Manta Interactive Ray Tracer Development Mailing List

Text archives Help


[Manta] Re: Re: Re: Re: Re: r2349 - in trunk: . CMake Engine Image Model Model/Readers UserInterface


Chronological Thread 
  • From: "James Bigler" < >
  • To:
  • Subject: [Manta] Re: Re: Re: Re: Re: r2349 - in trunk: . CMake Engine Image Model Model/Readers UserInterface
  • Date: Thu, 20 Nov 2008 14:37:23 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=Csfv+gKPmXM7NrslFyA7zLP1ZVEqm046uLj5gzKfbBkUwNbcm7FZW6gAsbdgVNitSW uDjNGtjbzu3eVfhOrmhB0bg4S0Bwi3EWFgqMxrSw6TWJeUV0p7AEYHSXXa/JGZseY9ou qdHrgh5MAjeooLHiDTO5FIEaPrEyEZpFYkykE=
Get CMake to handle what exactly?

I still think linking everything against Teem is only shadowing the
issue and not really solving it.

Which object has unresovled Teem symbols?  Is there some header file
that is pulling in symbols everywhere?

Can this problem be reproduced on any SCI machine?

James

On Thu, Nov 20, 2008 at 2:25 PM, J. Davison de St. Germain
< >
 wrote:
> Shared.
>
> updraft2:1010:updraft.arches/sys/pkg/scirun_tp> ls -l teem-install/lib/
> total 7984
> drwxr-xr-x  2 u0080076 uintah    4096 Nov 18 17:57 Teem-1.10.0
> -rw-r--r--  1 u0080076 uintah 4812692 Nov 18 17:23 libteem.a
> lrwxrwxrwx  1 u0080076 uintah      12 Nov 18 17:24 libteem.so ->
> libteem.so.1
> lrwxrwxrwx  1 u0080076 uintah      17 Nov 18 17:24 libteem.so.1 ->
> libteem.so.1.10.0
> -rwxr-xr-x  1 u0080076 uintah 3340962 Nov 18 17:25 libteem.so.1.10.0
>
> Note, though, you are absolutely right about problems with the static
> version... it will break.  Someone needs to update CMake to handle that
> correctly too.
>
>  - Dav
>
> James Bigler wrote:
>>
>> Was Teem compiled static or shared?  A static version of Teem would
>> explain the problems you were having.
>>
>> James
>>
>> On Thu, Nov 20, 2008 at 1:40 PM, J. Davison de St. Germain
>> < >
>>  wrote:
>>
>>>
>>> I'm surprised I had to do that too... I would have thought it would 'just
>>> work' too... but it didn't.  I'm not certain.  All I know is that when I
>>> typed "make", I got link errors due to missing Teem symbols.  And when I
>>> added the CMake stuff to link in Teem, they went away, and the compilation
>>> finished and manta ran.
>>>
>>>  - Dav
>>>
>>> James Bigler wrote:
>>>
>>>>
>>>> I'm surprised that you needed to do this.  Transitive linking should
>>>> happen for you.  Are you certain that you were linking against teem at 
>>>> the
>>>> right stage?
>>>>
>>>> Also, you could make this better backwards compatible by doing something
>>>> like this:
>>>>
>>>> if(NOT DEFINED Teem_BLAH_BLAH AND DEFINED TEEM_BLAH_BLAH)
>>>>  set(Teem_BLAH_BLAH TEEM_BLAH_BLAH)
>>>> endif()
>>>>
>>>> That way if you have an older version of teem you will still get the
>>>> right variable name.
>>>>
>>>> James
>>>>
>>>
>>> --
>>> -----------------------------------------------------------------------
>>> - J. Davison de St. Germain         
>>> 
>>>     (801) 581-4078 -
>>> - Chief Software Engineer           http://www.cs.utah.edu/~dav       -
>>> - SCI Institute, SE C-SAFE          University of Utah                -
>>> -----------------------------------------------------------------------
>>>
>>>
>
> --
> -----------------------------------------------------------------------
> - J. Davison de St. Germain         
> 
>     (801) 581-4078 -
> - Chief Software Engineer           http://www.cs.utah.edu/~dav       -
> - SCI Institute, SE C-SAFE          University of Utah                -
> -----------------------------------------------------------------------
>
>

Archive powered by MHonArc 2.6.16.

Top of page