Implement new faster version of Jar virtual file system
It's only enabled by default in FIR and might be turned on with a CLI flag The main idea is that default FarFS re-read ZIP file list each time when class file is requested that is quite slow. We read it once and them reading bytes from the known offset. Also, unlike the default version we don't perform attributes check on each access On the one hand, it works faster on the other it might not notice that one of the JAR has been changed during compilation process But looks like it's not supposed to be a frequent even during compilation of a single module
This commit is contained in:
committed by
TeamCityServer
parent
559e7d223a
commit
8f06e59d3b
+1
@@ -135,6 +135,7 @@ where advanced options include:
|
||||
Suppress the "cannot access built-in declaration" error (useful with -no-stdlib)
|
||||
-Xtype-enhancement-improvements-strict-mode
|
||||
Enable strict mode for some improvements in the type enhancement for loaded Java types based on nullability annotations,including freshly supported reading of the type use annotations from class files. See KT-45671 for more details
|
||||
-Xuse-fast-jar-file-system Use fast implementation on Jar FS. This may speed up compilation time, but currently it's an experimental mode
|
||||
-Xuse-ir Use the IR backend. This option has no effect unless the language version less than 1.5 is used
|
||||
-Xuse-javac Use javac for Java source and class files analysis
|
||||
-Xuse-old-backend Use the old JVM backend
|
||||
|
||||
Reference in New Issue
Block a user