| 1 |
On Wed, 22 Aug 2018 07:26:24 -0500 |
| 2 |
Ben Kohler <bkohler@g.o> wrote: |
| 3 |
|
| 4 |
> Thoughts? |
| 5 |
|
| 6 |
Is there a good reason we can't have a legacy profile for this? |
| 7 |
|
| 8 |
Or perhaps, a new (exp) arch entirely dedicated to legacy x86? |
| 9 |
|
| 10 |
The latter would be ideal for ensuring everything we *claim* works on |
| 11 |
i486 does indeed work there, and making it blatantly obvious what does |
| 12 |
and doesn't work on i486. |
| 13 |
|
| 14 |
And both strategies make it possible to restrict USE graphs to subsets |
| 15 |
that can only work on i486. |
| 16 |
|
| 17 |
The benefit of a new profile for x86 is it would assume everything |
| 18 |
currently x86 works on i486, and then exceptions can be shot on a |
| 19 |
case-by-case basis. |
| 20 |
|
| 21 |
But that's also the downside, you start with a huge set and play |
| 22 |
whack-a-mole with it, when ideally, you want to start with a *minimal* |
| 23 |
working subset and build it out ( and this is where the second option |
| 24 |
is better ) |
| 25 |
|
| 26 |
The arch-approach would also lend itself to compile-time switching in |
| 27 |
individual packages too. |
Archives