> I don't know if this qualifies as a bugs.
>
> Trying to make a binary smaller, I discovered that a few strings
> appear *many* times in it.
> Telling sh to `strings hhttpd | sort | uniq -c | sort -rn | head -n 30'
> gives me the top 30 included below. This is something like 200K out of
> the 1367996 bytes in hhttpd, which is a 15% aprox. hhttpd was built with
> ghc-3.01 with lots of optimization on (that is, -O -O2-for-C -fvia-C)
> That's too many `Urk's, isn't it?
Thank you for not only making a good point, but backing it with
numbers!
Here is another powerful motivation for doing simple common subexpression
analysis. (The duplications arise from inlinings, mostly of prelude
functions.)
Simon
