lua-users home
lua-l archive

[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index] [Thread Index]


On May 31, 2003 at 04:39 +0200, David Olofson wrote:
> 
> > There will be short pauses to traverse the stack(s),
> > to clean weak tables, etc.
> 
> How short? "Short enough" is fine by me. You need to use a CPU that's 
> fast enough for the job, no matter what.
> 
> Can the worst case delay be bounded? Yes, if the algorithm is right!

The problem with the current collector is that during a GC cycle, the
time is bounded, sure -- but it's bounded by the number of objects on
the heap.  IIRC, we had one report on the list of a group that was
using Lua, who observed an occasional half-second delay for GC (in a
robot application, of all things!).

Aha, here's the message from the archives:

http://lua-users.org/lists/lua-l/2002-08/msg00283.html

The point of the incremental work is to make it possible to convert
the occasional very long delay into a large number of relatively short
delays.

Dunno what that means in terms of hard/soft real-time.

-- 
Thatcher Ulrich
http://tulrich.com