Re: s6-ftrig-wait is not relocatable

From: Buck Evan <buck_at_yelp.com>
Date: Wed, 14 Oct 2015 17:02:12 -0700

On Wed, Oct 14, 2015 at 4:16 PM, Laurent Bercot <ska-skaware_at_skarnet.org>
wrote:

>
> Try the latest skalibs git. It doesn't install leapsecs.dat, the
> leap second table is directly in the code, as you suggested.
> That's a pretty massive simplification - at worst, something good
> will have come of this. :)
>


Very cool. Thanks. I'm trying it now.
I see from your documentation changes this removes whole modes of error.


Oh, definitely. But it's *so close* to fitting!
>>
>
> It will never fit entirely as is - you will have to patch stuff.
> It's doable, but that's a lot of effort for a benefit I still don't
> understand.


Are you idealogically opposed to the relpath strategy?
Under this scheme, no absolute paths would be encoded in any binary, only
relative paths.
To find files, binaries would use the known relative path and argv[0] (and
getcwd(3) if argv[0] is relative).

Obviously that's a lot of work that you may be unwilling to undertake given
questionable benefit.
I just want to know if you hate the idea.
Received on Thu Oct 15 2015 - 00:02:12 UTC