From rkennke at openjdk.org Thu Aug 10 22:14:00 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 10 Aug 2023 22:14:00 GMT Subject: [master] RFR: 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP registers Message-ID: The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. Testing: - [x] tier1 (x86_64, aarch64) - [x] tier2 (x86_64, aarch64) ------------- Commit messages: - 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register Changes: https://git.openjdk.org/lilliput/pull/102/files Webrev: https://webrevs.openjdk.org/?repo=lilliput&pr=102&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314130 Stats: 2 lines in 2 files changed: 0 ins; 0 del; 2 mod Patch: https://git.openjdk.org/lilliput/pull/102.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/102/head:pull/102 PR: https://git.openjdk.org/lilliput/pull/102 From rkennke at openjdk.org Thu Aug 10 22:43:28 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 10 Aug 2023 22:43:28 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8314131: [Lilliput/JDK17] Make loadNKlassCompactHeaders not use a TEMP register Message-ID: The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. Testing: - [x] tier1 +UseCompactObjectHeaders (aarch64, x86_64) - [x] tier2 +UseCompactObjectHeaders (aarch64, x86_64) ------------- Commit messages: - 8314131: [Lilliput/JDK17] Make loadNKlassCompactHeaders not use a TEMP register Changes: https://git.openjdk.org/lilliput-jdk17u/pull/55/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk17u&pr=55&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314131 Stats: 2 lines in 2 files changed: 0 ins; 0 del; 2 mod Patch: https://git.openjdk.org/lilliput-jdk17u/pull/55.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk17u.git pull/55/head:pull/55 PR: https://git.openjdk.org/lilliput-jdk17u/pull/55 From shade at openjdk.org Fri Aug 11 09:10:30 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Fri, 11 Aug 2023 09:10:30 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8314131: [Lilliput/JDK17] Make loadNKlassCompactHeaders not use a TEMP register In-Reply-To: References: Message-ID: On Thu, 10 Aug 2023 22:26:33 GMT, Roman Kennke wrote: > The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. > > Testing: > - [x] tier1 +UseCompactObjectHeaders (aarch64, x86_64) > - [x] tier2 +UseCompactObjectHeaders (aarch64, x86_64) This makes sense, like https://github.com/openjdk/lilliput/pull/102. GHAs are currently broken, because 17u GHA are broken. ------------- Marked as reviewed by shade (Reviewer). PR Review: https://git.openjdk.org/lilliput-jdk17u/pull/55#pullrequestreview-1573280327 From shade at openjdk.org Fri Aug 11 09:11:28 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Fri, 11 Aug 2023 09:11:28 GMT Subject: [master] RFR: 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register In-Reply-To: References: Message-ID: On Thu, 10 Aug 2023 21:57:14 GMT, Roman Kennke wrote: > The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. > > Testing: > - [x] tier1 +UseCompactObjectHeaders (x86_64, aarch64) > - [x] tier2 +UseCompactObjectHeaders (x86_64, aarch64) Okay, this makes sense. GHA are broken, because we have not picked up stuff from upstream recently. ------------- Marked as reviewed by shade (Committer). PR Review: https://git.openjdk.org/lilliput/pull/102#pullrequestreview-1573278708 From rkennke at openjdk.org Fri Aug 11 16:08:58 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Fri, 11 Aug 2023 16:08:58 GMT Subject: [lilliput-jdk17u:lilliput] Integrated: 8314131: [Lilliput/JDK17] Make loadNKlassCompactHeaders not use a TEMP register In-Reply-To: References: Message-ID: On Thu, 10 Aug 2023 22:26:33 GMT, Roman Kennke wrote: > The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. > > Testing: > - [x] tier1 +UseCompactObjectHeaders (aarch64, x86_64) > - [x] tier2 +UseCompactObjectHeaders (aarch64, x86_64) This pull request has now been integrated. Changeset: 93690bc8 Author: Roman Kennke URL: https://git.openjdk.org/lilliput-jdk17u/commit/93690bc86c5e9c3437b382e5779d729957b7ecc9 Stats: 2 lines in 2 files changed: 0 ins; 0 del; 2 mod 8314131: [Lilliput/JDK17] Make loadNKlassCompactHeaders not use a TEMP register Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput-jdk17u/pull/55 From rkennke at openjdk.org Fri Aug 11 16:09:28 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Fri, 11 Aug 2023 16:09:28 GMT Subject: [master] Integrated: 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register In-Reply-To: References: Message-ID: On Thu, 10 Aug 2023 21:57:14 GMT, Roman Kennke wrote: > The loadNKlassCompactHeaders template in the .ad files currently declares dst as TEMP_DEF. This has historic reasons: We needed to ensure that dst doesn't overlap with src, because we needed src for calling into the slow-path. In addition to that, there seems to be a related bug in C2's matcher which sometimes seems to create an oopmap entry for a temp reg, which would make GCs crash if they scan uninitialized slot (see [JDK-8051805](https://bugs.openjdk.org/browse/JDK-8051805)). That problem should also go away with this change. In addition to that, relaxing the type of dst might also lead to performance enhancements when register pressure is high, because we don't need two registers for loadNKlass - 1 register would be enough. > > Testing: > - [x] tier1 +UseCompactObjectHeaders (x86_64, aarch64) > - [x] tier2 +UseCompactObjectHeaders (x86_64, aarch64) This pull request has now been integrated. Changeset: db440d15 Author: Roman Kennke URL: https://git.openjdk.org/lilliput/commit/db440d154cb39bda84fbc8c129d5d9e57a243cd7 Stats: 2 lines in 2 files changed: 0 ins; 0 del; 2 mod 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput/pull/102 From rkennke at openjdk.org Fri Aug 11 16:07:28 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Fri, 11 Aug 2023 16:07:28 GMT Subject: [master] RFR: 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register In-Reply-To: References: Message-ID: On Fri, 11 Aug 2023 09:00:46 GMT, Aleksey Shipilev wrote: > Okay, this makes sense. > > GHA are broken, because we have not picked up stuff from upstream recently. Yeah, @tstuefe is working on merging newest JDK. ------------- PR Comment: https://git.openjdk.org/lilliput/pull/102#issuecomment-1675010587 From rkennke at openjdk.org Mon Aug 14 16:25:59 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Mon, 14 Aug 2023 16:25:59 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking [v4] In-Reply-To: References: Message-ID: <0jqbpOjKMZj3CE7dGLJlT97vzkwUchwO8c48aax2uhA=.231ef156-f42e-44a2-985b-d7abf146d7e8@github.com> > This ports the original https://github.com/openjdk/jdk/pull/7964 to Lilliput/JDK17. It also includes the follow-up (cosmetic) fixes https://bugs.openjdk.org/browse/JDK-8284687 and https://bugs.openjdk.org/browse/JDK-8284725. > > Testing: > - [x] tier1 > - [x] serviceability/jvmti > - [x] vmTestbase/nsk/jvmti Roman Kennke has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains seven commits: - Merge branch 'lilliput' into JDK-8308411 - Merge branch 'lilliput' into JDK-8308411 - Merge branch 'lilliput' into JDK-8308411 - Some follow-ups - 8283710: JVMTI: Use BitSet for object marking Reviewed-by: stuefe, coleenp - 8283710: JVMTI: Use BitSet for object marking Reviewed-by: stuefe, coleenp - Revert ------------- Changes: https://git.openjdk.org/lilliput-jdk17u/pull/21/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk17u&pr=21&range=03 Stats: 847 lines in 13 files changed: 392 ins; 434 del; 21 mod Patch: https://git.openjdk.org/lilliput-jdk17u/pull/21.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk17u.git pull/21/head:pull/21 PR: https://git.openjdk.org/lilliput-jdk17u/pull/21 From rkennke at openjdk.org Mon Aug 14 16:56:30 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Mon, 14 Aug 2023 16:56:30 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking [v5] In-Reply-To: References: Message-ID: > This ports the original https://github.com/openjdk/jdk/pull/7964 to Lilliput/JDK17. It also includes the follow-up (cosmetic) fixes https://bugs.openjdk.org/browse/JDK-8284687 and https://bugs.openjdk.org/browse/JDK-8284725. > > Testing: > - [x] tier1 > - [x] com/sun/jdi > - [x] serviceability/jvmti > - [x] vmTestbase/nsk/jdi > - [ ] vmTestbase/nsk/jdwp > - [x] vmTestbase/nsk/jvmti Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: Revert JVMTI changes and ProblemList ------------- Changes: - all: https://git.openjdk.org/lilliput-jdk17u/pull/21/files - new: https://git.openjdk.org/lilliput-jdk17u/pull/21/files/3f81c16c..d4f327f6 Webrevs: - full: https://webrevs.openjdk.org/?repo=lilliput-jdk17u&pr=21&range=04 - incr: https://webrevs.openjdk.org/?repo=lilliput-jdk17u&pr=21&range=03-04 Stats: 95 lines in 2 files changed: 0 ins; 95 del; 0 mod Patch: https://git.openjdk.org/lilliput-jdk17u/pull/21.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk17u.git pull/21/head:pull/21 PR: https://git.openjdk.org/lilliput-jdk17u/pull/21 From shade at openjdk.org Tue Aug 15 10:32:35 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Tue, 15 Aug 2023 10:32:35 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking [v5] In-Reply-To: References: Message-ID: <-8RNaeGDaV6lj5VNJVjiOgQGi5GKTRatkicFgZgQaPo=.61980a84-b172-4dbe-bf97-5b0bef70fda1@github.com> On Mon, 14 Aug 2023 16:56:30 GMT, Roman Kennke wrote: >> This ports the original https://github.com/openjdk/jdk/pull/7964 to Lilliput/JDK17. It also includes the follow-up (cosmetic) fixes https://bugs.openjdk.org/browse/JDK-8284687 and https://bugs.openjdk.org/browse/JDK-8284725. >> >> Testing: >> - [x] tier1 >> - [x] com/sun/jdi >> - [x] serviceability/jvmti >> - [x] vmTestbase/nsk/jdi >> - [x] vmTestbase/nsk/jdwp >> - [x] vmTestbase/nsk/jvmti > > Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: > > Revert JVMTI changes and ProblemList Looks okay! ------------- Marked as reviewed by shade (Reviewer). PR Review: https://git.openjdk.org/lilliput-jdk17u/pull/21#pullrequestreview-1578317100 From rkennke at openjdk.org Tue Aug 15 14:25:38 