Recent searches
Search options
The SystemV filesystem is getting removed from the #Linux kernel, as it had a bad bug proving no one was actually using it.
@atoponce In this case it probably doesn't matter, but this justification in general is bogus.
Often, it means the bug was introduced in like 5.4, and folks using the functionality are still using 4.x or even 3.x because that's what supports their devices or because there was no compelling reason to upgrade.
And then "feature they need was removed because the fact that it was broken on 5.4+ proved nobody was using it" ensures that they stay on the old version forever, in a self-reinforcing cycle.
@atoponce There's an analogue of XKCD 2501 for software versions, where people in software grossly overestimate the version numbers that are out there in the real world.
@atoponce "OMG you haven't updated to 6.x yet?? You're still on 5.15.178? or *gasp* 5.4.290?"
"Um, no. 2.6.12."
@dalias@hachyderm.io @atoponce@fosstodon.org no kidding - in kernel 4 days I had a realtek/mediatek card that only provided drivers for 2.6… luckily I was able to get an Intel based chipset but there’s a lot of random hardware servers use that may not be as lucky.
@dalias@hachyderm.io @atoponce@fosstodon.org Also the "proving nobody uses it"… see, I remember back in 5.15 they introduced NTFS support in kernel. It worked flawlessly on 5.15, a couple months ago on kernel 6 (I don’t even remember what version of 6) I tried it out again… thing has regressed and kept corrupting my NTFS filesystems to the point I just switched back to NTFS-3G. We can make the same argument of "so broken clearly nobody was using it" because I use a fallback instead of going through the whole process of reporting that.
Bug reporting could also be easier but it shouldn't be nearly as necessary.