Re: [Boost-build] gcc.jam weirdness (was: Boost-1.36.0 FreeBSD patches for review)

Subject: Re: [Boost-build] gcc.jam weirdness (was: Boost-1.36.0 FreeBSD patches for review)
From: Alexander Sack (pisymbol_at_[hidden])
Date: 2008-10-27 12:59:13

On Sun, Oct 26, 2008 at 11:31 AM, Steve M. Robbins <steve_at_[hidden]> wrote:
> On Sun, Oct 26, 2008 at 10:38:14AM +0300, Vladimir Prus wrote:
>> On Sunday 26 October 2008 08:32:03 Steve M. Robbins wrote:
>
>> > My take is that the cure is worse than the disease, and should
>> > be removed.
>> >
>> > Vladimir: is this solving a real problem, or a theoretical one?
>>
>> I believe the issue was that Martin was running tests with zillion
>> versions of gcc at the same time. Then, when running any given test
>> binary, you better point it at the lib dirs of the right compiler.
>> Global LD_LIBRARY_PATH is not going to work -- as you have 5 different
>> compilers, some of them using same sonames for libraries.
>
> OK.  So Martin can run his tests as follows:
>
>  LD_LIBRARY_PATH=/opt/gcc1/lib do_test ... configure with gcc1
>  LD_LIBRARY_PATH=/opt/gcc2/lib do_test ... configure with gcc2
>  LD_LIBRARY_PATH=/opt/gcc3/lib do_test ... configure with gcc3
>
> etc.  In other words: the test script can run with LD_LIBRARY_PATH set
> appropriately.  This solution keeps the complexity where it belongs:
> with the complex system.
>
> The current situation makes the Boost.Build maintainers second-guess
> the 99% of simple and correctly-set-up systems, which I submit is an
> impossible task.  Well, perhaps not impossible, but the costs outweigh
> the benefits from where I sit.

Guys in the interest of tracking this issue I filed:

https://svn.boost.org/trac/boost/ticket/2438

Again, I 100% agree with Steve that LD_LIBRARY_PATH should not be set
unless the user requests it.  99% of the builds done of boost are with
default paths and rtld should do the *right* thing by itself!

Let me know how I can help (this is a show stopper for any FreeBSD
64-bit build),

-aps