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Tue, 15 Aug 2023 14:25:38 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking [v5] In-Reply-To: References: Message-ID: On Mon, 14 Aug 2023 16:56:30 GMT, Roman Kennke wrote: >> This ports the original https://github.com/openjdk/jdk/pull/7964 to Lilliput/JDK17. It also includes the follow-up (cosmetic) fixes https://bugs.openjdk.org/browse/JDK-8284687 and https://bugs.openjdk.org/browse/JDK-8284725. >> >> Testing: >> - [x] tier1 >> - [x] com/sun/jdi >> - [x] serviceability/jvmti >> - [x] vmTestbase/nsk/jdi >> - [x] vmTestbase/nsk/jdwp >> - [x] vmTestbase/nsk/jvmti > > Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: > > Revert JVMTI changes and ProblemList Thanks! ------------- PR Comment: https://git.openjdk.org/lilliput-jdk17u/pull/21#issuecomment-1679017518 From rkennke at openjdk.org Tue Aug 15 14:25:40 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Tue, 15 Aug 2023 14:25:40 GMT Subject: [lilliput-jdk17u:lilliput] Integrated: 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking In-Reply-To: References: Message-ID: On Fri, 19 May 2023 13:13:57 GMT, Roman Kennke wrote: > This ports the original https://github.com/openjdk/jdk/pull/7964 to Lilliput/JDK17. It also includes the follow-up (cosmetic) fixes https://bugs.openjdk.org/browse/JDK-8284687 and https://bugs.openjdk.org/browse/JDK-8284725. > > Testing: > - [x] tier1 > - [x] com/sun/jdi > - [x] serviceability/jvmti > - [x] vmTestbase/nsk/jdi > - [x] vmTestbase/nsk/jdwp > - [x] vmTestbase/nsk/jvmti This pull request has now been integrated. Changeset: 9f3915d2 Author: Roman Kennke URL: https://git.openjdk.org/lilliput-jdk17u/commit/9f3915d2a4a0c48bdf31be8fb2d2fbd2bfd8dcfb Stats: 942 lines in 14 files changed: 392 ins; 529 del; 21 mod 8308411: [Lilliput/JDK17] Cherry-pick: 8283710: JVMTI: Use BitSet for object marking Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput-jdk17u/pull/21 From rkennke at openjdk.org Thu Aug 17 01:15:30 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 17 Aug 2023 01:15:30 GMT Subject: [lilliput-jdk21u:lilliput] RFR: 8314485: [Lilliput/JDK21] Cherry-pick: 8139457: Array bases are aligned at HeapWord granularity Message-ID: Let's cherry-pick the current state of the upstream PR https://github.com/openjdk/jdk/pull/11044 into Lilliput/JDK21 in preparation for the other Lilliput-related changes. The original change applies cleanly on top of lilliput/jdk21u. Testing: - [ ] tier1 -UCCP - [ ] tier1 +UCCP ------------- Commit messages: - 8314485: [Lilliput/JDK21] Cherry-pick: 8139457: Array bases are aligned at HeapWord granularity Changes: https://git.openjdk.org/lilliput-jdk21u/pull/1/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk21u&pr=1&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314485 Stats: 246 lines in 21 files changed: 134 ins; 59 del; 53 mod Patch: https://git.openjdk.org/lilliput-jdk21u/pull/1.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk21u.git pull/1/head:pull/1 PR: https://git.openjdk.org/lilliput-jdk21u/pull/1 From shade at openjdk.org Thu Aug 17 10:44:05 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Thu, 17 Aug 2023 10:44:05 GMT Subject: [lilliput-jdk21u:lilliput] RFR: 8314485: [Lilliput/JDK21] Cherry-pick: 8139457: Array bases are aligned at HeapWord granularity In-Reply-To: References: Message-ID: On Thu, 17 Aug 2023 01:09:07 GMT, Roman Kennke wrote: > Let's cherry-pick the current state of the upstream PR https://github.com/openjdk/jdk/pull/11044 into Lilliput/JDK21 in preparation for the other Lilliput-related changes. The original change applies cleanly on top of lilliput/jdk21u. > > Testing: > - [ ] tier1 -UCCP > - [ ] tier1 +UCCP Marked as reviewed by shade (Reviewer). ------------- PR Review: https://git.openjdk.org/lilliput-jdk21u/pull/1#pullrequestreview-1582337500 From stuefe at openjdk.org Thu Aug 17 11:24:46 2023 From: stuefe at openjdk.org (Thomas Stuefe) Date: Thu, 17 Aug 2023 11:24:46 GMT Subject: [master] RFR: Merge jdk Message-ID: This merges the Lilliput Repo to whats currently on JDK head, up to and including: commit e1fdef56135c2987b128884ef632b64c32dd674a (jdk-upstream/master) Author: Tom Rodriguez Date: Wed Aug 16 06:06:59 2023 +0000 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures It contains a preparatory pre-merge commit that rolls back the first prototype of 22bit nKlass IDs: a1c5f766276fe0c6782e2bf3aaaa41e0637223ad After that, merge-induced cleanups and fixes were minimal. ------------- Commit messages: - fix sa tests - Merge remote-tracking branch 'jdk-upstream/master' into merge-jdk-head - 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures - 4346610: Adding JSeparator to JToolBar "pushes" buttons added after separator to edge - 8314163: os::print_hex_dump prints incorrectly for big endian platforms and unit sizes larger than 1 - 8314242: Update applications/scimark/Scimark.java to accept VM flags - 8311591: Add SystemModulesPlugin test case that splits module descriptors with new local variables defined by DedupSetBuilder - 