(untitled)

Why is nobody talking about recent Intel leak and how a lot of CPU info like schematics, low level code, flashing tools, bootloaders and Intel ME info is out in the public? https://twitter.com/deletescape/status/1291405688204402689

People are talking about it, but mostly on /g/ and reddit, anyway here's Intel response to the leaks "We are investigating this situation. The information appears to come from the Intel Resource and Design Center, which hosts information for use by our customers, partners and other external parties who have registered for access. We believe an individual with access downloaded and shared this data."

The releaser says there's going to be more

backdoors aplenty

magnet:?xt=urn:btih:38f947ceadf06e6d3ffc2b37b807d7ef80b57f21&dn=Intel%20exconfidential%20Lake%20drop%201

Apparently no one understands what's inside (yet) aside from the fact they leaked a tool that allows to wipe Intel ME. That alone is good enough for a leak, but I wonder if anyone tested it.

>>11839 I've got a major erection, major Tom.

Does this mean free-software intel wifi drivers? Or more coreboot devices that support libre-/coreboot?

>>11839 >>11840 What does it mean? Is it a good thing, or are we going to start seeing worse viruses or something?

>>11844 >Is it a good thing, or are we going to start seeing worse viruses or something? yes and yes.

>>11841 If one where to use (or even look at) Intel code to write FOSS drivers they'd be breaking the law. But maybe someone will...

I have a funny feeling this is someone with a grudge who did this before investing in a competitor to make a proffit.

>>11847 Clean room reverse engineering is still a possibility.

>>11847 Someone in China or Russia probably will.

I want my intel me cleaner now! >>11841 >Does this mean free-software intel wifi drivers? Likely not, no one sane is going to use these documents and especially the source code to make any drivers/firmware, as it violates copyright. It is better to focus on making our own libre processor https://libre-soc.org/

>>11862 it would be very difficult to prove that open-source software uses the Intel documents and didn't reverse engineer the chips, especially if the drivers are produced anonymously.

>>11863 I'm sure companies with millions of dollars could sue the project into oblivion if it's public, and take it down on any clear net repository if they wanted, regardless of if it actually does or not. They have the resources to strongarm projects with millions of dollars of legal expenses.

>>11837 No seeds? I really would like to know what is going down in my cpu in depth. Thank you for linking. Has anyone had a chance to study these documents yet? Any interesting conclusions?

>>11900 Stop being corporate cocksucking larper. https://en.wikipedia.org/wiki/Clean_room_design This is how progress advanced in 20th century. Everyone stole each other's blueprints and reverse-engineered every device.

>information for use by our customers, partners and other external parties who have registered for access As an employee of one of those customers/partners, I can assure you it will contain only absolute garbage unless they've really been holding out on us.

>>11902 >Stop being corporate cocksucking larper. Not the lainon, but nothing in that post is in favor of big corporations. Remember to avoid using personal attacks. We live different times, when some made up bullshit like software patents can literally get you sued. Take GNOME vs some soykafty company doing nothing important as an example - they sued GNOME, because one of their applications sent some photos over the network. Eventually the case ended well, after the patent troll discovered the project is open source and they can't make much money from it. If Intel wants it, they can use their entire patent arsenal to jail someone. I think it is better to focus on developing our own, freedom-respecting hardware instead of reversing Intel's freedom-hostile crap.

>>11844 It means we have a very low chance of getting an almost universal Libreboot

>>11923 Why not just use the proprietary me killer for your own gain? Also are the seeds up and is there a repo for this somewhere?

>>11917 yes but intel definitely has some proprietary techniques that we could study, modify, and use in our own hardware.

>>11862 >Likely not, no one sane is going to use these documents and especially the source code to make any drivers/firmware, as it violates copyright. Only in some countries. Not everywhere copyright law is as retarded as in the USA.

>>11841 >Does this mean free-software intel wifi drivers? Or more coreboot devices that support libre-/coreboot? Yes, if someone would bother to write such software. They would just preferably have to be living in country that would not restrict such actions and would not render such code illegal to distribute (or just don't give a fuarrrk and publish in the way their identity remains hidden). There are countries where publication of company secret voids it's protection and everybody can make use of it without any risk of prosecution. Also there are countries that don't protect company secrets at all. Intel's lawyers could kiss asses of developers making use of leaked stuff in such countries. So we can't post generic answer to such question here - as it is a matter of interpretation of law of many countries.

>>11953 No significant software project is going to accept patches that aren't legal in USA. Doesn't really matter where the author happens to live, it would be a liability that nobody wants to take on.

>>11954 >Americans thinking that everything is about them name a better duo

>>11955 The best part of his comment was "NOBODY" word, it make me laugh hard here.

>>11955 >>11956 oh give me a break. Say you invest all the time to "improve" some driver by copy-pasting soykafty proprietary leaked code into it. What do you do with that? It's never going mainline. So maybe you host it on your personal website and it breaks every 3 months because some asshole changed a struct upstream and every other driver got fixed but yours didn't because it's some questionably licensed crap the maintainers don't care about. In a year (probably less) mainline has already merged patches that do the same thing but better and are either supported by the vendor or properly reverse engineered.

