RFR: 8276662: Scalability bottleneck in SymbolTable::lookup_common()

[Derek White]

On Mon, 15 Nov 2021 22:31:13 GMT, Derek White <drwhite at openjdk.org> wrote:

> Symbol table lookup had an optimization added when the symbol table was split into a shared table (for CDS?) and a local table. The optimization tries to track which table successfully found a symbol, so it can try that table first the next time.
> 
> Symbol table lookup is used in many JVM operations, including classloading, serialization, and reflection.
> 
> At startup time, more symbols will be from the shared table, but over time lookup can will be from a mix of local and shared symbols (eg user classes still have java.lang.String fields or subclass from java.lang.Object), resulting in multiple threads fighting over the value of this global variable.
> 
> With enough threads and cores, this can result in "true sharing" cache line contention.
> 
> This fix solves the scalability issue by checking the shared table first "early on", and when enough local symbols have been added, then check the local table first.  
> 
> Other options would also solve the the scaling problem, but may change the behavior that we're trying to optimize, or add more overhead or complexity than warranted, such as:
> - Statically preferring the shared or local table
> - Using a thread-local variable to track which table to search first
> - Using a NUMA-aware set of N variables distributed over M threads.

Yes, the thread-local should fix the scalability issue (I only tested on a small machine). And logically this performance hint makes more sense per-thread, not globally. 

On the other hand  HotSpot doesn't through around C++ thread locals everywhere. "_thread" is pervasive enough to make it worthwhile. If there are concerns about using too many thread locals I wasn't sure if it be worth burning one for this purpose.

If we want to use THREAD_LOCAL, we may need this:

#ifdef USE_LIBRARY_BASED_TLS_ONLY
static volatile bool _lookup_shared_first = false;
#else
static THREAD_LOCAL bool _lookup_shared_first = false;
#endif

(Not that I know what USE_LIBRARY_BASED_TLS_ONLY is really for :-)

-------------

PR: https://git.openjdk.java.net/jdk/pull/6400