On 28 May 2013 12:03, Matthew Gretton-Dann <matthew.gretton-d...@linaro.org> wrote: > So before we go any further I would like to see what the view of LEG is > about a better malloc. My questions boil down to: > > * Is malloc important - or do server applications just implement their own?
The Apache web server manages its heap allocations with differently scoped pools, i.e., per-request, per-session, per-thread. This allows it for instance to free() a whole request pool at once when it has finished serving a request instead of going through each of the allocations and freeing them one by one. PHP's Zend core has an elaborate memory management layer which does some similar tricks. That means I don't expect malloc() to be as dominant in a real-world *web* server application. > * Do you have any benchmarks that stress malloc and would provide us with > some more data points? > Not right now, but if we notice it anywhere near the top in the perf trace, we will let you know about it. -- Ard. > But any and all comments on the subject are welcome. > > Thanks, > > Matt > > -- > Matthew Gretton-Dann > Toolchain Working Group, Linaro > > _______________________________________________ > linaro-enterprise mailing list > linaro-enterpr...@lists.linaro.org > http://lists.linaro.org/mailman/listinfo/linaro-enterprise _______________________________________________ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-toolchain
