[MLton] Multicore CPU's and MLton

Wesley W. Terpstra wesley@terpstra.ca
Mon, 4 Jul 2005 12:11:01 +0200

On Sat, Jul 02, 2005 at 09:09:27PM -0700, Stephen Weeks wrote:
> As the C codegen is even slower than the native codegen, parallelization
> there could help even more.  But then we already have a native codegen for
> x86, so this only really helps on other platforms that are going multicore
> where we don't have a native codegen.

When I was building MLton on sparc/linux, the machine I used had _16_
processors all clocked at 400MHz. It was so slow building, that I eventually
solved the problem by writing a shell script which 'kill -9'ed MLton and
then ran a make -j16 to compile the binary.

Since we were talking about nearly a day of compiling, that was an
acceptable solution for me to gain a 16* speed-up. At the time, I thought
it would be nice to have -j16 in MLton too, but I was more focused on
getting it to work.

Wesley W. Terpstra