8330615: avoid signed integer overflows in zip_util.c readCen / hashN

[MBaesken]

In the hashN usages of readCen from zip_util.c we see a lot of signed integer overflows.
For example in the java/util jtreg tests those are easily reproducable when compiling with -ftrapv (clang/gcc toolchains).
While those overflows never seem to cause crashes or similar errors, they are unwanted because
signed integer overflows in C cause undefined behavior.
See
https://www.gnu.org/software/c-intro-and-ref/manual/html_node/Signed-Overflow.html

For signed integers, the result of overflow in C is in principle undefined, meaning that anything whatsoever could happen.
Therefore, C compilers can do optimizations that treat the overflow case with total unconcern.

So we might still get unwanted results (maybe bad/strange hashing, depending on compiler and optimization level).

Compilation with -ftrapv causes/triggers this SIGILL on macOS showing the issue :

Problematic frame:

C [libzip.dylib+0x6362] hashN+0x32

Stack: [0x000070000c496000,0x000070000c596000], sp=0x000070000c5957e0, free space=1021k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
C [libzip.dylib+0x6362] hashN+0x32
C [libzip.dylib+0x5d5e] readCEN+0xd2e
C [libzip.dylib+0x4ee0] ZIP_Put_In_Cache0+0x160
V [libjvm.dylib+0x544b1e] ClassLoader::open_zip_file(char const*, char**, JavaThread*)+0x3e
V [libjvm.dylib+0x543fec] ClassLoader::create_class_path_entry(JavaThread*, char const*, stat const*, bool, bool)+0x6c
V [libjvm.dylib+0x543833] ClassLoader::setup_bootstrap_search_path_impl(JavaThread*, char const*)+0xf3
V [libjvm.dylib+0x54819b] classLoader_init1()+0x1b
V [libjvm.dylib+0x92602a] init_globals()+0x3a
V [libjvm.dylib+0x12b3b74] Threads::create_vm(JavaVMInitArgs*, bool*)+0x314
V [libjvm.dylib+0xa848f4] JNI_CreateJavaVM+0x64
C [libjli.dylib+0x4483] JavaMain+0x123
C [libjli.dylib+0x7529] ThreadJavaMain+0x9
C [libsystem_pthread.dylib+0x68fc] _pthread_start+0xe0
C [libsystem_pthread.dylib+0x2443] thread_start+0xf

---

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-8330615: avoid signed integer overflows in zip_util.c readCen / hashN (Bug - P4)

Reviewers

• Lutz Schmidt (@RealLucy - Reviewer)
• Martin Doerr (@TheRealMDoerr - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/18908/head:pull/18908
$ git checkout pull/18908

Update a local copy of the PR:
$ git checkout pull/18908
$ git pull https://git.openjdk.org/jdk.git pull/18908/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 18908

View PR using the GUI difftool:
$ git pr show -t 18908

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/18908.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 Welcome back mbaesken! 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]

@MBaesken 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:

8330615: avoid signed integer overflows in zip_util.c readCen / hashN

Reviewed-by: lucy, mdoerr

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 56 new commits pushed to the master branch:

• 8031dab: 8331094: ZGC: GTest fails due to incompatible Windows version
• c9f8d0e: 8331050: Serial: Remove unused _saved_mark_word in DefNewGeneration and TenuredGeneration
• b1b953c: 8330972: Serial: Inline Generation::max_contiguous_available
• 87e864b: 8328703: Illegal accesses in Java_jdk_internal_org_jline_terminal_impl_jna_linux_CLibraryImpl_ioctl0
• ef745a6: 8331097: Tests build is broken after pr/18914
• d32f109: 8329797: Shenandoah: Default case invoked for: "MaxL" (bad AD file)
• ccc0d0f: 8325373: Improve StackCounter error reporting for bad switch cases
• e818ab6: 8330815: Use pattern matching for instanceof in KeepAliveCache
• d43654e: 8331030: langtools/tools/javac/tree tests fail with SOE with fastdebug and -Xcomp
• 21480a7: 8322135: Printing JTable in Windows L&F throws InternalError: HTHEME is null
• ... and 46 more: https://git.openjdk.org/jdk/compare/896107705615a3b9363b7a0a3e6703b20fedef70...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.

[openjdk]

@MBaesken 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

• 00: Full (256558e2)

[RealLucy]

LGTM.
Please have a look at the failing tests and verify the failure is not related to this PR.

[jaikiran]

Thank you Matthias, for this change. Lance has run some internal CI tests and they have come back fine. I am in the process of running some more CI tests with this change and I should have the results, very likely by tomorrow. Please wait for those results before integrating.

[TheRealMDoerr]

Undefined behavior should always get fixed. Thanks for doing it!

[MBaesken]

Hi Lutz and Martin, thanks for the reviews!
Jaikiran, are you done with your additional tests ?

[jaikiran]

Hello Matthias, the tests completed a couple of hours back and no failures related to this change have been observed. Thank you for waiting.

[MBaesken]

/integrate

[openjdk]

Going to push as commit 5af6b45.
Since your change was applied there have been 59 commits pushed to the master branch:

• 4dfaa9b: 8329805: Deprecate for removal ObjectOutputStream.PutField.write
• b9927aa: 8329194: Cleanup Type::cmp definition and usage
• c944201: 8331031: unify os::dont_yield and os::naked_yield across Posix platforms
• 8031dab: 8331094: ZGC: GTest fails due to incompatible Windows version
• c9f8d0e: 8331050: Serial: Remove unused _saved_mark_word in DefNewGeneration and TenuredGeneration
• b1b953c: 8330972: Serial: Inline Generation::max_contiguous_available
• 87e864b: 8328703: Illegal accesses in Java_jdk_internal_org_jline_terminal_impl_jna_linux_CLibraryImpl_ioctl0
• ef745a6: 8331097: Tests build is broken after pr/18914
• d32f109: 8329797: Shenandoah: Default case invoked for: "MaxL" (bad AD file)
• ccc0d0f: 8325373: Improve StackCounter error reporting for bad switch cases
• ... and 49 more: https://git.openjdk.org/jdk/compare/896107705615a3b9363b7a0a3e6703b20fedef70...master

Your commit was automatically rebased without conflicts.

[openjdk]

@MBaesken Pushed as commit 5af6b45.

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