| 1 |
On 01/08/13 19:54, Samuli Suominen wrote: |
| 2 |
> still, first the patch goes upstream and after upstream review and |
| 3 |
> commit to git it goes in tree otherwise we opt to the fallback and |
| 4 |
> disable udev from lvm2/cryptsetup when USE=static is enabled (like |
| 5 |
> cryptsetup upstream suggested to me) gentoo-specific patches mangling |
| 6 |
> namespace of udev, kmod, whatever doesn't sound good at all however |
| 7 |
> working it with upstream sounds great |
| 8 |
|
| 9 |
I just spent an evening introducing a friend willing to have a look the |
| 10 |
codebase. My solution to the problem of clashing symbols had been 3 fold: |
| 11 |
|
| 12 |
- many functions are small and already inline, they are using in the |
| 13 |
tool (bad practice IMHO) and in the library once. Making them static is |
| 14 |
easy and works. |
| 15 |
|
| 16 |
- some functions are using inside the library a couple of times, adding |
| 17 |
an (ugly) privkm_ is enough to avoid the problem. |
| 18 |
|
| 19 |
- some functions were used just in the tool and not in the library, |
| 20 |
moved where it belongs. |
| 21 |
|
| 22 |
Instead of running around in circles screaming static linking is unholy |
| 23 |
fixing it that way wouldn't had take much... |
| 24 |
|
| 25 |
I won't expect upstream picking up what Federico wrote anytime out of |
| 26 |
pride more than technical merit. |
| 27 |
|
| 28 |
lu |
Archives