[aarch64-port-dev ] RFR (XS) 8228767: Revert ResourceMark additions

[Andrew Dinn]

On 30/07/2019 10:50, Aleksey Shipilev wrote:
> RFE:
>   https://bugs.openjdk.java.net/browse/JDK-8228767
> 
> In aarch64-port, there is a divergence against upstream caused by this patch:
>   https://hg.openjdk.java.net/aarch64-port/jdk8u-shenandoah/hotspot/rev/80be8f688f7c
> 
> Inspecting code paths does not yield a firm reason to have those new ResourceMarks. Removing them
> does not regress hotspot_gc_shenandoah and the custom runs with/without specific method compilations
> disabled to trigger that path. This should be reverted. If we find it was needed on some path, we
> would need to introduce that change via 8u updates process.
> 
> Fix:
> 
> diff -r 53827a662858 src/share/vm/oops/method.cpp
> --- a/src/share/vm/oops/method.cpp      Tue Jul 30 10:29:02 2019 +0200
> +++ b/src/share/vm/oops/method.cpp      Tue Jul 30 11:49:57 2019 +0200
> @@ -757,5 +757,4 @@
>  void Method::print_made_not_compilable(int comp_level, bool is_osr, bool report, const char* reason) {
>    if (PrintCompilation && report) {
> -    ResourceMark rm;
>      ttyLocker ttyl;
>      tty->print("made not %scompilable on ", is_osr ? "OSR " : "");
> @@ -779,5 +778,4 @@
>    }
>    if ((TraceDeoptimization || LogCompilation) && (xtty != NULL)) {
> -    ResourceMark rm;
>      ttyLocker ttyl;
>      xtty->begin_elem("make_not_compilable thread='" UINTX_FORMAT "' osr='%d' level='%d'",
> 
> Testing: {x86_64, aarch64} hotspot_gc_shenandoah, manual tests
Yeah, those are not needed (I think they might once have been covering
calls to retrieve the method name for formatting into the string).

The patch 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