Would Phoronix do Gentoo Linux versus Ubuntu Linux benchmarks?

[duby229]

Personally I stopped using portage few years ago and never had problems with my systems There are better package managers for gentoo then old portage.

What package manager are you using right now? I've thought about giving paludis a try, but am still uncertain. Any advise or links how to get started?

[BlackStar]

Which optimizations should be done for Ubuntu, as you, after all, can compile your own package for Ubuntu (and Debian) too, using about the same optimizations as for Gentoo.

You don't understand, the point is to make Gentoo win the benchmark, not vice versa! I mean, otherwise the benchmarks will just show how pointless it is to turn to Gentoo purely for performance reasons.

I wonder why they didn't test 64bit though. Who in his right mind would use Gentoo on a 32bit (i.e. Atom) CPU anyway?

[dagger]

If you dont believe me then test it. I'm sure it is supposed to work the way you think it does, but it doesnt.

As far as preserved libs go, it is known to be broken in 2.2. Please dont recommend that people use it right now unless you tell them it is for bug testing only. I looked at the code, but I cant figure out the structure. I'm just not sure what they are trying to do with it.

Personally I stopped using portage few years ago and never had problems with my systems There are better package managers for gentoo then old portage.

[mirv]

I used to think that there wouldn't be much use to phoronix running gentoo tests, but actually it might be useful for one reason: it's not ubuntu. Running everything off ubuntu results in a heavy reliance upon the way that ubuntu does things - which may cause various conclusions to be ultimately flawed.
Gentoo, Fedora, openSUSE, something else to compare the tests against would provide much more useful information than only running tests against ubuntu.

[duby229]

revdep-rebuild recompiles library users. It is not always accurate, but cases where it fails are usually caught and documented long before the package becomes stable. duby229's characterization of revdep-rebuild is incorrect.

The library preservation feature which ferreira talked about is new to portage 2.2. It is not the same as slotting.

If you dont believe me then test it. I'm sure it is supposed to work the way you think it does, but it doesnt.

As far as preserved libs go, it is known to be broken in 2.2. Please dont recommend that people use it right now unless you tell them it is for bug testing only. I looked at the code, but I cant figure out the structure. I'm just not sure what they are trying to do with it.

[Shining Arcanine]

Upgrading Gentoo works fine for me. If it is possible that you are the first one with a configuration that upgrades will break, but like with all distributions, that is when you file a bug report.

[chithanh]

revdep-rebuild recompiles library users. It is not always accurate, but cases where it fails are usually caught and documented long before the package becomes stable. duby229's characterization of revdep-rebuild is incorrect.

The library preservation feature which ferreira talked about is new to portage 2.2. It is not the same as slotting.

[duby229]

Your application would continue to work, because Portage would preserve the older version of the library, so rebuild wouldn't be immediately needed anyway.

Thats only rarely true. There are only a few cases where portage will slot a package, and in those cases there is usually an *-config or a config-* or an eselect plugin to switch between the slotted versions. Besides portages slotting capability is too crude to be much use. It is necessary for some packages like gcc, others like python, or java are slotted, but I've come into problems where the switching application doesnt symlink something at all, or it symlinks to the wrong files. These scripts are better now then they were 5 years ago, but still not perfect.

Slotting is a solution, but it isnt the best solution.

[ferreira]

Your application would continue to work, because Portage would preserve the older version of the library, so rebuild wouldn't be immediately needed anyway.

[duby229]

No, revdep-rebuild would recompile AppA in your example.

That is how it should work I agree but it definitely doesnt. Set up an example to test it yourself. Upgrade a library for which you know some application depends on the older version, then run revdep-rebuild it always attempts to recompile the library.