| 1 |
On 07/07/2012 11:54 AM, Peter Stuge wrote: |
| 2 |
> Ian Stakenvicius wrote: |
| 3 |
>>> Is it unrealistic to assume that upstream ABI providers will mark |
| 4 |
>>> their ABIs by using sonames correctly? |
| 5 |
>>> |
| 6 |
>>> Maybe that is at least the common case, then ABI_SLOT could be set |
| 7 |
>>> automatically. |
| 8 |
>> |
| 9 |
>> Although we have a lot of this information available (which is why/how |
| 10 |
>> @preserved-libs works, for instance), there is no way for portage to |
| 11 |
>> know *prior to emerging the update* if abi has changed. |
| 12 |
> |
| 13 |
> Knowing that a library changes ABI before building is nice, but not |
| 14 |
> relying on a human to encode this is nicer still. |
| 15 |
> |
| 16 |
> After compile, before install, there is an opportunity to show the |
| 17 |
> effects of installing the package. |
| 18 |
> |
| 19 |
> I'm only talking about the context of the developer who is adding the |
| 20 |
> new ebuild. So ABI_SLOT (or / SLOT) would be set automatically just |
| 21 |
> once, in the process of committing the ebuild. |
| 22 |
|
| 23 |
Well, if you're talking about having portage automatically edit the |
| 24 |
ebuild, I don't think we want to do that. If developers use |
| 25 |
portage-2.2_alpha with preserve-libs, then they'll know automatically |
| 26 |
when there's an SONAME change that triggers preserve-libs. Part of the |
| 27 |
beauty of the approach used in EAPI 4-slot-abi is that that it can be |
| 28 |
used to trigger rebuilds in cases that don't even involve SONAME |
| 29 |
dependencies (consider a "pure" perl module that only needs to be |
| 30 |
rebuilt in order to install its interpreted *.pm files in a different |
| 31 |
directory for a new version of perl). |
| 32 |
|
| 33 |
> No need to repeat the |
| 34 |
> binary analysis - except if one source package has different ABI for |
| 35 |
> different ARCHes. Fun! |
| 36 |
|
| 37 |
It might be nice to add some binary analysis things beyond preserve-libs |
| 38 |
in the future. However, EAPI 4-slot-abi should work quite well even |
| 39 |
without that. It just automates rebuilds [1] that the user was |
| 40 |
previously required to handle manually, when prompted by elog messages, |
| 41 |
or by running tools like revdep-rebuild and perl-cleaner. It's being |
| 42 |
tested in the axs overlay [2], and it seems to be working pretty well. |
| 43 |
|
| 44 |
[1] |
| 45 |
http://blogs.gentoo.org/zmedico/2012/06/23/automatic-rebuilds-with-experimental-eapi-4-slot-abi/ |
| 46 |
[2] http://git.overlays.gentoo.org/gitweb/?p=dev/axs.git;a=summary |
| 47 |
-- |
| 48 |
Thanks, |
| 49 |
Zac |
Archives