Live data from Hacker News

Temptation of the Apple: Dolphin on macOS M1

dolphin-emu.org

321–330 of 390 posts

Re: Temptation of the Apple: Dolphin on macOS M1

#321
post #234

Earlier quoted context omitted.

> How would you prevent this from happening? Very simply - enforce security permissions at the OS level, rather than the app store level. There's no technical reason an app store also has to handle permissions. Leave the discoverability/reviews/curation functionality in the app store, and then just move the app installation functionality into the OS - the app store delivers an app package which the OS accepts, parses…

"Facebook wants access to your address book: In order to view your facebook timeline and newsfeed, we need access to your address book. Please allow access. Allow/Deny" There's no automated capabilities-based OS-level permissions model that can protect against this. Accessing the address book is a legitimate app request - just not for Facebook Inc. in my opinion. But they can gate access to your timeline and friends…

Deny shouldn't break the API, it just means the app gets garbage data instead.

Re: Temptation of the Apple: Dolphin on macOS M1

#322

Earlier quoted context omitted.

I am absolutely not an expert on microarchitecture, but I’ve had the same questions and tried my best to figure out answers. Here’s my understanding of the situation: > It isn't just because it is RISC, it's Apple magic? It’s both. We’ve known for decades that RISC was the “right” design, but x86 was so far ahead of everyone else that switching architectures was completely infeasible (even Intel themselves tried and…

> We’ve known for decades that RISC was the “right” design, but x86 was so far ahead of everyone else that switching architectures was completely infeasible All the experts I listened or read to, they told that instruction set doesn't matter and it is the insignificant thing. The part that matters is branch and data prediction, and caching. Also, even intel transforms an instruction into RISC like microinstructions i…

> All the experts I listened or read to, they told that instruction set doesn't matter and it is the insignificant thing. The part that matters is branch and data prediction, and caching. Also, even intel transforms an instruction into RISC like microinstructions internally.

I've heard this before, but I've also seen sources which indicate that x86 instruction decoding is definitely a bottleneck [1-5]. The M1 has a significantly wider pipeline/OoO window/reorder buffer than any other processor, and most sources seem to agree that this is because the simplicity of the ARM ISA allowed Apple to build an 8-wide instruction decoder (as compared to around 4-wide for x86 chips). [1] also mentions that Apple's impressive branch-prediction capabilities are at least partially because ARM's 4-byte-aligned instructions greatly simplify the design of the branch predictor.

So yes, it's true that an x86 processor really runs RISC-like uops under the hood. However, the best out-of-order execution pipeline in the world is limited by how far ahead it can see, and that depends on how fast the instruction decoder can feed it instructions.

Once again though, I am not a microarchitecture expert. I just read bits of information from people who do know what they're talking about and try to form it into a coherent mental model. If you have knowledge or sources that disagree with me, I would be happy to be proven wrong :)

[1]: https://news.ycombinator.com/item?id=25264384 [2]: https://www.agner.org/optimize/blog/read.php?i=25 [3]: https://news.ycombinator.com/item?id=26782213 [4]: https://www.quora.com/Why-dont-intel-or-AMD-design-an-x86-CP... [5]: https://www.anandtech.com/show/16226/apple-silicon-m1-a14-de...

Re: Temptation of the Apple: Dolphin on macOS M1

#323

Earlier quoted context omitted.

That restriction is already circumvented with e.g. React Native though, or any kind of webview - companies can change their whole apps without a re-review. They MIGHT get called out for it if someone reports it, but in theory it's very possible.

Sure; but the WebView component is Apple’s own secured little space, and it seems to be a fairly secure little sandbox.

React Native does not run within a WebView and has access to the full iOS runtime.

Re: Temptation of the Apple: Dolphin on macOS M1

#324
post #44

Earlier quoted context omitted.

Apple’s blanket ban on interpreted code is not arbitrary. It is there to prevent apps circumventing the review process and security model ie. apps pretending to do X during the review process and then doing Y when in use or obfuscating their use of private APIs. Now you can argue these restrictions are unreasonable but many of us don’t want our iPads or iPhones to be like our computers.

> many of us don’t want our iPads or iPhones to be like our computers And many of us do . I would never buy a locked-down piece of hardware like that. But I don't think it matters either way what either side wants, because it's what Apple wants that matters. They want to keep their walled garden's walls air tight, and there are apparently enough people that are OK living in that garden that it works. I'm positive tha…

The market has spoken. iPad revenues are higher than all the other tablets combined.

Re: Temptation of the Apple: Dolphin on macOS M1

#325
post #317

Earlier quoted context omitted.

