Boost logo

Boost-Build :

From: Ralf W. Grosse-Kunstleve (RWGK_at_[hidden])
Date: 2002-06-10 11:52:35


> If people try it, I would appreciate hearing how it works (or how
> it fails).

Platform:
IRIX64 pompano 6.5 04101930 IP27
MIPSpro Compilers: Version 7.3.1.3m

Using your file as posted, with two additional lines:

feature sgi-abi : n32 n64 ;
feature sgi-isa : mips3 mips4 ;

I am trying to compile the Boost.Python library:

cd boost/libs/python/test
bjam -sBOOST_ROOT=$HOME/boost -sPYTHON_ROOT=/yusr/pom1/rwgrosse/tools/irix -sTOOLS=mipspro -sPYTHON_VERSION=2.2 -sBUILD=debug -sALL_LOCATE_TARGET=/var/tmp/rwgk/bjam -d+2

With your "-INLINE:never" any attempt to compile results in:

Signal: Segmentation fault in Inliner phase.

-INLINE:none fixes this problem.

Then I get heaps of messages of this kind:

cc-3201 CC: REMARK File = /usr/include/CC/stl_ctype.h, Line = 105
The parameter "dfault" was never referenced.

virtual char do_narrow(char __c, char dfault) const { return __c; }

Or this:

cc-1506 CC: REMARK File = /u1/rwgrosse/boost_mpl_ublas/boost/python/converter/builtin_converters.hpp, Line = 92
There is an implicit conversion from "unsigned int" to "long"; rounding, sign
extension, or loss of accuracy may result.

Strangely, with my toolset I do not get any of these although the only
warning that is turned off is 1311.

Other than that the compilations seem to be OK, but the enormous number
of warning messages make it very difficult to spot real problems.

What platform are you using for testing?

Ralf

P.S.: Suggestions:
sgi-abi -> irix-abi (there is non-irix sgi hardware)
sgi-isa -> irix-isa
The command-line switch is -64, but your feature is names n64. This
is a bit confusing.

 


Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk