Neil Bothwick <n...@digimed.co.uk> writes:

> On Wed, 15 Jul 2015 12:42:00 +0200, hw wrote:
>
>> > Of course, there may be a very good reason why this flag is masked on
>> > no-multilib profiles, in which case you will not only see why but get
>> > to keep the resulting shrapnel.  
>> 
>> I would say it is wrong to disable this flag in the first place, 
>> especially so without giving any warning, like when emerging xen-tools, 
>> that that flag is required for hardware virtualisation.  The same goes 
>> for other flags which are alike as well.
>
> If it doesn't work without multilib, it doesn't work without multilib -
> adding a USE flag will not change that.

I think it doesn't need anything multilib other than for compiling some
parts.  Why can't the missing stuff, like stub-32.h, be pulled in as a
dependency, perhaps with a warning that some package(s) need to be
installed which you otherwise would only have with a multilib
installation?

In any case, I think equery should show masked use flags.  If it did, we
could see right away that we need to do something to enable something.
Wouldn't that be preferable over being left in the dark?

> But masking the USE flag will at least stop you trying to use it
> without multilib.

Not really:  You can change the profile to one that doesn't contain
'no-multilib' and enable the flag.

When you do that, what speaks against installing the missing packages
which you do not have in a no-multilib installation and need for xen
with hvm?  (After all, having to switch to a multilib installation just
for xen is a rather harsh requirement ...)

> ebuilds are only shell scripts, not magic spells.

Knowledge is volatile and fluid; software is power.


Anyway, I made a note about it on the wiki a couple days ago so ppl can
know that they need multilib.  How about a feature request/bug report on
equery so that it will show the masked use flags?  An 'equery uses ...'
is supposed to show the flags, yet it doesn't show them all.


-- 
Again we must be afraid of speaking of daemons for fear that daemons
might swallow us.  Finally, this fear has become reasonable.

Reply via email to