Re: Memory allocation problems glibc-2.2.2

[Stefan Hoffmeister]

: On 09 May 2001 14:39:01 +0200, Andreas Jaeger wrote:

>Stefan Hoffmeister <address@hidden> writes:
>
>> : On Wed, 9 May 2001 14:05:58 +0200 (MDT), Wolfram Gloger wrote:
>> 
>> >However, I believe there is an easy special-case solution in the glibc
>> >allocator at least for the code fragment above: Please try
>> >mallopt(M_MMAP_THRESHOLD, 128*128*3) at the start of your program.
>> 
>> Is there a tuning guide for the glibc malloc sub-allocator?
>
>That should be explained in the glibc manual,

Well, the parameters are documented - but there are no soft rules /
guidelines how the sub-allocator might behave with sets of tunable
parameters employed :-)

Perhaps rephrasing the question helps: Are there any data points how the
memory sub-allocator will perform (overhead/speed) in selected, exemplary
scenarios?

I hear you saying: "Find out yourself" - "It's specific to your allocation
patterns" - yes, sure. But having some (even informal) data points would
help to probe into the right direction :-)

Any pointers?