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

Freedom and security issues on x86 platforms

mail.fsfeurope.org

Post reply on HN
Print view View original

271–280 of 282 posts

Re: Freedom and security issues on x86 platforms

#271

Post by analognoise »

nickpsecurity wrote: post #269
analognoise wrote: post #268

Earlier quoted context omitted.

Most academic shuttle runs are still at 90nm, or bigger. There are a few reasons for this: cost of runs, cost of tooling (hundreds of k), and the fact that your assumptions about transistor action and modeling are exponentially more complicated at advanced process nodes. Academics also sign NDAs about the processes they use, and can only make certain things available; the most open is probably MOSIS, but that's absol…

I'd agree that most runs are at 90nm or above. Yet, the rest is confusing given I have quite a few papers with competitive stuff done at 45-65nm with some at 28 or 32nm. So, why you say forget about it or MOSIS below 90nm if academics are getting working chips done that low?

They sign NDAs for the process and get 100-thousand dollar layout packages for academic prices.

If you have a few spare million dollars, you still can't necessarily release a lot of data due to the NDAs - usually they give you models for the processes that are proprietary (and they invested a lot in developing, and so will consider any breach an act of war).

Re: Freedom and security issues on x86 platforms

#272

Post by 616c »

jensen123 wrote: post #261
616c wrote: post #94

The fight is increasingly political, so advocate and donate where you can. We lose when we give up, I suppose. I know what the Libreboot guy said before on his blog, alluded to here, but this is why, as crusty as some might find him, we most generally support Stallman's politics.

I wish the FSF was a bit more nuanced. For example, if DRM causes ordinary computers to come with proprietary code that is impossible to remove, then that is bad indeed. Then you no longer control your own computer. The same computer that you might use for political activities, for example. On the other hand, if entertainment computers, such as blu-ray players or gaming consoles are locked-down and full of DRM, then…

Disclosure: I became a member days ago, and did not want to mention it. Reading this article made me super proud to have chosen to fork the cash over before reading this crap as it gets worse all the time.

At the end of the day, I want a hardliner in this space pushing that line because I know, practically, he cannot win. But if consensus is drawn between him and the other extremes I find distasteful, I want him to pull the resolutions and positions as far left as possible, even if that is slightly left of center.

I worry, as current events show, anything less means that counter-forces to the free software movement will wear you down with abject greed by slowly going right of center and taking as much time as it takes to restore the balance back to their proprietary interests after the initial battle has gone to free software advocates. And that is how I see it. Very few of my friends understand the value of highly technical manuals, and that is what open source is about. My brother recently saw my side with automative hacking and experimentation countermeasures on the rise, as reported today on HN. But when I tell non-technical people these manufacturers hide secrets in their faulty designs and let you pay for their ineptitude, even if you want to fix it for yourself on your individual unit without harm or influence on them, they do not get the argument and ask why I think I know better than the compant. They only get the argument when they are locked out of a system they need for their very personal context.

Oh well. This is a very personal choice. I love GPL, I love MIT, and I smile when I think how all these hippies made a world for me in the 60s and 70s I could not live without today.

Re: Freedom and security issues on x86 platforms

#273

Post by LargoLasskhyfv »

analognoise wrote: post #271
nickpsecurity wrote: post #269

Earlier quoted context omitted.

I'd agree that most runs are at 90nm or above. Yet, the rest is confusing given I have quite a few papers with competitive stuff done at 45-65nm with some at 28 or 32nm. So, why you say forget about it or MOSIS below 90nm if academics are getting working chips done that low?

They sign NDAs for the process and get 100-thousand dollar layout packages for academic prices. If you have a few spare million dollars, you still can't necessarily release a lot of data due to the NDAs - usually they give you models for the processes that are proprietary (and they invested a lot in developing, and so will consider any breach an act of war).

From my amateurish POV this looks rather peacefully. No Cult of the NDA in sight. Nonetheless too expensive. OTOH, are you aware of the prices mentioned there? http://www.europractice-ic.com/ http://www.europractice.stfc.ac.uk/welcome.html http://cmp.imag.fr/ http://cmp.imag.fr/products/ic/?p=prices2014 http://cmp.imag.fr/products/ic/?p=prices2015 http://cmp.imag.fr/products/ic/?p=prices2016

ahh crap..."which are sent to customers upon signature of a Confidentiality and License Agreement"

anyways, prices...hope that helps

Re: Freedom and security issues on x86 platforms

#274

Post by hackuser »

5ilv3r wrote: post #222
hackuser wrote: post #197

Earlier quoted context omitted.

