From: Vladimir Prus (ghost_at_[hidden])
Date: 2006-09-20 02:19:14
On Wednesday 20 September 2006 06:55, David Abrahams wrote:
> "John Maddock" <john_at_[hidden]> writes:
> > Just when I thought everything was sweetnes and light, up pops another
> > one:
> > Testing with Intel on Win32, the libraries are being named for example:
> > Creating library
> > ..\..\..\..\..\bin.v2\libs\config\test\link\test\intel-win-8
> > .1\release\threading-multi\link_test-81-mt-1_35.lib and object
> > ..\..\..\..\..\bi
> > n.v2\libs\config\test\link\test\intel-win-8.1\release\threading-multi\lin
> >k_test- 81-mt-1_35.exp
> > So with a compiler-version-number, but without a "iw" toolset name.
> > The autolink code expects the reverse: for example in this case it would
> > be "link_test-iw-mt-1_35.lib".
> > I believe the rationale for not versioning the intel toolset, is that
> > what matters is the msvc compiler targeted, not the intel compiler
> > version as far as binary compatibility is concerned, so basically bbv1
> > chickened out of versioning the toolset names, and the auto-link code
> > followed.
> Except that libraries may have #ifdefs that change their ABIs based on
> which compiler/version is being used. I think you need the toolset
> name in there.
We need, but let's fix "V2 is different from V1" issues first, before
fixing "V1 is not quite right" issues. Can you file an issue for posterity at
-- Vladimir Prus http://vladimir_prus.blogspot.com Boost.Build V2: http://boost.org/boost-build2
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