8309335: Get rid of use of reflection to call Thread.isVirtual() in nsk/jdi/EventRequestManager/stepRequests/stepreq001t.java - 8314166: Update googletest to v1.14.0 - 8310134: NMT: thread count in Thread section of VM.native_memory output confusing with virtual threads - ... and 686 more: https://git.openjdk.org/lilliput/compare/db440d15...7461340a The webrevs contain the adjustments done while merging with regards to each parent branch: - master: https://webrevs.openjdk.org/?repo=lilliput&pr=104&range=00.0 - jdk: https://webrevs.openjdk.org/?repo=lilliput&pr=104&range=00.1 Changes: https://git.openjdk.org/lilliput/pull/104/files Stats: 160223 lines in 3359 files changed: 62206 ins; 77179 del; 20838 mod Patch: https://git.openjdk.org/lilliput/pull/104.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/104/head:pull/104 PR: https://git.openjdk.org/lilliput/pull/104 From rkennke at openjdk.org Thu Aug 17 11:24:47 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 17 Aug 2023 11:24:47 GMT Subject: [master] RFR: Merge jdk In-Reply-To: References: Message-ID: On Tue, 15 Aug 2023 08:14:16 GMT, Thomas Stuefe wrote: > This merges the Lilliput Repo to whats currently on JDK head, up to and including: > > > commit e1fdef56135c2987b128884ef632b64c32dd674a (jdk-upstream/master) > Author: Tom Rodriguez > Date: Wed Aug 16 06:06:59 2023 +0000 > > 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures > > > > It contains a preparatory pre-merge commit that rolls back the first prototype of 22bit nKlass IDs: a1c5f766276fe0c6782e2bf3aaaa41e0637223ad > > After that, merge-induced cleanups and fixes were minimal. I believe you need to change the title to just 'Merge jdk' if you merge from jdk head, or 'Merge jdk:$TAG' if you want to merge specific tag (See e.g. #94 or #97). Then please go ahead and integrate! ------------- PR Comment: https://git.openjdk.org/lilliput/pull/104#issuecomment-1678998629 From stuefe at openjdk.org Thu Aug 17 11:24:47 2023 From: stuefe at openjdk.org (Thomas Stuefe) Date: Thu, 17 Aug 2023 11:24:47 GMT Subject: [master] RFR: Merge jdk In-Reply-To: References: Message-ID: On Tue, 15 Aug 2023 14:09:52 GMT, Roman Kennke wrote: > I believe you need to change the title to just 'Merge jdk' if you merge from jdk head, or 'Merge jdk:$TAG' if you want to merge specific tag (See e.g. #94 or #97). Then please go ahead and integrate! Okay, thanks. Some tests fail though. All the new tests added for GenZ, hehe, they deadlock. I first thought this had to do with 8311981, but it still happens with Davids fix. Should I integrate nevertheless? ------------- PR Comment: https://git.openjdk.org/lilliput/pull/104#issuecomment-1679170107 From shade at openjdk.org Thu Aug 17 11:24:48 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Thu, 17 Aug 2023 11:24:48 GMT Subject: [master] RFR: Merge jdk In-Reply-To: References: Message-ID: <9Ng8Jxk_gxTFtNV1Scnn2C_2amoZt0MtJ9uvhIhA2K8=.e1927bb0-5ce0-4459-86d2-003ef8d276f8@github.com> On Tue, 15 Aug 2023 08:14:16 GMT, Thomas Stuefe wrote: > This merges the Lilliput Repo to whats currently on JDK head, up to and including: > > > commit e1fdef56135c2987b128884ef632b64c32dd674a (jdk-upstream/master) > Author: Tom Rodriguez > Date: Wed Aug 16 06:06:59 2023 +0000 > > 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures > > > > It contains a preparatory pre-merge commit that rolls back the first prototype of 22bit nKlass IDs: a1c5f766276fe0c6782e2bf3aaaa41e0637223ad > > After that, merge-induced cleanups and fixes were minimal. gc/stringdedup/ timeouts with GenZ look like: https://bugs.openjdk.org/browse/JDK-8311981 I wonder if every other failure is actually the manifestation of that one. Can you pick up another merge from master easily? If not, I would be okay with pushing the merge as is, and then dealing with the test failures separately, possibly with another merge. ------------- PR Comment: https://git.openjdk.org/lilliput/pull/104#issuecomment-1681803787 From stuefe at openjdk.org Thu Aug 17 11:24:48 2023 From: stuefe at openjdk.org (Thomas Stuefe) Date: Thu, 17 Aug 2023 11:24:48 GMT Subject: [master] RFR: Merge jdk In-Reply-To: <9Ng8Jxk_gxTFtNV1Scnn2C_2amoZt0MtJ9uvhIhA2K8=.e1927bb0-5ce0-4459-86d2-003ef8d276f8@github.com> References: <9Ng8Jxk_gxTFtNV1Scnn2C_2amoZt0MtJ9uvhIhA2K8=.e1927bb0-5ce0-4459-86d2-003ef8d276f8@github.com> Message-ID: On Thu, 17 Aug 2023 07:47:28 GMT, Aleksey Shipilev wrote: > gc/stringdedup/ timeouts with GenZ look like: https://bugs.openjdk.org/browse/JDK-8311981 > > I wonder if every other failure is actually the manifestation of that one. Can you pick up another merge from master easily? This is already with Davids fix. It only fails on Lilliput, not in head. I only see timeouts with compiler enabled (both c1 and c2). No errors with Xint. But on both x64 and arm. I tested with LockingMode 0 (HeavyMonitors) and 1 (default), all the same. I debugged the app, but could see no obvious deadlock. All relevant threads were waiting on unowned monitors. > If not, I would be okay with pushing the merge