monocasa

I'm a security researcher.

"Garbage" isn't the same thing as random noise.

I have looked at the entropy graphs of encrypted data commonly. It's one of the first things I look at when looking at a new firmware image I haven't seen before.

You didn't generally need to manage the ALAT as a programmer.

You can, in the same way that you can manually manage reservation station port residency as a programmer in an OoO core to get maximum throughput, but neither of those are required in normal programming.

If there's a difference visible, that's a weakness in the encryption algorithm.

Encrypted data should be statistically indistinct from random noise.

It's an information theory thing. Everything in the ciphertext should be sitting at maximum entropy.

Once again, Itanium isn't a normal, simple in order core that stalls on loads. It has a big table called the ALAT to allow you to start loads as soon as you know the address, and then finalize them later when you're out of other work to do.

It doesn't require determinism to work. And was designed by people that were quite aware of what the instruction stream looks like to an OoO core as it issues out of the rob.

Itanium had other sins than "magic compiler" woes, or even the inherent unpredictability of memory accesses. Mostly that it, like Cell, and Netburst was designed for a world where dennard scaling didn't end like a brick wall. As well as internal politics of Intel making it so that they were a bit loose and fast with die area.

I've heard about mathematicians going through kind of the same thing when they get a weird proof that ends up being right from some weird source or themselves.

Which is fair, they get inundated with kooky proofs from amateurs all the time and odds are incredibly good that there's some major fatal flaw that the amateur doesn't see. Or in the case of themselves, there's a certain blindness that makes it a little more difficult to critically evaluate your own leaps. In ether case the way it manifests is by going over it many times and many ways, each time more certain that you missed something until you just kind of break. Only then do you publicly start suggesting that there might be something to this new leap.

OoO cores have load stalls too after preemptions. Even something like a Apple M core basically has to stall if the load has to go out to the memory controller. The goal is to move far enough ahead in the stream that you can at least issue the next load, and advaned loads lets you do that. Some Intel cores will occasionally speculatively predict zero for a load, but AFAIK that got turned off as part of the Spectre mitigations.

Itanium had all sorts of specialized machinery so that you could issue a load that's required in the future, and then keep chugging away at the instruction stream, similar to what an OoO processor does with it's reordering logic.

One example was "advanced loads" which allowed you to issue a load as soon as you knew the address, even in the face of potential pointer aliasing in the future, and then later complete the load when you actually hit a data dependency that requires it. https://devblogs.microsoft.com/oldnewthing/20150805-00/?p=91...

Another example is "speculative loads" which lets you issue a load before you even know if it's valid, such as unrolling a loop for an array that you don't know if is a multiple of the unrolled loop chunk length (and therefore might trap with a page fault if fully resolved). https://devblogs.microsoft.com/oldnewthing/20150804-00/?p=91...

All the more reason to treat it as a commoditize your complement situation.

There's also at least the galactic orbit. There might be a very large scale orbit as well around the Great Attractor, but the jury's still bery much out on that one.

Anthropic isn't a hyperscaler, but instead a hyperscaler customer.

And I've seen first hand hyperscalers go to extremely large lengths to eradicate any use of Oracle (which mainly comes in these days through their acquisitions).

I think the hyperscalers are smart enough to not let Oracle be their landlord.

Yeah, LVGL is probably what you'd want in that environment.

On x86, basically whatever XSAVE writes out.

FWIW, I wouldn't be surprised if you only have a couple threads using this at a time max since it looks an awful lot Apple's AMX/SME stuff. Those Apple execution units only have single engines shared about per cluster.

XSAVE lets you not bother saving register state that user space hasn't changed at the granularity of each large feature.

Early utilities tend to not be built on standards. See how early electric utilities within a country or even a city couldn't agree on voltage, frequency, phase, or even AC vs DC.

Most utilities are monopolies.

AWS is essentially a utility and gets way better than 3%.

Microsoft never shipped Alpha support for win2k in the release builds, but only the betas and release candidates, so I doubt anyone ran it in "production".

If the cost to orbit does get low enough as is the assumption in this article, I would be shocked if we didn't start to see amateur space based telescopes.

They're talking about Doom, not Dune.

You're calling it CoWoS, but also said "this shift eliminates the need for silicon interposers", but also said "CoWoS uses a passive silicon interposer"

Something's got to give here. I think it's your original article that's wrong, and it's poorly trying to describe InFO-M.

So make the union for-profit instead? In other words the distinction isn't corporations vs unions, it's for-profit vs non-profit. This seems.. okay? What's the alternative, that organizations can get tax breaks and do whatever politicking they want?

The alternative is neither get to do politicking rather than only those who aren't formally working for some public good.

Walking off the job when you have an agreement not to walk off the job is "protected speech"?

Yeah, it's called a protest.

They're allowed to walk off for their own issues. It's solidarity strikes that are banned. Ie. the nature of the speech is what makes it illegal.

Doesn't that apply to corporations as well? If you click on a random company[1], you see a notice that says

No. For profit corporations can donate to PACs that are directly in support of a candidate rather than a broad swath of policy. It's a non profit versus profit distinction that creates this additional ratchet.

Are you sure that the reason why solidarity strikes are banned because union members are holding up a bunch of signs, and not say... walking off the job?

How is that not one of the most fundamental forms of protected speech?

Weaving was very much women's work. There are some exceptions that prove the rule, but society's labor expectations were extremely delineated along gender boundaries for adults in the middle ages. There's so many linguistic examples of this persisting to today. Another example being "spinster", ie. that an unmarried adult woman was expected to spend her time spinning thread.

And we don't have nearly the same amount of detail in women's work. That's why even today it's referred to as invisible labor.

???

It's complex, but those are not donations to a candidate, even more so than the normal PAC and Super PAC song and dance. You have to give up your non profit status to donate more directly to a candidates campaign.

Seems like a stretch to lump industrial action with political donations.

Seems like a stretch to say that political donations are speech and should be protected, but literal picketing isn't.

They already have. Unions generally can't donate to political campaigns, and can't do things like strike in solidarity with other unions which would be pretty clearly be speech if we're counting corporate donations to political campaigns as speech.

The point is that the data we have is on men's work. No one said that women were bored SAHM hobbies. Just that pointing out extra tasks that need to be done in addition to the record we have of men working 150 days a year doesn't make much sense when that additional work wasn't done by men.

It's still work. It doesn't magically have zero value because it was done by women. The methodology cited by the book counts only counts hours worked if it's farming or wage employment, but doesn't count any household labor, which means anything that's shifted from being made at home to being bought would could as extra hours worked, even if it saves time overall.

Nobody said it had zero value. The point is that labor roles were extremely gender striated, and the record is mostly kept only for men's labor. So saying that they worked more because of the example task you gave doesn't make sense when societally that work was put onto others.

The students and researchers weren't the customers, but instead the large institutions they belong to. In a sense, they should be the same from this context, but the amount of consolidation that has occurred in the academic publishing space means that large institutions don't really have the same plausible deniability that their individual members do.

Except that kind of work would have been split up by gender. The etymology of "woman" is literally "weaving person", and those kinds of tasks were up to the mostly adult women who also wouldn't have been working in the fields, so it ends up being a wash when painting with a wide brush.

As your article states, they're already using CoWoS; that's not a new technology for Apple to adopt.

Also, InFO-oS is a CoWoS technique.