[–]▶ No.912845>>912854 >>912863 >>912869 >>912924 >>912950 >>913003 >>913056 >>914264 >>915375 >>922163 >>922193 >>934603 [Watch Thread][Show All Posts]
> Purism has been working on reverse-engineering the Intel Firmware Support Package (FSP) module but it looks like that work may have taken a turn.
> A Phoronix reader tipped us off this morning that the Intel FSP reverse-engineering information made public by Purism has now been retracted. The past several months Purism has been working on reverse-engineering the Intel FSP to free the system further to run on only open-source code rather than still having the Intel binary-only module paired with Coreboot. Their big focus this year has been on figuring out the actual silicon initialization code inside the FSP. Purism's Youness Alaoui was very close to finding out this information at the start of April and he wrote a lengthy blog post outlining his reverse-engineering work.
> But as we were informed this morning by a sharp-eyed reader, that post has now been removed:
>> 2018-04-23 update: after receiving a courtesy request from Intel’s Director of Software Infrastructure, we have decided to remove this post’s technical contents while we investigate our options.
> That same individual also pointed out the repository where they had been working on their FSP code has also been taken offline. That's about all we know for now, but will update if/when we hear more.
Article: https://www.phoronix.com/scan.php?page=news_item&px=Purism-FSP-RE-Disappear
Archived sites taken from the forum comments: https://web.archive.org/web/20180421130723/https://code.puri.sm/kakaroto/FSP/src/master/fsp_m_init.c
https://web.archive.org/web/20180421130722/https://code.puri.sm/kakaroto/FSP/src/master/fsp_m.c
https://web.archive.org/web/20180421130720/https://code.puri.sm/kakaroto/FSP/src/master/fsp_common.c
https://web.archive.org/web/20180421130730/https://code.puri.sm/kakaroto/FSP
https://web.archive.org/web/20180407232908/https://puri.sm/posts/intel-fsp-reverse-engineering-finding-the-real-entry-point/
▶ No.912854>>912856 >>912860
>>912845 (OP)
The fact this had to happen gives them credibility, but the fact they complied throws that credibility right out the window. I wonder what Intel threatened them with.
▶ No.912855
They should comply and give evidence of compliance by releasing all research as Open Source.
Not-our-fault-the-project-grew-massively.jpg
▶ No.912856>>912918 >>912950
>>912854
I hope it just means they took the "smart" option for a while to plan better. I think they're gonna continue when they know how to do it without Intel getting too much serious legal power over it.
▶ No.912860>>912918 >>912962 >>922764
>>912854
How can you say no to a multibillionare megacorp though.
▶ No.912863
>>912845 (OP)
Go save these pages to your disk, too.
(((They))) will go after the archives, I bet my ass.
▶ No.912874
▶ No.912875>>912914 >>912918
buying purism still gives intel kikes money.
power9 is the clear better choice, even if all the important docs are locked behind a $100,000/year paywall that the average consumer goy has no access to.
▶ No.912885
>>912851
fucking cucks you mean
▶ No.912914>>912955 >>914499
>>912875
What’s a good non-x86 lappy??
▶ No.912918>>912933
>>912860
That's true, it's still disappointing. Fucking kikes. Hopefully they're going with >>912856
>>912875
Is RISC-V really our savior, or is it just a meme like POWER9? Even if they're both memes, then which one is better?
▶ No.912924>>912928 >>912931 >>912935 >>912936 >>912950 >>914216 >>914479 >>914504 >>915378 >>915394 >>920004
>>912845 (OP)
>>912851
Can't /tech/ make a single thread without spreading ITS THE JOOS bullshit?
▶ No.912928>>912968 >>916402
>>912924
>it's not the jews
▶ No.912931
>>912924
>"israel"i company does something wrong
>don't you dare call us out goy
How's the weather InTel-Aviv?
▶ No.912933>>913003 >>913086
>>912918
>Is RISC-V really our savior, or is it just a meme like POWER9?
RISC-V is a meme until there's an open-hardware, general purpose, 64-bit RISC-V-based SoC that's actually affordable and can run at least one mainstream open source OS (probably Linux).
That basically describes the lowRISC project, but they were supposed to have a crowdfunding campaign for their first SoC in 2017, and they've obviously blown right past that target. Looking at the lowRISC homepage, it looks like they're more involved with academic masturbation on FPGAs than actually producing a product.
