From thartmann at openjdk.java.net  Mon Feb  1 10:12:04 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 1 Feb 2021 10:12:04 GMT
Subject: [lworld] RFR: 8260363: [lworld] C2 compilation fails with
 assert(n->Opcode() != Op_Phi) failed: cannot match
In-Reply-To: <uM0jV7V4vtk8IG2xRt2129ZYnA7ck2Z_nEQUPo9hwgQ=.7df0b65a-6034-4e74-9fe2-7264bd8a4bc8@github.com>
References: <uM0jV7V4vtk8IG2xRt2129ZYnA7ck2Z_nEQUPo9hwgQ=.7df0b65a-6034-4e74-9fe2-7264bd8a4bc8@github.com>
Message-ID: <yNffCAuR4DI55Dx-SImbtTdgTPSpY8yhURVMaw5PvZ4=.3501b2a2-3eb9-4d6a-9fc8-9f05d93b2720@github.com>

On Thu, 28 Jan 2021 12:26:39 GMT, Roland Westrelin <roland at openjdk.org> wrote:

> This is caused by a StoreCM whose OopStore doesn't point to a Store
> node because it was optimized out but to a memory Phi. The Phi is for
> slice TypeAryPtr::INLINES. When
> Compile::adjust_flattened_array_access_aliases() runs, the
> TypeAryPtr::INLINES slice becomes dead and is replaced by independent
> slices for each field of flat array elements. Because the StoreCM
> doesn't point to a Store, it's impossible to tell what slice it's for
> so its input still points to the TypeAryPtr::INLINES Phi once
> Compile::adjust_flattened_array_access_aliases() is over.
> 
> Given the TypeAryPtr::INLINES slice is dead that's a problem: assuming
> we can end up with a StoreCM that points to a Phi for a non eliminated
> Store (the Store would be behind the Phi I suppose in that case), then
> after Compile::adjust_flattened_array_access_aliases() runs, the Store
> would have moved to a new Phi but the StoreCM wouldn't have been
> updated.
> 
> But that's not what causes the crash. The crash occurs because the
> TypeAryPtr::INLINES memory subgraph is partially killed by
> Compile::adjust_flattened_array_access_aliases() so when matching
> occurs the Phi is only reachable through a precedence edge from the
> StoreCM (final graph reshaping moves the OopStore input to precedence
> edges). It's unreachable from root at matching time (because that code
> doesn't follow precendence edge) and is not marked as shared
> by Matcher::find_shared().
> 
> When Compile::adjust_flattened_array_access_aliases() runs, the
> StoreCM OopStore is actually first set to a MergeMem node that
> captures slices created for flat array fields but when
> StoreCMNode::Ideal() executes it sets the input to the
> TypeAryPtr::INLINES input of the MergeMem. I think that last steps is
> what's incorrect after
> Compile::adjust_flattened_array_access_aliases(). So I changed that
> logic so it keeps the MergeMem input and also changed final graph
> reshape so it adds as many precedence edges as there are flat field
> inputs to the MergeMem. These changes are required to preserve
> correctness, I believe.
> 
> With that, i still hit the assert because the bottom memory input to
> the MergeMem can end up as a precedence edge to the StoreCM (if that
> entry to the MergeMem is top). That bottom memory in the case of the
> regression test is a Phi that was created by
> Compile::adjust_flattened_array_access_aliases() and is usually
> optimized out. It's only reachable from the StoreCM precedence edge. I
> don't think there's a fundamental problem here so I change matching so
> it looks at precedence edges when looking for shared node. I also had
> to change node scheduling in a similar way.

Looks scary but reasonable to me! :)

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

Marked as reviewed by thartmann (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/317

From rriggs at openjdk.java.net  Tue Feb  2 21:01:08 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Tue, 2 Feb 2021 21:01:08 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for methods
 related to primitive classes
Message-ID: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>

Update terminology used to refer to primitive classes (from inline classes).
Drop the descriptions using "projection" and use "conversion" as needed.
Updated dependencies on Class.isPrimitiveClass().

[Updating test names will be a separate task.]

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

Commit messages:
 - 8260606: Update Valhalla core-libs naming for methods related to primitive classes

Changes: https://git.openjdk.java.net/valhalla/pull/318/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=318&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8260606
  Stats: 206 lines in 28 files changed: 0 ins; 14 del; 192 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/318.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/318/head:pull/318

PR: https://git.openjdk.java.net/valhalla/pull/318

From dsimms at openjdk.java.net  Wed Feb  3 07:56:00 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Wed, 3 Feb 2021 07:56:00 GMT
Subject: [lworld] Integrated: Merge jdk
Message-ID: <8lQ8YOvS61VDmsQUBE-RtDLIXttFQUn7A1P5BpD5vM8=.5d6270b9-e87d-411b-af7a-3d7503d628b7@github.com>

Merge to 20210119

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

Commit messages:
 - Test fix
 - Test issue fixed
 - JDK-8257594 fix merged
 - Open java.base for reflective setAccessible
 - Valhalla adjust for JDK-8257772
 - 8260575: [lworld] Merge of JDK 17 causing javac type casting issues with reference projections
 - Version 17 on CheckFeatureGate tests
 - Tabs (still borked indentation)
 - Version 60 -> 61
 - Remove deprecated warnings
 - ... and 1075 more: https://git.openjdk.java.net/valhalla/compare/bd81ccfd...328b6125

The merge commit only contains trivial merges, so no merge-specific webrevs have been generated.