> Except modern baseband processors usually don't have direct access to main memory or peripherals - they are usually linked to the rest of the phone via a serial bus. Do you know where I can read more about that? A good, technical, authoritative resource? In my little bit of research, details are sparse and authoritative technical details even more sparse.

Paranoid android used to have a nice breakdown on which phones had isolated memory for the baseband and which used shared memory. I cannot find it now, and their site seems to have taken a very wrong turn in the design department.

fyi: As far as I know, Paranoid Android development stopped last summer, after OnePlus hired away key developers in February 2015. Here's an article with much more detail, including its prospects going forward:

http://www.androidauthority.com/paranoid-android-calling-it-...

Re: Freedom and security issues on x86 platforms

#275

Post by hackuser »

ashitlerferad wrote: post #255
vox_mollis wrote: post #196

Earlier quoted context omitted.

Except modern baseband processors usually don't have direct access to main memory or peripherals - they are usually linked to the rest of the phone via a serial bus. ME is very, very different - it transparently has access to everything.

serial attached basebands are much rarer than basebands with full or partial control of the CPU or RAM.

Do you know any good sources for this information? I can seem to find any (see my post above).

Re: Freedom and security issues on x86 platforms

#276

Post by nickpsecurity »

rcarmo wrote: post #193
nickpsecurity wrote: post #115

Earlier quoted context omitted.

There used to be a lot of competition between several types of RISC machine and several x86 vendors. I know about the consolidations on x86 side. I'm not sure why SPARC lost favor versus others, though. I wasn't able to afford RISC workstations back when all that was happening. It would be nice for one of older folks to chime in on what made SPARC unpopular back then.

It was the cost/performance ratio, not of the CPU itself, but of the entire thing, including software. Sun sold highly performing, but extremely overpriced hardware that rode in on the hype it built around the architecture but failed to deliver on flexibility and bang for the buck. I was involved in launching an ISP where the whole shebang ran on Sun boxes, and which was over-dimensioned to the point where I once ste…

Thanks for the enlightening comment. That makes plenty of sense. It's part of the reason I didn't try to buy one: price/performance numbers just didn't make sense.

Re: Freedom and security issues on x86 platforms

#277

Post by ashitlerferad »

hackuser wrote: post #275
ashitlerferad wrote: post #255

Earlier quoted context omitted.

serial attached basebands are much rarer than basebands with full or partial control of the CPU or RAM.

Do you know any good sources for this information? I can seem to find any (see my post above).

The Replicant wiki is a good place for devices that have evaluated, click the links on here:

https://redmine.replicant.us/projects/replicant/wiki/Replica...

Re: Freedom and security issues on x86 platforms

#278

Post by ashitlerferad »

ashitlerferad wrote: post #277
hackuser wrote: post #275

Earlier quoted context omitted.

Do you know any good sources for this information? I can seem to find any (see my post above).

The Replicant wiki is a good place for devices that have evaluated, click the links on here: https://redmine.replicant.us/projects/replicant/wiki/Replica...

Hmm, actually it isn't as good as I thought. I guess I was remembering IRC discussions. IIRC, the Replicant folks found Qualcomm based devices pretty commonly have this issue.

Re: Freedom and security issues on x86 platforms

#279

Post by krylon »

john61 wrote: post #98

There is also an additional possibility: Recycle old computers. A Intel 2008 laptop performs OK with a modern GNU/Linux with an efficient Desktop (for example XFCE4). This also helps avoiding CO2 emissions, saves rare earths and energy. And it is a statement against a unsustainable throwaway society.

The problem with such old devices is that some of them can be impressively reliable, at an age of ~8 years, one has to worry about the device starting to fail. If you want to keep the device going once something breaks, getting replacement parts can become interesting. Not impossible per se, but depending on how popular a particular device was in its day, finding spare parts can be very time-consuming.

Still, I agree. I have a 2008 netbook at home I still use regularly, and I hardly throw away a computer that still basically works.

Re: Freedom and security issues on x86 platforms

#280

Post by cesarb »

jensen123 wrote: post #261
616c wrote: post #94

The fight is increasingly political, so advocate and donate where you can. We lose when we give up, I suppose. I know what the Libreboot guy said before on his blog, alluded to here, but this is why, as crusty as some might find him, we most generally support Stallman's politics.

I wish the FSF was a bit more nuanced. For example, if DRM causes ordinary computers to come with proprietary code that is impossible to remove, then that is bad indeed. Then you no longer control your own computer. The same computer that you might use for political activities, for example. On the other hand, if entertainment computers, such as blu-ray players or gaming consoles are locked-down and full of DRM, then…

How long will "entertainment computers, such as blu-ray players or gaming consoles" remain separate from ordinary general-purpose computers?
Post reply on HN