April 2025 - leyden-dev

git: openjdk/leyden: premain: Add AOTCompileEagerly flag to control compilation after clinit
by duke 25 Apr '25

25 Apr '25

Changeset: b2bb5f25 Branch: premain Author: Igor Veresov <igor.veresov(a)oracle.com> Date: 2025-04-25 08:43:31 +0000 URL: https://git.openjdk.org/leyden/commit/b2bb5f2567e48dca57aee68e9d4929149680d… Add AOTCompileEagerly flag to control compilation after clinit ! src/hotspot/share/cds/cds_globals.hpp ! src/hotspot/share/compiler/compilationPolicy.cpp

1 0

segfault with discrepancies in FlightRecorder options in training run and AOT cache creation
by Olivier Bourgain 25 Apr '25

25 Apr '25

Hello, I stumbled upon another crash. When: * running a Spring Boot app with https://github.com/openjdk/leyden/tree/premain * FlightRecorder was enabled for the aotconf creation * FlightRecorder was disabled for the aotcache creation * FlightRecorder is enabled for the production run Then the app crashes with a segmentation fault. I wrote a reproducer https://github.com/obourgain/leyden-crash/tree/crash_jfr (notice the branch crash_jfr). Also, -XX:StartFlightRecording option for traning and production run are not the same, it prevents the use of the AOT cache. Regards.

1 0

git: openjdk/leyden: premain: 8355352: [premain] rename AOT Code classes and logging tags
by Vladimir Kozlov 24 Apr '25

24 Apr '25

Changeset: 5e57f1b0 Branch: premain Author: Vladimir Kozlov <kvn(a)openjdk.org> Date: 2025-04-24 16:19:41 +0000 URL: https://git.openjdk.org/leyden/commit/5e57f1b04b78edd89f69a35e770ee2a11e4f7… 8355352: [premain] rename AOT Code classes and logging tags Reviewed-by: shade, asmehra ! make/hotspot/lib/JvmFeatures.gmk ! src/hotspot/cpu/aarch64/c1_LIRAssembler_aarch64.cpp ! src/hotspot/cpu/aarch64/gc/g1/g1BarrierSetAssembler_aarch64.cpp ! src/hotspot/cpu/aarch64/macroAssembler_aarch64.cpp ! src/hotspot/cpu/aarch64/macroAssembler_aarch64.hpp ! src/hotspot/cpu/aarch64/stubGenerator_aarch64.cpp ! src/hotspot/cpu/x86/c1_LIRAssembler_x86.cpp ! src/hotspot/cpu/x86/gc/g1/g1BarrierSetAssembler_x86.cpp ! src/hotspot/cpu/x86/gc/shared/cardTableBarrierSetAssembler_x86.cpp ! src/hotspot/cpu/x86/macroAssembler_x86.cpp ! src/hotspot/cpu/x86/runtime_x86_64.cpp ! src/hotspot/cpu/x86/stubGenerator_x86_64.cpp ! src/hotspot/cpu/x86/x86_64.ad ! src/hotspot/share/adlc/main.cpp ! src/hotspot/share/asm/codeBuffer.hpp ! src/hotspot/share/c1/c1_Compiler.cpp ! src/hotspot/share/c1/c1_LIRGenerator.cpp ! src/hotspot/share/cds/aotLinkedClassBulkLoader.hpp ! src/hotspot/share/cds/archiveBuilder.cpp ! src/hotspot/share/cds/cdsConfig.cpp ! src/hotspot/share/cds/metaspaceShared.cpp ! src/hotspot/share/ci/ciEnv.cpp ! src/hotspot/share/ci/ciEnv.hpp ! src/hotspot/share/ci/ciField.cpp ! src/hotspot/share/ci/ciMethod.cpp ! src/hotspot/share/ci/ciObject.cpp - src/hotspot/share/code/SCCache.cpp - src/hotspot/share/code/SCCache.hpp + src/hotspot/share/code/aotCodeCache.cpp + src/hotspot/share/code/aotCodeCache.hpp ! src/hotspot/share/code/codeBlob.cpp ! src/hotspot/share/code/codeBlob.hpp ! src/hotspot/share/code/codeCache.cpp ! src/hotspot/share/code/exceptionHandlerTable.hpp ! src/hotspot/share/code/nmethod.cpp ! src/hotspot/share/code/nmethod.hpp ! src/hotspot/share/code/relocInfo.cpp ! src/hotspot/share/code/relocInfo.hpp ! src/hotspot/share/compiler/compilationPolicy.cpp ! src/hotspot/share/compiler/compilationPolicy.hpp ! src/hotspot/share/compiler/compileBroker.cpp ! src/hotspot/share/compiler/compileBroker.hpp ! src/hotspot/share/compiler/compileTask.cpp ! src/hotspot/share/compiler/compileTask.hpp ! src/hotspot/share/compiler/oopMap.hpp ! src/hotspot/share/compiler/precompiler.cpp ! src/hotspot/share/compiler/recompilationPolicy.cpp ! src/hotspot/share/gc/g1/c1/g1BarrierSetC1.cpp ! src/hotspot/share/gc/g1/c2/g1BarrierSetC2.cpp ! src/hotspot/share/gc/shenandoah/c1/shenandoahBarrierSetC1.hpp ! src/hotspot/share/gc/z/c1/zBarrierSetC1.hpp ! src/hotspot/share/jvmci/jvmciRuntime.cpp ! src/hotspot/share/logging/logTag.hpp ! src/hotspot/share/oops/method.cpp ! src/hotspot/share/oops/method.hpp ! src/hotspot/share/opto/c2compiler.cpp ! src/hotspot/share/opto/graphKit.cpp ! src/hotspot/share/opto/output.cpp ! src/hotspot/share/opto/output.hpp ! src/hotspot/share/opto/rootnode.cpp ! src/hotspot/share/opto/runtime.hpp ! src/hotspot/share/runtime/arguments.cpp ! src/hotspot/share/runtime/deoptimization.cpp ! src/hotspot/share/runtime/init.cpp ! src/hotspot/share/runtime/java.cpp ! src/hotspot/share/runtime/sharedRuntime.cpp ! src/hotspot/share/runtime/stubRoutines.hpp ! src/hotspot/share/runtime/threads.cpp ! test/hotspot/jtreg/premain/javac/javac-test.sh ! test/hotspot/jtreg/premain/javac_new_workflow/run.sh ! test/hotspot/jtreg/premain/jmh/run.sh ! test/hotspot/jtreg/premain/lib/bench-lib-new-workflow.sh ! test/hotspot/jtreg/premain/lib/bench-lib.sh ! test/hotspot/jtreg/premain/lib/premain-run.sh ! test/hotspot/jtreg/premain/spring-petclinic/WarmupMakefile ! test/hotspot/jtreg/premain/spring-petclinic/bench.sh ! test/hotspot/jtreg/runtime/cds/appcds/AOTFlags.java ! test/hotspot/jtreg/runtime/cds/appcds/applications/HelidonQuickStartSE.java ! test/hotspot/jtreg/runtime/cds/appcds/applications/MicronautFirstApp.java ! test/hotspot/jtreg/runtime/cds/appcds/applications/QuarkusGettingStarted.java ! test/hotspot/jtreg/runtime/cds/appcds/applications/SpringPetClinic.java ! test/hotspot/jtreg/runtime/cds/appcds/methodHandles/JDKMethodHandlesTestRunner.java

1 0

RFR: 8355352: [premain] rename AOT Code classes and logging tags
by Vladimir Kozlov 24 Apr '25

24 Apr '25

Rename AOT Code classes and logging tags to match changes for mainline: https://github.com/openjdk/jdk/pull/24740 I also searched and renamed `sc_` to `ac_` and `scc_` to `aot_code_` patterns, also `SC`, `SCA`. I tried to build without CDS locally and found few places we missed `NOT_CDS_RETURN` macro or `#if INCLUDE_CDS` check. This is first part of changes in premain branch. AOT code flags renaming and more code changes will be done in separate PRs. Tested with premain-tier1 ------------- Commit messages: - 8355352: [premain] rename AOT Code classes and logging tags Changes: https://git.openjdk.org/leyden/pull/62/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=62&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8355352 Stats: 11536 lines in 82 files changed: 5574 ins; 5573 del; 389 mod Patch: https://git.openjdk.org/leyden/pull/62.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/62/head:pull/62 PR: https://git.openjdk.org/leyden/pull/62

3 4

Mainline PR for Soft/Weak Reference Support
by ioi.lam＠oracle.com 24 Apr '25

24 Apr '25

Folks, I am upstreaming the Soft/Weak Reference Support from leyden to mainline https://github.com/openjdk/jdk/pull/24757 The implementation of References in the JDK is complex. In this PR, we added a new "prepareForAOTCache()" pattern to simplify the kind of References that can be cached. Please see the PR for detail. Any comments/suggestions about this new pattern will be greatly appreciated. - Ioi

1 0

RFR: 8355352: [premain] rename AOT Code classes and logging tags
by Vladimir Kozlov 23 Apr '25

23 Apr '25

Rename AOT Code classes and logging tags to match changes for mainline: [#24740](https://github.com/openjdk/jdk/pull/24740) This is first part of changes in premain branch. AOT code flags renaming and more code changes will be done in separate PRs. Tested with premain-tier1 ------------- Commit messages: - 8355352: [premain] rename AOT Code classes and logging tags Changes: https://git.openjdk.org/leyden/pull/60/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=60&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8355352 Stats: 11458 lines in 79 files changed: 5562 ins; 5571 del; 325 mod Patch: https://git.openjdk.org/leyden/pull/60.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/60/head:pull/60 PR: https://git.openjdk.org/leyden/pull/60

2 9

RFR: 8355334: [leyden] Missing type profile info in archived training data
by Vladimir Ivanov 23 Apr '25

23 Apr '25

There's a regression from #56 which causes pruning of all type profiles from archived training data. It severely affects quality of compiled code. Proposed fix removes only excluded classes. Testing: hs-tier1 - hs-tier2, JavacBench ------------- Commit messages: - current_thread_is_vm_or_dumper - handle shared classes during dynamic archive dumping - is_excluded - 8355334: [leyden] Missing type profile info in archived training data Changes: https://git.openjdk.org/leyden/pull/59/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=59&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8355334 Stats: 34 lines in 2 files changed: 22 ins; 7 del; 5 mod Patch: https://git.openjdk.org/leyden/pull/59.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/59/head:pull/59 PR: https://git.openjdk.org/leyden/pull/59

4 8

git: openjdk/leyden: premain: 8355334: [leyden] Missing type profile info in archived training data
by Vladimir Ivanov 23 Apr '25

23 Apr '25

Changeset: 08822752 Branch: premain Author: Vladimir Ivanov <vlivanov(a)openjdk.org> Date: 2025-04-23 18:09:37 +0000 URL: https://git.openjdk.org/leyden/commit/08822752b408fa8b2e068bea490d7537f1f9a… 8355334: [leyden] Missing type profile info in archived training data Reviewed-by: iveresov, shade, iklam ! src/hotspot/share/cds/archiveBuilder.cpp ! src/hotspot/share/oops/methodData.cpp

1 0

git: openjdk/leyden: premain: 8355296: [leyden] Some methods are stuck at level=0 with -XX:-TieredCompilation
by Aleksey Shipilev 23 Apr '25

23 Apr '25

Changeset: 98f45324 Branch: premain Author: Aleksey Shipilev <shade(a)openjdk.org> Date: 2025-04-23 17:26:26 +0000 URL: https://git.openjdk.org/leyden/commit/98f4532416e1b2218d734aded01730f3a8c41… 8355296: [leyden] Some methods are stuck at level=0 with -XX:-TieredCompilation Reviewed-by: iveresov ! src/hotspot/share/compiler/compilationPolicy.cpp

1 0

RFR: 8355296: [leyden] Some methods are stuck at level=0 with -XX:-TieredCompilation
by Aleksey Shipilev 23 Apr '25

23 Apr '25

Discovered this originally when testing Persistent Profiles PR in Leyden. In short, the transition from level=0 would not go to level=4 ever for some methods, if we disable intermediate levels. I believe this is one the causes for at least some performance artifacts I seen in Leyden testing last week. There are two bugs: 1. In `CompilationPolicy::common`: Trivial/native method shortcut was lost, so trivial methods enter through the normal profiled policy (without a good reason?) 2. In `CompilationPolicy::should_create_mdo`: When training data is used we _only_ create MDOs for methods that are referenced in TD. This has a major knock-on effect on _normal_ JIT compilations that want to see MDO invocation/backedge counters. Instead, MDO is missing, so interpreter only increments MCS counters, and then compilation policy misses the transitions. (1) is really subsumed by (2), so we can really only do (2). See reproducers and logs in the issue. More performance data in comments. Additional testing: - [x] Linux x86_64 server fastdebug, `runtime/cds` ------------- Commit messages: - Fix Changes: https://git.openjdk.org/leyden/pull/57/files Webrev: https://webrevs.openjdk.org/?repo=leyden&pr=57&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8355296 Stats: 32 lines in 1 file changed: 10 ins; 14 del; 8 mod Patch: https://git.openjdk.org/leyden/pull/57.diff Fetch: git fetch https://git.openjdk.org/leyden.git pull/57/head:pull/57 PR: https://git.openjdk.org/leyden/pull/57

2 6