[9] RFR (S): 8177963: Parallel GC fails fast when per-thread task log overflows

Kim Barrett kim.barrett at oracle.com
Wed Apr 5 15:59:28 UTC 2017


> On Apr 5, 2017, at 7:26 AM, Thomas Schatzl <thomas.schatzl at oracle.com> wrote:
> Continuing to count gives an indication on how large the user needs to
> size this buffer without too many retries.

A different approach to reporting needed size when overflow occurs
would be to limit _time_stamp_index by GCTaskTimeStampEntries
and add an overflow counter.  But what you have is okay with me too.

> New webrevs:
> http://cr.openjdk.java.net/~tschatzl/8177963/webrev.0_to_1 (diff)
> http://cr.openjdk.java.net/~tschatzl/8177963/webrev.1 (full)

Looks good.




More information about the hotspot-gc-dev mailing list