The New Microsoft exFAT File-System Driver Has Landed In Linux 5.7

[CommunityMember]

we need ex4 support in windows

3rd parties have been offering file system drivers for Windows for ext[n], and hfs+, and apfs for some quite some time.

[angrypie]

we need ex4 support in windows

You spelled XFS wrong.

[CommunityMember]

Is there a legal issue on that?

Well, yes, and no. In order to get a code signing certificate for Windows in order to sign the kernel mode driver to be able to load it you need to spend a few hundred dollars, and provide various proofs of identities. Typically only corporations have processes to do that, and they want to make money on their development efforts, but in any case, if one wishes to do so as an independent, and if one lives in a litigious location (i.e. US) one has to be prepared for some future failure of the driver to corrupt/destroy data, or a bug in the kernel mode driver is used to exfiltrate data, and to be sued by some person who is claiming to be harmed (costs next to zero to sue, costs a lot to defend), so the individual would likely need (variously) Professional Liability Insurance, Directors and Officers Insurance (for their LLC/Corp), and Produce Liability Insurance in order to cover your assets (you like that house you live in, don't you? Hate to see it go to the lawyers to pay the bills....). Companies do these things all the time (individual developers rarely see or even understand all those overheads, but they are there).

[Lycanthropist]

@CommunityMember
The GPL already excludes any liability:

"This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details."

There is also btrfs support with WinBtrfs.

[zoomblab]

Long story short, with Linux 5.7 is a much better Microsoft exFAT file-system implementation that is more reliable and with more functionality than the older driver

How do we know all that? Are there QA tests that prove it?

while it will continue to receive improvements by Samsung and others.

Again how can we know that?

Something being newer does not automatically mean better IMO.

[Danny3]

we need ex4 support in windows

No, we don't need Windows spyware to be able to read our files in Linux.

[johannesburgel]

Let me get this straight, because it is so mind-boggling I can't really believe it:

Microsoft shoehorned their crappy FAT successor into the SDXC standard, forcing the whole world to license their trivial patents, something which should have never happened in the first place. Then they made gazillions from royalties, to the point where they are making more money from every Android device than Google does on the whole OS. They don't even maintain a reference implementation of exFAT, so besides having to pay royalties everybody also has to buy or maintain their own exFAT drivers.

15 years later, shortly before the patents run out anyway, they pull a marketing stunt. Linux now ships a "Microsoft" exFAT driver that wasn't written by them, wasn't contributed by them and isn't maintained by them. They haven't even contributed the patents to the OIN yet, more than half a year after "looking into it".

Objectively speaking Microsoft got billions for trivial patents and didn't give back or do anything. Zero. Nada. They can't even be bothered to put a simple signature under the OIN contract. Yet the get the praise for "having changed" and "loving Linux".

Yes, they have changed. This is not the Microsoft from the 90s. This is much worse.

[oiaohm]

How do we know all that? Are there QA tests that prove it?

In this case the new exfat driver the same exfat driver used in every samsung phone that has been modified by samsung to be mainlined forever more. So this is a driver that has been actively used in production for over a decade to take the rough edges off.

The case is the prior mainline exfat driver that the samsung exfat driver is replacing is the young new kid on the block. So yes newer is not always better in this case the new exfat driver that went mainline first is way worse than the samsung exfat driver that has over a decade of development under it belt already.

Again how can we know that?.

https://lkml.org/lkml/2020/1/18/114

The merge into mainline kernel request and samsung developer is the maintainer of the exfat driver going forwards after 5.7.

The annoying part about is this driver from Samsung could have been merged mainline over a decade ago if Microsoft was not being jackass with patents.

[Guest]

we need ex4 support in windows

Not at all,... Or, exploits could easily access yours root partition in unprotected access,... Or, secure it by some mapping between windows user accounts and Linux UID(s) / GID(s), and still it can damage yours files (ie. create an exploit alias for su/sudo),... And, still allow spyware to collect yours private, sensitive and DNA-covered, data,...

No, missing ext4 FS driver is not blocker/slowdown of productivity for anybody,...

If you need (are forced to have) windows for corporate work, then ext4 would probably complicate yours life (as it's not very compatible to windows user management)... Though, there might be some rage edge cases, where it might make sense, it's not worth it for such extreme minority,..

[board]

and Ubuntu loves enabling broken or buggy things on the kernel, just ask those bricked laptops.

An OS shouldn't be able to brick a computer. A computer that can get bricked this way has a defective design. This is not the OS's fault. It's like taking a step in your own home and causing the entire building's structure to collapse onto itself. Is it your fault for being so clumsy and taking the step or is it the fault of whoever built your home? Same thing.