I've attached two patches.

The first replaces the definition of string-for-each in (rnrs
base). R6RS's version of string-for-each is not the same as srfi 13's
string for each (which guile provides by default). Rather, it is more
closely analogous to the usual multi-list definition of for-each. The
R6RS specifies that all arguments must have the same length, and so I've
thrown an &assertion in this case.

The second one is a change to resolve-r6rs-interface. Previously
mark-weaver [0], changes this so that it would correctly look up
submodules under the srfi namespace, but in doing so took into account
the srfi 97[1] library name, which it should not have done. I have added a
comment to this effect in the source.

I should have noticed this at the time, but I didn't until I rebuilt and my
.guile broke :)

0. https://lists.gnu.org/archive/html/guile-devel/2012-11/msg00011.html
1. http://srfi.schemers.org/srfi-97/srfi-97.html

-- 
Ian Price -- shift-reset.com

"Programming is like pinball. The reward for doing it well is
the opportunity to do it again" - from "The Wizardy Compiled"