On Fri, 15 Oct 2021 15:09:23 GMT, Peter Levart wrote:
>> It's not driven by performance data. It's part of Peter's contribution.
>> I also prefer it without the packing. I propose to keep the parameter
>> count as a separate field and examine it when there is footprint issue.
>
> The
On Thu, 14 Oct 2021 16:05:19 GMT, Mandy Chung wrote:
>> src/java.base/share/classes/jdk/internal/reflect/DirectMethodHandleAccessor.java
>> line 86:
>>
>>> 84: }
>>> 85:
>>> 86: private static final int PARAM_COUNT_MASK = 0x00FF;
>>
>> Is this packing logic required? I get it that it
On Thu, 14 Oct 2021 19:53:46 GMT, Claes Redestad wrote:
> In my runs the relative regression for `instanceFieldPoly` is roughly the
> same as `staticFieldPoly` (0.61x vs 0.62x).
I was looking at this:
https://github.com/openjdk/jdk/pull/5027#issuecomment-939185418
(e.g. I didn't run
On Wed, 13 Oct 2021 18:53:22 GMT, Mandy Chung wrote:
>> This reimplements core reflection with method handles.
>>
>> For `Constructor::newInstance` and `Method::invoke`, the new implementation
>> uses `MethodHandle`. For `Field` accessor, the new implementation uses
>> `VarHandle`.For
On Thu, 14 Oct 2021 19:36:02 GMT, Mandy Chung wrote:
>> Mandy Chung has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Minor cleanup. Improve javadoc in CallerSensitiveAdapter
>
> The static Const/Poly/Var is doing better than the
On Wed, 13 Oct 2021 18:53:22 GMT, Mandy Chung wrote:
>> This reimplements core reflection with method handles.
>>
>> For `Constructor::newInstance` and `Method::invoke`, the new implementation
>> uses `MethodHandle`. For `Field` accessor, the new implementation uses
>> `VarHandle`.For
On Thu, 14 Oct 2021 00:23:29 GMT, Maurizio Cimadamore
wrote:
> I noted that in the static case, Poly does regress for fields - do we know
> why instance Poly is better than static Poly? That seems surprising.
I think you're asking why the regression of instance Poly is smaller than that
of
On Wed, 13 Oct 2021 23:53:17 GMT, Maurizio Cimadamore
wrote:
>> Mandy Chung has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Minor cleanup. Improve javadoc in CallerSensitiveAdapter
>
>
On 14/10/2021 07:05, Rémi Forax wrote:
On Thu, 14 Oct 2021 00:54:57 GMT, Mandy Chung wrote:
src/java.base/share/classes/jdk/internal/reflect/MethodHandleAccessorFactory.java
line 151:
149: var setter = isReadOnly ? null : JLIA.unreflectField(field,
true);
150:
On Thu, 14 Oct 2021 00:54:57 GMT, Mandy Chung wrote:
>> src/java.base/share/classes/jdk/internal/reflect/MethodHandleAccessorFactory.java
>> line 151:
>>
>>> 149: var setter = isReadOnly ? null :
>>> JLIA.unreflectField(field, true);
>>> 150: Class type =
On Wed, 13 Oct 2021 23:49:19 GMT, Maurizio Cimadamore
wrote:
>> Mandy Chung has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Minor cleanup. Improve javadoc in CallerSensitiveAdapter
>
>
On Thu, 14 Oct 2021 00:10:50 GMT, Maurizio Cimadamore
wrote:
>> Mandy Chung has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Minor cleanup. Improve javadoc in CallerSensitiveAdapter
>
>
On Wed, 13 Oct 2021 18:53:22 GMT, Mandy Chung wrote:
>> This reimplements core reflection with method handles.
>>
>> For `Constructor::newInstance` and `Method::invoke`, the new implementation
>> uses `MethodHandle`. For `Field` accessor, the new implementation uses
>> `VarHandle`.For
On Wed, 13 Oct 2021 18:53:22 GMT, Mandy Chung wrote:
>> This reimplements core reflection with method handles.
>>
>> For `Constructor::newInstance` and `Method::invoke`, the new implementation
>> uses `MethodHandle`. For `Field` accessor, the new implementation uses
>> `VarHandle`.For
> This reimplements core reflection with method handles.
>
> For `Constructor::newInstance` and `Method::invoke`, the new implementation
> uses `MethodHandle`. For `Field` accessor, the new implementation uses
> `VarHandle`.For the first few invocations of one of these reflective
>
15 matches
Mail list logo