The hardware situation is so sad, though, that RISC-V is about all we have to look forward to at the moment, unless the open-hardware reimplementation of the SuperH family goes anywhere.
▶ No.912950>>913113
>>912845 (OP)
>>912851
>>912924
Live feed from Intel HQ
>>912856
I hope that's the case, because botnet-free x86 would be nice
>>912869
Oy vey! Is that the real thing?
▶ No.912955>>913086 >>914507
>>912914
Librebootable arm chromebook
Mips laptops(if you can find one)
The powerpc notebook(real future product)
ppc macbooks
▶ No.912962>>912964 >>913054
>>912860
>how can you throw truth and freedom out of the window for some legal troubles
▶ No.912964>>912978
>>912962
How could you you make a strategic decision that ultimately fulfills your main agenda and the longevity of the careers of all your employees when you can sacrifice it all for the opportunity to sabotage yourself in a gesture of feigned rebellion that benefits no one?
▶ No.912978>>912983 >>913117
>>912964
>benefits no one
If you mean freeing all the x86 hardware with the FSP in the future, benefits no one. You are a liar.
▶ No.912983
>>912978
It will not benefit his employer and his paycheck.
▶ No.913003>>913008 >>914190 >>934615
>>912845 (OP)
Why even bother at this point? the best way to screw over Intel would be to focus on a decent x86 re-interpreter/re-compiler toolchain so that the last 30 years of software written for x86 and x86-64 can be run on non-x86 systems.
All these people are doing is furthering x86 dominance, and x86 is AIDS.
>>912933
>RISC-V is a meme until there's an open-hardware, general purpose, 64-bit RISC-V-based SoC
I wouldn't hold out on that being a thing anytime soon, there will be decent and affordable 64 bit RISC-V SoCs available but they wont be open in the sense that you will be able to view the RTL for them.
>Looking at the lowRISC homepage, it looks like they're more involved with academic masturbation on FPGAs than actually producing a product.
But that's what RISC-V has always been, its taken Google throwing their weight on the project to get it moving in a focused way toward becoming a viable alternative to ARM.
▶ No.913008>>913016
>>913003
You make a great point, I wonder why someone hasn't written a QEMU patch for other architechures into wine yet? You can run x86 windows .EXE's on ARM with wine. Why not just make it so you can do it on other architectures, like powerpc and risc-V, also? Granted the patchset was dropped because of (((reasons))). But it could be re-created and ressurected in wine-staging now and the kernel now.
▶ No.913016>>914477
>>913008
>QEMU
What I suggested is different from QEMU, QEMU is emulating the hardware whereas what I suggested is more of a decompiler+compiler toolchain. Such a project will never get upstreamed into the kernel for a myriad of legal reasons but in reality it wouldn't need to be.
▶ No.913045
That is interesting though. Intel has shown by doing this they are aware people want a botnet free system. So why don't they deliver what the market wants? My bet is they are receiving federal funding not to.
▶ No.913054>>913069 >>914412
▶ No.913056
>>912845 (OP)
Thats what they get for using proprietary hardware. Perhaps purism will pick shit that respects our freedom next time. Retards.
▶ No.913069>>913209
>>913054
And these people are examples on why you do this privately and then release all the information all at once anynomylously as possible.
>Doctors studying nagalese
I have heard of all the others. But what the fuck is nagalese?
▶ No.913086>>914501
>>912955
>Librebootable arm chromebook
Thats just the C201, right? I'm not sure if I really like the idea of sticking with depthcharge as payload though. I heard it has signature verification of the kernel or something, but what if you're on a rolling release and you get a kernel update? Wouldn't that mess things up? Also, I heard theres like, a factory reset available by hitting space at some point, but idk if that's a part of depthcharge or a part of ChromeOS.
Also someone find that manager whos keeping the Mali GPU driver away from us and tell him to stop being such a meanie >_<
https://lwn.net/Articles/738225/
>Mips laptops(if you can find one)
outside of the old yeeloong, I don't know of any, and that lemote is actually slower today than your average SBC, at least judging by clockspeed.
>The powerpc notebook(real future product)
I saw that! I really hope it's good! Seems very early to say though.
>>912933
;_; it really does make me sad how little good freedom-respecting hardware is out there these days..
▶ No.913097>>913113
>>912869
Well done boy, any idea of a commit date on that? The zip says 1984...