Changes: https://git.openjdk.java.net/valhalla/pull/316/files
  Stats: 143804 lines in 1307 files changed: 137167 ins; 2045 del; 4592 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/316.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/316/head:pull/316

PR: https://git.openjdk.java.net/valhalla/pull/316

From dsimms at openjdk.java.net  Wed Feb  3 07:56:04 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Wed, 3 Feb 2021 07:56:04 GMT
Subject: [lworld] Integrated: Merge jdk
In-Reply-To: <8lQ8YOvS61VDmsQUBE-RtDLIXttFQUn7A1P5BpD5vM8=.5d6270b9-e87d-411b-af7a-3d7503d628b7@github.com>
References: <8lQ8YOvS61VDmsQUBE-RtDLIXttFQUn7A1P5BpD5vM8=.5d6270b9-e87d-411b-af7a-3d7503d628b7@github.com>
Message-ID: <AF10ZVdBjzGWPrb1Wk6rwntjJRaqQ5ZAddScCMTJxEc=.7424c682-2071-4409-8eb8-dfe61d6276f9@github.com>

On Thu, 28 Jan 2021 08:29:09 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge to 20210119

This pull request has now been integrated.

Changeset: 85690886
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/85690886
Stats:     118958 lines in 3159 files changed: 64244 ins; 32945 del; 21769 mod

Merge jdk

Merge to 20210119

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

PR: https://git.openjdk.java.net/valhalla/pull/316

From roland at openjdk.java.net  Wed Feb  3 12:06:06 2021
From: roland at openjdk.java.net (Roland Westrelin)
Date: Wed, 3 Feb 2021 12:06:06 GMT
Subject: [lworld] RFR: 8260363: [lworld] C2 compilation fails with
 assert(n->Opcode() != Op_Phi) failed: cannot match
In-Reply-To: <yNffCAuR4DI55Dx-SImbtTdgTPSpY8yhURVMaw5PvZ4=.3501b2a2-3eb9-4d6a-9fc8-9f05d93b2720@github.com>
References: <uM0jV7V4vtk8IG2xRt2129ZYnA7ck2Z_nEQUPo9hwgQ=.7df0b65a-6034-4e74-9fe2-7264bd8a4bc8@github.com>
 <yNffCAuR4DI55Dx-SImbtTdgTPSpY8yhURVMaw5PvZ4=.3501b2a2-3eb9-4d6a-9fc8-9f05d93b2720@github.com>
Message-ID: <pJAzj3QMpgkqfcXnNvejbKF-jbsGZZDAGiPc3bbrVE4=.387b2ec4-abe4-4187-bdc2-e3a2e4bc256b@github.com>

On Mon, 1 Feb 2021 10:09:23 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> Looks scary but reasonable to me! :)

right! Thanks for the review.

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

PR: https://git.openjdk.java.net/valhalla/pull/317

From roland at openjdk.java.net  Wed Feb  3 12:06:09 2021
From: roland at openjdk.java.net (Roland Westrelin)
Date: Wed, 3 Feb 2021 12:06:09 GMT
Subject: [lworld] Integrated: 8260363: [lworld] C2 compilation fails with
 assert(n->Opcode() != Op_Phi) failed: cannot match
In-Reply-To: <uM0jV7V4vtk8IG2xRt2129ZYnA7ck2Z_nEQUPo9hwgQ=.7df0b65a-6034-4e74-9fe2-7264bd8a4bc8@github.com>
References: <uM0jV7V4vtk8IG2xRt2129ZYnA7ck2Z_nEQUPo9hwgQ=.7df0b65a-6034-4e74-9fe2-7264bd8a4bc8@github.com>
Message-ID: <J87q6fv2mtc1rVnfluh5E4ZsWj353bZR_n6Kf1y9D5w=.df4180d9-1468-486a-b398-e7c6e3a67cb7@github.com>

On Thu, 28 Jan 2021 12:26:39 GMT, Roland Westrelin <roland at openjdk.org> wrote:

