[aarch64-port-dev ] RFR (XS) 8228746: Revert incorrect StubRoutines::contains change

[Andrew Dinn]

On 29/07/2019 20:22, Aleksey Shipilev wrote:
> Bug:
>   https://bugs.openjdk.java.net/browse/JDK-8228746
> 
> Bad merge, very old, such shame. All code paths I found affect logging (e.g. demangling in hs_errs)
> only, which must be why it is not catastrophic.
> 
> Fix:
> 
> diff -r f82f2008d4cf src/share/vm/runtime/stubRoutines.hpp
> --- a/src/share/vm/runtime/stubRoutines.hpp     Mon Jul 29 13:10:07 2019 +0200
> +++ b/src/share/vm/runtime/stubRoutines.hpp     Mon Jul 29 21:15:35 2019 +0200
> @@ -253,11 +253,11 @@
>    static bool is_stub_code(address addr)                   { return contains(addr); }
> 
>    static bool contains(address addr) {
>      return
>        (_code1 != NULL && _code1->blob_contains(addr)) ||
> -      (_code2 != NULL && _code1->blob_contains(addr)) ;
> +      (_code2 != NULL && _code2->blob_contains(addr)) ;
>    }
> 
>    static CodeBlob* code1() { return _code1; }
>    static CodeBlob* code2() { return _code2; }
> 
> Testing: x86_64 build, hotspot_gc_shenandoah
Hmm, looking at the merge change set you mention in the JIRA it was
already broked before it was borked!

       (_code1 != NULL && _code1->blob_contains(addr)) ||
-      (_code2 != NULL && _code1->blob_contains(addr)) ||
-      (_code3 != NULL && _code2->blob_contains(addr)) ;
+      (_code2 != NULL && _code1->blob_contains(addr)) ;

Anyway, this fix looks good.

regards,


Andrew Dinn
-----------
Senior Principal Software Engineer
Red Hat UK Ltd
Registered in England and Wales under Company Registration No. 03798903
Directors: Michael Cunningham, Michael ("Mike") O'Neill, Eric Shander