RFR: 8266950: Remove vestigial support for non-strict floating-point execution

David Holmes david.holmes at oracle.com
Tue May 18 03:08:06 UTC 2021


Hi Vladimir,

Thanks for the review.

On 18/05/2021 8:27 am, Vladimir Kozlov wrote:
> On Fri, 14 May 2021 16:29:41 GMT, Vladimir Ivanov <vlivanov at openjdk.org> wrote:
> 
>>> As part of JEP 306, the vestiges of HotSpot support for non-strict floating-point execution can be removed. All methods implicitly have strictfp semantics so the explicit checks for is_strict() can be replaced by true and the code reformulated accordingly.
>>>
>>> There are still some names that include "strict" that could potentially be renamed to remove it, but the fact we have to have strict fp semantics is still important on some platforms, so the names help reinforce that IMO.
>>>
>>> Testing: tiers 1-3
>>>
>>> Thanks,
>>> David
>>
>> Overall, it looks very good.
>> Thanks for taking care of compiler part, David.
>>
>> I think it makes sense to remove lir_div_strictfp and lir_mul_strictfp in C1 as well:
>>    https://github.com/openjdk/jdk/pull/4027
>>
>> Feel free to incorporate the patch into the current PR if you agree with the change.
>> (Passed hs-tier1 - hs-tier4 testing and x86_32 build.)
>>
>> Otherwise, I'll handle it as a separate PR.
> 
> @iwanowww I agree with your suggestion but lets do it in separate RFE.

I can pull it into this change as there is no rush to integrate this 
given I need to wait for JEP 306 to be targeted.

Thanks,
David

> -------------
> 
> PR: https://git.openjdk.java.net/jdk/pull/3991
> 


More information about the hotspot-runtime-dev mailing list