> Back in 2.4 time there seems to be a different kernel memory > allocator which does not use buddy system. That one probably gives > less fragmentation. I have been thinking of porting that to 2.6. > Maybe you can try. > > Or perhaps someone else already has tried and has the patch? >
I've read a paper recently that describe how to use C++ mixins to easily combine different allocation strategies for userland application mallocators. To convert that into a kernel allocator (and in plain C!) would be nice, but probably more fun that I can afford at the moment :) > In any case we also noticed that busybox does a lot of malloc's which > worsen the memory fragmentation. We have got XIP working, but that > does not help on the heavy malloc and heavy fragmentation issue much. > I managed to downsize the busybox below 256K, and with some other hacks in my shell scripts, the memory allocation failures disappeared; that revealed that I have a second issue, read/write/execute operations outside of addressable space (Exception 12, Bus error and such); now I have to find out where it comes from. However, the solution for the former OOM issue I came to is probably fragile, and I'm interested in how to have XIP working. Thanks for your answer. _______________________________________________ uClinux-dev mailing list uClinux-dev@uclinux.org http://mailman.uclinux.org/mailman/listinfo/uclinux-dev This message was resent by uclinux-dev@uclinux.org To unsubscribe see: http://mailman.uclinux.org/mailman/options/uclinux-dev
