[8u] RFR: 8247469: getSystemCpuLoad() returns -1 on linux when some offline cpus are present and cpusets.effective_cpus is not available

[Severin Gehwolf]

On Tue, 2021-07-13 at 17:41 +0100, Andrew Dinn wrote:
> On 13/07/2021 17:28, Severin Gehwolf wrote:
> > This is a prerequisite backport for JDK-8265836, which I'd like to
> > bring to OpenJDK 8u as the bug is present there too[2]. Note that the
> > OperatingSystemMXBean has been made container aware with JDK-8226575[1]
> > in 8u272 and better.
> > 
> > The OpenJDK 11u patch doesn't apply cleanly due to restructuring of the
> > code in later JDKs. The gist of the patch is still the same. One
> > addition of note is the mapfile-vers change for the new native method.
> > 
> > Bug: https://bugs.openjdk.java.net/browse/JDK-8247469
> > webrev: https://cr.openjdk.java.net/~sgehwolf/webrevs/JDK-8247469/jdk8/02/webrev/
> > 
> > Testing: Builds on Linux x86_64, Solaris Sparc, AIX.
> >           Manually via the reproducer for JDK-8265836. Tier 1 on Linux x86_64
> > 
> > Thoughts?
> The backport looks good to me and it certainly looks necessary as a 
> prerequisite for backport for JDK-8265836. So count this as a review.

Thanks for the review, Andrew!

Cheers,
Severin