On Mon, 11 Apr 2022 10:03:58 GMT, Claes Redestad wrote:
>>> How would the performance change if the `isDirect` and
>>> `checkExactAccessMode` merger was reverted?
>>
>> Add around 15-20ns/op for these micros.
>
> Restructuring so that we only check `direct` once sounds reasonable at face
>
On Mon, 11 Apr 2022 09:53:31 GMT, Claes Redestad wrote:
>> src/java.base/share/classes/java/lang/invoke/VarHandle.java line 2075:
>>
>>> 2073:
>>> 2074: @DontInline
>>> 2075: final void
>>> throwWrongMethodTypeException(VarHandle.AccessDescriptor ad) {
>>
>> This can actually be
On Mon, 11 Apr 2022 09:57:42 GMT, Claes Redestad wrote:
>> src/java.base/share/classes/java/lang/invoke/VarHandles.java line 719:
>>
>>> 717: //
>>> MethodHandle.linkToStatic();
>>> 718: //} else {
>>> 719: //MethodHandle mh =
On Mon, 11 Apr 2022 09:34:17 GMT, ExE Boss wrote:
>> Claes Redestad has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Simplified as suggested by @ExE-Boss
>
> src/java.base/share/classes/java/lang/invoke/VarHandles.java line 719:
>
>>
On Mon, 11 Apr 2022 09:26:12 GMT, ExE Boss wrote:
>> Claes Redestad has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Simplified as suggested by @ExE-Boss
>
> src/java.base/share/classes/java/lang/invoke/VarHandle.java line 2075:
>
>>
On Fri, 8 Apr 2022 12:20:32 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
On Fri, 8 Apr 2022 12:20:32 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
On Fri, 8 Apr 2022 12:20:32 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
On Fri, 8 Apr 2022 12:20:32 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
On Fri, 8 Apr 2022 15:05:54 GMT, Maurizio Cimadamore
wrote:
> TestAdaptVarHandles
Thanks for reviewing, Maurizio. I ran the micros you suggested and I detect no
change whatsoever at peak performance. With `-Xint` there are some small
improvements in line with the results above.
On Fri, 8 Apr 2022 12:20:32 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
On Fri, 8 Apr 2022 12:12:06 GMT, ExE Boss wrote:
>> Claes Redestad has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Simplified as suggested by @ExE-Boss
>
> src/java.base/share/classes/java/lang/invoke/IndirectVarHandle.java line 104:
>
On Fri, 8 Apr 2022 12:16:17 GMT, Claes Redestad wrote:
>> A few additional enhancements aiming to improve VH performance in the
>> interpreter:
>>
>> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
>> 40->48) but removes an object and an indirection on any
> A few additional enhancements aiming to improve VH performance in the
> interpreter:
>
> - Flatten `TypeAndInvokers`: adds a pointer to `VarHandle` (a small increase
> 40->48) but removes an object and an indirection on any instance actually
> used - and might avoid allocating the
14 matches
Mail list logo