> This is caused by a StoreCM whose OopStore doesn't point to a Store
> node because it was optimized out but to a memory Phi. The Phi is for
> slice TypeAryPtr::INLINES. When
> Compile::adjust_flattened_array_access_aliases() runs, the
> TypeAryPtr::INLINES slice becomes dead and is replaced by independent
> slices for each field of flat array elements. Because the StoreCM
> doesn't point to a Store, it's impossible to tell what slice it's for
> so its input still points to the TypeAryPtr::INLINES Phi once
> Compile::adjust_flattened_array_access_aliases() is over.
> 
> Given the TypeAryPtr::INLINES slice is dead that's a problem: assuming
> we can end up with a StoreCM that points to a Phi for a non eliminated
> Store (the Store would be behind the Phi I suppose in that case), then
> after Compile::adjust_flattened_array_access_aliases() runs, the Store
> would have moved to a new Phi but the StoreCM wouldn't have been
> updated.
> 
> But that's not what causes the crash. The crash occurs because the
> TypeAryPtr::INLINES memory subgraph is partially killed by
> Compile::adjust_flattened_array_access_aliases() so when matching
> occurs the Phi is only reachable through a precedence edge from the
> StoreCM (final graph reshaping moves the OopStore input to precedence
> edges). It's unreachable from root at matching time (because that code
> doesn't follow precendence edge) and is not marked as shared
> by Matcher::find_shared().
> 
> When Compile::adjust_flattened_array_access_aliases() runs, the
> StoreCM OopStore is actually first set to a MergeMem node that
> captures slices created for flat array fields but when
> StoreCMNode::Ideal() executes it sets the input to the
> TypeAryPtr::INLINES input of the MergeMem. I think that last steps is
> what's incorrect after
> Compile::adjust_flattened_array_access_aliases(). So I changed that
> logic so it keeps the MergeMem input and also changed final graph
> reshape so it adds as many precedence edges as there are flat field
> inputs to the MergeMem. These changes are required to preserve
> correctness, I believe.
> 
> With that, i still hit the assert because the bottom memory input to
> the MergeMem can end up as a precedence edge to the StoreCM (if that
> entry to the MergeMem is top). That bottom memory in the case of the
> regression test is a Phi that was created by
> Compile::adjust_flattened_array_access_aliases() and is usually
> optimized out. It's only reachable from the StoreCM precedence edge. I
> don't think there's a fundamental problem here so I change matching so
> it looks at precedence edges when looking for shared node. I also had
> to change node scheduling in a similar way.

This pull request has now been integrated.

Changeset: fbb8a6bd
Author:    Roland Westrelin <roland at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/fbb8a6bd
Stats:     111 lines in 5 files changed: 98 ins; 0 del; 13 mod

8260363: [lworld] C2 compilation fails with assert(n->Opcode() != Op_Phi) failed: cannot match

Reviewed-by: thartmann

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

PR: https://git.openjdk.java.net/valhalla/pull/317

From mchung at openjdk.java.net  Wed Feb  3 19:10:54 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Wed, 3 Feb 2021 19:10:54 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
Message-ID: <fttGPJHbXoASjelY8T0i18rA2WS6XVmjYCzQRL4r6ts=.7dea78e4-c96a-49d2-a4dc-f1bda80cf399@github.com>

On Tue, 2 Feb 2021 20:55:07 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Update terminology used to refer to primitive classes (from inline classes).
> Drop the descriptions using "projection" and use "conversion" as needed.
> Updated dependencies on Class.isPrimitiveClass().
> 
> [Updating test names will be a separate task.]

src/java.base/share/classes/java/lang/Class.java line 558:

> 556:      * Returns {@code true} if this class is a primitive class.
> 557:      *
> 558:      * @return {@code true} if this class is a primitive class, otherwise false

Nit: s/false/`{@code false}`/

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From mchung at openjdk.java.net  Wed Feb  3 19:23:00 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Wed, 3 Feb 2021 19:23:00 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
Message-ID: <WdzHNLUipbAauTOrBKusF5UG48NihCM7EVXVzDAeWy0=.f142122e-c6c7-4f6d-9da4-ddde689e2105@github.com>

On Tue, 2 Feb 2021 20:55:07 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Update terminology used to refer to primitive classes (from inline classes).
> Drop the descriptions using "projection" and use "conversion" as needed.
> Updated dependencies on Class.isPrimitiveClass().
> 
> [Updating test names will be a separate task.]

src/java.base/share/classes/java/lang/Class.java line 666:

> 664:      * is returned.
> 665:      */
> 666:     private Class<?>[] newPrimitiveTypeArray() {

perhaps rename to `newTypeArrayForPrimitiveClass`

src/java.base/share/classes/java/lang/invoke/ValueBootstrapMethods.java line 555:

> 553: 
> 554:     /**
> 555:      * Invoke the bootstrap methods hashCode for the given primitive classobject.

a space is missing in `classobject`

src/java.base/share/classes/java/lang/invoke/ValueBootstrapMethods.java line 557:

> 555:      * Invoke the bootstrap methods hashCode for the given primitive classobject.
> 556:      * @param o the instance to hash.
> 557:      * @return the hash code of the given primitive classobject.

space is missing in the word `classobject`

src/java.base/share/classes/java/lang/invoke/ValueBootstrapMethods.java line 584:

> 582:      * Invoke the bootstrap methods hashCode for the given primitive classobject.
> 583:      * @param o the instance to hash.
> 584:      * @return the string representation of the given primitive classobject.

same typo `classobject`

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From mchung at openjdk.java.net  Wed Feb  3 19:27:55 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Wed, 3 Feb 2021 19:27:55 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
Message-ID: <o47kgPSXddcRM4FEjRkPBwY_2An7BefrHMePZkE5QSc=.3b454206-fdc4-47f6-8cc1-73af7a217aff@github.com>

On Tue, 2 Feb 2021 20:55:07 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Update terminology used to refer to primitive classes (from inline classes).
> Drop the descriptions using "projection" and use "conversion" as needed.
> Updated dependencies on Class.isPrimitiveClass().
> 
> [Updating test names will be a separate task.]

