Following yesterday’s Linux 6.18 kernel release, GNU Linux-libre 6.18-gnu is out today as the latest release of this free software purist kernel that will drop/block drivers from loading microcode/firmware considered non-free-software and other restrictions in the name of not pushing binary blobs even when needed for hardware support/functionality on otherwise open-source drivers.
With Linux 6.18 there are more upstream kernel drivers dependent upon binary-only firmware/microcode. Among the drivers called out this cycle are the open-source NVIDIA Nova-Core Rust driver as well as the modern Intel Xe driver. Nova-Core is exclusively designed around the NVIDIA GPU System Processor (GSP) usage and thus without its firmware the driver is inoperable. Similarly, with the newer Intel Xe driver depending upon the GuC micro-controller without its firmware the support is also rendered useless.
@Scoopta @cm0002 The point is that if everything was open Linux support would be so much better as we would understand the working of the hardware so much better, and we should do everything we can to discourage manufacturers from adopting this stance. FOSS has the great benefit that anyone in the world can improve it, and then share their improvements with everyone else. That makes a better world. Just better.
To be clear, I’m not saying I don’t want open hardware, what I’m saying is I don’t get the point of allowing closed hardware that doesn’t require a firmware blob as opposed to closed hardware that does. That’s a very arbitrary and silly line that does nothing useful. They’re going on this crusade of “no blobs.” But why? There’s lots of hardware that already has closed blobs on the HW, but because it’s not uploaded by the driver those blobs are ok? You either have to say all closed firmware is bad and we’re going to take a stance against any devices which have any amount of closed firmware, even when shipped on ROM in the HW. Or, closed firmware is tolerable so long as the driver is fully FOSS. I love the idea of not having closed firmware but I just don’t get the intellectual inconsistency here.
Keep up the good fight Scoopia.
It is clear from the responses you get that people just do not understand what you are saying. But keep trying.
Imagine the FSF argument at the level of the whole computer. If the computer lets you run software, it has to be free software. But if the computer has all its software in hardware that cannot be programmed or updated, then it is not only totally fine but superior to a Windows computer because Windows is closed software even though the Windows computer let’s you run free software and the non-programmable one does not.
Running on Windows is evil because it is software. But running on closed non-updatable firmware baked into your closed, proprietary hardware is good. That is what the FSF has to say. Programmability bad.
There are two ways to fix the FSF position:
1 - they demand that everything be free including all hardware and all hardware running on it. But that means the purists have to stop using all the closed hardware they enjoy today.
2 - the FSF continues with the position that they are ok with closed hardware and defines “free software” to exclude firmware.
There are three words for a a reason, they are three different things: hardware, firmware, and software.
Defining firmware as software makes no more sense than defining it as hardware. In fact, the latter makes more sense to me and would fix the FSF silliness. But what makes the most sense is to acknowledge that they are all different.
Now that I type this, we need a Free Software Foundation, a Free Hardware Foundation, and Free Firmware Foundation.
The FSF mission makes sense if you exclude firmware. The FFF could preach that free firmware is superior to closed firmware. No argument there. The FHF can push for free hardware. That would be great and we could all push for it. But the FHF could also acknowledge that programmable hardware is superior to non-programmable hardware even when only closed firmware exists. The hardware itself is more open and more free. Basically both the FSF and the FHF could be more relaxed about the other. The FHF could be ok with closed software on open hardware just like the FSF is ok with running free software on closed hardware (their stance today).
Honestly, the above is maybe the only sane solution.
That would allow all of us to:
I bet every one of us would agree to the above. Or at least we could choose which of the three missions to endorse. At least they would all be sane and consistent.
Instead, we get the FSF telling us to ignore the closed hardware behind the curtain while choosing hardware that is more restrictive and less free in the name of avoiding binary blobs. We are being forced to fight a religious war and nobody ever wins those.
Yeah, that would be a much more consistent setup and I agree with everything you said here. I just don’t understand how being less programmable is good, it isn’t, I don’t see any world in which it is unless there is truly NO firmware involved and it’s pure HW.
The line is basically programmability. If a device runs “firmware” that can’t be changed, that’s really just a part of the design of the hardware you bought, so the fact that you can’t see or modify the source code is irrelevant- even if you could, it wouldn’t give you any more control over the hardware. If it runs firmware that can be changed, it’s a programmable computer, and by running proprietary firmware, you’re giving up control you would otherwise have over your computer.
Exactly. It is about programmability. And that is dumb.
Imagine you have a piece of hardware that is not programmable. The FSF says that hardware is fine. Buy the shit out it and “be free” apparently.
Version 2 of that hardware is released with the new feature that the firmware is upgradable. Of course, only closed firmware updates are available initially. According to the FSF, this programmable hardware must now be avoided. Keep buying the original “more free” version that cannot be programmed.
And if you do have hardware version 2, the FSF says you should at least never update your firmware. Nevermind new features. Security fixes are to be avoided. Because the baked in firmware is more free than the firmware update. It is not that you are not using closed firmware. Of course you are. But you did not change it. So that is better?
It is total nonsense.
If there was a Free Hardware Foundation, a device whose hardware was programmable and whose firmware could be upgraded would clearly be seen as superior to one that was completely closed. It is definitely more open, “more free” hardware even if only closed firmware is available. The hardware is obviously more free. Self-evidently.
But the FSF position is that this “more free” hardware is less free than fully closed options when only closed firmware exists. There is no way for that to make sense unless you move “firmware” into the software bucket and completely ignore the concept of hardware all together. Sorry, but that is dumb.
It is also a good way to roadblock progress towards open hardware. Please stop.
An organization which exists exclusively to advocate for a type of program caring about programmability is not dumb. That seems… kinda obvious? They don’t exist to rate the technical superiority or inferiority of hardware devices, they exist to advocate for the simple position that: if a device can be programmed, the user of that device ought to control the program on the device, not some company which happens to hold the copyright over the on-device program.
Um… absolutely not? They say that running proprietary firmware represents an injustice (perpetrated by the copyright holders of the firmware, btw, not the user). Updating the firmware to free software would obviously be great in the eyes of the FSF; upgrading to proprietary firmware would be simply continuing the existing, unjust status quo. You appear to have completely made up this particular position.
I think you missed the point. Imagine 2 devices, device A has a chip with flash memory that contains a binary blob with firmware. Device B doesn’t have built-in flash storage so it requires the driver to load the same binary blob during boot. Both devices are reprogrammable and both contain the same closed source firmware. However device A would be allowed but device B would not. From my point of view they are the same device. The fact that you don’t know how to reprogram device A doesn’t make it more or less proprietary.
The very silly argument the FSF is trying to make is that device A is not programmable because the firmware is baked into the HW effectively making it part of the HW rather than a separate entity. Therefore it’s a HW limitation and not proprietary software. Device B on the other hand has proprietary software uploaded to it which is not to be allowed under any circumstances and therefore must be neutered. I call it silly because as you so rightfully point out, the firmware blob could be literally the same exact blob, just stored differently