▶ No.913113
>>912950
>Oy vey! Is that the real thing?
It is.
>>913097
>The zip says 1984...
I did that on purpose while getting rid of all metadata and zipping the files. I don't know the exact date but they were commited on April.
▶ No.913117>>913132
>>912978
Good luck using all the wonderful software derived from an unfinished study that no one is legally allowed to write or maintain, moron.
▶ No.913122>>913123
(((intel))) is the real enemy of freedom. who is (((RMS))) kidding here?
▶ No.913123>>913127 >>914409 >>914501
>>913122
umm RMS has no connections to Purism, aside from saying bad things about them back when they used to have a proprietary BIOS.
▶ No.913127>>913131
>>913123
>not getting it
AMD did nothing wrong
▶ No.913130>>913133
>>912869
4354c9a424f4f4e9bb48509809e64f374558d9ac47833c86e6e8b788c273a5c5
is that correct?
▶ No.913131>>914501
▶ No.913132
>>913117
>no one is legally allowed to write or maintain
this was not really a problem for I2P, or for DeCSS, etc.
▶ No.913133>>913147 >>913173
>>913130
Yes. Individual SHA-256 file hashes are:
c8564ddfd50a61330743e8734c25f3d0e12696150dcc94a683b08107df77a97b fsp_common.c
6cbd3e1216b33c47c3224b69d9f6478364f784b6f81958afb95c0cb958df526c fsp.h
f14017bfd9e761811b0da216addfda915b6c3da1fd825607b87fa46fef530964 fsp_m.c
f5cdde3e2d75ef3c5594e514e68c0bab28e8707dd61c63fb2729c7310f9aa327 fsp_m_init.c
6ad4d1494d35105f4c0631b215dbc12ee5888c50395db2a9777999f4a175bcb0 fsp_m_pei.c
f4740c73fc0b870a8070fd24b5e1d2a249a69e0e4840ef0ea620d4933330e285 Fsp_uncleaned.c
e947a019adf4060ccabc5c7bb8784a41be110a74b5281e4472fd9df48b736457 fsp_utils.c
▶ No.913147>>913170
>>913133
So now the idea is to spread this everywhere with links, etc.?
I don't think I am able to do any other meaningful actions as I'm far from being a qualified OS/firmware developer.
▶ No.913170
>>913147
Someone should tweet it at Purism and/or (((Intel))).
▶ No.913173
▶ No.913183
▶ No.913209>>913210
>>913069
It was related to cancer
This might explain it, same situation don't know if it's about nagalase though
▶ No.913210
▶ No.913282
>>912869
A backup if the kikes get to 0x0.st
mediagoblin.xyz/sj6a.zip
▶ No.913286>>913287
>Purism's Youness Alaoui
purism is a front for pan arabic nationalism!
▶ No.913287
▶ No.914190>>914195
>>913003
>I wouldn't hold out on that being a thing anytime soon, there will be decent and affordable 64 bit RISC-V SoCs available but they wont be open in the sense that you will be able to view the RTL for them.
You can have a near fully open-source RISC-V PC now if you're willing to spend $3K for performance similar to a Raspberry Pi:
https://www.sifive.com/products/hifive-unleashed/
https://www.crowdsupply.com/microsemi/hifive-unleashed-expansion-board
I say "near" because the first generation SOC still contains some 3rd party proprietary IP cores for peripheral functions where open-source cores were not available. SiFive's business model is based on offering chip customization and fabrication services. They're trying to get an open-source ecosystem going.
https://youtu.be/rzUVu07xX_s?t=30m23s
▶ No.914195>>915172
>>914190
Those were discussed in the Talos thread, they don't have open RTL since SiFive want to sell you a license for their core designs (complete with a small mountain of NDAs and other legal documents). Only severly cut down version is available as RTL (of course only after signing a bunch of legal docs) and the full core is only available as a pre-synthesized bitstream for the FPGA board.
That video may say otherwise but I suspect they aren't being entirely honest, I would love to be proven wrong and to have the entire RTL for the chip publicly available but I am extremely doubtful of it happening. I judge people by what they do rather than what they say and as it stands there is some behind the scenes hand rubbing going on at SiFive.
▶ No.914216
>>912924
Jewtel strikes again
▶ No.914237>>914242
>>913338
Gross. If you're going to suck another dudes cock, at least do a decent job.
