Hi, for some time on develop, top-level Jamroot used to deduce address-model and architecture from compiler. The only issue was that both properties would be added to targets paths when not necessary. Fix for that has been just committed, for develop: https://github.com/boostorg/boost/commit/945e3c0bbdd31ce8c297e0aff5340f0adb1... and I performed these operations on master: - Merged all of Boost.Build changes: https://github.com/boostorg/build/commit/2e764707e03d25b6928501c25a717e195c0... - Cherry-picked one Boost.Context change: https://github.com/boostorg/context/commit/9b392c68b59021aa4801dcb245bc64ad5... - Cherry-picked one Boost.Log change: https://github.com/boostorg/log/commit/27a822a6b82f8bf34cbc681ac6583412bdf27... - Cherry-picked two root project changes and updated pointers to Context and Log: https://github.com/boostorg/boost/commit/e3bc35f7891e72ad74e57dc216db899d162... https://github.com/boostorg/boost/commit/beb53b6b9574f95309384bfc5afffb40d46... https://github.com/boostorg/boost/commit/b69a072401695c9978085158af4c1ea84ce... With that, "./b2 -n" in up-to-date checkout of superproject works fine on Linux. Also, "./b2 -n --with-context --with-system" works fine, no address-model elements in path. Putting explicit address-model=64 does not change path. Putting explicit address-model=32 does add this element to path, as expected. Note that: - I did not merge any other Boost.Context or Boost.Log changes. - Nothing was required for Boost.Config, since it was merged earlier. I'd appreciate if people test current state of master branch, and report any concerns. Thanks, -- Vladimir Prus CodeSourcery / Mentor Embedded http://vladimirprus.com