the 4 series-era) Yubico's marketing materials, their Android app works fine with USB (in fact better than with NFC), so if you have and Android phone with a USB-C port, you can plug in a USB-C Yubikey directly (o
145.
by hsivonen 8y ago
C's standard library is completely broken for text handling (depends on external locale instead of caller-specified locale) and on Windows broken for file system access if your paths contain non-ASCII. (At least until the latest dev ve
146.
by hsivonen 8y ago
Does anyone who isn't a Ubuntu dev themselves get a good outcome from filing bugs on Launchpad?
147.
by hsivonen 8y ago
I paid Canonical for single-seat support for a while years ago. It was useless. The main problem isn't either of the above. It's that there's no way the annual single-seat price can pay for the time of a competent person inve
148.
by hsivonen 8y ago
clang is doing pretty nicely compared to MSVC these days: https://news.ycombinator.com/item?id=17504197
149.
by hsivonen 8y ago
Note that herokuapp.com is a public suffix, so subdomains under it have separate Let's Encrypt quotas.
150.
by hsivonen 8y ago
Does this generation still lack the kind of performance counters that are needed for rr?

More ›