as is, and then dealing with the test failures separately, possibly with another merge. Okay, will do that. ------------- PR Comment: https://git.openjdk.org/lilliput/pull/104#issuecomment-1682105101 From rkennke at openjdk.org Thu Aug 17 11:24:48 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 17 Aug 2023 11:24:48 GMT Subject: [master] RFR: Merge jdk In-Reply-To: References: Message-ID: On Tue, 15 Aug 2023 15:46:30 GMT, Thomas Stuefe wrote: > > I believe you need to change the title to just 'Merge jdk' if you merge from jdk head, or 'Merge jdk:$TAG' if you want to merge specific tag (See e.g. #94 or #97). Then please go ahead and integrate! > > Okay, thanks. > > Some tests fail though. All the new tests added for GenZ, hehe, they deadlock. I first thought this had to do with 8311981, but it still happens with Davids fix. > > Should I integrate nevertheless? Are the failures only in Lilliput repo, or do they also deadlock in upstream? ------------- PR Comment: https://git.openjdk.org/lilliput/pull/104#issuecomment-1679176136 From stuefe at openjdk.org Thu Aug 17 19:08:00 2023 From: stuefe at openjdk.org (Thomas Stuefe) Date: Thu, 17 Aug 2023 19:08:00 GMT Subject: [master] RFR: Merge jdk [v2] In-Reply-To: References: Message-ID: <-JIrsTUxfkxR1fKVjezPLQCp3nrTt9-9phFOYW4fnQo=.1fb10c9f-00a4-4ba8-a138-435d964dd9d3@github.com> > This merges the Lilliput Repo to whats currently on JDK head, up to and including: > > > commit e1fdef56135c2987b128884ef632b64c32dd674a (jdk-upstream/master) > Author: Tom Rodriguez > Date: Wed Aug 16 06:06:59 2023 +0000 > > 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures > > > > It contains a preparatory pre-merge commit that rolls back the first prototype of 22bit nKlass IDs: a1c5f766276fe0c6782e2bf3aaaa41e0637223ad > > After that, merge-induced cleanups and fixes were minimal. Thomas Stuefe has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 109 commits: - fix sa tests - Merge remote-tracking branch 'jdk-upstream/master' into merge-jdk-head - remove remnant header from first nKlass prototype - Post-Merge fixes - Merge JDK - [Lilliput] Rollback 22-bit KlassPointers - 8314130: [Lilliput] Make loadNKlassCompactHeaders not use a TEMP register Reviewed-by: shade - 8311875: [Lilliput] Disallow accessing oop metadata vmStructs with +UCOH Reviewed-by: rkennke - 8310010: [Lilliput] SA: Fix oop array element alignment Reviewed-by: shade - 8310528: [Lilliput] Provide infrastructure for Lilliput-specific ProblemList Reviewed-by: rkennke - ... and 99 more: https://git.openjdk.org/lilliput/compare/e1fdef56...7461340a ------------- Changes: https://git.openjdk.org/lilliput/pull/104/files Webrev: https://webrevs.openjdk.org/?repo=lilliput&pr=104&range=01 Stats: 3180 lines in 172 files changed: 2471 ins; 365 del; 344 mod Patch: https://git.openjdk.org/lilliput/pull/104.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/104/head:pull/104 PR: https://git.openjdk.org/lilliput/pull/104 From stuefe at openjdk.org Thu Aug 17 19:08:03 2023 From: stuefe at openjdk.org (Thomas Stuefe) Date: Thu, 17 Aug 2023 19:08:03 GMT Subject: [master] Integrated: Merge jdk In-Reply-To: References: Message-ID: On Tue, 15 Aug 2023 08:14:16 GMT, Thomas Stuefe wrote: > This merges the Lilliput Repo to whats currently on JDK head, up to and including: > > > commit e1fdef56135c2987b128884ef632b64c32dd674a (jdk-upstream/master) > Author: Tom Rodriguez > Date: Wed Aug 16 06:06:59 2023 +0000 > > 8314324: "8311557: [JVMCI] deadlock with JVMTI thread suspension" causes various failures > > > > It contains a preparatory pre-merge commit that rolls back the first prototype of 22bit nKlass IDs: a1c5f766276fe0c6782e2bf3aaaa41e0637223ad > > After that, merge-induced cleanups and fixes were minimal. This pull request has now been integrated. Changeset: d85adb60 Author: Thomas Stuefe URL: https://git.openjdk.org/lilliput/commit/d85adb60eb20c275dd86b5f113ad1e1b5949f2c4 Stats: 160223 lines in 3359 files changed: 62206 ins; 77179 del; 20838 mod Merge jdk ------------- PR: https://git.openjdk.org/lilliput/pull/104 From rkennke at openjdk.org Thu Aug 17 20:53:10 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 17 Aug 2023 20:53:10 GMT Subject: [master] RFR: 8314545: [Lilliput] Revert changes in zRelocate.cpp Message-ID: We have a bunch of STS additions in zRelocate.cpp . The guidance that I've got from @fisk is that those are not needed and may in-fact cause deadlocks. Let's remove those changes and figure out if anything else is needed to deal with ZGC sync with ObjectMonitor deflation. Testing: - [ ] hotspot_gc +UCOH - [ ] tier1 +UCOH ------------- Commit messages: - 8314545: [Lilliput] Revert changes in zRelocate.cpp Changes: https://git.openjdk.org/lilliput/pull/105/files Webrev: https://webrevs.openjdk.org/?repo=lilliput&pr=105&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314545 Stats: 5 lines in 1 file changed: 0 ins; 5 del; 0 mod Patch: https://git.openjdk.org/lilliput/pull/105.