From duke at openjdk.org  Thu May  8 00:53:38 2025
From: duke at openjdk.org (Anton Rybochkin)
Date: Thu, 8 May 2025 00:53:38 GMT
Subject: RFR: Added ZGC executors and test for object alignment detection
Message-ID: <5lqb7DyrDtr7gMrAcIeboP-Eje9UPCi23mfHa_shVBQ=.4aafe9be-7623-47bf-9c68-4ec44bd7ddf3@github.com>

This test reveals incorrect object alignment detection when using generational ZGC which results in high object size reported (512k on aarch64 and 64k otherwise)

-------------

Commit messages:
 - Added test executions with ZGC
 - Added test for object alignment detection

Changes: https://git.openjdk.org/jol/pull/64/files
  Webrev: https://webrevs.openjdk.org/?repo=jol&pr=64&range=00
  Stats: 108 lines in 2 files changed: 108 ins; 0 del; 0 mod
  Patch: https://git.openjdk.org/jol/pull/64.diff
  Fetch: git fetch https://git.openjdk.org/jol.git pull/64/head:pull/64

PR: https://git.openjdk.org/jol/pull/64

From duke at openjdk.org  Thu May  8 01:42:52 2025
From: duke at openjdk.org (Anton Rybochkin)
Date: Thu, 8 May 2025 01:42:52 GMT
Subject: RFR: Uncolorize addresses when using Generational ZGC
Message-ID: <jCG3M7nv8LRyiiTKLJgU_XDzBPKzcCdZEnGspj4E0ow=.101b53b6-df43-4181-af99-57fa887089f4@github.com>

Because of color bits in lower positions, object alignment size heuristics does not work with generational ZGC. This PR implements address uncolorizing so alignment size and hence object size is calculated correctly

-------------

Commit messages:
 - Uncolorize ZGC addresses

Changes: https://git.openjdk.org/jol/pull/65/files
  Webrev: https://webrevs.openjdk.org/?repo=jol&pr=65&range=00
  Stats: 76 lines in 4 files changed: 73 ins; 0 del; 3 mod
  Patch: https://git.openjdk.org/jol/pull/65.diff
  Fetch: git fetch https://git.openjdk.org/jol.git pull/65/head:pull/65

PR: https://git.openjdk.org/jol/pull/65

From duke at openjdk.org  Wed May 14 18:09:09 2025
From: duke at openjdk.org (Anton Rybochkin)
Date: Wed, 14 May 2025 18:09:09 GMT
Subject: RFR: Uncolorize addresses when using Generational ZGC
In-Reply-To: <jCG3M7nv8LRyiiTKLJgU_XDzBPKzcCdZEnGspj4E0ow=.101b53b6-df43-4181-af99-57fa887089f4@github.com>
References: <jCG3M7nv8LRyiiTKLJgU_XDzBPKzcCdZEnGspj4E0ow=.101b53b6-df43-4181-af99-57fa887089f4@github.com>
Message-ID: <HWpiXrIsmR4E9PihgAOH0vBVBgsanCpnXO0CJa903NM=.52a40372-fa53-4b42-afa0-da3ee378a41b@github.com>

On Thu, 8 May 2025 01:37:35 GMT, Anton Rybochkin <duke at openjdk.org> wrote:

> Because of color bits in lower positions, object alignment size heuristics does not work with generational ZGC. This PR implements address uncolorizing so alignment size and hence object size is calculated correctly

@shipilev Could you review, please?

-------------

PR Comment: https://git.openjdk.org/jol/pull/65#issuecomment-2881100184

From duke at openjdk.org  Sat May 17 17:41:01 2025
From: duke at openjdk.org (Mateusz Serafin Gajewski)
Date: Sat, 17 May 2025 17:41:01 GMT
Subject: RFR: 7903905: JOL: Simplify Lilliput headump-estimates
In-Reply-To: <nhHB7U48pJJAMCPZp8JbCbfILzffv9N-yA8N-yM3ALs=.9d7834e4-7676-41a4-84db-57dc8d03ced6@github.com>
References: <nhHB7U48pJJAMCPZp8JbCbfILzffv9N-yA8N-yM3ALs=.9d7834e4-7676-41a4-84db-57dc8d03ced6@github.com>
Message-ID: <RpFmSU8ZjguCaiHG7tu4FHDSAKyNh6VU-hXhJ0AIxy4=.09963130-efc7-4fa1-a47a-f37aa501aec1@github.com>

On Wed, 11 Dec 2024 17:17:13 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> Lilliput was shipped in JDK 24 as experimental option. Current heapdump-esitmates provides *two* estimates: one with array base improvements, and one without. The version of Lilliput in upstream does implement array base improvements unconditionally ([JDK-8139457](https://bugs.openjdk.org/browse/JDK-8139457)). Therefore, we do not need one of the estimates.

Hi @shipilev. Do you plan to make a new JOL release since JDK 24 is out?

-------------

PR Comment: https://git.openjdk.org/jol/pull/61#issuecomment-2888508009

From duke at openjdk.org  Thu May 29 16:14:02 2025
From: duke at openjdk.org (Ivo Anjo)
Date: Thu, 29 May 2025 16:14:02 GMT
Subject: RFR: Uncolorize addresses when using Generational ZGC
In-Reply-To: <jCG3M7nv8LRyiiTKLJgU_XDzBPKzcCdZEnGspj4E0ow=.101b53b6-df43-4181-af99-57fa887089f4@github.com>
References: <jCG3M7nv8LRyiiTKLJgU_XDzBPKzcCdZEnGspj4E0ow=.101b53b6-df43-4181-af99-57fa887089f4@github.com>
Message-ID: <L_TXGGhcO8cYCsKrM861mdhfrZcm_sj-UWGB6Yj2IBQ=.332089d7-ad40-4ffa-8d5e-8e3a5781af13@github.com>

On Thu, 8 May 2025 01:37:35 GMT, Anton Rybochkin <duke at openjdk.org> wrote:

> Because of color bits in lower positions, object alignment size heuristics does not work with generational ZGC. This PR implements address uncolorizing so alignment size and hence object size is calculated correctly

I was running https://github.com/corretto/heapothesys on JDK 24 with ZGC and it was breaking with 


Caused by: java.lang.NegativeArraySizeException: -1047953
	at com.amazon.corretto.benchmark.hyperalloc.AllocObject.<init>(AllocObject.java:29)


due to the size fetched from jol. Running heapothesys with jol from this PR fixed it! ?

-------------

PR Comment: https://git.openjdk.org/jol/pull/65#issuecomment-2919893879