swetland

8 ms·

31.
by swetland 5y ago
Probably challenging given that the GBA's ARM7TDMI core runs at about 17MHz. Perhaps doable if the graphics hardware as accessed from the ARM core can operate in the same modes the Z80 core uses (I forget if it's a strict superse
32.
by swetland 5y ago
Pointers are 128bit (to be large enough to contain the range, permission, type, etc, as well as the address), so existing code will at a minimum need recompilation, but likely much C/C++ code that makes assumptions about sizes of point
33.
by swetland 5y ago
I'm a bit puzzled by "reducing binary footprint (surprisingly) by removing bloated dependencies" since linenoise has no dependencies (beyond libc) that I'm aware of. $ size linenoise_example text data bs
34.
by swetland 5y ago
Sure, but it made a lot less sense on a device with a much faster data network and full featured web browser... even so carriers had their lists of all the features required for phones to be certified on their networks and it was very fea
35.
by swetland 5y ago
At the time (very very early) there was only a small amount of code and small amount of native bindings, so it only took a couple days to rebuild the lower layers and I recall Chris got the "framework" running again in js pretty q
36.
by swetland 5y ago
I dunno, touch-only had been done before (Newton, Palm), and I think it would have surfaced again. The UI/UX Apple built on top of the technology that was in everyone's (manufacturers) hands, I still think, was the biggest factor
37.
by swetland 5y ago
Yup, Apple made software and UX important. And had the clout to ship their device with their software and ignore the carriers' obsession with random checkbox features. Which, in the end, helped us a ton with Android -- post iPhone ann
38.
by swetland 5y ago
Fun fact: The original, original demo (pre-acquisition) was Lua based. Andy was skeptical that enough people would know what Lua was so I shifted it over to Javascript. So for a while we were Javascript + 2d render engine. Smells a bit l
39.
by swetland 6y ago
Definitely some rough edges, and like most CAD systems it has a pretty steep initial learning curve, but I've been surprised at how useful it is. Finding a recent youtube tutorial or two can help with getting started (sadly the wiki&#
40.
by swetland 6y ago
I'm not aware of anybody reproducing the exact softdevice design, but there are open source BLE stacks out there that run on the nRF528xx family, for example: https://github.com/apache/mynewt-nimble
41.
by swetland 6y ago
Congrats on the promo! It didn't retain me (though I didn't get promoted to Principal, so maybe I don't count) but I did enjoy working on Fuchsia. Nice to see much of the syscall design seems to have survived.
42.
by swetland 6y ago
Two observations about "unsafeness" of your threading system: 1. As a matter of paranoia I would likely write schedule() entirely in assembly -- it's a simple function and if you write it by hand you can be certain the compil
43.
by swetland 7y ago
You're incorrect. This document is describing languages used to develop the platform itself (and a bit about priority of external support for languages). The Fuchsia platform was designed to be language agnostic, bring-your-own-runti
44.
by swetland 7y ago
This document is about the use of languages within the project itself. Fuchsia, as a fundamental principle , supports "bring your own runtime" -- if you are a PIE ELF executable and can dynamically link against libzircon.so (the
45.
by swetland 7y ago
Apart from having a pretty generous chunk of RAM and flash, the '840 is a 64MHz Cortex M4F -- there's a pretty huge variety of ARM M MCUs out there, especially if you don't care about the radio.
46.
by swetland 7y ago
Nordic provides binary blobs for BLE, but also provides full register level documentation for the radio peripheral, so there's nothing in the way of an open source radio solution with the NRF52 family parts. Here's one: https:&#x
47.
by swetland 7y ago
The only hardware that I know of that they've EOL'd are their original (pretty clunky) remotes. I have a ZP100 from 2005 still happily playing music in my home office alongside more modern units elsewhere in my home. I am annoyed
48.
by swetland 7y ago
It's always fun seeing different approaches to these things. I've been tinkering with a simple RV32I/RV32IM simulator lately -- less fancy, not bothering with ELF loading or emulating linux syscalls or the like yet. I did h
49.
by swetland 7y ago
If folks want to experiment with a tiny, bare-bones unix-y OS on relatively modern platforms, xv6 (used as a teaching OS at MIT) is ~10kloc of C for the kernel and a handful of userland programs: https://github.com/mit-pdos&
50.
by swetland 7y ago
The Datasheet has a diagram indicating that but the User Manual claims there's a bus matrix which provide SRAM connectivity from the CPU I/D ports and DMA units. The documentation is... not amazing.
51.
by swetland 9y ago
That's a bit outdated (magenta was renamed zircon a while back and that mirror is overdue for deletion). Try: https://fuchsia.googlesource.com/zircon/+/master/docs/syscal... https://fuch
52.
by swetland 9y ago
Verified boot support will arrive prior to 1.0, and as an earlier reply speculated, leaving the ipv6 boot path enabled at all is a development feature.
53.
by swetland 9y ago
Some more detailed info here: http://zircon.fyi/syscalls.md http://zircon.fyi/concepts.md The PCI syscalls and a lot of the other DDK scaffolding stuff will be going away in time (and are not accessible to
54.
by swetland 9y ago
Yes, handles are partitioned per-process. Having them be unlikely to be reused or collide with small numbers (locally) helps avoid some common failure cases (somebody uses a fd after close but somebody else has opened something else that g
55.
by swetland 9y ago
The creation syscalls operated under the execution constraints of the Job in which your Process is contained. The APIs aren't fully landed, but it will (very soon) be possible to create Jobs where-in none of the kernel object creation
56.
by swetland 9y ago
The sample output on the github page looks a lot like the sort of information you get from "readelf -a ...", but colorized.
57.
by swetland 9y ago
Yeah, the evolution of Windows and OSX towards integrated "App Stores" and less control over app installation bums me out too. I grew up with a C64 and BASIC at the READY prompt, and hope the world of personal computing allowing
58.
by swetland 9y ago
The stats are a little unfair there, as I've done a lot of rearranging and housecleaning, which tends to inflate github's scoreboards. I suspect more of my work is outside the kernel than in (driver model/hosting/managi
59.
by swetland 9y ago
I'd suggest that it's possible we're using simple, but functional placeholder implementations of things so the entire system can boot and run, and then various parts of it can be worked on at various paces, in parallel, etc.
60.
by swetland 9y ago
Most Android devices also run some OS on the baseband processor (on-die or separate), and often the Wifi, Codecs, Bluetooth, Power Management blocks, etc also run their own little OSes. It's getting a little crazy, how many distinct c

More ›