Looks good.   Do you want to add `IdentityObject` in a separate PR?  It's okay with me to add this in this PR.

src/java.base/share/classes/jdk/internal/misc/Unsafe.java line 276:

> 274:     }
> 275: 
> 276:     public Object getReferenceVolatile(Object o, long offset, Class<?> vc) {

`vc` parameter should also be renamed to `pc` to be consistent with other methods.

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

Marked as reviewed by mchung (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/318

From rriggs at openjdk.java.net  Wed Feb  3 19:52:54 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Wed, 3 Feb 2021 19:52:54 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <o47kgPSXddcRM4FEjRkPBwY_2An7BefrHMePZkE5QSc=.3b454206-fdc4-47f6-8cc1-73af7a217aff@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
 <o47kgPSXddcRM4FEjRkPBwY_2An7BefrHMePZkE5QSc=.3b454206-fdc4-47f6-8cc1-73af7a217aff@github.com>
Message-ID: <Tfrz0yqES72d3j3lXUuDMzGIzNp-n0Nd_LeJRPoe0iY=.530f7390-b9e8-4e65-983d-c17f2da7b8b9@github.com>

On Wed, 3 Feb 2021 19:24:46 GMT, Mandy Chung <mchung at openjdk.org> wrote:

> Looks good. Do you want to add `IdentityObject` in a separate PR? It's okay with me to add this in this PR.

IdentityObject is already in java.lang.

java.lang.PrimitiveObject has not yet been added.

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From mchung at openjdk.java.net  Wed Feb  3 20:00:55 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Wed, 3 Feb 2021 20:00:55 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <Tfrz0yqES72d3j3lXUuDMzGIzNp-n0Nd_LeJRPoe0iY=.530f7390-b9e8-4e65-983d-c17f2da7b8b9@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
 <o47kgPSXddcRM4FEjRkPBwY_2An7BefrHMePZkE5QSc=.3b454206-fdc4-47f6-8cc1-73af7a217aff@github.com>
 <Tfrz0yqES72d3j3lXUuDMzGIzNp-n0Nd_LeJRPoe0iY=.530f7390-b9e8-4e65-983d-c17f2da7b8b9@github.com>
Message-ID: <59Yvzwa1YlEtxZl-aXNYIwoN-pr0l_8fR48T7W4L9ow=.eabbdb99-d7ad-42ba-bfa4-2be49b842912@github.com>

On Wed, 3 Feb 2021 19:50:30 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> > Looks good. Do you want to add `IdentityObject` in a separate PR? It's okay with me to add this in this PR.
> 
> IdentityObject is already in java.lang.
> 
> java.lang.PrimitiveObject has not yet been added.

I meant `PrimitiveObject` (sorry for the typo).

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From rriggs at openjdk.java.net  Wed Feb  3 21:23:03 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Wed, 3 Feb 2021 21:23:03 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes [v2]
In-Reply-To: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
Message-ID: <n7FZnlVuvWMPv53WTD_PcrOZf3ytS3ZW0KFMN03KidE=.ed8df81c-a670-449c-964e-ea8fbab21aac@github.com>

> Update terminology used to refer to primitive classes (from inline classes).
> Drop the descriptions using "projection" and use "conversion" as needed.
> Updated dependencies on Class.isPrimitiveClass().
> 
> [Updating test names will be a separate task.]

Roger Riggs has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains three commits:

 - Merge branch 'lworld' of https://git.openjdk.java.net/valhalla into 8251554-rename-primitiveclass
 - Updates per review comments
 - 8260606: Update Valhalla core-libs naming for methods related to primitive classes

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

Changes: https://git.openjdk.java.net/valhalla/pull/318/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=318&range=01
  Stats: 209 lines in 28 files changed: 0 ins; 14 del; 195 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/318.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/318/head:pull/318

PR: https://git.openjdk.java.net/valhalla/pull/318

From mchung at openjdk.java.net  Wed Feb  3 21:28:54 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Wed, 3 Feb 2021 21:28:54 GMT
Subject: [lworld] RFR: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes [v2]
In-Reply-To: <n7FZnlVuvWMPv53WTD_PcrOZf3ytS3ZW0KFMN03KidE=.ed8df81c-a670-449c-964e-ea8fbab21aac@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
 <n7FZnlVuvWMPv53WTD_PcrOZf3ytS3ZW0KFMN03KidE=.ed8df81c-a670-449c-964e-ea8fbab21aac@github.com>
Message-ID: <W8e_ci971DjJJwgGTchHX4L6pbymWY1ltf_2hlqiRSQ=.74652435-53cc-4c63-9480-6c2b62943640@github.com>

On Wed, 3 Feb 2021 21:23:03 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

>> Update terminology used to refer to primitive classes (from inline classes).
>> Drop the descriptions using "projection" and use "conversion" as needed.
>> Updated dependencies on Class.isPrimitiveClass().
>> 
>> [Updating test names will be a separate task.]
>
> Roger Riggs has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains three commits:
> 
>  - Merge branch 'lworld' of https://git.openjdk.java.net/valhalla into 8251554-rename-primitiveclass
>  - Updates per review comments
>  - 8260606: Update Valhalla core-libs naming for methods related to primitive classes

Marked as reviewed by mchung (Committer).

