asmtools bug: ldc constant dynamic that returns a long or a double should use ldc2_w instead of ldc

[Lois Foltan]

On 9/2/2018 10:40 AM, Remi Forax wrote:

> Hi all, hi Leonid,
> while fixing a bug in ASM [1] when a ldc to a condy that returns a long uses the wrong bytecode,
> i've discovered that autotools has the same issue.
>
> The following code should generate a ldc2_w and not a ldc.
>
> public static Method primitiveExample:"()J"
>    stack 2 locals 0
> {
>      ldc  Dynamic REF_invokeStatic
>          :jdk11/AllInstructions.bsm
>          :"(Ljava/lang/invoke/MethodHandles$Lookup;Ljava/lang/String;Ljava/lang/Class;)J"
>          :_
>          :"J" {};
>      lreturn;
> }
>
> in case you wonder why we use asmtools, for testing ASM we try to not use bytecodes generated by ASM as source :)
>
> regards,
> Rémi
>
> [1] https://gitlab.ow2.org/asm/asm/issues/317850
Hi Remi,

Since your code looks very much like jasm, for my clarification, by 
autotools do you mean the asmtools within the code-tools repo?  The test 
test/hotspot/jtreg/runtime/CondyLDCTest.java tries to execute 
CondyBadLDC.jasm that has the similar instruction sequence as yours 
above, an ldc instruction of a condy returning a double.

public static Method D:"()D"
         stack 2 locals 0
{
                 // ldc of a double will yield a VerifyError, should be 
an ldc2_w instruction
                 ldc     Dynamic 
REF_invokeStatic:CondyBadLDC.intConversion:"(Ljava/lang/invoke/MethodHandles$Lookup;Ljava/lang/String;Ljava/lang/Class;I)Ljava/lang/Object;":D:"D" 
i
nt 2147483647;
                 dreturn;
}

which yields a VerifyError.  So my interest in clarification on this.

Thanks,
Lois