kimixa

39 ms·

511.
by kimixa 3y ago
But it unconditionally calls a memory barrier, which is most of the cost of an un-contended futex-style lock already. And I'd be interested in your ideas on removing the lock - as I can't see any paths that don't change seman
512.
by kimixa 3y ago
It might to the originating company who dealt with said shell company.
513.
by kimixa 3y ago
What are you trying to say using that example? The getenv() function shown calls initenv(), which can (clearly) take a lock?
514.
by kimixa 3y ago
Yeah, a 4k density no more than 27" 16:10 screen would be my dream. I recently moved to 4k 27" from 1440p 27" (both 16:9) and the text clarity is noticably better, but definitely in diminishing returns in clarity and not real
515.
by kimixa 3y ago
And that's just 1080p, quickly adding the survey results shows that only ~27% of people have larger screens than 1080p, so ~73% if people have screens of 1080p or less. Though arguably people on older screens aren't likely in th
516.
by kimixa 3y ago
Yeah, and you could set the process cpu masks at init time in your own code and do the equivalent. Hell, Intel could add that init step as a flag for their compiler. As mentioned elsewhere, I'm not sure this is hardware design being fo
517.
by kimixa 3y ago
Generally it seems things that benefit from super wide simd also benefit from threading. If productivity apps want to leave performance on the table, so be it. But really I think the entire ecore/pcore split for avx512 is academic, as
518.
by kimixa 3y ago
Executable formats like ELF and PE already allow pretty much arbitrary tags - just define one that says "This executable uses instruction set extension X" and you don't even need the cost of a trap or modifying the file/
519.
by kimixa 3y ago
It's also a massive PITA trying to debug or profiling performance on it - no native tools work right. While "ok" for porting an existing game as a secondary platform, you don't want to be using it for primary development
520.
by kimixa 3y ago
The speed of light in an optical fiber is still only about 2*10^8 m/s - ddr5 6000's 3ghz clock means it still only travels 6.7cm/clock - so there'll still be difficulties around latency if it's not "close"
521.
by kimixa 3y ago
You don't even need unusual architectures - e.g. it's pretty easy to find a constant that might be fast to encode a value as 2 immediate and store the value in memory as two halfword stores on ARM - see https://alisdair
522.
by kimixa 3y ago
I'm still saying that treating this like a lawsuit isn't really correct, as "Is this design inherently" unsafe isn't a legal issue, so not answerable by the courts. And honestly if a car company is willing to subver
523.
by kimixa 3y ago
The first step in any legal fight would be to root cause it to determine liability. Arguably the best people to figure that out are those intimately familiar with the design of the vehicle - like those who designed and built it in the first
524.
by kimixa 3y ago
Also many encoders have things like a lowpass filter even on their higher presets - some people (often younger) may be able to hear those frequencies. It may be that two different people have different results of being able to tell the diff
525.
by kimixa 3y ago
No problem - though virtual memory was required by the DX10 spec, so it's not like it's a new thing. And even before that, many GPUs had (more limited) memory management units/memory protection.
526.
by kimixa 3y ago
Unless the traffic has some correlation to the contents - e.g. the data is compressed so "simpler" data has less traffic - it doesn't seem directly relevant. As far as I'm aware, there's no data dependance on CPU c
527.
by kimixa 3y ago
All modern GPUs have isolated memory contexts and full CPU-like virtual memory address space and page tables for each user context. It's not possible to (directly) read data from another context. As far as I can tell from what you&#x27
528.
by kimixa 3y ago
There's already a number of apps in the iOS and android ecosystems that allow this - as far as I can tell this is moving that to default in the OS rather than a third party app.
529.
by kimixa 3y ago
"Infinite Growth or Death" is an impossible goal but still people seem to strive for it. Though most VC funding seems to have the explicit goal of "Burn cash to drive out competition so you have a monopoly, then abuse your ma
530.
by kimixa 3y ago
Sure, there's places where it's good, but it's always a balance - and large mailing lists are often a poor place for actual discussion and instead just get flooded with what ends up being "spam" to many viewers - be
531.
by kimixa 3y ago
Directly voicing views in a forthright manner is useful if the communication is meaningful and informed - which often isn't correlated with how "loud" it is.
532.
by kimixa 3y ago
Group emails are also places where the "loudest" get heard - it's not just physically having a strong voice, but being the sort of person who feels the need to speak even if they're unsure they're correct, or this i
533.
by kimixa 3y ago
To this day the term "UFO" tends to be related to "Alien Flying Saucers", rather than "Exaggerated reports of military testing", which seems to be the most related facts we know decades later.
534.
by kimixa 3y ago
This would only work if every source of atmospheric carbon was equally taxed - otherwise you get incentives to have indirect carbon usage, be it in costs of manufacture, or energy generation through other means. Otherwise you get BS like
535.
by kimixa 3y ago
The idea of cars themselves as we know them may be an issue too - tyres are one of the largest sources of microplastics in the environment, and the infrastructure around them (roads, parking etc.) is a big contributor to disruption of local
536.
by kimixa 3y ago
Also DLSS has pretty strict requirements in advertising, requiring nvidia logo & adverts on splash screens and give nvidia any rights to use your app in their own advertising. [0] It's certainly not "free". A completely
537.
by kimixa 3y ago
ROCm already supports HMM. You're not helping anything by going off on some rant based on an assumption and falsehood - this sort of comment is exactly the sort of thing the phrase "FUD" is used to describe.
538.
by kimixa 3y ago
I think Intel having in-house manufacturing was one of the big causes of them getting "Too fat and slow to respond" - from the outside much of their fall from grace was due to massively delayed production improvements rather than
539.
by kimixa 3y ago
Tech design and development seems, to me at least, pretty much naturally opposed to the "being kept in check with competition" state - as design isn't really a cost that scales per-unit, the company that sells slightly more c
540.
by kimixa 3y ago
If you don't initialize the initial iterator state, or compare with something that isn't a valid iterator of the same container, you kinda end up with the same issues. The complaint here is that the warnings/errors for the in

More ›