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/105/head:pull/105 PR: https://git.openjdk.org/lilliput/pull/105 From rkennke at openjdk.org Thu Aug 17 21:22:54 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Thu, 17 Aug 2023 21:22:54 GMT Subject: [master] RFR: 8314545: [Lilliput] Revert changes in zRelocate.cpp In-Reply-To: References: Message-ID: On Thu, 17 Aug 2023 20:44:11 GMT, Roman Kennke wrote: > We have a bunch of STS additions in zRelocate.cpp . The guidance that I've got from @fisk is that those are not needed and may in-fact cause deadlocks. Let's remove those changes and figure out if anything else is needed to deal with ZGC sync with ObjectMonitor deflation. > > Testing: > - [ ] hotspot_gc +UCOH > - [ ] tier1 +UCOH Awww, this results in the following crash. This is exactly the symptom of missing sync between ZGC worker and monitor deflation. Moving this PR back to draft until we figured something out. # # A fatal error has been detected by the Java Runtime Environment: # # Internal Error (/home/rkennke/src/openjdk/lilliput/src/hotspot/share/oops/markWord.inline.hpp:39), pid=19182, tid=19184 # assert(!CompressedKlassPointers::is_null(narrow_klass())) failed: narrow klass must not be null: 0x0000000000000001 # # JRE version: OpenJDK Runtime Environment (22.0) (fastdebug build 22-internal-adhoc.rkennke.lilliput) # Java VM: OpenJDK 64-Bit Server VM (fastdebug 22-internal-adhoc.rkennke.lilliput, mixed mode, tiered, compressed class ptrs, z gc, linux-amd64) # Problematic frame: # V [libjvm.so+0x2f6d4f] markWord::klass() const [clone .part.36]+0x21 # # No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again # # An error report file with more information is saved as: # /local/home/rkennke/src/openjdk/lilliput/build/linux-x86_64-server-fastdebug/test-support/jtreg_test_hotspot_jtreg_hotspot_gc/scratch/15/hs_err_pid19182.log # # If you would like to submit a bug report, please visit: # https://bugreport.java.com/bugreport/crash.jsp # ------------- PR Comment: https://git.openjdk.org/lilliput/pull/105#issuecomment-1682995966 From rkennke at openjdk.org Tue Aug 22 10:57:21 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Tue, 22 Aug 2023 10:57:21 GMT Subject: [lilliput-jdk21u:lilliput] Integrated: 8314485: [Lilliput/JDK21] Cherry-pick: 8139457: Array bases are aligned at HeapWord granularity In-Reply-To: References: Message-ID: On Thu, 17 Aug 2023 01:09:07 GMT, Roman Kennke wrote: > Let's cherry-pick the current state of the upstream PR https://github.com/openjdk/jdk/pull/11044 into Lilliput/JDK21 in preparation for the other Lilliput-related changes. The original change applies cleanly on top of lilliput/jdk21u. > > Testing: > - [x] tier1 -UCCP > - [x] tier1 +UCCP This pull request has now been integrated. Changeset: 98e7eddc Author: Roman Kennke URL: https://git.openjdk.org/lilliput-jdk21u/commit/98e7eddc3008f5c650d04b0e3bb4dc2e9b49a24f Stats: 246 lines in 21 files changed: 134 ins; 59 del; 53 mod 8314485: [Lilliput/JDK21] Cherry-pick: 8139457: Array bases are aligned at HeapWord granularity Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput-jdk21u/pull/1 From rkennke at openjdk.org Wed Aug 23 09:26:56 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 09:26:56 GMT Subject: [master] RFR: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput Message-ID: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. Testing: - [ ] hotspot_gc +UCOH - [ ] hotspot_gc -UCOH (default) ------------- Commit messages: - 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput Changes: https://git.openjdk.org/lilliput/pull/107/files Webrev: https://webrevs.openjdk.org/?repo=lilliput&pr=107&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314863 Stats: 14 lines in 2 files changed: 9 ins; 5 del; 0 mod Patch: https://git.openjdk.org/lilliput/pull/107.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/107/head:pull/107 PR: https://git.openjdk.org/lilliput/pull/107 From shade at openjdk.org Wed Aug 23 09:27:03 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Wed, 23 Aug 2023 09:27:03 GMT Subject: [master] RFR: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput In-Reply-To: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> References: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> Message-ID: On Wed, 23 Aug 2023 08:42:01 GMT, Roman Kennke wrote: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [ ] hotspot_gc +UCOH > - [ ] hotspot_gc -UCOH (default) Minor nit, otherwise looks good. src/hotspot/share/runtime/arguments.cpp line 3101: > 3099: FLAG_SET_DEFAULT(UseCompactObjectHeaders, false); > 3100: } else { > 3101: fatal("Incompatible options: +UseZGC can not currently be used together with +UseCompactObjectHeaders"); I think this should be `vm_exit_during_initialization`, not `fatal`. ------------- Marked as reviewed by shade (Committer). PR Review: https://git.openjdk.org/lilliput/pull/107#pullrequestreview-1591088442 PR Review Comment: https://git.openjdk.org/lilliput/pull/107#discussion_r1302698970 From rkennke at openjdk.org Wed Aug 23 09:46:57 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 