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From dsimms at openjdk.java.net  Thu Feb  4 12:47:41 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 4 Feb 2021 12:47:41 GMT
Subject: [lworld] RFR: Merge jdk
Message-ID: <uvKf1QE9t1uTbdaA2b2UHcvBQmqrfNRWwf0D3UQeQZo=.1a0364fa-b8bf-45ac-b082-5cf2e2ba8503@github.com>

Merge tag 'jdk-17+8' into lworld_merge_jdk_17_8
Logical merge compilation fixes

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

Commit messages:
 - Logical merge compilation fixes
 - Merge tag 'jdk-17+8' into lworld_merge_jdk_17_8
 - 8260301: misc gc/g1/unloading tests fails with "RuntimeException: Method could not be enqueued for compilation at level N"
 - 8257858: [macOS]: Remove JNF dependency from libosxsecurity/KeystoreImpl.m
 - 8258799: [Testbug] RandomCommandsTest must check if tested directive is added via jcmd
 - 8241995: Clarify InetSocketAddress::toString specification
 - 8259265: Refactor UncaughtExceptions shell test as java test.
 - 8260998: Shenandoah: Restore reference processing statistics reporting
 - 8260617: Merge ZipFile encoding check with the initial hash calculation
 - 8260581: IGV: enhance node search
 - ... and 222 more: https://git.openjdk.java.net/valhalla/compare/fbb8a6bd...604b0cb2

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=319&range=00.0
 - jdk: https://webrevs.openjdk.java.net/?repo=valhalla&pr=319&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/319/files
  Stats: 46740 lines in 1143 files changed: 18239 ins; 15300 del; 13201 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/319.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/319/head:pull/319

PR: https://git.openjdk.java.net/valhalla/pull/319

From dsimms at openjdk.java.net  Thu Feb  4 13:23:22 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 4 Feb 2021 13:23:22 GMT
Subject: [lworld] RFR: Merge jdk [v2]
In-Reply-To: <uvKf1QE9t1uTbdaA2b2UHcvBQmqrfNRWwf0D3UQeQZo=.1a0364fa-b8bf-45ac-b082-5cf2e2ba8503@github.com>
References: <uvKf1QE9t1uTbdaA2b2UHcvBQmqrfNRWwf0D3UQeQZo=.1a0364fa-b8bf-45ac-b082-5cf2e2ba8503@github.com>
Message-ID: <Cie05WdN1nYYpS2a6kj4cW6pZLGLdEByz8GGUKBO_f4=.63e4caa2-0f78-4e72-88e7-0c0cfdcbbaed@github.com>

> Merge tag 'jdk-17+8' into lworld_merge_jdk_17_8
> Logical merge compilation fixes

David Simms has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 1083 commits:

 - Logical merge compilation fixes
 - Merge tag 'jdk-17+8' into lworld_merge_jdk_17_8
   
   Added tag jdk-17+8 for changeset f025bc1d
   
   # Conflicts:
   #	src/hotspot/cpu/x86/c1_MacroAssembler_x86.cpp
   #	src/hotspot/share/c1/c1_Compilation.cpp
   #	src/hotspot/share/c1/c1_LIRAssembler.cpp
   #	src/hotspot/share/ci/ciInstanceKlass.cpp
   #	src/hotspot/share/classfile/classFileParser.cpp
   #	src/hotspot/share/classfile/placeholders.hpp
   #	src/hotspot/share/classfile/systemDictionary.cpp
   #	src/hotspot/share/classfile/systemDictionary.hpp
   #	src/hotspot/share/classfile/verificationType.cpp
   #	src/hotspot/share/compiler/compileBroker.cpp
   #	src/hotspot/share/memory/metaspaceClosure.hpp
   #	src/hotspot/share/memory/universe.cpp
   #	src/hotspot/share/oops/arrayKlass.cpp
   #	src/hotspot/share/oops/objArrayKlass.cpp
   #	src/hotspot/share/opto/parse2.cpp
   #	src/hotspot/share/prims/jvm.cpp
   #	test/hotspot/jtreg/vmTestbase/jit/tiered/Test.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessBoolean.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessByte.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessChar.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessDouble.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessFloat.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessInt.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessLong.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessShort.java
   #	test/jdk/java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessString.java
   #	test/jdk/java/lang/invoke/VarHandles/X-VarHandleTestMethodHandleAccess.java.template
 - 8260363: [lworld] C2 compilation fails with assert(n->Opcode() != Op_Phi) failed: cannot match
   
   Reviewed-by: thartmann
 - Merge jdk
   
   Merge to 20210119
 - [lworld] code cleanup reverting some old leftover change
 - [lworld] Field::get should return default value instead of null on non-flattened inline field
   
   Reviewed-by: rriggs
 - 8260375: [lworld] ValueTearing fails with -XX:FlatArrayElementMaxSize=0 -XX:InlineFieldMaxFlatSize=0
   
   Reviewed-by: thartmann
 - Remove unused field from Method
   
   Reviewed-by: hseigel, lfoltan
 - 8260283: [lworld] C1's EliminateFieldAccess optimization fails with "wrong types"
   
   Reviewed-by: fparain
 - 8260225: [lworld] C1's delayed load indexed optimization sets incorrect type
   
   Reviewed-by: fparain
 - ... and 1073 more: https://git.openjdk.java.net/valhalla/compare/f025bc1d...604b0cb2

