| 1 |
Santiago M. Mola a écrit : |
| 2 |
> On Fri, May 30, 2008 at 9:16 AM, Mike Auty <ikelos@g.o> wrote: |
| 3 |
>> I've been using it for a pretty long time now (probably a couple weeks |
| 4 |
>> after Diego first blogged about it) and don't have many problems at all |
| 5 |
>> (now), but every once in a while a version bump or a new package will |
| 6 |
>> just fail to compile properly and the problem leads back to as-needed. |
| 7 |
>> I'm not sure whether ~arch users would be able to catch all the |
| 8 |
>> as-needed bugs before they hit stable, so I couldn't say whether it |
| 9 |
>> should be enabled by default or not. |
| 10 |
|
| 11 |
Our experience in the gnome herd is that most --as-needed issues come up |
| 12 |
*very* quickly: either the build fails, or application plugins don't get |
| 13 |
loaded. As such, they're quite easy to catch. Fixing them can be trickier. |
| 14 |
|
| 15 |
> --as-needed breaking legitimate code is a problem, though. I wonder if |
| 16 |
> we have that kind of code in any application in the tree and if we |
| 17 |
> have some way to detect it. |
| 18 |
|
| 19 |
--as-needed breaks legitimate C++ code, I have yet to see it break plain |
| 20 |
C code (but I could be wrong). Ciaran posted an example code a while |
| 21 |
ago, either here or on his blog. |
| 22 |
|
| 23 |
His example is valid C++ code (basically it was about implicitly loading |
| 24 |
.so plugins) but IMHO, it's just not a good engineering practice. Plugin |
| 25 |
loading should be explicit (readdir + dlopen). But that's my opinion. |
| 26 |
There could be a couple other odd cases but I don't remember them. |
| 27 |
|
| 28 |
All I can say is that --as-needed *today* does much more good than harm. |
| 29 |
|
| 30 |
Cheers |
| 31 |
|
| 32 |
-- |
| 33 |
Rémi Cardona |
| 34 |
LRI, INRIA |
| 35 |
remi.cardona@×××.fr |
| 36 |
remi@g.o |
| 37 |
-- |
| 38 |
gentoo-dev@l.g.o mailing list |
Archives