▶ No.914242
>>914237
>He saw the thumbnail and still watched it
You can't hide from us faggot
▶ No.914264
>>912845 (OP)
Is it hard to "leak" those FSP files from some irrelevant chinese OEM factory and then do the development communicating and publishing anonymously over darknets like i2p?
Fucking normies can't live without namefagging and good goy paychecks.
▶ No.914408
>"our botnet is mandatory, reverse-engineering and/or removing it is not allowed"
How is this allowed? With this kind of morality displayed constantly by corporations and state institutions and agencies, am I supposed to believe newborns aren't secretly fitted with tiny microchip implants soon after birth these days?
▶ No.914409>>914425
>>913123
Could it be because Purism are competitors to the EFF/FSF selling "gluglugs" (i.e. vastly overpriced and vastly outdated trannybooted Thinkpads)?
▶ No.914412
>>913054
>Julian Assange
What is his status? Is he alive in the Schroedinger type of sense?
▶ No.914425>>914496 >>914501
>>914409
no it was because they were saying they were selling a "total freedom" laptop, or something like that, when they still hadn't gotten coreboot or me_clean, or anything. They were basically just another system76 or whatever at that point.
Even now they still only do coreboot, not libreboot, so theres still a bit of binary blobbiness and the ME is still kinda there (even though they say it's disabled/removed/neutralized, there was a blackhat presentation that said it can still be an exploit)
FSF did endorse their distro, PureOS, so its not like theres bias against the company.
See this article...
https://puri.sm/learn/avoiding-intel-amt/
...and the part about wanting to be 100% certain rather than 99%. FSF will only give the RYF certification if Purism can get to that "100%".
I mean, Purism is probably the best option if you can't live with an outdated lappy, and is most likely not botnet in the slightest, but it's not "100%"
and this Intel news troubles me. Like, are they still working on the FSP at all? They took down their repo and their update in the phoronix article points to Intel's FSP, not theirs. are they compromised now? O_O
▶ No.914477>>914479
>>913016
Why would a compiler need to be in the kernel regardless?
▶ No.914479
>>912924
>OY VEY THE SOULS OF THE 6 TRILLION ECHO IN MY NOSE
>>914477
Same reason a dbus implementation needs to be in the kernel.
▶ No.914496
>>914425
>the cianigger A.I asks why the kikes do what they do, like it doesn't already know why
Puzzling at best is this question. Suffice it to say the FSP must have something to do with the redirection of x86 assembly to whatever architecture is on the die now. Or there are hidden backdoors in the FSP. Or both.
▶ No.914501
▶ No.914504
>>912924
guaranteed_replies.bmp
▶ No.914507
>>912955
>The powerpc notebook(real future product)
[desire to know more intensifies]
▶ No.914515
▶ No.915055>>915057
Purism are googlekiked controllled opposition themselves.
▶ No.915172>>915206
>>914195
>they don't have open RTL since SiFive want to sell you a license for their core designs
I'm no expert on this, but I think you're missing something here. They've fully open sourced their work, but it's not in RTL, they work at a higher level of abstraction. This stuff is written in Chisel:
https://chisel.eecs.berkeley.edu/
https://github.com/freechipsproject/chisel3/
>Chisel3 is much more modular than Chisel2, and the compilation pipeline looks like:
>
>Chisel3 (Scala) to Firrtl (this is your "Chisel RTL").
>Firrtl to Verilog (which can then be passed into FPGA or ASIC tools).
>Verilog to C++ for simulation and testing using Verilator.[/code]
>I would love to be proven wrong and to have the entire RTL for the chip publicly available.
Their Chisel source is available here:
https://github.com/sifive
You can compile it yourself to obtain the Verilog. The actual RISC-V CPU cores used in the u500 were already an open design, the Rocket Chip, which is also written in Chisel.
>Only severly cut down version is available as RTL (of course only after signing a bunch of legal docs) and the full core is only available as a pre-synthesized bitstream for the FPGA board.
You don't need to sign anything to obtain the above. Of course this isn't the complete chip, but they've been totally open about the 3rd party proprietary cores they've used (pics related). That's what you have to buy the license and sign the NDAs for. Their justification is that there weren't open cores available for these functions, and they wanted to get RISC-V hardware out of the door and into people's hands. This seems logical to me. To get an open-source hardware ecosystem going requires some compromise at the start, just like how GNU had to be bootstrapped from proprietary Unix systems, replacing utilities one by one until they had a fully free environment.
