Stephen Hemminger <step...@networkplumber.org> 于2023年2月16日周四 01:16写道： > > On Wed, 15 Feb 2023 12:10:23 +0100 > Morten Brørup <m...@smartsharesystems.com> wrote: > > > Looking isolated at the patch itself... > > > > I agree with the way the patch modifies the ranges of the free list, and > > the consequential removal of the "- 1" from the calculation of log2. > > > > Intuitively, the lists should cover ranges such as [0x100..0x3FF], which > > this patch does, not [0x101..0x400], how it was previously... The ranges > > with this patch make much more sense. > > > > So if the existing code is otherwise correct, i.e. handles the size > > with/without MALLOC_ELEM_HEADER_LEN correctly, my gut feeling says this > > patch is an improvement. > > > > Acked-by: Morten Brørup <m...@smartsharesystems.com> > > It would be good to have a malloc performance test. > Possibly something reused from some other project.
I have done some performance tests in SPDK before，maybe available for your reference: https://bytedance.feishu.cn/wiki/wikcnKWXQmN4qhyxLcWJqP4vrOs In one word, the performance of 4k malloc has improved, and other cases almost the same.
