bonzini

16 ms·

361.
by bonzini 1y ago
Do I understand correctly that it's brute forcing a small grid rather than learning the algorithm?
362.
by bonzini 1y ago
Yes the code for the alternative is awful. However I tried rewriting it with a better alternative (basically the same as the lock free code, but with a mutex around it) and was still 40% slower. See FixedVec in https://play.rust-
363.
by bonzini 1y ago
Yes, as I said I am looking at it exclusively from the point of view of replacing C for new code in existing programs, with a smaller cost than for example adding safe Rust bindings.
364.
by bonzini 1y ago
Anything that produces object files is a lot harder to debug and maintain than a transpiler.
365.
by bonzini 1y ago
It doesn't even distinguish integer and floating point?
366.
by bonzini 1y ago
There's a lot less complexity in Rust than C++. Compilation times are mostly due to lack of compiled code reuse, which is bad but not unsolvable.
367.
by bonzini 1y ago
You can do it in multiple passes (analysis/dumping IR followed by code generation, with dep files to connect them). It's what Rust does internally with rmeta files. Mine is not necessarily a criticism of Zig, mind. It's a cri
368.
by bonzini 1y ago
Back when it still had async support, I looked at using Zig in QEMU. My main qualm with it is that it doesn't try to integrate with existing C programs. The ideal C replacement for me would be a two-way per-file C transpiler, in the sa
369.
by bonzini 1y ago
It gets rid of the clock bit, synchronizing the two CPUs so that the 1541 can transmit 2 bits at a time. It was still slower due to GCR decoding. Over the last 10 years people have started to write streaming GCR decoders too.
370.
by bonzini 1y ago
> Microsoft adopted the most popular one That's still consolidation, and it also needs time. Even in Rust crates like hashbrown or parkinglot have been basically subsumed in the standard library.
371.
by bonzini 1y ago
More time enables more consolidation.
372.
by bonzini 1y ago
It's the same that we're doing for external crates in QEMU's experiments with Rust. Each new dependency is added to the build by hand.
373.
by bonzini 1y ago
That's true, but the fact that NSA wanted to make brute force cheaper also suggests that they didn't have any particular offensive tricks up their sleeve (they had differential cryptanalysis but they used their knowledge defensive
374.
by bonzini 1y ago
In the early 2000s I was working in a webdev team where editors worked on the raw HTML (some with the help of Macromedia Dreamweaver, some simply with Notepad++) and pushed it to the httpd root via SFTP[1]. None of them had a developer back
375.
by bonzini 1y ago
It's worth noting that when the NSA invented DES, they took a cipher from IBM and made it more resistant (to differential cryptanalysis, a technique that at the time wasn't known outside the NSA itself).
376.
by bonzini 1y ago
If you program a register at a moment and in a way that causes two signals to "collide", the result effectively depends on transistor behavior. That in turn can be temperature dependent. For an example on the PC see https:/&
377.
by bonzini 1y ago
Nothing prevents you, you just get uglier code and more possibility of confusion.
378.
by bonzini 1y ago
How does that compare to "this error for libraries and anyhow for applications"?
379.
by bonzini 1y ago
Those are tests, it seems. The actual sources are at https://github.com/OFFTKP/felix86/tree/master/src/felix86 and the disassembler is https://zydis.re .
380.
by bonzini 1y ago
You'd be surprised, this is one of the ways in which Red Hat stays independent from their mothership. IBM prefers to donate to OSUOSL and Red Hat uses it for Fedora. Then there are private labs, internal to Red Hats, that are used by d
381.
by bonzini 1y ago
> Could they collect every phone before school and hand them out after? Sure.. but that's a nightmare. I am pretty sure that no one would know if she just puts it on silent or airplane mode? But in any case my son's class did t
382.
by bonzini 1y ago
So called "Liberation Day" and consequent mini-crash was on April 2, so this is just based on perception (of stupidity) rather than an anything real. No matter how well-grounded that perception is, the actual reaction to the tarif
383.
by bonzini 1y ago
As an Italian who has tried (only) the Israeli Bamba, I can certify that it is pretty addictive.
384.
by bonzini 1y ago
Hardening focuses first on not damaging equipment and second on providing energy. If things go wrong quickly enough you don't have time to react, because after a power plant disconnects you get sudden bumps in load that can trigger a c
385.
by bonzini 1y ago
> How about an X11 config file? Nginx? Puppet? Bind? Fstab? Vimrc? Rsyslog? Netrc? Cups? Pppd? Iptables? Apt? Cron? Sysctl? SSH? X11, ssh, Cups are 100% section-key-value or key-value and could be served by TOML easily. Some of these are
386.
by bonzini 1y ago
> These days TOML (which more or less just works) just works The second "just works" should have been "is almost always enough".
387.
by bonzini 1y ago
> When config files were written not to be easy to parse, but to make it easier for human beings to configure software Config files have always been a variant of key-value or section-key-value, except that we used to have ad hoc (and pro
388.
by bonzini 2y ago
Release candidates are just time-based for many projects. For Linux, in addition, the rhythm of stabilization can be different for various subsystems.
389.
by bonzini 2y ago
GCC 2.96 lasted a year or more and even after GCC 3.0 was released it wasn't able to compile a working kernel. This lasted two weeks and the issue is just a new warning; it's just bad timing across the release cycles of two projec
390.
by bonzini 2y ago
Yeah, admitting he's wrong is certainly not his strong suit. He will do so years down the road but not in the heat of the argument.

More ›