RFR (L) JDK-8195100: Use a low latency hashtable for SymbolTable

[Kim Barrett]

> On Aug 10, 2018, at 4:58 PM, Gerard Ziemski <gerard.ziemski at oracle.com> wrote:
> 
> hi all,
> 
> After some offline discussion with Kim and Coleen, we concluded that some of the issues noted in rev5 would be better dealt with as followup RFEs, so I’m abandoning rev6. Some of those are issues shared with StringTable and need to be addressed for both the SymbolTable and StringTable together, possibly after additional discussion. Rolling all these changes into this fix seemed to be expanding the scope too much.
> 
> New webrev at http://cr.openjdk.java.net/~gziemski/8195100_rev7
> 
> rev7 vs rev5 at http://cr.openjdk.java.net/~gziemski/8195100_rev7_5
> 
> Passes Mach5 hs-tier1,2,3,4,5,6 with a few unrelated issues in tier 6 (mostly timeouts), and one unrelated issue in tier4 (timeout)

Looks good.