Integrated: 8288594: Address possibly lossy conversions in java.base FloatToDecimal

Raffaello Giulietti raffaello.giulietti at oracle.com
Thu Jun 16 22:15:47 UTC 2022


Ah, no, it's a P4 bug.
Sorry for the noise


On 2022-06-16 23:57, Raffaello Giulietti wrote:
> Although the conversion is mathematically sound and never lossy, this 
> change should probably be backported to 19 as well.
> 
> 
> Greetings
> Raffaello
> 
> 
> On 2022-06-16 19:01, Adam Sotona wrote:
>> On Thu, 16 Jun 2022 15:26:37 GMT, Adam Sotona <asotona at openjdk.org> 
>> wrote:
>>
>>> One new case of possibly lossy conversions recently appeared in 
>>> jdk.internal.math.FloatToDecimal.
>>> This patch avoids the warning by explicit cast from long to int.
>>>
>>> Please review.
>>>
>>> Thanks,
>>> Adam
>>
>> This pull request has now been integrated.
>>
>> Changeset: e7d52e25
>> Author:    Adam Sotona <asotona at openjdk.org>
>> URL:       
>> https://git.openjdk.org/jdk/commit/e7d52e25a273a534c3195ea58745fd4a34963454 
>>
>> Stats:     1 line in 1 file changed: 0 ins; 0 del; 1 mod
>>
>> 8288594: Address possibly lossy conversions in java.base FloatToDecimal
>>
>> Reviewed-by: rriggs
>>
>> -------------
>>
>> PR: https://git.openjdk.org/jdk/pull/9188


More information about the core-libs-dev mailing list