ljhsiung

13 ms·

61.
by ljhsiung 5y ago
Even putting aside security aspects aside, in general I've been seeing research pop up over the years criticizing SMT's performance claims of ~30%. Hell, even Amazon's Graviton CPUs don't have it (though I'm sure th
62.
by ljhsiung 6y ago
Two economists are walking down the street. One sees a $100 bill on the ground and says, "Hey, isn't that a $100 bill?" The other laughs-- "Of course not. If it was, someone else would have taken it by now." While I
63.
by ljhsiung 6y ago
I think there's 2 arguments here. One is, as you pointed out, binge drinking is bad for you. I think most people would agree. The second argument, primarily in the second half of the article, is that creating the conditions to easily p
64.
by ljhsiung 6y ago
Also worth noting is that several large tech companies are on <100 feet away from her apartment, literally across the street. AMD, Hitachi, Applied Materials Again, not doubting her story. I very much think her problem needs to be addres
65.
by ljhsiung 6y ago
-- "now let me sit in my owner's pants as he sits on the toilet"
66.
by ljhsiung 6y ago
>> One can simply "check off the list" of attacks >> The more difficult/interesting problem is being future-proof
67.
by ljhsiung 6y ago
Run this [1]. If it works, flush your caches or just update your kernel. If it doesn't, you checked off an item on your list. I misspoke, silicon should be more like "system". Said it in this comment more about silicon fixes
68.
by ljhsiung 6y ago
Incredibly doubtful. Current attacks are, by this point, pretty well understood. One can simply "check off the list" of attacks and simulate the code gadgets to see what happens (EDIT: e.g. https://github.com/IAIK&
69.
by ljhsiung 6y ago
Current mitigations are still largely software based, and frankly, aren't fundamentally any different than the ones proposed just a couple days after 1/3/2018. Additionally, current academic literature still hasn't add
70.
by ljhsiung 6y ago
Google built one of their TPUs in Chisel [1]. TL;DW: Chisel is beautiful/fun to write in, with a definite productivity bonus, but has a pretty large learning curve and had a much greater verification cost, partly because it's an H
71.
by ljhsiung 6y ago
That exact link is already included.
72.
by ljhsiung 6y ago
Their link to "Computer Architecture: A Quantitative Approach (5th edition)" ( https://booksite.elsevier.com/9780123838728/references.php ) is broken. Here's a couple more [1],[2]. I'm also surprised
73.
by ljhsiung 6y ago
Firstly, this isn't direct at you, more at the title. I think the title of "47 events" is greatly misleading. While the simulations are definitely incredibly valuable, it's misleading to use the figures from simulations
74.
by ljhsiung 6y ago
Couple things-- 1) You underestimate how critical prototyping has become, again likely since you say it's been a couple decades. Time to market has become more important, and verification has become harder as CPUs have gotten even more
75.
by ljhsiung 6y ago
True, but also note that a 52b mode (i.e. 12 bit tag) is also supported. I can't imagine anyone needing more than 4PB in the next 30 years, and the security gains now is (in my opinion) worth the potential refactoring in 30+ years th
76.
by ljhsiung 6y ago
When I ask the question, I do it out of genuine curiosity. I like my potential coworkers (since that's kinda half the point of interviews-- see if he/she is a feasible coworker) to be easy to talk with. I've had interviews wh
77.
by ljhsiung 6y ago
The ISA design has very little impact on actual energy consumption rather than an actual CPU's uArch. This is a conception that hasn't been true for at least 20 years. It was somewhat true in the days where memory was expensive,
78.
ARM server startup NUVIA raises $240M in Series B(anandtech.com)
2 points by ljhsiung 6y ago|0 comments
79.
by ljhsiung 6y ago
Any particular reason this was article was posted on multiple sites, and multiple URLs with their dates redirect to this one? Seems a little weird that it "was posted" today, when really it was posted multiple times. Yes, I realiz
80.
by ljhsiung 6y ago
I'd say the fact that 1) it's very lightweight/portable and 2) a bit more constrained in terms of hardware. I can't open 50 tabs and watch YouTube in the background. Additionally, I have some side projects on ARM code th
81.
by ljhsiung 6y ago
It's pretty astounding to me how much progress that Pine64 has made in the last year. I just got my Pinebook Pro last week (and typing this comment on it), and funny enough, I've been more productive on it for side projects than I
82.
by ljhsiung 6y ago
I agree that it was amazing and super cool! Rather it's just a comment on the relevance of the name to the actual attack, contorted to the purpose of a cool acronym. Just to stoke some fires, I think poodle is also in that bag.
83.
by ljhsiung 6y ago
I thought heartbleed was a decent name. A callback to the heartbeat protocol I think DROWN was kiiinda pushing it.
84.
by ljhsiung 6y ago
Did you read the article? > It is an essential hormone, definitely, but not a magical pill that will ward off disease. There are conditions that benefit from Vitamin D supplementation - for hypoparathyroidism, a condition that can cause
85.
by ljhsiung 6y ago
I think removing the capital gains tax is a horrible idea. Firstly, there are 2 types of capital gains. Short term gains are just taxed as normal income and long term gains are taxed from 0,15,20% depending on your bracket. The whole purpos
86.
by ljhsiung 6y ago
It might also depend on how AMD clears this temporal register across contexts, and the method it's updated. The first I could think of (in the 5 minutes I've read this) is that it could be potentially ASLR breaking. As Agner says,
87.
by ljhsiung 6y ago
I think the jury's still out on whether the link is causal or correlated, and that study in particular is under controversy. Personally, I think its just because higher Vitamin D levels mean you generally go out more, which is associat
88.
by ljhsiung 6y ago
Minor nitpick, but "three years ago" the crypto world still did "worry" about Spectre and general transient execution attacks [1][2][3][4]. To me, the biggest thing about Spectre et al. wasn't the proof that CPU uAr
89.
by ljhsiung 6y ago
It's more about false pretenses, and thereby racism (or, at best, massive negligence and lack of self awareness) I could agree that more recruiting resources should be dedicated towards higher ranking schools, but food scraps and half-
90.
by ljhsiung 6y ago
If a company visits a school, I 1000% expect the same treatment as any other school. To provide special treatment is elitist at best, and racist and exploitative for PR points at worst. And if the entire school doesn't pass the "h

More ›