On Wed, 1 Nov 2023 10:13:03 GMT, Thomas Stuefe <stu...@openjdk.org> wrote:
>> src/hotspot/share/nmt/memMapPrinter.cpp line 105: >> >>> 103: _ranges[_count - 1].to = to; >>> 104: return true; >>> 105: } >> >> I'm pretty sure that the virtual memory tracker already gives you the >> minimal set of regions, have you observed this branch being taken? > > Yes, it resulted in a significant performance gain for a test run where I > interleaved non-committed and committed memory of the same tag. I may have > been mistaken, of course. It being committed or reserved shouldn't matter (I assume non-committed = reserved), as committed just means that there's a pointer to the committed memory region in the reserved memory region. Well, all I can say is that I'm surprised. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/16301#discussion_r1378643276