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

Changes: https://git.openjdk.java.net/valhalla/pull/319/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=319&range=01
  Stats: 144118 lines in 1307 files changed: 137496 ins; 2036 del; 4586 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/319.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/319/head:pull/319

PR: https://git.openjdk.java.net/valhalla/pull/319

From dsimms at openjdk.java.net  Thu Feb  4 13:23:26 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 4 Feb 2021 13:23:26 GMT
Subject: [lworld] Integrated: Merge jdk
In-Reply-To: <uvKf1QE9t1uTbdaA2b2UHcvBQmqrfNRWwf0D3UQeQZo=.1a0364fa-b8bf-45ac-b082-5cf2e2ba8503@github.com>
References: <uvKf1QE9t1uTbdaA2b2UHcvBQmqrfNRWwf0D3UQeQZo=.1a0364fa-b8bf-45ac-b082-5cf2e2ba8503@github.com>
Message-ID: <uzIAfVEC-15VbfKdhkaRJJEWKt8WnnluMvvyzSyqg2s=.2dec5c0f-8bf9-486e-b414-0b20e127171b@github.com>

On Thu, 4 Feb 2021 09:55:30 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge tag 'jdk-17+8' into lworld_merge_jdk_17_8
> Logical merge compilation fixes

This pull request has now been integrated.

Changeset: 33dc4fb9
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/33dc4fb9
Stats:     46740 lines in 1143 files changed: 18239 ins; 15300 del; 13201 mod

Merge jdk

Merge tag 'jdk-17+8'

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

PR: https://git.openjdk.java.net/valhalla/pull/319

From github.com+5010047+kelthuzadx at openjdk.java.net  Thu Feb  4 16:02:06 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Thu, 4 Feb 2021 16:02:06 GMT
Subject: [lworld] RFR: 8261019: valhalla/valuetypes/QTypeDescriptorTest.java
 "Illegal class name "QLine; " in class file $Proxy5"
Message-ID: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>

Hi all,

QTypeDescriptorTest.java[0] also fails on my machine, this is occurred after merging JDK[1], I did some investigation. The root cause of the problem is that after the JDK was merged[1], the value of the constant CONSTANT_CLASS_DESCRIPTORS was changed from 61 to 62, but the major version of the class bytes generated dynamically by java.lang.reflect.ProxyGenerator was not updated simultaneously, it remained at the value 60:
private byte[] generateClassFile() {
  // V16 == 60
  visit(V16, accessFlags, dotToSlash(className), null,
          JLR_PROXY, typeNames(interfaces));
...
}
ClassFileParser::verify_legal_class_name would go to an unexpected branch, which causes ClassFormatError.
void ClassFileParser::verify_legal_class_name(...) const {
  ...
  if (length > 0) {
    ...
    // read _major_version(60) from dynamically generated class bytes,
    // compare with updated CONSTANT_CLASS_DESCRIPTORS(61)
    else if (_major_version >= CONSTANT_CLASS_DESCRIPTORS && bytes[length - 1] == ';' ) {
      // EXPECTED HERE
      legal = verify_unqualified_name(bytes + 1, length - 2, LegalClass);
    } else {
      // GO TO HERE
      legal = verify_unqualified_name(bytes, length, LegalClass);
    }
  }
  if (!legal) {
    ResourceMark rm(THREAD);
    assert(_class_name != NULL, "invariant");
    Exceptions::fthrow(
      THREAD_AND_LOCATION,
      vmSymbols::java_lang_ClassFormatError(),
      "Illegal class name "%.*s" in class file %s", length, bytes,
      _class_name->as_C_string()
    );
    return;
  }
}
JDK-826101 suspects that JDK-8261021[2]  has the same problem since their exception messages are relatively similar. But according to their stack traces, they might be caused by different but related problems.

[0] https://bugs.openjdk.java.net/browse/JDK-8261019
[1] https://github.com/openjdk/valhalla/pull/316
[2] https://bugs.openjdk.java.net/browse/JDK-8261021

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

Commit messages:
 - [lworld] valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;" in class file $Proxy5

Changes: https://git.openjdk.java.net/valhalla/pull/320/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=320&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261019
  Stats: 2 lines in 1 file changed: 0 ins; 0 del; 2 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/320.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/320/head:pull/320

PR: https://git.openjdk.java.net/valhalla/pull/320

From mchung at openjdk.java.net  Thu Feb  4 18:14:56 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Thu, 4 Feb 2021 18:14:56 GMT
Subject: [lworld] RFR: 8261019: [lworld]
 valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;
 " in class file $Proxy5"
In-Reply-To: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
References: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
Message-ID: <lbmQqZ2GR1uRR9VyZlNbGotTl24J_ch658zZazPWkRo=.49c3ceb5-0559-4973-a9aa-8de5d6c0936d@github.com>

On Thu, 4 Feb 2021 15:57:30 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> QTypeDescriptorTest.java[0] also fails on my machine, this is occurred after merging JDK[1], I did some investigation. The root cause of the problem is that after the JDK was merged[1], the value of the constant CONSTANT_CLASS_DESCRIPTORS was changed from 61 to 62, but the major version of the class bytes generated dynamically by java.lang.reflect.ProxyGenerator was not updated simultaneously, it remained at the value 60:
> private byte[] generateClassFile() {
>   // V16 == 60
>   visit(V16, accessFlags, dotToSlash(className), null,
>           JLR_PROXY, typeNames(interfaces));
> ...
> }
> ClassFileParser::verify_legal_class_name would go to an unexpected branch, which causes ClassFormatError.
> void ClassFileParser::verify_legal_class_name(...) const {
>   ...
>   if (length > 0) {
>     ...
>     // read _major_version(60) from dynamically generated class bytes,
>     // compare with updated CONSTANT_CLASS_DESCRIPTORS(61)
>     else if (_major_version >= CONSTANT_CLASS_DESCRIPTORS && bytes[length - 1] == ';' ) {
>       // EXPECTED HERE
>       legal = verify_unqualified_name(bytes + 1, length - 2, LegalClass);
>     } else {
>       // GO TO HERE
>       legal = verify_unqualified_name(bytes, length, LegalClass);
>     }
>   }
>   if (!legal) {
>     ResourceMark rm(THREAD);
>     assert(_class_name != NULL, "invariant");
>     Exceptions::fthrow(
>       THREAD_AND_LOCATION,
>       vmSymbols::java_lang_ClassFormatError(),
>       "Illegal class name "%.*s" in class file %s", length, bytes,
>       _class_name->as_C_string()
>     );
>     return;
>   }
> }
> JDK-826101 suspects that JDK-8261021[2]  has the same problem since their exception messages are relatively similar. But according to their stack traces, they might be caused by different but related problems.
> 
> [0] https://bugs.openjdk.java.net/browse/JDK-8261019
> [1] https://github.com/openjdk/valhalla/pull/316
> [2] https://bugs.openjdk.java.net/browse/JDK-8261021

Thanks for fixing it.

src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 456:

> 454:      */
> 455:     private byte[] generateClassFile() {
> 456:         visit(V17, accessFlags, dotToSlash(className), null,

Alternatively, we can use the value of the system property `java.class.version` to prepare for future class file version change.  But this requires parsing it to extract the major version.

I'm okay with this simple approach and remember to update it if the class file version is bumped.

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

Marked as reviewed by mchung (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/320

From github.com+5010047+kelthuzadx at openjdk.java.net  Fri Feb  5 02:33:10 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Fri, 5 Feb 2021 02:33:10 GMT
Subject: [lworld] RFR: 8261019: [lworld]
 valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;
 " in class file $Proxy5" [v2]
In-Reply-To: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
References: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
Message-ID: <QZ33FvLLhC5074w11QRAV7MLyDbDq_uKeHmP68WrjT4=.35503576-2053-436f-ad92-ca74571fe3fb@github.com>

> Hi all,
> 
> QTypeDescriptorTest.java[0] also fails on my machine, this is occurred after merging JDK[1], I did some investigation. The root cause of the problem is that after the JDK was merged[1], the value of the constant CONSTANT_CLASS_DESCRIPTORS was changed from 61 to 62, but the major version of the class bytes generated dynamically by java.lang.reflect.ProxyGenerator was not updated simultaneously, it remained at the value 60:
> private byte[] generateClassFile() {
>   // V16 == 60
>   visit(V16, accessFlags, dotToSlash(className), null,
>           JLR_PROXY, typeNames(interfaces));
> ...
> }
> ClassFileParser::verify_legal_class_name would go to an unexpected branch, which causes ClassFormatError.
> void ClassFileParser::verify_legal_class_name(...) const {
>   ...
>   if (length > 0) {
>     ...
>     // read _major_version(60) from dynamically generated class bytes,
>     // compare with updated CONSTANT_CLASS_DESCRIPTORS(61)
>     else if (_major_version >= CONSTANT_CLASS_DESCRIPTORS && bytes[length - 1] == ';' ) {
>       // EXPECTED HERE
>       legal = verify_unqualified_name(bytes + 1, length - 2, LegalClass);
>     } else {
>       // GO TO HERE
>       legal = verify_unqualified_name(bytes, length, LegalClass);
>     }
>   }
>   if (!legal) {
>     ResourceMark rm(THREAD);
>     assert(_class_name != NULL, "invariant");
>     Exceptions::fthrow(
>       THREAD_AND_LOCATION,
>       vmSymbols::java_lang_ClassFormatError(),
>       "Illegal class name "%.*s" in class file %s", length, bytes,
>       _class_name->as_C_string()
>     );
>     return;
>   }
> }
> JDK-826101 suspects that JDK-8261021[2]  has the same problem since their exception messages are relatively similar. But according to their stack traces, they might be caused by different but related problems.
> 
> [0] https://bugs.openjdk.java.net/browse/JDK-8261019
> [1] https://github.com/openjdk/valhalla/pull/316
> [2] https://bugs.openjdk.java.net/browse/JDK-8261021

Yang Yi has updated the pull request incrementally with one additional commit since the last revision:

  Update full name

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/320/files
  - new: https://git.openjdk.java.net/valhalla/pull/320/files/65d376dd..04836bb7

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=320&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=320&range=00-01

  Stats: 0 lines in 0 files changed: 0 ins; 0 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/320.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/320/head:pull/320

