| 1 |
On Mon, Apr 6, 2015 at 8:02 AM, Martin Vaeth <martin@×××××.de> wrote: |
| 2 |
> My suggestion is something in between - less invasive |
| 3 |
> (and, in particular, less time consuming) |
| 4 |
> than your suggestion to recalculate the USE-settings |
| 5 |
> with every emerge, but more automatic than the current state. |
| 6 |
|
| 7 |
Keep in mind that keeping track of past decisions made by portage does |
| 8 |
not require user-editable config files in /etc. It just requires a |
| 9 |
cache of some kind, much as we do with installed packages/etc. |
| 10 |
|
| 11 |
That said, portage still has to spend time basically re-validating the |
| 12 |
consistency of the entire system because we allow the use of overlays |
| 13 |
and other situations that don't guarantee that portage will have some |
| 14 |
kind of consistent pre-calculated depgraph handed to it. If we |
| 15 |
required all repositories to have some kind of pre-generated cache in |
| 16 |
them and ensured that this was always up-to-date and better controlled |
| 17 |
the kinds of dependency changes we made, then maybe there might be an |
| 18 |
opportunity to offload some of the work to the repository level |
| 19 |
instead of doing it on every Gentoo system. Still, unless we banned |
| 20 |
overlays I'm not sure how much even this would buy you, since you'd |
| 21 |
have many of these that need to be merged somehow. |
| 22 |
|
| 23 |
From Zac's email and other discussions in the past it seems like we're |
| 24 |
basically committed to doing all these calculations all the time |
| 25 |
anyway, so we shouldn't be too shy about taking advantage of them. |
| 26 |
|
| 27 |
-- |
| 28 |
Rich |
Archives