8253409: Double-rounding possibility in float fma

[jddarcy]

In floating-point, usually doing an operation to double precision and then rounding to float gives the right result in float precision. One exception to this is fused multiply add (fma) where "a * b + c" is computed with a single rounding. This requires the equivalent of extra intermediate precision inside the operation. If a float fma is implemented using a double fma rounded to float, for some well-chosen arguments where the final result is near a half-way result in float, an incorrect answer will be computed due to double rounding. In more detail, the double result will round up and then the cast to float will round up again whereas a single rounding of the exact answer to float would only round-up once.

The new float fma implementation does the exact arithmetic using BigDecimal where possible, with guard to handle the non-finite and signed zero IEEE 754 details.

---

Progress

• Change must not contain extraneous whitespace
• Commit message must refer to an issue
• Change must be properly reviewed

Issue

• JDK-8253409: Double-rounding possibility in float fma

Reviewers

• Brian Burkhalter (@bplb - Reviewer)

Download

$ git fetch https://git.openjdk.java.net/jdk pull/2684/head:pull/2684
$ git checkout pull/2684

[bridgekeeper]

👋 Welcome back darcy! 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]

@jddarcy The following label will be automatically applied to this pull request:

• core-libs

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

• 01: Full - Incremental (ee2ea23)
• 00: Full (9d26b31)

[bplb]

Looks fine. Presumably the updated test fails without the source change.

[openjdk]

@jddarcy 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:

8253409: Double-rounding possibility in float fma

Reviewed-by: bpb

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 63 new commits pushed to the master branch:

• c30a90b: 8261976: Normalize id's used by the standard doclet
• 53b1545: 8223355: Redundant output by javadoc
• d2b9c22: 8262011: [JVMCI] allow printing to tty from unattached libgraal thread
• 0257caa: 8261397: Try Catch Method Failing to Work When Dividing An Integer By 0
• 8a2f589: 8260637: Shenandoah: assert(_base == Tuple) failure during C2 compilation
• 67762de: 8262197: JDK-8242032 uses wrong contains_reference() in assertion code
• 9d9bedd: 8262094: Handshake timeout scaled wrong
• 29c7263: 8252709: Enable JVMCI when building linux-aarch64 at Oracle
• 12f6ba0: 8262087: Use atomic boolean type in G1FullGCAdjustTask
• a5c4b9a: 8260223: Handling of unnamed package in javadoc pages
• ... and 53 more: https://git.openjdk.java.net/jdk/compare/c4664e64463aa9d3f231e1115857b1a20213d8ab...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.

[jddarcy]

Looks fine. Presumably the updated test fails without the source change.

Right; the added test case is the failing one from the bug report. It will fail if the old non-intrinsic implementation, that is the Java implementation is used.

[jddarcy]

/integrate

[openjdk]

@jddarcy Since your change was applied there have been 64 commits pushed to the master branch:

• 3132b1c: 8261665: Clean up naming of StringContent and FixedStringContent
• c30a90b: 8261976: Normalize id's used by the standard doclet
• 53b1545: 8223355: Redundant output by javadoc
• d2b9c22: 8262011: [JVMCI] allow printing to tty from unattached libgraal thread
• 0257caa: 8261397: Try Catch Method Failing to Work When Dividing An Integer By 0
• 8a2f589: 8260637: Shenandoah: assert(_base == Tuple) failure during C2 compilation
• 67762de: 8262197: JDK-8242032 uses wrong contains_reference() in assertion code
• 9d9bedd: 8262094: Handshake timeout scaled wrong
• 29c7263: 8252709: Enable JVMCI when building linux-aarch64 at Oracle
• 12f6ba0: 8262087: Use atomic boolean type in G1FullGCAdjustTask
• ... and 54 more: https://git.openjdk.java.net/jdk/compare/c4664e64463aa9d3f231e1115857b1a20213d8ab...master

Your commit was automatically rebased without conflicts.

Pushed as commit e5304b3.

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

[jddarcy]

/backport 16u

[openjdk]

@jddarcy Unknown command backport - for a list of valid commands use /help.