| 1 |
On Sun, 7 Apr 2013 11:13:10 +0100 |
| 2 |
Markos Chandras <hwoarang@g.o> wrote: |
| 3 |
|
| 4 |
> On 7 April 2013 08:27, Ciaran McCreesh <ciaran.mccreesh@××××××××××.com> wrote: |
| 5 |
> > On Sat, 6 Apr 2013 20:05:11 -0600 |
| 6 |
> > Ryan Hill <dirtyepic@g.o> wrote: |
| 7 |
> >> In short: |
| 8 |
> >> Toolchain packages, for better or worse, are built by eclass. We are |
| 9 |
> >> not forward-porting toolchain.eclass every time someone decides there |
| 10 |
> >> are too many EAPIs in the tree. Every change to that eclass breaks |
| 11 |
> >> something (the trick is to break things people don't care about any |
| 12 |
> >> more and hope no one notices). I don't know the ins and outs of |
| 13 |
> >> glibc's eblits but I doubt they would be simple to port either. I |
| 14 |
> >> also don't know much about toolchain-binutils.eclass, but it seems |
| 15 |
> >> like it would be doable. |
| 16 |
> > |
| 17 |
> > Sounds like a good opportunity to replace toolchain.eclass with |
| 18 |
> > something clean and understandable. |
| 19 |
> |
| 20 |
> I see no reason to break something that works just fine as it is. |
| 21 |
|
| 22 |
I'm afraid this is quite short-sighted. |
| 23 |
|
| 24 |
Well, I have to believe that it works since I don't even try to get |
| 25 |
nearby those ebuilds. Those are simply impossible to understand without |
| 26 |
detailed study of all involved files, and those are huge. |
| 27 |
|
| 28 |
The problem is that we can't assume it will work forever, and we can't |
| 29 |
assume that if it stops working, there will be somebody around who |
| 30 |
knows what is going on in there. |
| 31 |
|
| 32 |
I see that the key components of Gentoo system are complex, fragile |
| 33 |
and understood only by the few people who maintain them *now*. That |
| 34 |
doesn't sound like a bright future to me. |
| 35 |
|
| 36 |
-- |
| 37 |
Best regards, |
| 38 |
Michał Górny |
Archives