I guess speeding up your code by a factor of 10+ and no longer breaking the memory limit just doesn't appeal to some people.
Don't get me wrong, I've applied the patch, even pushed it to the repo. Better code is always welcome.
But your stress test is on range calculation, mine is on the requests/second. Generating HTML is what takes long and eats memory, not generating numerical posts range. And by a huge factor.
On a DoS test, the code in the master branch is slowed down to an average of 1.51 req/s while the code in the optimized filter branch manage a whopping 1.7 req/s. It's good, not negligible. But it's not enough, there's much more optimization to be done.
If you'd like to test it yourself, the only thing missing thing to boot SchemeBBS was data directories and index files for a dummy board. I've added a board named "foo" in the repo and a shell script to create as many boards as you want. SchemeBBS works standalone and out of the box (well, provided you've patched the files mit-scheme-9.2/src/lib/runtime/http-syntax.scm and mit-scheme-9.2/src/runtime/httpio.scm which use outdated RFCs)