On Sat, 26 Jul 2003 15:54:11 +0200 CEST "Axel Dörfler" <axeld@pinc- software.de> wrote: > "Ingo Weinhold" <bonefish@xxxxxxxxxxxxxxx> wrote: > > 1) Use a cleverer strategy to place the stacks in the virtual > > address > > space, e.g. have heap and stack growing towards each other. > > I am afraid that's not that easy to do. The heap/stack growing > towards > each other doesn't work well in a multi-threaded environment - there, > you'll need a reserved address space for each thread's stack. And > that > pretty much cuts off the growing possibility of a stack. Sorry, I wasn't clear. I actually meant, that the stacks for new threads are allocated at descending addresses (e.g. starting at 0xf0000000, then 0xeffc0000, 0xeff80000, 0xeff40000,...) while the heap grows towards these addresses. So you have for heap and stacks the full range from 0x80000000 to 0xf0000000. > Even if I also think that one thread per application might be too > little, I also don't seen the need to have one thread per window - > with > a single CPU machine, there is always only one thread to run anyway. > But I also think that we shouldn't care about that right now - that > looks like something we should play with at a later point, when > everything is working well. Agreed. > Then, I am in favour of Ingo's solution to have a maximum thread > limit > per application at which point one thread can take over the duties of > more than one window (even if he has proposed that for bitmaps). *sigh* ;-) CU, Ingo