“Blob-free OpenBSD kernel needed”
marc.info
marc.info
I mean how many leading Linux/Open Source developers are ready to say 'we don't really like it, but if you want to use their hardware you will have to use their blobs'
How hard is it for open source designers to design their own hardware and open the firmware for it? How difficult is it for hardware developers to release the specs for dated hardware to allow software improvements, and keep the firmware for new more performant hardware closed.
The complainer has a point. It is disingenious for Linux to claim commitment to open source and neglect be virtually silent on the threat closed firmware blobs pose in this era of security issues.
If you have to ask, you are not qualified to comment on the issue.
Modern devices are complex, and the firmware often comprises the majority of the functionality of the devices and implements the interfaces exposed to OS drivers. It would not be an exaggeration to suggest companies spend just as much or more on development of firmware, than on the hardware.
Who are you to make such demands of them, and the people working on creating open drivers for such devices?
Who are you to make such demands of them
Since I realized that money is the driving factor in this world - it's way more powerful than any other vote you have - I buy devices as open as possible, like the libreboots from http://shop.gluglug.org.uk/ . I don't mind if they are more expensive and slower than other devices. If enough people think like that, eventually the open products will become better because there'd be a commercial incentive for the manufacturers.
If they didn't care, there wouldn't be open-source drivers for GPUs, projects like coreboot, ...
And the lifetime of these devices is short. Projects that try to replace firmware on consumer hardware often have this problem that they spend a fair amount of work on something and once they have reasonable support the device is no longer sold (e.g. the rockbox project constantly strugles with this).
I think the only possibility to have widespread free firmwares would be if this could be made a selling point for vendors. However to do that you'd need a widespread demand for these things, much larger than the open source / free software / hacker community is able to create right now. If it'd be considered "cool" to release the code of your firmware that could change things.
It's also quite disingenious to rant about the rant without noticing that it has nothing to do with Linux (except for the last sentence where Theo points to That Other Community)
As more and more functions are being baked into software this issue is going to become more critical, and all talk about tackling the planting of secret backdoors into software becomes moot, as it can all be done in firmware without detection. This does not even deal with the issue of unintended bugs which are just as exploitable.
Depending on the type of hardware we're talking about, it's either: already done, hard or very hard.
If you want to go open with hardware, you either:
- decide that you can go with some obsolete specs. Then you only reach the small niche really interested in freedom and openness and pretty much nobody else. The costs and prices skyrocket, because you cannot rely on economy of scale.
- device that you want to provide competitive hardware to what's already on the market, but free. In most of the fields, you will have a hard time to catch up to competitors, as you most likely won't be able to get any funding in the way they did. You'll have to resort to crowdfunding and probably fail anyway, as a very small fraction of projects like that manages to do anything at all - and they're usually the smaller and easier ones anyway.
Also, closed firmware blobs don't pose any threats - any additional ones that you already have even when not using them, that is. If you have a device that doesn't need a firmware to be loaded in order to run, then most likely it already has a non-free firmware integrated on board. For instance: HDD, SSD, SD cards, some cameras, most of controllers like USB or RAID... virtually anything.
By calling people like Theo de Raadt conspirators, you only hurt the cause you think you are fighting for. You won't change the world by that. You can only change it with small steps, by supporting niche projects like Neo900, even if they had to go for some compromises. What differentiates them from others is that their compromise lies far further than what's common in mainstream devices, and only if this iteration succeeds, a new one could in the future, moving that compromise even further towards the ideal state.
Having been involved in Openmoko and related communities for quite a long time exactly for the reasons of hardware freedoms and openness, I really don't see any other way.
Also, just so you're aware: you cannot "wait to decide", as it's basically built to order and financed almost completely that way. The moment it goes to production you will have to find at least another 200-300 people interested in buying it, so the second production batch will be possible to happen.
"You don't know jack shit about computers or electronics, that much is obvious."
You might be sensitive to profanity (I am not), but it seems is a waste of time and energy to try to explain them the obvious.
"Fact is, modern hardware simply "is what it is". Telling people that OpenBSD should not run on such hardware is a gigantic illusion."
A good, realistic view, as opposed to that by Richard Stallman
I believe that firmware should be free, but refusing to ship blobs for this reason, while still attempting to be compatible with hardware that buries them under the covers (even when the blobs not being shipped are updates for the very same hardware - e.g. all major x86 CPUs), is burying one's head in the sand. Would be more principled to avoid hardware with proprietary firmware entirely, and advocate for both documented hardware and 'free' (i.e. open HDL etc.) hardware.
