On Tue, 25 Jan 2022 09:06:49 GMT, Thomas Stuefe <stu...@openjdk.org> wrote:
>> Tyler Steele has updated the pull request with a new target base due to a
>> merge or a rebase. The pull request now contains two commits:
>>
>> - Merge branch 'master' into JDK-8203290
>> - Implements JFR on AIX
>>
>> - Implements interfaces from os_perf.hpp in os_perf_aix.cpp
>> - Updates libperfstat_aix to contain functionality needed by os_perf_aix
>> - Implements missing functionality in loadlib_aix (Fixes failure noted
>> by TestNativeLibraies.java)
>> - Removes platform checks for --enable-feature-jfr (Now enable-able on
>> all platforms)
>> - Enables TestNetworkUtilizationEvent.java which now passes on AIX
>> - Updates AIX JavaThread::pd_get_top_frame_for_profiling with changes
>> from Linux
>
> src/hotspot/os/aix/os_perf_aix.cpp line 214:
>
>> 212: // populate cpu_stats && check that the expected number of records
>> have been populated
>> 213: if (ncpus != libperfstat::perfstat_cpu(&name_holder, all_lcpu_stats,
>> sizeof(perfstat_cpu_t), ncpus)) {
>> 214: FREE_RESOURCE_ARRAY(perfstat_cpu_t, all_lcpu_stats, ncpus);
>
> ResourceArray are stack-based arenas. No need to free them manually (freeing
> is often a noop anyway unless the allocation is a the top of the arena). Very
> similar to those AutoReleasePools in Objective-C, if you know that.
>
> The way to use them is don't release them. If those allocations don't escape
> the function (so, you don't return RA memory from the function), you can set
> a ResourceMark at the start of the function. That is an RAII object which
> will clear all RA objects allocated in and under this function upon return.
>
> Arguably, you also could just use malloc and free here (NEW_C_HEAP_ARRAY and
> FREE_C_HEAP_ARRAY)
This is interesting, and definitely good to know. I have removed all
FREE_RESOURCE_ARRAY calls and added a ResourceMark to the top of all procedures
which use RESOURCE_ARRAYs.
-------------
PR: https://git.openjdk.java.net/jdk/pull/6885
