This is the mail archive of the java@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: compacting _Jv_Utf8Const


Tom Tromey wrote:

"Per" == Per Bothner <per@bothner.com> writes:


Per> Any suggestions for easily counting the number of _Jv_Utf8Const
Per> objects in libgcj?

I think you can run nm on libgcj.so and look at all the _Utf* symbols.
I got:

fleche. nm .libs/libgcj.so|grep _Utf|wc -l
  66488

I assume this is without duplicate removal. The _UtfNN symbols are local symbols; I'm not sure how to count unique symbols.

If we take this number as accurate, we're taking about a saving
of around 227kB.  That's substantial.

If you're considering touching every place that uses a Utf8Const, how
about going one step further and write some C++ classes to abstract
out the representation from the uses?

Yes, that makes sense. -- --Per Bothner per@bothner.com http://per.bothner.com/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]