| 1 |
Donnie Berkholz wrote: |
| 2 |
> If you turn off features you need, things break. There's nothing new |
| 3 |
> about that. If you disable ext3 support in your kernel, you can't mount |
| 4 |
> an ext3 partition and you'll get an error during boot about not finding |
| 5 |
> the root. |
| 6 |
|
| 7 |
I see your point, but extended attributes aren't as common as ext3, are |
| 8 |
they. And stuff that got broken because Portage suddenly started to |
| 9 |
require new features on the kernel side is bad. |
| 10 |
|
| 11 |
> The idea of the sqlite-based fallback is what's interesting here. |
| 12 |
|
| 13 |
If it is a fallback that must be supported (because of NFS), then there |
| 14 |
isn't much point in using xattrs. What benefits do they provide? There's |
| 15 |
no speed constraint here as we already cache metadata somewhere. |
| 16 |
|
| 17 |
>> Also note that in some circumstances like when running in a |
| 18 |
>> virtualized environment, imposing additional requirements on the kernel |
| 19 |
>> might be problematic. |
| 20 |
> |
| 21 |
> Why's that? |
| 22 |
|
| 23 |
Things with Xen got better than they were, but I can imagine a situation |
| 24 |
where some hosting provider offers their customers a virtual Xen box and |
| 25 |
their kernel configuration doesn't include extended attributes. You |
| 26 |
can't use your own kernel without access to dom0. |
| 27 |
|
| 28 |
>> It wouldn't be great to require extended attributes for each and every |
| 29 |
>> Gentoo box... |
| 30 |
> |
| 31 |
> Why not? |
| 32 |
|
| 33 |
Because they aren't so common, NFS doesn't support them and we haven't |
| 34 |
ever required them. |
| 35 |
|
| 36 |
Cheers, |
| 37 |
-jkt |
| 38 |
|
| 39 |
-- |
| 40 |
cd /local/pub && more beer > /dev/mouth |
Archives