Rich Freeman posted on Tue, 06 Oct 2015 21:52:27 -0400 as excerpted: > On Tue, Oct 6, 2015 at 8:27 PM, Zac Medico <zmed...@gentoo.org> wrote: >> >> Yeah, there can be a benefit, as long as you're not one of the people >> who uses --changed-deps for all updates (revbumps for dependency >> changes are basically irrelevant to these people). >> >> > Presumably those inclined to do this would be likely to stop if our QA > standards eliminated the need for it.
As a user, I use --changed-deps, and I know I'd like to stop using it, were I able to do it due to updated QA standards. [TL;DR folks stop there, or skip to last paragraph summary.] ATM, it's just simpler to run with --changed-deps, even with all the extra rebuilds (particularly so because quick-cycle rebuilds means ccache still has them cached), than it is to manually sort out all not automatically resolved blockers that I get without it. It's a lot more rebuilds, but I can look at them, see that big R and nicely colored --newuse changes, and not worry too much about it. And because I can do other things during the rebuilds and with the metadata-cached git repos I'm updating much more frequently again so they don't stack up, it's some extra hardware-CPU cycles at little cost other than electricity, vs much higher cost wetware-CPU cycles. But if sane revbumps on deps-change becomes the norm, portage should resolve most of those blockers automatically because affected packages should all get at least a revbump together, so I won't be left with an unresolved blockage because some affected installed packages weren't revbumped, and thus portage won't be left with stale deps blocking the upgrade of the others. Right now it's a simple matter of extra raw hardware CPU cycles making far more expensive wetware CPU cycles avoidable. =:^( If those wetware CPU cycles were avoided with proper revbumps in the first place, I wouldn't need to make that choice. =:^) With --newuse, because they may affect the way I use the package, I want to be informed of the USE changes in any case, while with deps, particularly simple subslot deps, I'd prefer to avoid worrying about it in the first place, and the easiest way to do that right now is with --changed-deps, tho a proper revbump policy should avoid that. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman
