[MLton-devel] pinned large arrays

Andrew Wright akwright@acm.org
Wed, 22 May 2002 19:09:12 -0400


Henry Cejtin writes:
 > I don't deny that manually listing the tests out of loops is inelegant.
 > You are thinking of the file itself as consisting of data structures, which I
 > tend not to since you lose portability (byte order, word size, etc.).  In my

In this case I think speed is going to rule the day.  I want to
*completely* avoid any marshalling/unmarshalling and even copying
between user and kernel space at all.

 > The problems of pinning objects with fragmenting the virtual address space
 > are really quite serious.  A friend of mine is pimped by this (in C) all the
 > time.  Once 64-bit machines are here (perhaps in under a year of AMD does ok
 > with Sledge-hammer) then that problem will go away, but for now it is going
 > to make any solution rather tenuous.

It is quite possible that serious uses of the applications I have in
mind would only be run on seriously big 64 bit machines.

Andrew


_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm

_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel