8323667: Library debug files contain non-reproducible full gcc include paths

[andrew-m-leonard]

For gcc toolchains in ALLOW_ABSOLUTE_PATHS_IN_OUTPUT=False builds, this PR finds the location of the gcc system include paths, and adds -fdebug-prefix-map flags to map them to a standard location. Thus making the debuginfo and resulting libraries reproducible when using DevKits in different path locations.

---

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-8323667: Library debug files contain non-reproducible full gcc include paths (Bug - P4)

Reviewers

• Erik Joelsson (@erikj79 - Reviewer)
• Magnus Ihse Bursie (@magicus - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/17399/head:pull/17399
$ git checkout pull/17399

Update a local copy of the PR:
$ git checkout pull/17399
$ git pull https://git.openjdk.org/jdk.git pull/17399/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 17399

View PR using the GUI difftool:
$ git pr show -t 17399

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/17399.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back aleonard! 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.

[openjdk]

@andrew-m-leonard The following label will be automatically applied to this pull request:

• build

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

[mlbridge]

Webrevs

• 05: Full - Incremental (3dcfcc5d)
• 04: Full - Incremental (4f46bc17)
• 03: Full - Incremental (1b167c12)
• 02: Full - Incremental (adcb0fd5)
• 01: Full - Incremental (587bdc41)
• 00: Full (7c8e81cc)

[andrew-m-leonard]

I have tested this change successfully at Eclipse Adoptium across all the various Linux gcc platforms: x64Linux, ppc64leLinux, aarch64Linux, alpineAarch64Linux, s390xLinux

[andrew-m-leonard]

@erikj79 when you get a moment, can you re-review please? thanks

[magicus]

Have you checked with strings if there are any other hard-coded paths, or are these really the only ones left?

[andrew-m-leonard]

Have you checked with strings if there are any other hard-coded paths, or are these really the only ones left?

I have been doing some separate machine/location builds with DevKits located in different paths and build workspace in different location, and they are now "identical" with this change.
So yes, I believe this is complete.

[erikj79]

I think this looks ok but could you split the longer lines to make side by side comparisons easier in the future?

[magicus]

So $CC $CFLAGS did not work?

[andrew-m-leonard]

I think this looks ok but could you split the longer lines to make side by side comparisons easier in the future?

Yep, done, thanks Erik

[andrew-m-leonard]

So $CC $CFLAGS did not work?

@magicus
So I purposely chose SYSROOT_CFLAGS thinking that would all that would be needed, but I think you're correct, I should use the complete CFLAGS in case a custom user extension include may have pointed at some other include path.
I have updated and tested successfully with CFLAGS_JDKLIB.
Ready for re-review, Thanks

[erikj79]

So $CC $CFLAGS did not work?

@magicus So I purposely chose SYSROOT_CFLAGS thinking that would all that would be needed, but I think you're correct, I should use the complete CFLAGS in case a custom user extension include may have pointed at some other include path. I have updated and tested successfully with CFLAGS_JDKLIB. Ready for re-review, Thanks

Magnus specifically meant $CFLAGS. It's set to "$MACHINE_FLAG $SYSROOT_CFLAGS $USER_CFLAGS" in flags.m4 and meant to contain relevant flags for running checks in configure. I don't think $CFLAGS_JDKLIB is appropriate here even if it should work.

[andrew-m-leonard]

So $CC $CFLAGS did not work?

@magicus So I purposely chose SYSROOT_CFLAGS thinking that would all that would be needed, but I think you're correct, I should use the complete CFLAGS in case a custom user extension include may have pointed at some other include path. I have updated and tested successfully with CFLAGS_JDKLIB. Ready for re-review, Thanks

Magnus specifically meant $CFLAGS. It's set to "$MACHINE_FLAG $SYSROOT_CFLAGS $USER_CFLAGS" in flags.m4 and meant to contain relevant flags for running checks in configure. I don't think $CFLAGS_JDKLIB is appropriate here even if it should work.

Ah, I thought CFLAGS was only set in the gmk files...? i'll give it a try thanks.

