8299616: [11u] Bootcycle build fails after JDK-8257679 backport

[shipilev]

Currently, JDK 11u make bootcycle-images fails with:

 bootcycle-spec.gmk:32: *** non-numeric first argument to `word' function: ''. Stop.

The apparent reason is JDK-8257679 that did the following change:
 40f4fc2#diff-935527b393cb4625a8874febf5fe39d43479a2228ef437f4e104ba3001e1e30fL31-L40

Note that the left-over hunk uses JAVA_EXEC_POS, which is defined in the hunk that was removed. This PR reinstantiates that hunk.

Additional testing:

• Linux x86_64 fastdebug make bootcycle-images
• Windows x86_64 fastdebug make bootcycle-images

---

Progress

• Change must be properly reviewed (1 review required, with at least 1 Reviewer)
• Change must not contain extraneous whitespace
• Commit message must refer to an issue

Issue

• JDK-8299616: [11u] Bootcycle build fails after JDK-8257679 backport

Reviewers

• Andrew John Hughes (@gnu-andrew - Reviewer) ⚠️ Review applies to b0ad92ed
• Christoph Langer (@RealCLanger - Reviewer) ⚠️ Review applies to b0ad92ed

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk11u-dev pull/1639/head:pull/1639
$ git checkout pull/1639

Update a local copy of the PR:
$ git checkout pull/1639
$ git pull https://git.openjdk.org/jdk11u-dev pull/1639/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 1639

View PR using the GUI difftool:
$ git pr show -t 1639

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk11u-dev/pull/1639.diff

[bridgekeeper]

👋 Welcome back shade! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

[gnu-andrew]

Thanks for tracking this down so quickly. After you identified the problematic patches, I came to the same conclusion that this is down to JAVA_EXEC_POS still being used in bootcycle-spec.gmk.in, but now undefined.

I think the reason for the change was two-fold:

1. $(FIXPATH) was removed from JAVA in spec.gmk.in and only added to JAVA_CMD in bootcycle-spec.gmk.in so the second branch would always be true.
2. The block as a whole was removed as JAVA_EXEC_POS is unused after JDK-8244036 (not backported to 11u)

The safest option is to restore what we had before, as your patch does, but it may be that JAVA_EXEC_POS can now always be 1 (or just removed and a hardcoded 1 used in bootcycle-spec.gmk.in).

I'm happy with this as the lowest risk fix, assuming Windows builds fine.

[mlbridge]

Webrevs

• 01: Full - Incremental (97b192f0)
• 00: Full (b0ad92ed)

[openjdk]

@shipilev This change now passes all automated pre-integration checks.

After integration, the commit message for the final commit will be:

8299616: [11u] Bootcycle build fails after JDK-8257679 backport

Reviewed-by: andrew, clanger

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been no new commits pushed to the master branch. If another commit should be pushed before you perform the /integrate command, your PR will be automatically rebased. If you prefer to avoid any potential automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

[RealCLanger]

Ah, yes, JDK-8244036, is not in jdk11u, so removing this block was wrong. Thanks for finding & fixing. Please integrate this and I'll cherry-pick it into the bundle for 11.0.18, too.

[shipilev]

The safest option is to restore what we had before, as your patch does, but it may be that JAVA_EXEC_POS can now always be 1 (or just removed and a hardcoded 1 used in bootcycle-spec.gmk.in).

True. But, I could not convince myself there is a guarantee for fixpath to be absent in either $SJAVAC_SERVER_JAVA or $JAVA, so kept the most pessimistic fix.

I did massage the comments a bit, so that they would match the original code shape.

[shipilev]

Windows and Linux bootcycle builds are fine. GHAs are clean. Push approval is there. I am integrating.

/integrate

[openjdk]

Going to push as commit 6073296.

[openjdk]

@shipilev Pushed as commit 6073296.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

[gnu-andrew]

The safest option is to restore what we had before, as your patch does, but it may be that JAVA_EXEC_POS can now always be 1 (or just removed and a hardcoded 1 used in bootcycle-spec.gmk.in).

True. But, I could not convince myself there is a guarantee for fixpath to be absent in either $SJAVAC_SERVER_JAVA or $JAVA, so kept the most pessimistic fix.

I did massage the comments a bit, so that they would match the original code shape.

Yes, that was my thinking too. Better safe than sorry.