Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A preview of x64 emulation on Windows for ARM

blogs.windows.com

Post reply on HN
Print view View original

291–300 of 418 posts

Re: A preview of x64 emulation on Windows for ARM

#291

Post by annilt »

sharpneli wrote: post #124
oblio wrote: post #67

Earlier quoted context omitted.

Well, two main reasons: 1. Innovator's dilemma: https://en.wikipedia.org/wiki/The_Innovator%27s_Dilemma ARM in this case is the underdog, attacking the incumbent x86. It is the lesser tech because it started from "below" (lower value niches not taken by the incumbent which prefers higher profit margins). 2. RISC vs CISC has never been settled. Until Apple (and Amazon super recently) produced their super recent archit…

Another major difference is the memory model. In X86 other CPU’s must always see the writes of a core exactly in the right order. This limits the ability to reorder store ops significantly. ARM requires a memory barrier for this. This is a major reason why X86 emulation is so slow. One must basically issue a memory barrier after every store op. M1 actually also implements the X86 memory model too in HW. It’s only usa…

Just want to add one thing, x86 has stronger memory “semantics”. So, it doesn’t have to work that way behind the scenes, just at the end of the block, it has to appear it worked that way. So, x86 does reordering, store combining etc a lot. IMHO, performance difference between arm vs x86 barely related with ISA, in M1 case, it’s definitely not, a lot more going on than just taking advantage of weaker memory model.

Re: A preview of x64 emulation on Windows for ARM

#292

Post by vidarh »

ChuckNorris89 wrote: post #272
anthk wrote: post #270

Earlier quoted context omitted.

Not in Spain.

Lucky you. Still, it's not a UK only tax, as you originally claimed.

They're not so lucky. In Spain the public broadcasters are paid from government grants from general taxation, so any tax payer pays towards it whether or not they have a TV and watch it.

In any case, you're right, most of Europe has TV licenses of some kind or other [1]

Thought notably in the UK, you can certainly own a TV without paying TV license [2]

[1] https://en.wikipedia.org/wiki/Television_licence

[2] https://en.wikipedia.org/wiki/Television_licensing_in_the_Un...

Re: A preview of x64 emulation on Windows for ARM

#293

Post by socialdemocrat »

dan-robertson wrote: post #77
oblio wrote: post #67

Earlier quoted context omitted.

Well, two main reasons: 1. Innovator's dilemma: https://en.wikipedia.org/wiki/The_Innovator%27s_Dilemma ARM in this case is the underdog, attacking the incumbent x86. It is the lesser tech because it started from "below" (lower value niches not taken by the incumbent which prefers higher profit margins). 2. RISC vs CISC has never been settled. Until Apple (and Amazon super recently) produced their super recent archit…

The reinvention that gave x86 the advantage over (workstation) RISC was OoO execution in the Pentium. It’s a bit silly to think of x86 and ARM as being so different these days. Most x86 code looks a lot like it was produced for a RISC chip and ARM has been gaining some more complex instructions and addressing modes. Academics may have felt that RISC was better than CISC for a long time but I don’t think they were pre…

But ARM64 instructions are all fixed width still right?

And how specifically are these instructions complex? Do they produce an unusually high number of micro-ops when decided?

Re: A preview of x64 emulation on Windows for ARM

#294

Post by rsynnott »

NelsonMinar wrote: post #242
whatever1 wrote: post #132

Now we need an ARM processor that can compete with M1. Seems that Apple caught everyone with the pants down in the PC industry. Microsoft SQ2 processor that ships in their $1400 Surface Pro X is almost 50% slower than M1.

Also you need the rest of the integrated system-on-a-chip, which no one else has designed and poses fundamental challenges to the PC hardware industry. Also you need to secure 5nm chip fabrication capabilities, which is awkward since Apple just bought all of the world's capacity. Then you get to start working on making the software work.

Eh? Effectively all high-end ARM cores today dwell inside some sort of SoC.

Re: A preview of x64 emulation on Windows for ARM

#295

Post by socialdemocrat »

Wildgoose wrote: post #249
bserge wrote: post #105

Earlier quoted context omitted.

Wasn't that simply because there was less software available for ARM? And the first ARM chips were quite anemic in performance. That's not the case anymore, but the stereotype will persist for a while.

The first ARM chips were blazingly fast and made their competitors performance look anaemic - they have been around since the mid-1980s. The real difference is they went the low-power route while retaining as much performance as possible, rather than depending on desktop power supplies and industrial cooling. One of the reasons this new chip is so performant is the "headroom" that this approach has given them.