How can you know that firmware is "completely non-updateable"? You trust the vendor with that, there's no way to know for sure. What's more - recently (~1.5y ago) somebody figured out how to update firmware of microSD cards [1]. Thanks to him, if we want to apply FSF's point of view, microSD cards suddenly became unacceptable, non-free hardware, while before that blog post they were just fine.
I'm involved in development of Neo900 [2] phone and already had fair share of discussions about freedom and openness of components with various people. I was a part of Openmoko community in the past and remember the TI Calypso modem used inside Neo1973 and Neo Freerunner. At first they were non-updateable (for consumer), so were acceptable to RMS. Some time later Openmoko Inc. got firmware fixes for pretty annoying bugs like SIM compatibility and started trying to figure out how to distribute the updates. They obtained a license to redistribute the updater tool, provided the instructions and the firmware image. Suddenly, Neo Freerunner became non-free hw.
Then, some time later, partially thanks to Openmoko who provided the updater and instructions, OsmocomBB [3] was born (Harald Welte, who initiated OsmocomBB, also worked earlier in Openmoko). Suddenly Neo Freerunner became free again because there's a free firmware available for Calypso now.
So my device I was carrying in my pocket, without any my intervention, went from free to non-free and to free back again. Fun!
Also, as a user who cares about his freedom, I prefer to have a hardware with updateable or even entirely user provided non-free firmware than a black-box with the same non-free firmware inside, but not updateable. Of course over that I'll prefer a hw with open firmware, but in the case of the black box, there is even no hope that such a firmware will be provided in the future. And when at the same time you cannot be 100% sure that it really is non-updateable, then what's the advantage? I can't see any.
In theory, I can agree with FSF and RMS. In practice, hardware doesn't really work that way.
And today, chances are the firmware is updatable, even if it's not documented/supported.
Or just use theo(1) from 9front: https://plan9front.googlecode.com/hg/lib/theo
I can typically blacklist things, but I need to know they are there to begin with so, to me, the problem boils down to the kernel, and other software, potentially doing things that I don't want. Since closed kernel blobs are clearly controversial, something as simple as highlighting their loading and making it as easy as possible to prevent it from happening on future boots may be enough. If the process is in place, it can be a simple lack of orientation - but Theo's response certainly doesn't help with that.
As for running blob-free in general... It's not that hard, really. Often the only loss is fast 3D. Debian does a good job of separating free from non-free - and really, that is enough, since adding non-free items then becomes a conscious act (there can be other ways of making the distinction though).
If you give "loss of fast 3D" as an example of running blob-free, you didn't understand what Theo is talking about. 3D drivers run in user or kernel space, there is a clear policy about how to deal with stuff like that in basically any serious operating system that labels itself free and open.
This mail was about firmwares. The little blobs your OS has to upload into memory of your WiFi card, Bluetooth controller, sensors, camera... and that's all it has to do with it. The rest happens on completely different CPUs - the ones that live inside your WiFi card, Bluetooth controller, sensors, camera...
Even if your hardware seemingly doesn't require the blob to be loaded, it's most likely because it simply has a on board memory to store it. Some don't (obviously, that's cheaper way) and require user to load their firmware before they can operate.
And nowadays, basically anything has such a firmware. Even microSD cards.
From OS point of view, the only sensible approach to differentiate between free and non-free firmwares is to ask a question "is it freely redistributable?". If you go deeper than that, you enter the Open Hardware territory - which comes with its own set of principles and pandora boxes.
> "the only sensible approach to differentiate between free and non-free firmwares is to ask a question 'is it freely redistributable?'"
Is it? That's the question. It may not be when not differentiating between licenses, as shown by the current conversation, is differently important to different people. I'm still convinced that it's just a disagreement over log verbosity. It is perfectly appropriate for a kernel to support loading non-free anything... It's even a feature. I just want to know what it's doing, and show me a line based on a license that I fully acknowledge the kernel doesn't need to differentiate between other than for clear logging. I am simply hoping, as a user, to get the information I need to make informed choices, including the ability to turn feature X on or off as I see fit. I can only know to buy different hardware when I am told about the blob, and I can only see the effects of my goals if I can progress towards them (by disabling binary loading). I shouldn't have to guess, the feedback and levers should simply be presented clearly.
I will stop tilting at this one, for lack of OpenBSD-specific examples, but I'm really just advocating for better information - not reducing hardware support or anything so nonsensical.