09:46:57 GMT Subject: [master] RFR: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput [v2] In-Reply-To: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> References: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> Message-ID: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [ ] hotspot_gc +UCOH > - [ ] hotspot_gc -UCOH (default) Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: Replace fatal with vm_exit... ------------- Changes: - all: https://git.openjdk.org/lilliput/pull/107/files - new: https://git.openjdk.org/lilliput/pull/107/files/b7bd5a47..eaed4926 Webrevs: - full: https://webrevs.openjdk.org/?repo=lilliput&pr=107&range=01 - incr: https://webrevs.openjdk.org/?repo=lilliput&pr=107&range=00-01 Stats: 1 line in 1 file changed: 0 ins; 0 del; 1 mod Patch: https://git.openjdk.org/lilliput/pull/107.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/107/head:pull/107 PR: https://git.openjdk.org/lilliput/pull/107 From shade at openjdk.org Wed Aug 23 10:08:39 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Wed, 23 Aug 2023 10:08:39 GMT Subject: [master] RFR: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput [v3] In-Reply-To: References: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> Message-ID: On Wed, 23 Aug 2023 10:03:49 GMT, Roman Kennke wrote: >> The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. >> >> Testing: >> - [ ] hotspot_gc +UCOH >> - [ ] hotspot_gc -UCOH (default) > > Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: > > Never exit hard, only change to -UCOH with warning Marked as reviewed by shade (Committer). ------------- PR Review: https://git.openjdk.org/lilliput/pull/107#pullrequestreview-1591234544 From rkennke at openjdk.org Wed Aug 23 10:08:38 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 10:08:38 GMT Subject: [master] RFR: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput [v3] In-Reply-To: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> References: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> Message-ID: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [ ] hotspot_gc +UCOH > - [ ] hotspot_gc -UCOH (default) Roman Kennke has updated the pull request incrementally with one additional commit since the last revision: Never exit hard, only change to -UCOH with warning ------------- Changes: - all: https://git.openjdk.org/lilliput/pull/107/files - new: https://git.openjdk.org/lilliput/pull/107/files/eaed4926..4cfd48c6 Webrevs: - full: https://webrevs.openjdk.org/?repo=lilliput&pr=107&range=02 - incr: https://webrevs.openjdk.org/?repo=lilliput&pr=107&range=01-02 Stats: 6 lines in 1 file changed: 0 ins; 4 del; 2 mod Patch: https://git.openjdk.org/lilliput/pull/107.diff Fetch: git fetch https://git.openjdk.org/lilliput.git pull/107/head:pull/107 PR: https://git.openjdk.org/lilliput/pull/107 From rkennke at openjdk.org Wed Aug 23 14:35:01 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 14:35:01 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8314894: [Lilliput/JDK17] Revert changes in zRelocate, prevent ZGC with Lilliput Message-ID: The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. Testing: - [x] hotspot_gc +UCOH - [x] hotspot_gc -UCOH (default) ------------- Commit messages: - 8314894: [Lilliput/JDK17] Revert changes in zRelocate, prevent ZGC with Lilliput Changes: https://git.openjdk.org/lilliput-jdk17u/pull/56/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk17u&pr=56&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8314894 Stats: 11 lines in 2 files changed: 5 ins; 6 del; 0 mod Patch: https://git.openjdk.org/lilliput-jdk17u/pull/56.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk17u.git pull/56/head:pull/56 PR: https://git.openjdk.org/lilliput-jdk17u/pull/56 From rkennke at openjdk.org Wed Aug 23 14:35:56 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 14:35:56 GMT Subject: [master] Integrated: 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput In-Reply-To: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> References: <56cc7MH72rH8mDmO8JWDgTB2jtLDYvg_OypPIplhxf4=.7c2c6a9b-8819-4909-ba37-c08afdcb8c64@github.com> Message-ID: On Wed, 23 Aug 2023 08:42:01 GMT, Roman Kennke wrote: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [x] hotspot_gc +UCOH > - [x] hotspot_gc -UCOH (default) This pull request has now been integrated. Changeset: 0a73377a Author: Roman Kennke URL: https://git.openjdk.org/lilliput/commit/0a73377af005082941396c1eaca60056d1bbe890 Stats: 10 lines in 2 files changed: 5 ins; 5 del; 0 mod 8314863: [Lilliput] Revert changes in zRelocate, prevent ZGC with Lilliput Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput/pull/107 From shade at openjdk.org Wed Aug 23 14:45:02 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Wed, 23 Aug 2023 14:45:02 GMT Subject: [lilliput-jdk17u:lilliput] RFR: 8314894: [Lilliput/JDK17] Revert changes in zRelocate, prevent ZGC with Lilliput In-Reply-To: References: Message-ID: On Wed, 23 Aug 2023 14:26:58 GMT, Roman Kennke