Hi Matthias Klose schrieb: > Arthur Korn writes: > > So bash doesn't release memory even if the system is under > > memory pressure, which is _BAD_. > > do you propose that applications should have a memory allocation > strategy depending on the system state?
Well in the meantime I learned quite a bit more about virtual memory :) Now I'm puzzled why linux doesn't page out those unused pages allocated by bash before killing it. > It looks like you could file this kind of bug to any program, > which doesn't free memory immediately. My reasoning was flawed, granted. But after all, since there is enough memory in the system to keep it working perfectly fine, the system shouldn't go OOM. I did another test now. While running ": $(seq 100000)", the resident memory size of the bash process grows to about 10M, then when it is done the virtual and resident memory size drop to much lower values immediately. Thus I guess this problem has been resolved by some modification to bash's malloc or such. This is ii bash 3.0-6 The GNU Bourne Again SHell whatever, 2ri
signature.asc
Description: Digital signature
