| 1 |
On Fri, Apr 19, 2013 at 10:11:50AM -0400, Rick "Zero_Chaos" Farina wrote: |
| 2 |
> On 04/16/2013 03:42 PM, W. Trevor King wrote: |
| 3 |
> > From: "W. Trevor King" <wking@×××××××.us> |
| 4 |
> > |
| 5 |
> > The current approach to avoiding problems due to stale binary packages |
| 6 |
> > with untracked ABI dependencies is to disable binpkg use during |
| 7 |
> > troublesome sections of the build (e.g. seed updates). I think that a |
| 8 |
> > cleaner solution would be to use a configurable spec option |
| 9 |
> > blacklisting binpkgs for troublesome packages. For example, in a |
| 10 |
> > stage1 with update_seed enabled, the Portage emerge (before the seed |
| 11 |
> > update) has: |
| 12 |
> |
| 13 |
> This needs to remain optional. |
| 14 |
|
| 15 |
It is optional. Just set `binpkg_blacklist` explicitly to an empty |
| 16 |
string if you don't want to blacklist the default packages (currently |
| 17 |
only two). |
| 18 |
|
| 19 |
> I'm not going to nack a patch that some people may find useful but |
| 20 |
> in my personal opinion this is a terrible solution that should not |
| 21 |
> be used. We need to find a way to rebuild packages as needed (like |
| 22 |
> EAPI 5) not force a rebuild everytime. |
| 23 |
|
| 24 |
Agreed, but in the absence of one of the following: |
| 25 |
|
| 26 |
* a tree full of EAPI-5+ packages that correctly use ABI sub-slotting |
| 27 |
(if I live long enough to see this, I'll be very happy ;), |
| 28 |
* local overlays fixing ABI sub-slotting (maintained by folks without |
| 29 |
much experience in the packages in question?), or |
| 30 |
* Portage tweaks to work around packages that don't use EAPI-5+ (or |
| 31 |
that do, but don't use ABI sub-slots appropriately). |
| 32 |
|
| 33 |
I'm fairly confident that none of those are happening in the next six |
| 34 |
months, and there seems to be agreement that catalyst needs some sort |
| 35 |
of local hack to work around the problem until one of the real |
| 36 |
solutions lands. |
| 37 |
|
| 38 |
The `binpkg_blacklist` option lets you name (on a per-package level) |
| 39 |
troublesome packages that get ABI sub-slots wrong (or don't even try). |
| 40 |
I think that this approach will force *fewer* rebuilds than the |
| 41 |
current approach (from e7ea409 and 6c0a577) which blacklists |
| 42 |
everything from the seed update. |
| 43 |
|
| 44 |
Another (non-exclusive) possibility is to put a big warning on the |
| 45 |
binpkgs option [1] and leave it up to folks to use at their own risk. |
| 46 |
|
| 47 |
Cheers, |
| 48 |
Trevor |
| 49 |
|
| 50 |
[1]: http://mid.gmane.org/cover.1366075786.git.wking@×××××××.us |
| 51 |
|
| 52 |
-- |
| 53 |
This email may be signed or encrypted with GnuPG (http://www.gnupg.org). |
| 54 |
For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy |
Archives