Or maybe you need some custom, proprietary solution to compile it that even the company that developed this hardware have no rights to redistribute?
Or maybe even the company you bought that hardware from has no idea about the source code or firmware, because they simply used some chipset available on the market?
Or...
It's really counterproductive to differentiate it in this way. It doesn't help the cause at all. The license of firmware blob is really not a kernels issue. It's a lower layer - you can go for that when you start requiring all your hardware to be open. Otherwise, it's like bashing Mozilla due to Firefox not providing an option to load only Creative Commons or similarly free content from the Web.
The kernel is here just to upload a bitstream to memory of some device. For kernel, it doesn't matter if it's open or not. It's just set of bits needed to be sent somewhere, nothing more, nothing less. It not like a typical software, as it's not even executed on your system.
For example, I would appreciate a "noblob" kernel option. If my system doesn't boot, I may reverse that choice, but I can also (still) hope for, and request, useful information along the way so I can choose the closest to ideal (for me) from the options available.
At some point it becomes a "if you don't like it, submit a patch" scenario, which is fine. It is more likely to lead to that when the response isn't overwhelmingly dismissive or insulting.
However, ideologically it doesn't make any sense, since when your kernel is booted you're most likely already past the loading of lots of non-free blobs in lots of chips. Kernel deals only with the ones that don't have internal memory to store them.
That's clearly not what you were talking about. It's true that drivers usually also contain firmwares, or microcodes for GPU. However, removing them will most likely prevent you from using GPU at all, not just "fast 3D".
And even if it doesn't, because the most important parts of that non-free firmware are already sealed in memory on the card, it's no different from, let's take the easier example, any HDD out there. So either your blob-free system works almost perfectly, cause you only take the software running on main CPU into account, or your blob-free system doesn't work at all, cause you actually removed all firmware blobs and basically nothing works anymore, including HDD, CD drive and motherboard.
On the contrary, the mere fact that it is a series of bits under user control - pure software for which there is a user-initiated loading mechanism - is precisely what makes it a the subject of peoples' libre software goals. Furthermore, it is because it is required that the user perform the initialization to use the main features that the user is constantly reminded about the proprietary dependency. Hard drives, for example, aren't a big focus for advocates because the drives generally don't forget they are hard drives between boots. It is true that they CAN be re-flashed, but it is not conventionally necessary to run, and that places them on a much more remote wish list. The main responsibility of the kernel in such cases is to prevent unauthorized firmware changes coming from userspace (ex, the NSA, over the network).
Boot-time firmware is qualitatively different from hardware that ships with working firmware, just as the latter is a different beast than ROM-based blobs.
Sure, it's harder to mess with reflashable firmware stored inside the device. However, it's only just like you said - a change on priority list, nothing else. Differentiating this in terms of freedom is a hypocrisy.
Anyway. Kernel just provides you an interface to load such firmware into your device. When the firmware is freely redistributable, it's also included in the kernel source tree. Nothing more, nothing less. It's up to you to decide what to do with it - you can easily look up all the needed info and remove unwanted firmware files.
For me there's no difference where it's stored. Closed firmware is closed firmware. We either have to accept it or not use PCs at all.
In the meantime, we should rather focus on producing stuff that increasingly stops requiring it, instead of aimlessly fighting with kernel developers.
Only, this process could be made more clear and more readily configurable (as I suggest, for example, a libre verbosity level, and noblob params). It is also not hypocrisy to fight battles in order of proximity - rather, it is strategy.
I went down the road recently to try and build a binary firmware free desktop. I gave up. I did manage a no non-free firmware build, which runs OpenBSD spectacularly, but I couldn't get rid of all firmware blobs.
It seems like open hardware is very hard to come by in this respect. Even swapping out your bios for coreboot is not possible for many (most?) systems--never mind all the other umpteen specialty computers running in your computer. A computer isn't one computer these days, and you will have a hard time taking a principled stand on all your software being free.
I'm not an expert, but I won't be surprised if it's completely unrealistic.
One usually rarely flashes the motherboard firmware (BIOS/EFI) or the graphic card firmware. A lot can go wrong, if the power goes down or the computer crash during a firmware flash-process, then the hardware is dead (except there is a backup-chip/logic like on more expensive motherboard e.g. Asus and Android devices).
