[8u-backport] RFR (S/M): 8180048: Interned string and symbol table leak memory during parallel unlinking

Thomas Schatzl thomas.schatzl at oracle.com
Mon May 15 17:39:13 UTC 2017


Hi all,

  can I have reviews for this backport of 8180048 from jdk9 to 8u?

Unfortunately this change does not apply cleanly:
- in jdk8 both the string and symbol table are located in the same file
- symbol table uses an extra "memory_total" size that needs to be
passed around (used for some log message removed in 9).

Nevertheless I kept the backport as similar to the original as
possible, i.e. I kept memory_total extra, and did not fix indentation
in vmstructs.

So I need reviewers before asking to be allowed to push it to jdk8u-
dev.

CR:
https://bugs.openjdk.java.net/browse/JDK-8180048
Webrev (9):
http://cr.openjdk.java.net/~tschatzl/8180048/webrev.2/
Webrev:
http://cr.openjdk.java.net/~tschatzl/8180048-8u/webrev/
Testing:
jprt, attached test case

Thanks,
  Thomas




More information about the hotspot-gc-dev mailing list