Re: [Jython-dev] A few micro-benchmarks for Jython 2 and two other ideas

[Andrew H. <ap...@re...>]

On 1/6/21 4:28 PM, Jeff Allen wrote:
>> And to see if intrinsic methods are being utilized or not and where in 
>> compiled code, you can add:
>>     -XX:+PrintCompilation -XX:+UnlockDiagnosticVMOptions 
>> -XX:+PrintInlining
>>
> Unlock has to come first, it seems. I've experimented with those options 
> and found what they produced was pretty incomprehensible. I never made 
> the disassembly option work.

If you're interested in persisting a bit more, I can help with this.

git clone https://github.com/openjdk/jdk

cd jdk/src/utils/hsdis/
mkdir build
git clone git://sourceware.org/git/binutils-gdb.git build/binutils
make ARCH=amd64
ls build/linux-amd64/

bfd/  config.log  config.status*  hsdis-amd64.so*  intl/  libiberty/  Makefile	opcodes/  serdep.tmp  zlib/

That hsdis-amd64.so should be installed in your library path somewhere:
on every machine I have access to it's in /usr/lib64. Then all of the
PrintAssembly stuff should work.

> Going back and trying a little harder, thanks to your suggestion, I got 
> further this morning. The output remains too complex for me to follow 
> (so many jumps!), but a superficial inspection supports the conjectures 
> I made based only on timing.

WRT intrinsics: if there's something important, we can do it.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671