A fully free chip won't happen until there are free replacements for the aforementioned proprietary IP cores. They've stated their intention to work with the community towards that, and they're done their bit by releasing their work. I don't see any hand rubbing here - this is real progress.
▶ No.915206>>915519
>>915172
>the cells, pads, PLL, OTP, DDR, GbE, and ROM are propietary botnet
>it has jtag/hardware backdoor/debugging
>it has propiatary blobs for the ROM directly in the CPU bootstrap code
Literally botnet. Don't buy this and let them release something with the bootstrap for the CPU bringup being FOSS and hopefully but not neccessary the removal of JTAG. They could use DDR2 if they wanted FOSS code for the bringup. They could remove the Gigabit ethernet controller and let us use our own via GPIO FOSS ethernet modems. The OTP chips are no big deal as they can just be isolated. The PLL needs to be FOSS, they have no excuse for the loop cycle not being FOSS other then its probably apart of the boot ROM. And I have no idea what cells and pads are for this chip as I am too lazy to study it.
▶ No.915375
>>912845 (OP)
WTF? This is a shitty state of affairs. Intel's already being attacked by the Russians and they're making more progress than they're letting on.
▶ No.915378
>>912924
When it isn't the Jews we'll all be as shocked as possible, trust us, we'll let you know.
Kike.
▶ No.915519>>915697 >>915746
>>915206
Ok, you appear to be more knowledgeable about this than me, so what's to stop you taking their work as a base and working on these things yourself? I still don't get what there is to criticize them for, when they've released their own work. This is considerable progress in the direction of a fully open system.
▶ No.915669
b96b60e37c8e7714c6d561ad567a9f899c7848a5
▶ No.915688>>924066
>>915394
>/leftypol/ are shilling philosemitism
While you won't find le happy merchant in every thread, it's /pol/ that at this point consists of Zionists pretending to hate jews while supporting American foreign policies that favour the creation of Greater Israel.
(sage for off-topic)
▶ No.915697
>>915519
Nothing except I am lazy and too poor to purchase one to reverse engineer the bootrom via jtag for R/W testing, osciliscope for finding memory adresses, and time for mapping it all. It isn't exactly difficult to remove the gigabit ethernet controller, just desolder it. The DDR2 thing I couldn't do shit about short of desoldering the DDR3 ports and attaching DDR2 ports and then in the proccess of writing the bootrom to make firmware to communicate with the previous adress of the DDR3 ports. But that's alot of work for literally nothing but a slow risc-V board with a jtag attached which is shit for security.
▶ No.915698>>915875 >>921354
I'm more hopeful for the EOMA68 than Purism at this point. I kinda get the feeling Purism is just showboating for popularity still.
▶ No.915746
>>915519
lowrisc are trying to make a fully open chip. They currently support 100MB ethernet.
▶ No.915875>>916297
>>915698
This. I was more or less interesting in the group since they appeared to be all about security and protection while trying to make more modern systems, but it seems more and more like they're just a shady outfit trying to ride the security wave to get rich, while producing subpar equipment that isn't what they marketed it as. I recall at one point they had all the info on their phone hardware-wise, but now it's just bullet points stating "it's super great" and all the radio bands are a mystery, despite being so far into their crowd funding. I'm not putting money into something that's just gonna be a wifi telephony device. If verizon doesn't support it, I won't have it.
▶ No.916297>>916361
>>915875
Doesn't it seem probable that CIA and NSA glow darks would infiltrate an operation like Purism?
▶ No.916361>>917071
>>916297
Incredibly so is it likely. That's why when you do sensitive research you do it for free, you do it anonymously, you do it on the most secure hardware possible i.e not windows 10, and you release the full research all at once so that everything is in the open on multiple websites yourself and don't trust middlemen like reporters. Purism immidiatelly failed when they weren't anonymous and their devs could be (((losing confidence in the product))) or commit (((suicide))) by two gunshots to the back of the head.
▶ No.916402
>>912928
Pure cohencidence.
▶ No.916423
▶ No.917071
>>916361
I wonder how many security researchers have already been killed by the US government? Is this why those Russians working on Intel ME don't ever travel out of the country together? Somebody always stays behind in Russia.