I wish I understood better why Acorn Archimedes did not do better in the market. It looked like they had many great computer models.

Re: A preview of x64 emulation on Windows for ARM

#296

Post by rsynnott »

0xffff2 wrote: post #255
DaiPlusPlus wrote: post #50

Earlier quoted context omitted.

No, it's patent-law, not copyright law. Interestingly while the US has Fair Use (which is an open-ended legal defence), I don't think any countries in the EU (or "Western Europe" if Brexit happens) have anything beyond the closed and enumerated "Fair Dealing" defence).

>or "Western Europe" if Brexit happens Wait a second, one of us is behind the times. I thought Brexit happened earlier this year?

Well, that depends what you mean by Brexit :)

The UK ceased being in the EU in January, and transitioned into a temporary EEA membership, subject to most EU law. That comes to an end in a few days; at that point what happens next depends on the outcome of negotiations. There's certainly a world in which EU patent rules continue to apply to the UK; in fact, that would probably have been the default assumption until the British government went off the deep end.

So, if you use 'Brexit' to mean purely the legalistic act of leaving the EU, then yes, it's happened. However, effectively none of the _effects_ of Brexit have happened yet.

Re: A preview of x64 emulation on Windows for ARM

#297

Post by rsynnott »

kalleboo wrote: post #35
scoopertrooper wrote: post #15

Earlier quoted context omitted.

While that's true, Intel has steadily extended x86 over the years and many programs depend on those extensions, which are still patent protected.

Rosetta 2 doesn't support the SIMD extensions AVX (2011) and newer, but compatibility still seems pretty good

Very little software would _depend_ on AVX being there; it'd be too great a limit on backwards compatibility.

Re: A preview of x64 emulation on Windows for ARM

#298

Post by slashdev »

px43 wrote: post #278
slashdev wrote: post #234

Earlier quoted context omitted.

Can I boot Linux on an iPhone? Install android instead? I don't think we ended up in a place much better off from a freedom perspective. Android is a little better because parts of it are open. Google has been steadily moving things into play services though, which is not open. There is now an open source replacement for play services - I don't think it's very compatible yet though.

> Google has been steadily moving things into play services though, which is not open. This is a good thing, they've been de-googling the base OS and making it as stripped down and bare-bones as possible. This is a huge win for security, privacy, and freedom in general, and exactly what they should be doing. Android should be a small, secure Linux based operating system with an API that makes it easy to build service…

Fair enough, but android without play services is also barely usable and most apps will not run.

It's also an effective way of building the walls around the garden.

Re: A preview of x64 emulation on Windows for ARM

#299

Post by andy_ppp »

acdha wrote: post #227
andy_ppp wrote: post #56

Earlier quoted context omitted.

This is because very little uses AVX instructions, hasn't it always been better to optimise software for the GPU rather than AVX instructions, even though they are not perfectly interchangeable.

You might be thinking of the most recent AVX-512 extensions which had significant power management complications delaying adoption. Many common programs have optimized SIMD paths because GPU computing is significantly more complicated and the setup overhead makes it slower for uses which don’t process tons of data - one of the interesting points about the M1 is that the unified memory design avoids the overhead of co…

Yes sorry so am thinking of AVX-512 you are correct.

Re: A preview of x64 emulation on Windows for ARM

#300

Post by AsyncAwait »

Hamuko wrote: post #244
AsyncAwait wrote: post #180

Earlier quoted context omitted.

Bootcamp doesn't run on ARM Macs so your point is moot.

That's not really because Apple hates allowing you to install Windows, but rather because you can't actually acquire a Windows for ARM license. Once again, Apple has stated that Bootcamp for Apple Silicon is "up to Microsoft". And ARM Macs are still just a minority of the computers that Apple sells. The rest have Bootcamp, which allows you to just turn your Mac into a Windows machine. Saying that "Apple will never al…

> but rather because you can't actually acquire a Windows for ARM license

You'd still have to develop all the device drivers required for this custom SoC. After all, Linux doesn't need a license but without extensive documentation and support from Apple it's like saying you're free to challenge your boss openly, it's pedantically true, but it is impossibly impractical for most.

Unless Apple provides extensible documentation for 3rd party drivers to be developed for their SoCs it's just cheap talk from their side.

Post reply on HN