Boost logo

Boost :

From: Kevin Wheatley (hxpro_at_[hidden])
Date: 2003-12-11 13:47:13


Kevin Wheatley wrote:
>
> David Abrahams wrote:
> >
> > I just noticed that our como-win32 toolset (and I don't know how many
> > other EDG-based toolsets) are set to do link-time instantiation. That
> > means we may be getting misleading results for compile and
> > compile-fail tests. What should we do about it?

In case somebody knows better than myself, I'm currently running with
the following patch which eliminates most of the IRIX regression test
failures when using the 7.41 MIPSPro compilers. I'm basically ignoring
some pf the warnings as well as using the SGI supplied headers rather
than the boost ones which were needed with 7.3x. I'm not sure how the
jam files need tweaking to only do this for versions >7.4x

Kevin

cvs diff mipspro-tools.jam
Index: mipspro-tools.jam
===================================================================
RCS file: /cvsroot/boost/boost/tools/build/v1/mipspro-tools.jam,v
retrieving revision 1.10
diff -r1.10 mipspro-tools.jam
56c56,58
< flags mipspro LINKFLAGS : -nostdlib ;

---
> flags mipspro LINKFLAGS : -nostdlib
>                           -dont_warn_unused -demangle
>                           ;
60c62,65
< # We better link the maths library unconditionally.
---
> # We better link the maths library unconditionally, but ignore any
> # warnings from doing so (84).
> # Also (47) "branch instruction(s) that might degrade performance on an
> # older version (rev. 2.2) R4000 processor."
61a67
> flags mipspro LINKFLAGS : -woff 47,84 ;
67a74
>                        -quiet_prelink
94a102
> # 1552: The variable "test" is set but never used.
96a105
> flags mipspro CFLAGS : -woff 1021,1375,1424,1234,1552 ;
119d127
<                           
"$(BOOST_ROOT)/boost/compatibility/cpp_c_headers"
-- 
| Kevin Wheatley                   | These are the opinions of |
| Senior Do-er of Technical Things | nobody and are not shared |
| Cinesite (Europe) Ltd            | by my employers           |

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk