July 2025 - leyden-dev

git: openjdk/leyden: premain: 8363978: [Leyden] Crash in C1 AOT compiled java.lang.Thread.<init> method
by duke 24 Jul '25

24 Jul '25

Changeset: 2c812df7 Branch: premain Author: Vladimir Kozlov <vladimir.kozlov(a)oracle.com> Date: 2025-07-24 14:06:11 +0000 URL: https://git.openjdk.org/leyden/commit/2c812df78710301738b91bb9bf0f3d4fe4dd6… 8363978: [Leyden] Crash in C1 AOT compiled java.lang.Thread.<init> method ! src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp ! src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp

1 0

RFR: 8362559: [leyden] assert(!InstanceKlass::cast(receiver_klass)->is_not_initialized()) failed: receiver_klass must be initialized
by Vladimir Kozlov 24 Jul '25

24 Jul '25

Fix is proposed by @iklam: add check that we precompiling AOT code in C1 where we check for initialized klass. I renamed `is_compiled()` to `is_compile()` because we are in pricing of compiling. Tested tier1-4,tier8-comp ------------- Commit messages: - 8362559: [leyden] assert(!InstanceKlass::cast(receiver_klass)->is_not_initialized()) failed: receiver_klass must be initialized Changes: https://git.openjdk.org/leyden/pull/88/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=88&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8362559 Stats: 27 lines in 9 files changed: 6 ins; 1 del; 20 mod Patch: https://git.openjdk.org/leyden/pull/88.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/88/head:pull/88 PR: https://git.openjdk.org/leyden/pull/88

3 5

git: openjdk/leyden: premain: 8362559: [leyden] assert(!InstanceKlass::cast(receiver_klass)->is_not_initialized()) failed: receiver_klass must be initialized
by Vladimir Kozlov 24 Jul '25

24 Jul '25

Changeset: 2d6b3b12 Branch: premain Author: Vladimir Kozlov <kvn(a)openjdk.org> Date: 2025-07-24 16:30:31 +0000 URL: https://git.openjdk.org/leyden/commit/2d6b3b126c3b65c502cbce51da1d3b7dc6a5c… 8362559: [leyden] assert(!InstanceKlass::cast(receiver_klass)->is_not_initialized()) failed: receiver_klass must be initialized Reviewed-by: iklam, shade ! src/hotspot/share/c1/c1_GraphBuilder.cpp ! src/hotspot/share/c1/c1_LIRGenerator.cpp ! src/hotspot/share/ci/ciEnv.cpp ! src/hotspot/share/ci/ciEnv.hpp ! src/hotspot/share/ci/ciInstanceKlass.cpp ! src/hotspot/share/ci/ciMethod.cpp ! src/hotspot/share/code/aotCodeCache.cpp ! src/hotspot/share/compiler/compileBroker.cpp ! src/hotspot/share/compiler/compileTask.hpp

1 0

git: openjdk/leyden: premain: AOT code caching fixes:
by duke 24 Jul '25

24 Jul '25

Changeset: 24c18701 Branch: premain Author: Vladimir Kozlov <vladimir.kozlov(a)oracle.com> Date: 2025-07-24 07:58:36 +0000 URL: https://git.openjdk.org/leyden/commit/24c187011bc04dd1b6a5dd64712681903603b… AOT code caching fixes: - added Xlog info when AOT code caching starts and finishsX - call CDSConfig::disable_dumping_aot_code() before we close AOT code cache: to avoid AOT compilation after cache is closed - small code reordering in AOTCodeAddressTable::init_extrs() - fixed Zero build on Aarch64 (relocation.hpp changes) - adjusted compiler threads count setting for AOT assembly phase - fixed merge mismatch: double call to TrainingData::initialize() ! src/hotspot/share/cds/metaspaceShared.cpp ! src/hotspot/share/code/aotCodeCache.cpp ! src/hotspot/share/code/relocInfo.hpp ! src/hotspot/share/compiler/compilationPolicy.cpp ! src/hotspot/share/runtime/init.cpp

1 0

RFR: 8362657: Make tables used in AOT assembly phase GC-safe
by Ioi Lam 22 Jul '25

22 Jul '25

This fixes crashes when using `-Xlog:aot+map=trace,aot+map+oops=trace:file=aot.oops.txt:none:filesize=0` during the assembly phase. I basically changed a few raw `oop` pointers to `OopHandle`. I also simplified `ArchiveHeapWriter::is_marked_as_native_pointer()`. Note that for `HeapShared::archived_object_cache()`, I calculate the hashcode using the raw oop. This way, we can avoid calling `oopDesc::identity_hash()` for every archived oops. As a result, when we use this table after exiting the CDS safepoint, we must rehash it. See changes in `ArchiveBuilder::CDSMapLogger::log()`. - It might be alright to pre-calculate the identity hash of the archived oops in the assembly phase, but I think we should do that only if we find a reason to do so. We shouldn't do it just because it's convenient for implementing an internal table, ------------- Commit messages: - Fixed white spaces - 8362657: Make tables used in AOT assembly phase GC-safe Changes: https://git.openjdk.org/leyden/pull/86/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=86&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8362657 Stats: 144 lines in 8 files changed: 89 ins; 21 del; 34 mod Patch: https://git.openjdk.org/leyden/pull/86.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/86/head:pull/86 PR: https://git.openjdk.org/leyden/pull/86

4 11

git: openjdk/leyden: premain: 8362657: Make tables used in AOT assembly phase GC-safe
by Ioi Lam 22 Jul '25

22 Jul '25

Changeset: a529aa73 Branch: premain Author: Ioi Lam <iklam(a)openjdk.org> Date: 2025-07-22 17:41:17 +0000 URL: https://git.openjdk.org/leyden/commit/a529aa731d4daf2744d3ea5e94a112030c34c… 8362657: Make tables used in AOT assembly phase GC-safe Reviewed-by: kvn, asmehra, shade ! src/hotspot/share/cds/archiveBuilder.cpp ! src/hotspot/share/cds/archiveHeapWriter.cpp ! src/hotspot/share/cds/archiveHeapWriter.hpp ! src/hotspot/share/cds/cdsHeapVerifier.cpp ! src/hotspot/share/cds/cdsHeapVerifier.hpp ! src/hotspot/share/cds/heapShared.cpp ! src/hotspot/share/cds/heapShared.hpp ! src/hotspot/share/cds/metaspaceShared.cpp

1 0

RFR: 8362558: [Leyden] fatal error: New workflow: should not compile code for unarchived class: UnsupportedBSMs$$Lambda/0...
by Ioi Lam 22 Jul '25

22 Jul '25

We should not AOT-compile indy call sites that cannot be archived in the resolved state. ------------- Commit messages: - 8362558: [Leyden] fatal error: New workflow: should not compile code for unarchived class: UnsupportedBSMs$$Lambda/0... Changes: https://git.openjdk.org/leyden/pull/87/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=87&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8362558 Stats: 11 lines in 2 files changed: 9 ins; 1 del; 1 mod Patch: https://git.openjdk.org/leyden/pull/87.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/87/head:pull/87 PR: https://git.openjdk.org/leyden/pull/87

2 12

git: openjdk/leyden: premain: 8362558: [Leyden] fatal error: New workflow: should not compile code for unarchived class: UnsupportedBSMs$$Lambda/0...
by Ioi Lam 22 Jul '25

22 Jul '25

Changeset: 45288e04 Branch: premain Author: Ioi Lam <iklam(a)openjdk.org> Date: 2025-07-22 03:46:17 +0000 URL: https://git.openjdk.org/leyden/commit/45288e047583bd68ed91cdb604aa302830dec… 8362558: [Leyden] fatal error: New workflow: should not compile code for unarchived class: UnsupportedBSMs$$Lambda/0... Reviewed-by: kvn ! src/hotspot/share/ci/ciEnv.cpp ! src/hotspot/share/classfile/systemDictionaryShared.cpp

1 0

Using CDS and AOT with the Eclipse IDE
by Daniel Schmid 22 Jul '25

22 Jul '25

Hi there, I made a few experiments with CDS and AOT archives on the Eclipse IDE and wanted to share my results here. I ran 4 Eclipse installations (my personal main Eclipse installation is in a comment further down) with AppCDS and -XX:AOTCache and I have written down my results here: https://github.com/eclipse-platform/eclipse.platform/discussions/2060 I think the Eclipse IDE is an interesting application to test as it is a fairly big codebase with many classes loaded at runtime (so it doesn't make it easy for Leyden's improvements) via OSGi. I have used the latest EA build of JDK 26 (26-ea+6-582). Should I use a different JDK build to test it? Essentially my findings (which are relevant to this mailing list) are the following: - There has been an error when creating the AOT archive but it was still created and it was usable. I want to make sure that the people here are aware of that error. To be honest, I did not expect -XX:AOTCache to work that well with Eclipse. - While CDS seems to come with a noticeable improvement for Eclipse startup time, -XX:AOTCache seems to have similar startup times as CDS (not faster). This might be because of Eclipse's class loading shenanigans. - It worked with an agent attached (The Lombok tooling for Eclipse attaches an agent into the IDE, I think that's necessary to inject itself into the Eclipse Java Compiler). The error I got: |[0.074s][error][aot] An error has occurred while processing the AOT configuration file. Run with -Xlog:aot for details. [0.074s][error][aot] archivedBootLayer not available, disabling full module graph AOTCache creation is complete: app.aot 106397696 bytes | I provided the logs here: https://github.com/user-attachments/files/21313886/AOT_logs_26ea.txt (technically not the same run but it also has that error). Yours, Daniel Schmid

2 4

git: openjdk/leyden: premain: 8362518: [Leyden] assert(_ref_cnt > 0) failed: precond when run AOT tests with -Xcomp
by duke 21 Jul '25

21 Jul '25

Changeset: f227d85e Branch: premain Author: Vladimir Kozlov <vladimir.kozlov(a)oracle.com> Date: 2025-07-21 08:35:26 +0000 URL: https://git.openjdk.org/leyden/commit/f227d85e89356d2e17a691538139f003d8517… 8362518: [Leyden] assert(_ref_cnt > 0) failed: precond when run AOT tests with -Xcomp ! src/hotspot/share/asm/codeBuffer.hpp ! src/hotspot/share/code/codeBlob.cpp ! test/hotspot/jtreg/ProblemList-AotJdk.txt

1 0