Safer type casting #363

zhengyu123 · 2026-02-10T21:43:29Z

What does this PR do?:

Capture uninitialized field offsets
Capture invalid memory accesses
Capture bad type castings
Cleanup, remove unused code
Renamed some classes to avoid name collisions with JVM, this improves debugability.

Motivation:
Profiler code heavily depends on vmStructs to decode hotspot JVM internal data structures, but vmStructs evolves version-by-version, that can results in misinterpreting the data structures and leads to crashes.

This PR is intended to capture the problem early, by:

check for uninitialized field offsets - likely due to vmStructs name changes, etc.
check underlaid memory is mapped before access.
check memory range that covers the destination type, is valid before casts to destination type.
cleanup: removed unused code, such as:
permGen - no long presented in supported JVM
CollectedHeap - no use
use macro to generate type declarations and corresponding initializations to avoid boilerplate code.

Additional Notes:
This PR also provides a way to mitigate the tedious tasks to define cast function for a type, which you have to:

define a type size variable
initialize the size variable
populate the size from vmStructs
define cast and load_then_cast functions for the type

This PR introduces a macro:

#define DECL_TYPES_DO(f) \
    f(VMClassLoaderData,    MATCH_SYMBOLS("ClassLoaderData")) \
    f(VMConstantPool,       MATCH_SYMBOLS("ConstantPool")) \
    f(VMConstMethod,        MATCH_SYMBOLS("ConstMethod")) \
    f(VMFlag,               MATCH_SYMBOLS("JVMFlag", "Flag")) \
    f(VMJavaFrameAnchor,    MATCH_SYMBOLS("JavaFrameAnchor")) \
    f(VMKlass,              MATCH_SYMBOLS("Klass")) \
    f(VMMethod,             MATCH_SYMBOLS("Method")) \
    f(VMNMethod,            MATCH_SYMBOLS("nmethod")) \
    f(VMSymbol,             MATCH_SYMBOLS("Symbol")) \
    f(VMThread,             MATCH_SYMBOLS("Thread"))

For each row, it defines a type and the symbol(s) used in vmStructs to populate type size. Notice, there can be more than one symbol for the type, due to different JVM versions that export the same value under different names.

To add a new type, you only need to add a row, then use following macros to define the type in profiler, e.g.

DECL_TYPE(VMClassLoaderData)
....
DECL_TYPE_END

The definition automatically creates three methods, with safety checks:

        // size of the type - sizeof(T)
        static uint64_t type_size();
        // cast from a pointer
        static name * cast(const void* ptr); 
        // load a pointer from given location, then cast the pointer
        static name * load_then_cast(const void* ptr);

How to test the change?:
Running tests on different version of JDKs, make sure that newly added checks do not fail.

For Datadog employees:

If this PR touches code that signs or publishes builds or packages, or handles
credentials of any kind, I've requested a review from @DataDog/security-design-and-guidance.
This PR doesn't touch any of that.
JIRA: PROF-13721

Unsure? Have a question? Request a review!

dd-octo-sts · 2026-02-10T21:46:06Z

Scan-Build Report

User:	runner@runnervmwffz4
Working Directory:	/home/runner/work/java-profiler/java-profiler/ddprof-lib/src/test/make
Command Line:	make -j4 all
Clang Version:	Ubuntu clang version 18.1.3 (1ubuntu1)
Date:	Thu Feb 12 14:05:01 2026

Bug Summary

Unused code
Bug Type	Quantity	Display?
All Bugs	1
Dead assignment	1

Reports

Bug Group	Bug Type ▾	File	Function/Method	Line	Path Length
Unused code	Dead assignment	libraryPatcher_linux.cpp	patch_library_unlocked	94	1

pr-commenter · 2026-02-10T21:51:09Z

Integration Tests

✅ All 32 integration tests passed

📊 Dashboard · 👷 Pipeline · 📦 9b576012

pr-commenter · 2026-02-10T21:57:15Z

Benchmarks [aarch64 memleak,alloc]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	memleak,alloc	memleak,alloc
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

dd-octo-sts · 2026-02-10T22:00:54Z

CI Test Results

Run: #21949711057 | Commit: c1900bb | Duration: 20m 36s (longest job)

✅ All 40 test jobs passed

Status Overview

JDK	glibc-aarch64/debug	glibc-amd64/debug	musl-aarch64/debug	musl-amd64/debug
8	-	✅	-	-
8-ibm	-	✅	-	-
8-j9	✅	✅	-	-
8-librca	-	-	✅	✅
8-orcl	-	✅	-	-
8-zing	✅	✅	-	-
11	-	✅	-	-
11-j9	✅	✅	-	-
11-librca	-	-	✅	✅
11-zing	✅	✅	-	-
17	✅	✅	-	-
17-graal	✅	✅	-	-
17-j9	✅	✅	-	-
17-librca	-	-	✅	✅
17-zing	✅	✅	-	-
21	✅	✅	-	-
21-graal	✅	✅	-	-
21-librca	-	-	✅	✅
21-zing	✅	✅	-	-
25	✅	✅	-	-
25-graal	✅	✅	-	-
25-librca	-	-	✅	✅

Legend: ✅ passed | ❌ failed | ⚪ skipped | 🚫 cancelled

Summary: Total: 40 | Passed: 40 | Failed: 0

Updated: 2026-02-12 14:27:37 UTC

pr-commenter · 2026-02-11T00:12:27Z

Benchmarks [x86_64 cpu,wall,alloc,memleak]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	cpu,wall,alloc,memleak	cpu,wall,alloc,memleak
wall	on	on

Summary

Found 1 performance improvements and 0 performance regressions! Performance is the same for 14 metrics, 23 unstable metrics.

scenario	Δ mean execution_time	Δ mean rss
scenario:renaissance:scala-doku	better [-889.952ms; -446.048ms] or [-3.412%; -1.710%]	unstable [-254.860MB; +257.112MB] or [-22.665%; +22.865%]

pr-commenter · 2026-02-11T00:12:31Z

Benchmarks [x86_64 wall]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	wall	wall
wall	on	on

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 15 metrics, 23 unstable metrics.

pr-commenter · 2026-02-11T00:12:41Z

Benchmarks [x86_64 cpu,wall]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	cpu,wall	cpu,wall
wall	on	on

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 14 metrics, 24 unstable metrics.

pr-commenter · 2026-02-11T00:12:54Z

Benchmarks [x86_64 cpu]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	cpu	cpu
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 14 metrics, 24 unstable metrics.

pr-commenter · 2026-02-11T00:12:58Z

Benchmarks [x86_64 alloc]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	alloc	alloc
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

pr-commenter · 2026-02-11T00:13:04Z

Benchmarks [aarch64 wall]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	wall	wall
wall	on	on

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 15 metrics, 23 unstable metrics.

pr-commenter · 2026-02-11T00:13:12Z

Benchmarks [aarch64 cpu,wall,alloc,memleak]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	cpu,wall,alloc,memleak	cpu,wall,alloc,memleak
wall	on	on

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

pr-commenter · 2026-02-11T00:13:36Z

Benchmarks [aarch64 cpu,wall]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	cpu,wall	cpu,wall
wall	on	on

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 15 metrics, 23 unstable metrics.

pr-commenter · 2026-02-11T00:13:37Z

Benchmarks [x86_64 memleak,alloc]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	memleak,alloc	memleak,alloc
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 14 metrics, 24 unstable metrics.

pr-commenter · 2026-02-11T00:13:56Z

Benchmarks [aarch64 cpu]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	on	on
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	cpu	cpu
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

pr-commenter · 2026-02-11T00:14:03Z

Benchmarks [aarch64 alloc]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	on	on
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	off	off
modes	alloc	alloc
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

pr-commenter · 2026-02-11T00:14:16Z

Benchmarks [aarch64 memleak]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	memleak	memleak
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 15 metrics, 23 unstable metrics.

pr-commenter · 2026-02-11T00:15:05Z

Benchmarks [x86_64 memleak]

Parameters

	Baseline	Candidate
config	baseline	candidate
ddprof	1.37.0	1.38.0-zgu_vmstruct_safety-SNAPSHOT

See matching parameters

	Baseline	Candidate
alloc	off	off
cpu	off	off
iterations	5	5
java	"11.0.28"	"11.0.28"
memleak	on	on
modes	memleak	memleak
wall	off	off

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 16 metrics, 22 unstable metrics.

ddprof-lib/src/main/cpp/stackFrame_ppc64.cpp

ddprof-lib/src/main/cpp/vmStructs.h

ddprof-lib/src/main/cpp/vmStructs.cpp

ddprof-lib/src/main/cpp/vmStructs.h

ddprof-lib/src/main/cpp/vmStructs.inline.h

ddprof-lib/src/main/cpp/vmStructs.h

jbachorik

Good effort to get the vmstructs more organized. I wonder, if we can get to a state where the VMStructs class does not need to hold all possible offset holders.

I would really appreciate more cleanup before requesting review. There was a bunch of low-hanging typos and miss-spels here :/

zhengyu123 added 2 commits February 9, 2026 14:19

v0

61ce326

v0

641d18c

v1

9b57601

zhengyu123 added 4 commits February 11, 2026 15:49

v2

dc1c85a

v3

feaa4d3

v4

d4de63a

Add missing file

b7a7a8e

zhengyu123 changed the title ~~WIP:~~ WIP: Safer type casting Feb 11, 2026

zhengyu123 added 5 commits February 11, 2026 20:00

Allows nullptr to cast

0101624

Merge branch 'main' into zgu/vmstruct_safety

d45ad11

Remove dead code

645ccfa

Improve macro

e33499e

update

2751736

zhengyu123 changed the title ~~WIP: Safer type casting~~ Safer type casting Feb 12, 2026

zhengyu123 marked this pull request as ready for review February 12, 2026 00:24

zhengyu123 requested a review from a team as a code owner February 12, 2026 00:24

jbachorik requested changes Feb 12, 2026

View reviewed changes

zhengyu123 requested a review from rkennke February 12, 2026 13:21

Review comments

48f7306

Safer type casting #363

Are you sure you want to change the base?

Safer type casting #363

Conversation

zhengyu123 commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dd-octo-sts bot commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Scan-Build Report

Bug Summary

Reports

Uh oh!

pr-commenter bot commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Integration Tests

Uh oh!

pr-commenter bot commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmarks [aarch64 memleak,alloc]

Parameters

Summary

Uh oh!

dd-octo-sts bot commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

CI Test Results

Status Overview

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 cpu,wall,alloc,memleak]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 wall]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 cpu,wall]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 cpu]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 alloc]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [aarch64 wall]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [aarch64 cpu,wall,alloc,memleak]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [aarch64 cpu,wall]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [x86_64 memleak,alloc]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [aarch64 cpu]

Parameters

Summary

Uh oh!

pr-commenter bot commented Feb 11, 2026

Benchmarks [aarch64 alloc]

zhengyu123 commented Feb 10, 2026 •

edited

Loading

dd-octo-sts bot commented Feb 10, 2026 •

edited

Loading

pr-commenter bot commented Feb 10, 2026 •

edited

Loading

pr-commenter bot commented Feb 10, 2026 •

edited

Loading

dd-octo-sts bot commented Feb 10, 2026 •

edited

Loading

jbachorik left a comment •

edited

Loading