Bug#675577: Bug#718094: mpfr4: FTBFS: configure: error: gmp.h can't be found, or is unusable.

[Steve M. Robbins]

On July 30, 2013 12:42:09 PM Vincent Lefevre wrote:
> On 2013-07-29 23:46:38 -0500, Steve M. Robbins wrote:
> > OK, how about "with standard abi, the compiler will find it";i.e.
> > without -m32. The lib32gmp package had to place its header in an
> > unusual location which means, yes, the compiler won't find it
> > without help.
> 
> But in the past, the header was placed in the usual location.

Yes.  Things were changed to accomodate multi-arch.


> > Note, however, that the lib32gmp packages are going to be removed in
> > the near future.  I had sent a note to this effect to
> > mpfr4 at packages.debian.org and received a positive response from
> > Laurent Fousse.  My suggestion is to simply remove the lib32mpfr
> > packages now.
> 
> That's bad because in the past, would could use the -m32 option
> for programs using MPFR, and that's no longer possible

Yes.  The general idea is to take advantage of multi-arch to use the variant of your architecture.


> (note that libmpfr-dev:i386 is not installable).

I'm surprised you say that as it appears to have worked here.  To my surprise, it appears even to be co-installable with the amd64 version:

steve at riemann{2013}dpkg --list libmpfr-dev
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name                                              Version                       Architecture                  Description
+++-=================================================-=============================-=============================-=======================================================================================================
ii  libmpfr-dev:amd64                                 3.1.1-1                       amd64                         multiple precision floating-point computation developers tools
ii  libmpfr-dev:i386                                  3.1.1-1                       i386                          multiple precision floating-point computation developers tools
steve at riemann{2013}ls -l /usr/lib/*/*mpfr*
-rw-r--r-- 1 root root 692110 Feb 20 01:17 /usr/lib/i386-linux-gnu/libmpfr.a
lrwxrwxrwx 1 root root     16 Feb 20 01:17 /usr/lib/i386-linux-gnu/libmpfr.so -> libmpfr.so.4.1.1
lrwxrwxrwx 1 root root     16 Feb 20 01:17 /usr/lib/i386-linux-gnu/libmpfr.so.4 -> libmpfr.so.4.1.1
-rw-r--r-- 1 root root 386044 Feb 20 01:17 /usr/lib/i386-linux-gnu/libmpfr.so.4.1.1
-rw-r--r-- 1 root root 917162 Feb 19 22:13 /usr/lib/x86_64-linux-gnu/libmpfr.a
lrwxrwxrwx 1 root root     16 Feb 19 22:13 /usr/lib/x86_64-linux-gnu/libmpfr.so -> libmpfr.so.4.1.1
lrwxrwxrwx 1 root root     16 Feb 19 22:13 /usr/lib/x86_64-linux-gnu/libmpfr.so.4 -> libmpfr.so.4.1.1
-rw-r--r-- 1 root root 382208 Feb 19 22:13 /usr/lib/x86_64-linux-gnu/libmpfr.so.4.1.1

Regards,
-Steve
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 190 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.alioth.debian.org/pipermail/debian-science-maintainers/attachments/20130730/5b42988d/attachment-0001.sig>