RFR: JDK-8322141: SequenceInputStream.transferTo should not return as soon as Long.MAX_VALUE bytes have been transferred [v4]
Markus KARG
duke at openjdk.org
Sat Dec 16 14:07:53 UTC 2023
On Fri, 15 Dec 2023 22:40:17 GMT, Brian Burkhalter <bpb at openjdk.org> wrote:
>>> > [...] maybe we need to come up with tests that quickly check the handling at this limit.
>>
>>> @mkarg Do you intend to provide any such test as part of this request?
>>
>> @bplb I quickly drafted an absolute minimal test for this in [2aaac63](https://github.com/openjdk/jdk/pull/17119/commits/2aaac630a3e6d1df0fd77f6b34c0f0d3a3b50292). IMHO this is sufficient for the current case.
>
>> I quickly drafted an absolute minimal test for this in [2aaac63] [...].
>
> This test does not fail for me when run against my local build of the current mainline. Usually, unless randomness is involved, a test should fail before the proposed patch is applied but succeed thereafter.
@bplb Test is fixed now. Sorry for the delay. I have updated the test (contained in https://github.com/openjdk/jdk/pull/17119/commits/556a8dc4d3d8ff157e8374625eb129a08926e0dd) and successfully tried it with the original code (test fails) and the fixed code (test succeeds). IMHO this should do the job.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/17119#issuecomment-1858826626
More information about the core-libs-dev
mailing list