RFR: 8265711: C1: Intrinsify Class.getModifier method [v2]
Yi Yang
yyang at openjdk.java.net
Tue Apr 27 12:59:57 UTC 2021
On Tue, 27 Apr 2021 12:52:56 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:
>> With the new IR test framework, these kinds of problem would never happen, we can do verification at IR level.
>
>> Because is_intrinsic_support is not aware of the fact that getModifiers is an intrinsic method. So canonicalizer is even not applied.
>
> But your first version (https://github.com/openjdk/jdk/pull/3616/commits/7f0be4e8a2cb4f6a9ef650b1b31c16f46716a257) already had the change to `is_intrinsic_supported`? Shouldn't that version have triggered test failures because it does not handle primitive types?
>
>> With the new IR test framework, these kinds of problem would never happen, we can do verification at IR level.
>
> Unfortunately not. The IR test framework currently only supports C2.
> But your first version (7f0be4e) already had the change to is_intrinsic_supported? Shouldn't that version have triggered test failures because it does not handle primitive types?
The first version only changes Canonicalizer::do_Intrinsic :_
-------------
PR: https://git.openjdk.java.net/jdk/pull/3616
More information about the hotspot-compiler-dev
mailing list