RFR: JDK-8374395 : Improve MemoryUsage.toString() and constructor error messages for better clarity [v2]

[Kevin Walls]

On Wed, 31 Dec 2025 09:38:10 GMT, Eunbin Son <duke at openjdk.org> wrote:

>> ### Summary
>> MemoryUsage.toString() now displays "N/A" for undefined init and max values (-1) instead of "-1(-1K)".
>> 
>> This addresses the first of three issues in JDK-8374395.
>> 
>> ### Description
>> When init or max is -1 (undefined), toString() outputs "-1(-1K)". The Javadoc states that -1 means undefined, but the output does not clearly indicate this.
>> 
>> toString() checks if init or max equals -1 and outputs "init = N/A" or "max = N/A" respectively. Output format for valid values remains unchanged.
>> 
>> Added test cases to the existing jtreg test file. All tests pass.
>> 
>> ### Bug-ID : JDK-8374395 
>> https://bugs.java.com/bugdatabase/view_bug?bug_id=JDK-8374395
>
> Eunbin Son has updated the pull request incrementally with one additional commit since the last revision:
> 
>   8374395: Add JUnit test for MemoryUsage.toString() undefined values

> When `init` or `max` is `-1` (undefined), `MemoryUsage.toString()` outputs `"-1(-1K)"`, which is confusing. While Javadoc states that `-1` means undefined, the output does not clearly indicate this, making it difficult for users to distinguish undefined values from actual negative values.

Hi. When does this happen?  When is a MemoryUsage created with init or max as -1 ?  When is it then printed using toString() for a user to understand, and when would the user then have to distinguish between the -1 which suggests a problem, and "actual negative values"?

-------------

PR Review: https://git.openjdk.org/jdk/pull/29009#pullrequestreview-3630703891