The only regression deserving to note is cl-murmurhash, and libraries using it (cl-bloom).
New version of cl-murmurhash depends on fast-io, which depends on static-vectors,
which is not portable. Therefore new version of cl-murmurhash doesn't build on some lisps.
Authors are informed.
Several other false failures were caused by QL-DIST:BADLY-SIZED-LOCAL-ARCHIVE conditions.
I think it was caused by temporary Internet outage during testing on my Windows machine.
I have added a TODO item for cl-test-grid to handle this condition by
invoking QL-DIST:DELETE-AND-RETRY restart.
The quicklisp 2013-04-20 had some regressions comparing to the previous quicklisp
(and also improvements). Unfortunately I didn't have time to investigate all the regressions.
ECL seems to be the lisp impl. most affected by the regressions. I still didn't have time
to investigate them.
BTW, strange that this month I do not receive test results from any contributors,
so quicklisp 2013-06-15 is represented only by results from my machines.
If you have cl-test-grid, please run your agent.