[jdk11u-dev] RFR: 8360816: [11u] Use default value for ProgramFiles(x86) in GHA [v3]

[Severin Gehwolf]

On Fri, 27 Jun 2025 12:51:21 GMT, Antonio Vieiro <duke at openjdk.org> wrote:

>> A fix for [JDK-8360816](https://bugs.openjdk.org/browse/JDK-8360816). 
>> 
>> The `ProgramFiles(x86)` environment variable is not properly propagated to JTReg tests in JDK11, which makes the AOT compiler tests fail on the new `windows-2025` GHA runners, as manifested in the backport of [https://bugs.openjdk.org/browse/JDK-8358538](JDK-8358538).  
>> 
>> Propagating this environment variable to JTRegs may require creating devkits for Windows 2022, but previous attempts were discarded [JDK-8283723](https://bugs.openjdk.org/browse/JDK-8283723). 
>> 
>> This fix prints a warning message when this environment variable is not set, and then uses a default value (`C:\Program Files (x86)`) that makes it possible to run the AOT tests with these new `windows-2025` GHA runners.  If the `ProgramFiles(x86)` env variable is set the fix has no effect, ensuring the previous behaviour. If the proposed default value is incorrect the AOT tests will continue to fail as before.
>> 
>> The PR is on top of  https://github.com/openjdk/jdk11u-dev/pull/3052 to verify the fix works as intended with the new `windows-2025` GHA runners.
>
> Antonio Vieiro has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains four commits:
> 
>  - Merge master
>  - 8360816: [11u] Use default value for ProgramFiles(x86) in GHA
>  - Backport 72679c94ee00c87b9b51233938e5ffa97ef825b1
>  - Backport b787ff6def08a050b690b60e4a0ceb3aec2b73c8

Why don't we set the expected `ProgramFiles(x86)` environment variable to value `C:\\Program Files (x86)` in the GHA job? This would be a less invasive fix as the code in this PR changes product code.

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

PR Review: https://git.openjdk.org/jdk11u-dev/pull/3056#pullrequestreview-2966504131