29 Dec
2005
29 Dec
'05
10:21 a.m.
On Thu, Dec 29, 2005 at 11:22:11AM +0300, Bulat Ziganshin wrote:
stack overflows AFAIK is never occur because of laziness, but only because your recursion is not tail-optimized.
AFAIK, evaluating a thunk in GHC-compiled code does use the stack - update frames are being put on it. There is a nice description in this paper, at the bottom of page 3: http://research.microsoft.com/Users/simonpj/papers/parallel/multiproc.pdf Best regards Tomasz -- I am searching for a programmer who is good at least in some of [Haskell, ML, C++, Linux, FreeBSD, math] for work in Warsaw, Poland