I think you're right when it comes to security, but in order for that to work Apple would have to admit that checking apps for security issues is not inseparable from imposing rules that are supposed to benefit their own business model. That said, here's a challenge for everybody (including myself) who doesn't like Apple's app store monopoly and side-loading ban: Apple now requires apps to ask for permission before t…

The security comes from the sandbox, Apple's curation is pretty mediocre and they only remove malware once it's popular or if it's extremely obvious during the short review. iOS malware authors tend to publish their binaries as unsigned dylibs and don't need Apple ID accounts so they aren't even banned when they're caught. instead they convince other developers to ship it in seperate apps (there are various ways, mon…

I can guarantee you that the sandbox can be circumvented if you can just run an IPA on the device. iOS has a humongous set of APIs and that attack surface is impossible to protect properly.

Don’t get me wrong, I’d love to have an iPhone I can install anything on - but there is _no_ way I would ever install anything from the open internet on the same device use to read my email or log in to my bank.

Re: Temptation of the Apple: Dolphin on macOS M1

#326

Earlier quoted context omitted.

I still dont get how people pretend Apple cares about privacy. By default Apple devices phone home and collect data on you, and this is not optional, and they will hand that data to law enforcement. They also have the ability to change their data handling policieson a whim since there is nobody holding them accountable. THIS IS NOT PRIVACY. Not that windows or android devices are better as they come out of the box, b…

So what's the deal? Any company that makes users opt out of diagnostic data is just as bad as the companies whose entire business model is predicated around collecting as much data as possible about you and selling it to people whose sole ambition is to manipulate you? Are these two things really equally bad? We can't acknowledge that one of those is considerably more in favor of privacy than the other--they must bot…

It seems like it is. It’s like the relativity of wrong all over again. Two things being wrong does not mean that one of them is not better than the other. I am so tired of this fallacy, particularly in an otherwise well read community like here.

Re: Temptation of the Apple: Dolphin on macOS M1

#327

Earlier quoted context omitted.

I still dont get how people pretend Apple cares about privacy. By default Apple devices phone home and collect data on you, and this is not optional, and they will hand that data to law enforcement. They also have the ability to change their data handling policieson a whim since there is nobody holding them accountable. THIS IS NOT PRIVACY. Not that windows or android devices are better as they come out of the box, b…

You’re in an echo chamber. By default, iPhones do send diagnostic data, but that data can be disabled in Settings. They also upload your files in an encrypted connection (but without end to end encryption) into your iCloud backup, and that can be handed over with a subpoena, but you can disable iCloud backup or simply not pay for it.

> By default, iPhones do send diagnostic data, but that data can be disabled in Settings.

They also ask during setup and regularly after major updates. I know because I refuse every time. Also, quite often it is opt in, with the box to send information unticked by default.

Re: Temptation of the Apple: Dolphin on macOS M1

#328
post #286

Earlier quoted context omitted.

Why would they? It is more work. It is 2 code paths to write, to test and plan for. There is a cost associated with this that will not likely be recovered in any reasonable price difference on the phone versions.

I would be surprised if Apple don’t already have something like this internally…

It's public.[1]

They're just asses about it.

[1] https://developer.apple.com/programs/security-research-devic...

Re: Temptation of the Apple: Dolphin on macOS M1

#329
post #317

Earlier quoted context omitted.

The security comes from the sandbox, Apple's curation is pretty mediocre and they only remove malware once it's popular or if it's extremely obvious during the short review. iOS malware authors tend to publish their binaries as unsigned dylibs and don't need Apple ID accounts so they aren't even banned when they're caught. instead they convince other developers to ship it in seperate apps (there are various ways, mon…

I can guarantee you that the sandbox can be circumvented if you can just run an IPA on the device. iOS has a humongous set of APIs and that attack surface is impossible to protect properly. Don’t get me wrong, I’d love to have an iPhone I can install anything on - but there is _no_ way I would ever install anything from the open internet on the same device use to read my email or log in to my bank.

Most of us don't want to run everything we find on the internet, we want to be able to run one or two apps that Apple doesn't like.

Re: Temptation of the Apple: Dolphin on macOS M1

#330

Earlier quoted context omitted.

Your comment made me realize I’m waiting for the next generation Apple Silicon SoC as a second data point. Even if I don’t buy it, it will tell us something about the expected performance trajectory.

I'm waiting for it too, I want to see how they're going to scale (more processors? faster processors?) as some of the design decisions could be arbitrary or actually inherent (can they make an M2 with more PCI lanes?).

I think this is the more interesting stuff. We know how the trajectory looks from the A-series chips that have rolled out over the last decade. What's currently a concern is if this sort of design can compete with Xeons or Threadrippers, or if it can power multiple displays or have a multitude of USB4 ports. The M1 is great but it's really only great for these entry level devices they've put them in so far.
Post reply on HN