▶ No.918514
So the Libreboot tranny hasn't updated its code in eons, Purism's apparently now finished, is there any actual progress on defeating ME?
▶ No.918523
▶ No.920004
>>912924
nu/pol/ overran /tech/ and turned it into cancer. /tech/ died a long time ago.
▶ No.920595
I thought this was a thread about pure nudism getting a letter from intel
That would be interesting
▶ No.921333>>921342 >>922162
>want to pre-order Librem 5
>expensive and first batches will probably have big issues so want to wait before buying
>but if don't buy, they have less funding and progress will stagnate
dilemma
▶ No.921342
>>921333
The only reason they got the funds was that a few, or one, donator made giant donations at the last minute to get the requested funding. Don't feel as if it all falls on you, there are many people who want it who are willing to work through the early pains(that's why they put money forward in the first place).
▶ No.921354
>>915698
>I kinda get the feeling Purism is just showboating for popularity still.
Welcome to the world of marketing anon, they have to shill.
I like the EOMA68 as well but where is it? The crowdsupply campaign got funded nearly 2 years ago, last update we had shows he still has issues sourcing DRAM and is shifting focus to some imaginary RISC-V device.
https://www.crowdsupply.com/eoma68/micro-desktop/updates/ddr3-ram-and-a-libre-risc-v-soc
Purism did some downright dishonest stuff initially but right now they have 2 products that get yearly refreshes, at this rate they're going to have the 5, 11, 13 & 15 in production before EOMA68 even exists.
▶ No.922154
Who's even surprised here?
(((Intel))) has a clear history of Divide & Counquer policies. If someone even tries to reverse engineer their botnet, they WILL strike back.
I honestly can't see why anyone's surprised.
▶ No.922162
>>921333
Buy it. Worst case scenario you'll encounter software issues for the first month or two.
▶ No.922163>>922184
>>912845 (OP)
Why don't they just move to AMD? AMD is basically just as good as Intel now just slightly cheaper and not as back doored.
▶ No.922184>>922191
>>922163
While having a smaller back door is objectively more attractive, it doesn't change the fact that the back door shouldn't exist in the first place.
▶ No.922191>>922280
>>922184
All processors are backdoored. I doubt AMD is as corrupt as Intel when it comes to removing backdoors so they should have less issues.
▶ No.922193>>922210 >>922359
>>912845 (OP)
easy
>actually not stop and continue the reverse engineer project sekritly and privately (if they got snapshots and you use windows with rdp then you deserve it!)
>then make or use existing cracker release group release your "reverse engineered crap" to make it seem "not my fault but thanks to them (you)" so everyone can pick the code up
>???
>!!!
Just follow the same concepts the console crackers always use.
>(((stop)))
maybe if they dig deeper, the rabbit hole will appear
>telling crackers/hackers/reverse engineers to stop
▶ No.922210
>>922193
Find rabbit hole
Anonymously post research in some random tech forum/imageboard.
"Oops well i guess we didnt do that so it's okay"
▶ No.922280>>922326
>>922191
but not all processors!
▶ No.922326>>922328 >>922758
>>922280
All modern x86 most likely are. You're probably safe if you go back to 80's tech, like 80386 chips. Those don't have enough margin to waste on botnet, and besides most of them were never so much as connected to a modem. But I'd still much rather have a 68030, or even an eZ80.
In theory, OpenSPARC and RISCV are also safe, but it depends who manufactures them. Don't buy from anyone connected to Israel.
Also, FPGA but it's more limited than those other two.
▶ No.922328
>>922326
What about MIPS, ARM, or IBM POWER? Where do those fit in?
▶ No.922359>>922758
>>922193
wasnt this just about intel wanting some of their documents removed from the site?
▶ No.922758
>>922359
(((they're))) telling them to stop because there will be another MINIX-tier findings from reverse engineering modern jewish CPUs.
>>922326
sure enough tech before 1986 would be free of all that crap that is due to the fact that around 1986 is when the illerminaty group decided to design the humancattle aka goyim system.
The templar-assassin-brotherhood-knights uses the ZOG as their method of enforcing new world botnet.
Kill the ZOG then kill these retards who have obsession with the cross = crucifixion = death of savior or christ (symbological loosh that mocks all the christ candidates).
▶ No.922764
▶ No.923883
how libre is Librem at this point in time? (libre by power, not by volume)
how much hard botnet is still remaining and hypothetically what is it capable of?
eg. if intel really wanted to backdoor your new Librem 13 to steal your company secrets, could they?
