qsLI commented on issue #788: URL: https://github.com/apache/fory/issues/788#issuecomment-3451156735
> You can dump jvm c2 generated code, you will see all code are compiled and inlined. And if you dump a flame graph, you will see the difference too. Other serialization framework will iterate fields by loop and dispatch serializer by hash lookup if it's not final. Then invoke virtual method of the serializer. All those operation are avoided in fury. The actual optimization are much more complicated than this. We will write a blog for it. Currently you can reference https://medium.com/@shawn.ck.yang/fury-a-blazing-fast-multi-language-serialization-framework-powered-by-jit-and-zero-copy-fdd16f3215cb For bigger bean , will fory cause cpu instruction cache miss with aggressive inline by split codec method smaller? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