wrote: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [x] hotspot_gc +UCOH > - [x] hotspot_gc -UCOH (default) Marked as reviewed by shade (Reviewer). ------------- PR Review: https://git.openjdk.org/lilliput-jdk17u/pull/56#pullrequestreview-1591773668 From rkennke at openjdk.org Wed Aug 23 17:59:53 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 17:59:53 GMT Subject: [lilliput-jdk17u:lilliput] Integrated: 8314894: [Lilliput/JDK17] Revert changes in zRelocate, prevent ZGC with Lilliput In-Reply-To: References: Message-ID: On Wed, 23 Aug 2023 14:26:58 GMT, Roman Kennke wrote: > The additions of SuspendibleThreadSet in [zRelocate.cpp](https://builds.shipilev.net/patch-openjdk-lilliput/src/hotspot/share/gc/z/zRelocate.cpp.udiff.html) can cause deadlocks. This problem can not easily be solved without major work in Lilliput. Let's revert those changes and prevent usage of ZGC together with Lilliput. > > Testing: > - [x] hotspot_gc +UCOH > - [x] hotspot_gc -UCOH (default) This pull request has now been integrated. Changeset: 8c0c160f Author: Roman Kennke URL: https://git.openjdk.org/lilliput-jdk17u/commit/8c0c160f87d4822690a97146663bcf7afe10d8fe Stats: 11 lines in 2 files changed: 5 ins; 6 del; 0 mod 8314894: [Lilliput/JDK17] Revert changes in zRelocate, prevent ZGC with Lilliput Reviewed-by: shade ------------- PR: https://git.openjdk.org/lilliput-jdk17u/pull/56 From rkennke at openjdk.org Wed Aug 23 18:18:52 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Wed, 23 Aug 2023 18:18:52 GMT Subject: [master] Withdrawn: 8314545: [Lilliput] Revert changes in zRelocate.cpp In-Reply-To: References: Message-ID: On Thu, 17 Aug 2023 20:44:11 GMT, Roman Kennke wrote: > We have a bunch of STS additions in zRelocate.cpp . The guidance that I've got from @fisk is that those are not needed and may in-fact cause deadlocks. Let's remove those changes and figure out if anything else is needed to deal with ZGC sync with ObjectMonitor deflation. > > Testing: > - [ ] hotspot_gc +UCOH > - [ ] tier1 +UCOH This pull request has been closed without being integrated. ------------- PR: https://git.openjdk.org/lilliput/pull/105 From rkennke at openjdk.org Fri Aug 25 18:23:22 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Fri, 25 Aug 2023 18:23:22 GMT Subject: [lilliput-jdk21u:lilliput] RFR: 8315047: [Lilliput/JDK21] Some additions to 8139457: Array bases are aligned at HeapWord granularity Message-ID: The original cherry-pick for 8139457 missed a couple of new tests. Also, I'd also bring in some recent updates from the PR (in response to reviews). Testing: - [x] tier1 -CCP - [x] tier1 +CCP ------------- Commit messages: - 8315047: [Lilliput/JDK21] Some additions to 8139457: Array bases are aligned at HeapWord granularity Changes: https://git.openjdk.org/lilliput-jdk21u/pull/2/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk21u&pr=2&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8315047 Stats: 399 lines in 15 files changed: 349 ins; 33 del; 17 mod Patch: https://git.openjdk.org/lilliput-jdk21u/pull/2.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk21u.git pull/2/head:pull/2 PR: https://git.openjdk.org/lilliput-jdk21u/pull/2 From shade at openjdk.org Fri Aug 25 18:26:36 2023 From: shade at openjdk.org (Aleksey Shipilev) Date: Fri, 25 Aug 2023 18:26:36 GMT Subject: [lilliput-jdk21u:lilliput] RFR: 8315047: [Lilliput/JDK21] Some additions to 8139457: Array bases are aligned at HeapWord granularity In-Reply-To: References: Message-ID: <-18bUh_uU6DlKCMRKLJ4oU4Ow-BO2Ny8fogaK3wzERA=.7fedb5e6-d7dc-469e-b255-7b7d3f273eee@github.com> On Fri, 25 Aug 2023 18:17:03 GMT, Roman Kennke wrote: > The original cherry-pick for 8139457 missed a couple of new tests. Also, I'd also bring in some recent updates from the PR (in response to reviews). > > Testing: > - [x] tier1 -CCP > - [x] tier1 +CCP Marked as reviewed by shade (Reviewer). ------------- PR Review: https://git.openjdk.org/lilliput-jdk21u/pull/2#pullrequestreview-1596289386 From rkennke at openjdk.org Fri Aug 25 18:42:21 2023 From: rkennke at openjdk.org (Roman Kennke) Date: Fri, 25 Aug 2023 18:42:21 GMT Subject: [lilliput-jdk21u:lilliput] RFR: 8315046: [Lilliput/JDK21] Cherry-pick: 8305896: Alternative full GC forwarding Message-ID: Let's cherry-pick the alternative full GC forwarding PR in its current state into Lilliput-21. I needed to make some adjustments because PreservedMarks stuff has been refactored between 21 and 22. Testing: - [ ] hotspot_gc +UseAltGCForwarding - [ ] hotspot_gc -UseAltGCForwarding ------------- Commit messages: - 8315046: [Lilliput/JDK21] Cherry-pick: 8305896: Alternative full GC forwarding Changes: https://git.openjdk.org/lilliput-jdk21u/pull/3/files Webrev: https://webrevs.openjdk.org/?repo=lilliput-jdk21u&pr=3&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8315046 Stats: 1196 lines in 39 files changed: 1077 ins; 10 del; 109 mod Patch: https://git.openjdk.org/lilliput-jdk21u/pull/3.diff Fetch: git fetch https://git.openjdk.org/lilliput-jdk21u.git pull/3/head:pull/3 PR: https://git.openjdk.org/lilliput-jdk21u/pull/3