[andrew-m-leonard]

So $CC $CFLAGS did not work?

@magicus So I purposely chose SYSROOT_CFLAGS thinking that would all that would be needed, but I think you're correct, I should use the complete CFLAGS in case a custom user extension include may have pointed at some other include path. I have updated and tested successfully with CFLAGS_JDKLIB. Ready for re-review, Thanks

Magnus specifically meant $CFLAGS. It's set to "$MACHINE_FLAG $SYSROOT_CFLAGS $USER_CFLAGS" in flags.m4 and meant to contain relevant flags for running checks in configure. I don't think $CFLAGS_JDKLIB is appropriate here even if it should work.

Thanks @erikj79 yep, that works great, thank you for pointing that out.
Updated.

[erikj79]

Ah, I thought CFLAGS was only set in the gmk files...? i'll give it a try thanks.

No, we specifically only set the variable CFLAGS in configure because there are autoconf macros that relies on that variable. In our makefiles we avoid the standard variable CFLAGS and are instead choosing to be explicit with various CFLAGS_* and *_CFLAGS variables.

[erikj79]

This looks good to me now.

[openjdk]

@andrew-m-leonard This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

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

8323667: Library debug files contain non-reproducible full gcc include paths

Reviewed-by: erikj, ihse

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 59 new commits pushed to the master branch:

• ff8cc26: 8323694: RISC-V: Unnecessary ResourceMark in NativeCall::set_destination_mt_safe
• dadd9cd: 8316497: ColorConvertOp - typo for non-ICC conversions needs one-line fix
• de237fb: 8323554: The typos in Javadoc: "@return if "
• 51dbd36: 8322100: Fix GCMIncrementByte4 & GCMIncrementDirect4, and increase overlap testing
• ffa33d7: 8317771: [macos14] Expand/collapse a JTree using keyboard freezes the application in macOS 14 Sonoma
• 4e53235: 8323710: (fc) FileChannel.lock creates a FileKey with a poor hashCode after JDK-8321429 (win)
• 19287ee: 8321561: (fs) Clarify non-atomic behavior of Files.move
• 8b29e12: 8323794: Remove unused jimage compressor plugin configuration
• 7be9f1d: 8321137: Reconsider ICStub alignment
• b8dafa6: 8323612: IOVecWrapper should be changed to be TerminatingThreadLocal
• ... and 49 more: https://git.openjdk.org/jdk/compare/8643cc21333c6b51242ed3b9295b25f372244755...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this 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.

[magicus]

Looks good now.

Thanks Erik for helping me clarify what I meant.

[andrew-m-leonard]

/integrate

[openjdk]

Going to push as commit 57fad67.
Since your change was applied there have been 59 commits pushed to the master branch:

• ff8cc26: 8323694: RISC-V: Unnecessary ResourceMark in NativeCall::set_destination_mt_safe
• dadd9cd: 8316497: ColorConvertOp - typo for non-ICC conversions needs one-line fix
• de237fb: 8323554: The typos in Javadoc: "@return if "
• 51dbd36: 8322100: Fix GCMIncrementByte4 & GCMIncrementDirect4, and increase overlap testing
• ffa33d7: 8317771: [macos14] Expand/collapse a JTree using keyboard freezes the application in macOS 14 Sonoma
• 4e53235: 8323710: (fc) FileChannel.lock creates a FileKey with a poor hashCode after JDK-8321429 (win)
• 19287ee: 8321561: (fs) Clarify non-atomic behavior of Files.move
• 8b29e12: 8323794: Remove unused jimage compressor plugin configuration
• 7be9f1d: 8321137: Reconsider ICStub alignment
• b8dafa6: 8323612: IOVecWrapper should be changed to be TerminatingThreadLocal
• ... and 49 more: https://git.openjdk.org/jdk/compare/8643cc21333c6b51242ed3b9295b25f372244755...master

Your commit was automatically rebased without conflicts.

[openjdk]

@andrew-m-leonard Pushed as commit 57fad67.

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