Given the following rationale; Huge files in common desktops, even servers are rare. The only situation on a desktop a hugefile (2TB+) is used is in the case of an extreme size virtualmachine image. Why would we want to format our ext4 filesystem with the huge_file flag? Which basically brings us more pain and misery because huge_file actually requires an extra kernel option (CONFIG_LBDAF) for R/W mounting. The standard configuration of mke2fs.conf includes huge_file for ext4. My suggestion would be: Introduce the useflag huge_file, this useflag would control the contents and default configuration of mke2fs.conf. And basically adds or removes the option appropriately.
I would like to see this too
if we had a common USE flag that people were commonly setting system-wide, i'd be inclined to tweak this. but we're talking about a local use flag which someone has to read and manually add for their build. i dont see how that is any different from editing mke2fs.conf yourself. further, this flag only affects initial fs creation ... it doesnt have any bearing on mounting/fscking existing ones.
The question here is whether the default makes sense? And as the original reporter stated, it doesn't as people rarely want to have partitions for huge_files in reality. We should probably then switch the default and let the users alter that setting if needed.
if the default doesn't make sense, then get it changed upstream. a USE flag doesn't make sense either way for the reasons i already stated.