▶ No.923888>>923890 >>923926
>2019
>Intel release new line of CPUs with completely different FSP and ME, even harder to disable
What do?
▶ No.923890
▶ No.923926>>923996
>>923888
Shit, if you're still thinking of buying x86 system after all the ME fiasco and endless bugs, I just don't know what to say. Party on, dude! I guess.
▶ No.923996>>924216
>>923926
I grew up in a cave. It's hard to see the light.
Is there a way to get a decent ARM or whatever without forking out $3k per build? (plus won't it be incompatible with lots of good OSs and software?)
▶ No.924021
>Time for real nerds to re it themselves. Don't share, but maybe open source em.
▶ No.924066
>>915688
>pretending to hate jews
Calm your tits Ahmed, /pol/ supports Assad and since the kike mods are gone, criticism of Trump's foreign policy is quite widespread.
▶ No.924176
where can I tell Intel I won't buy their ME shit?
▶ No.924216
>>923996
>Is there a way to get a decent ARM or whatever without forking out $3k per build?
Tough to say. There's an overwhelming number of Pi-clone SBCs out there that are fairly cheap. For laptops I think you can do some stuff to get GNU/Linux running on a chromebook, and you can also reflash the coreboot on them or get Libreboot on the C201.
There's really no desktop/workstation class stuff yet sadly. We do have 48-core Cavium ARM server chips, which can be found in offerings from PogoLinux and System76.
>(plus won't it be incompatible with lots of good OSs and software?)
Many GNU/Linux distros and I think pretty much all BSDs support ARM fully. You're gonna have issues if you need anything proprietary for whatever reason, but if it's Free Software/Open Source, it should be perfectly available on ARM.
Also, the GPU driver for the most common ARM GPU (Mali) is still proprietary. There is literally one dude in upper management at ARM who is holding it back. Everyone else is on board with releasing it.
https://lwn.net/Articles/738225/
>It was said that everybody within ARM is in favor of solving the problem by open-sourcing ARM's driver --- except for one recalcitrant high-level manager.
Hope this helped! ^.^
▶ No.934603
>>912845 (OP)
(((courtesy request)))
Black Cube came to their offices and roughed them up?
▶ No.934615>>934616
▶ No.934616>>934654
▶ No.934654>>934657
>>934616
Encapsulate a summary for us please.
▶ No.934657>>934706 >>934719
>>934654
x86 and x86_64 -> llvm intermediate
>McSema is an executable lifter. It translates ("lifts") executable binaries from native machine code to LLVM bitcode. LLVM bitcode is an intermediate representation form of a program that was originally created for the retargetable LLVM compiler, but which is also very useful for performing program analysis methods that would not be possible to perform on an executable binary directly.
>McSema enables analysts to find and retroactively harden binary programs against security bugs, independently validate vendor source code, and generate application tests with high code coverage. McSema isn’t just for static analysis. The lifted LLVM bitcode can also be fuzzed with libFuzzer, an LLVM-based instrumented fuzzer that would otherwise require the target source code. The lifted bitcode can even be compiled back into a runnable program! This is a procedure known as static binary rewriting, binary translation, or binary recompilation.
>McSema supports lifting both Linux (ELF) and Windows (PE) executables, and understands most x86 and amd64 instructions, including integer, X87, MMX, SSE and AVX operations. AARCH64 (ARMv8) instruction support is in active development.
▶ No.934706
>>934657
Someone fork this before it gets shoahed.
▶ No.934719>>934723
>>934657
Nice, this is great. I thought that this was the whole point of the LLVM anyway though, kinda surprised that this is a new thing.
▶ No.934723>>934869 >>937431 >>937780
>>934719
The point jewgle created LLVM for is to be a bytecode generator for GPU's to have interoperability for pajeets writing GPU code and their opencl/cuda implementations. Using it to further free software on the CPU side is not a goal.
▶ No.934869
>>934723
but the point was to free poorly planned java from depreciation so it could be re implemented with 'muh tensorflow', no? its somewhat close
▶ No.937431>>937780
>>934723
>Jewgle created LLVM
I thought it was done by some grad student who was hired by Apple.
▶ No.937780
>>937431
It was, Google has nothing to do with it. This poster: >>934723 is a liar.