Ext4 is intended for development, not mainstream use. The patches for it are summarily being included in 1.39-r1 per bug 156697. These patches causes problems. For one, when I tried to initialize a 2+ TB filesystem, the 64bit flag was applied, but ext3 in the kernel does not about this flag and it was not able to use the filesystem. Presumably, there could be other problems. Please see messages on the ext3 mailing list posted by the programmer: https://www.redhat.com/archives/ext3-users/2007-February/msg00014.html and https://www.redhat.com/archives/ext3-users/2007-February/msg00017.html I suggest providing a separate ebuild for a version containing the patches and masking that version everywhere.
*** Bug 168315 has been marked as a duplicate of this bug. ***
With ext4 patches the e2fsprogs does not even pass its own tests, see bug #159834. Could we at least have an ext4 USE flag to control applying of the patches? And if enabled print a warning that the tool is not fully compatible with ext3.
That sounds reasonable to me.
*** Bug 159834 has been marked as a duplicate of this bug. ***
fixed in 1.39-r2
It's not fixed. You didn't provide ext4 flag or ebuild. I have to mask the newer version just to use ext4 - and this means I won't get the newest fixes.
nor do i plan to if you want ext4, use 1.39-r1 until it gets merged into a real version
(In reply to comment #7) > nor do i plan to > > if you want ext4, use 1.39-r1 until it gets merged into a real version > Grreate! And now when you've droped ext4 support and my system at work doesn't boot I must search in bug list, why. Ofcause, i fix my trouble, but would you possible to do more correct fixes next time.
maybe some day if you are a developer you can make these sort of calls ... until then, you're not having to maintain the package, so you can deal with the reality
*** Bug 212099 has been marked as a duplicate of this bug. ***