>>11959 ah yes, because the only thing you could ever do with proprietary code is copy/paste it, never study it to aid in reverse engineering or improve it.

>>11839 Seriously has anyone tested this? I've wanted to remove ME for years but coreboot is too iffy.

Honestly, fuarrrk copyright. Torrent indexes are huge for a reason. Having used some shady tool from some corporate leak seems much safer than installing a random game repacked by some anon online. I'm sure it'll be a matter of time until places pop up that'll guide you relatively easily to exterminate the botnet on your system. Be patient and dare to take risks because you're already at risk anyway. Posted from my librebooted X200.

>>11961 >I've wanted to remove ME for years but coreboot is too iffy. ME_cleaner is for that, not coreboot

>>11963 ME_Cleaner only recently added my model of CPU and last I checked it said something like "not tested" so same issue.

>>11959 >oh give me a break. Say you invest all the time to "improve" some driver by copy-pasting soykafty proprietary leaked code into it. Give me one reason to treat you seriously after these 2 sentences... >What do you do with that? It's never going mainline. So maybe you host it on your personal website and it breaks every 3 months because some asshole changed a struct upstream and every other driver got fixed but yours didn't because it's some questionably licensed crap the maintainers don't care about. ... I'm speechless... >In a year (probably less) mainline has already merged patches that do the same thing but better and are either supported by the vendor or properly reverse engineered. Do what?

>>11960 I mean yeah, that's what people usually do. >study it to aid in reverse engineering or improve it One thing to keep in mind is that this sort of proprietary code never meant for public consumption is usually absolutely terrible. It will be full of mystery register write sequences and weird workarounds that may be of some use, but won't be fun to decipher. Intel has over 100,000 employees with the usual siloing and other corporate dysfunction, so you shouldn't assume that just because the author worked for intel that they are privy to all the dirty secrets. They may not even have worked for intel, but for some contractor working off poorly written documents and 20 page email chains. Another thing to note is that a lot of the people contributing to open source projects already have access to this stuff. So for this to matter some individual who doesn't already have access would have to care enough to do whatever work is necessary. I don't even know what sort of improvements people are expecting to get from this. >>11966 You haven't actually said anything.

>>11969 >You haven't actually said anything. Neither did poster that it was a reply to

>>11970 shut up your nagging and go exchange your contact info so you stop soykafting up this thread

>>11969 still infinitely easier than reverse engineering assembler code.

Any symptoms of people making use of it so far?

There are many different legal systems around the world. In the USA license can ban disassembling and changing binary executable. In Poland for example law explicitly allows such things for user needs and license paragraphs that forbid such things are void in that country. In the USA there are software patents. In Europe there is no software patents and you are free to stuff like algorithms etc covered by software patents in the US without any licensing or other bullshit. and many more examples

source: They have a server hosted online by Akami CDN that wasn't properly secure. After an internet wide nmap scan I found my target port open and went through a list of 370 possible servers based on details that nmap provided with an NSE script. source: I used a python script I made to probe different aspects of the server including username defaults and unsecure file/folder access. source: The folders were just lying open if you could guess the name of one. Then when you were in the folder you could go back to root and just click into the other folders that you didn't know the name of. deletescape: holy soykaf that's incredibly funny source: Best of all, due to another misconfiguration, I could masqurade as any of their employees or make my own user. deletescape: LOL source: Another funny thing is that on the zip files you may find password protected. Most of them use the password Intel123 or a lowercase intel123 or intel@123 or "I accept" source: Security at it's finest.

>>12105 Fuсking never gets old. Security will never work lmao people are nuts

I can't wait to need to buy a new CPU if I get malware.

>>12122 I can't wait to get a riscv machine. When the hell is that gonna drop with a good loadout and support from a company with good economy of scale? It's taking forever, idk what the holdup is. Probably some corrupt corporate crap. Maybe they're developing their own filthy cancerous version of Intel ME to shoehorn into the die and ruin it. I know there's a couple obscure dev boards that are astronomically expensive considering their specs. But damn. Maybe I should just find an old ibm power cpu of some kind. Just seems like the hardware should be ready by now. Some fab shop should get in there, hurry the fuarrrk up. I still have a powerpc lying around somewhere with some old Linux distro installed but it's ooooold.

>>12124 Get a team, write a business plan, register Ltd company and find a venture capital or start founding campaign. Et volia, hasta la vista, baby!

>>12124 I don't get why people seem to care so much about what ISA they're using. What are you expecting it to change, aside from having to recompile everything with buggy beta gcc? I think the main appeal of RISC-V is that corporations don't have to pay licensing fees to other corporations in order to use it, but for the consumer there's not much advantage compared to ARM or whatever.

Honestly speaking theere is no reason why coreboot may not benefit from leak in matter of finding new exploits that would allow bypassing BIOS write protection. As BIOSes on for example Thinkpads starting from T61/X61 have write protected region that makes it impossible to write to CHIP w/o external programmer unless you know some dark magic of disarming that block. That block may be reverse engineered also by analizying Intels' firmware update code - as it does not require external programmer for SPI chips, because it speaks in a proper protocol that enables writing on software level for the time of writing updated version of the BIOS.