> hard disk controllers
One usually never flash the HDD firmware. The same goes for the GSM/3G modem in your smartphone which is a separate real time OS on a separate ARM CPU. It's a blob and one buys it and cannot change it. It's also a black-box for Apple and Google and they cannot change them. There only less than a handful real completely open source smartphones with open source firmware ("open source hardware"), one example is OpenMoko (2011): http://en.wikipedia.org/wiki/Openmoko
The Linux/*BSD/Android world has a historical deficit in the availability of suitable drivers. Also the binary ABI isn't stable like on WindowsNT which means older drivers don't work with never kernels (e.g. Android upgrade means no camera). There are open source projects that could make a difference, if they get more donations like: ReactOS which aims to be binary compatible with Windows NT series hardware drivers. And NDISwrapper, that recreated a partial Windows NT kernel (using ReactOS code) inside a Linux kernel making it possible to execute modern Windows drivers in Linux. It consists of a NTOSKRNL API, a basic WDM controller, and a set of Windows call wrappers such as Wireless/NDIS/USB/PnP to Linux APIs.
The best you can get regarding openness of devices that can be practically used is to completely separate the main CPU from baseband processor (sans some generic interface like USB or serial) and this is exactly what Openmoko, and now also Neo900 are doing. Neo900 project goes one step further - it assumes the modem is a rogue black box by default, so deploys systems that allow the user to monitor the modem.
Unfortunately, mainstream devices go further and further away from that separation and now most of the time use fully integrated designs, where the non-free firmware of the baseband has full access to user OS via shared RAM or even shared CPU.
http://www.malwaretech.com/2015/05/hard-disk-firmware-hackin...
But, oddly enough, I have yet to hear anyone calling for open-source hard drives...
I think controllers used to be dedicated circuitry, but now they are more like CPUs with firmware.
http://en.wikipedia.org/wiki/Disk_controller
http://en.wikipedia.org/wiki/Controller_(computing)
In a modern computer, there are dozens of these little chips, each having security implications. Nobody knows the provenance of all the code since they are mostly manufactured overseas.
Fruitless debates about HDD firmware (and many more) are an undying hobby of some people in the wider coreboot community. These days the topic comes up every week or two.
Perhaps being a maintainer of a significant infrastructure project like an OS kernel wears on a person and builds a bit of a short-fuse temper :)
A reason IMHO that OpenBSD has never reached the level of notoriety of FreeBSD is Theo's childish behaviour. I can understand a 20-something insulting someone like that, but not Theo.
ps. I didn't quite understand the Linus reference, but anyway.
This can be interpreted both ways, however. Where Linus will be more elusive, Theo will have no qualms with just barging in and telling you to read the source code, ivory tower be damned. Almost as if he sort of respects even newbies enough to slap some quick advice to their face.
A funny example is here: http://comments.gmane.org/gmane.os.openbsd.misc/212565
A minor reason, perhaps. There are other, more serious concerns.
"Hi guys, I know you've been working for years on this! I haven't really bothered to look at your code, or understand the principles of operation, but I have strong opinions about How Things Should Be Done and I'm here to tell you how to run your project."
The reason Theo himself replies to this sort of thing, and in this sort of fashion, is to make it absolutely clear (both to this imbecile and anyone who seeks to emulate him) that the people who want this sort of thing are wrong and will be wasting their effort in asking for it.
I loved this mail, and it's actually making OpenBSD stand out as a human project, because it's not riddled with indecipherable political correctness. You can feel there are people involved, that they didn't copy-pasted a template but actually thought out their words, you can feel the involvement of the people. On the other end of the spectrum a smiley would also make the project more human, closer to the people. This mail also serves as a warning to future people who will want to engage a similar path of "I don't know anything, but I think things should be done This Way" to actually think about it. A net advantage.
Yes, the mail is harsh; that doesn't mean countrygeek should have a problem with it. The latter can (and should) absolutely go back to the research part, try to understand what he asks for, and then come back with a plan. That is what is really asked from the kind of people in the same position.
Why should fools be suffered? OpenBSD developers have an OS to develop.
Also, they can be very different in their nature. Firmwares can contains fairly complex operating systems that run on some microcontroller inside your hard drive, they can be hand-crafted for particular piece of hardware (so there's no source code) or they may only consist of some data structures needed to properly initialize the registers or other stuff like that.
The property of being handcrafted for a particular piece of hardware does not imply that there is no source code.
A few other firmwares are slightly less free. Meaning they are free
for money, but they try to stipulate subtle rules we do not want to
impact the freedom of our source tree with. ... OpenBSD
has a tool built in which will download those, so that our base source
tree remains full of freedom.