PR: https://git.openjdk.java.net/valhalla/pull/320

From github.com+5010047+kelthuzadx at openjdk.java.net  Fri Feb  5 02:33:11 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Fri, 5 Feb 2021 02:33:11 GMT
Subject: [lworld] RFR: 8261019: [lworld]
 valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;
 " in class file $Proxy5" [v2]
In-Reply-To: <lbmQqZ2GR1uRR9VyZlNbGotTl24J_ch658zZazPWkRo=.49c3ceb5-0559-4973-a9aa-8de5d6c0936d@github.com>
References: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
 <lbmQqZ2GR1uRR9VyZlNbGotTl24J_ch658zZazPWkRo=.49c3ceb5-0559-4973-a9aa-8de5d6c0936d@github.com>
Message-ID: <Qu1bKShZ03ZXok8w0kSKHaPoKGcwjWsUdsSyrAJW_HQ=.54034dc9-d9ed-4a45-986f-ba92acaba8e0@github.com>

On Thu, 4 Feb 2021 18:12:05 GMT, Mandy Chung <mchung at openjdk.org> wrote:

>> Yang Yi has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Update full name
>
> src/java.base/share/classes/java/lang/reflect/ProxyGenerator.java line 456:
> 
>> 454:      */
>> 455:     private byte[] generateClassFile() {
>> 456:         visit(V17, accessFlags, dotToSlash(className), null,
> 
> Alternatively, we can use the value of the system property `java.class.version` to prepare for future class file version change.  But this requires parsing it to extract the major version.
> 
> I'm okay with this simple approach and remember to update it if the class file version is bumped.

It seems reasonable. When I fix another related fail, I can tweak this part.

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

PR: https://git.openjdk.java.net/valhalla/pull/320

From github.com+5010047+kelthuzadx at openjdk.java.net  Fri Feb  5 03:29:08 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Fri, 5 Feb 2021 03:29:08 GMT
Subject: [lworld] RFR: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
Message-ID: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>

Hi all,

Can I have a review of this trivial fix?

The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.

Thanks,
Yang Yi

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

Commit messages:
 - fix JDK-8261021

Changes: https://git.openjdk.java.net/valhalla/pull/321/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=321&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261021
  Stats: 2 lines in 1 file changed: 0 ins; 0 del; 2 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/321.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/321/head:pull/321

PR: https://git.openjdk.java.net/valhalla/pull/321

From thartmann at openjdk.java.net  Fri Feb  5 08:38:02 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 5 Feb 2021 08:38:02 GMT
Subject: [lworld] RFR: 8261037: [lworld] Assert during C2 compilation due to
 inconsistent JVMState at safepoint
Message-ID: <CLA94NkaNJi6ILR_4q1pDygTFp4cDPug3GV1jG_D6HA=.2ddafc47-fa50-4ec0-b2b7-511475ce994f@github.com>

We assert when scalarizing an inline type in a safepoint because `debug_end()` (-> `_endoff`) of the attached `JVMState` is larger than `SafePointNode::_max`. I don't think JVMStates should be shared between safepoint nodes and I intend to fix this in mainline with [JDK-8261158](https://bugs.openjdk.java.net/browse/JDK-8261158).

Gory details:
In `PhaseIdealLoop::is_counted_loop`, we `// Check for immediately preceding SafePoint and remove` by cloning _SafePointNode1_ to _SafePointNode2_ without cloning the attached _JVMState_. Loop unswitching then clones the loop and therefore also clones _SafePointNode2_ to _SafePointNode3_, again without cloning the _JVMState_. All three safepoint nodes now share the same _JVMState_. An inline type is scalarized in _SafePointNode2_, increasing _JVMState_`::_endoff` to account for the fields. Once another inline type is scalarized in _SafePointNode3_,  _JVMState_`::_endoff ` is out of bounds of _SafePointNode3_`::_max`.

This fix refactors the code (which also serves as workaround because we now access `debug_end()` only after we updated it to the correct value for the current safepoint) and adds a regression test.

Best regards,
Tobias

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

Commit messages:
 - 8261037: [lworld] Assert during C2 compilation due to inconsistent JVMState at safepoint

Changes: https://git.openjdk.java.net/valhalla/pull/322/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=322&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261037
  Stats: 35 lines in 3 files changed: 30 ins; 3 del; 2 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/322.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/322/head:pull/322

PR: https://git.openjdk.java.net/valhalla/pull/322

From nick.gasson at arm.com  Fri Feb  5 08:41:57 2021
From: nick.gasson at arm.com (Nick Gasson)
Date: Fri, 05 Feb 2021 16:41:57 +0800
Subject: Status of AArch64 lworld port
Message-ID: <85a6sidhyi.fsf@nicgas01-pc.shanghai.arm.com>

Hi,

I tried building the lworld branch on AArch64 but it seems to have
bit-rotted since the port was originally done. I also noticed
JDK-8258038 was raised to track this but it's not assigned at the
moment. Just wondering if anyone is planning to work on this?

--
Thanks,
Nick

From thartmann at openjdk.java.net  Fri Feb  5 09:14:55 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 5 Feb 2021 09:14:55 GMT
Subject: [lworld] Integrated: 8261037: [lworld] Assert during C2 compilation
 due to inconsistent JVMState at safepoint
In-Reply-To: <CLA94NkaNJi6ILR_4q1pDygTFp4cDPug3GV1jG_D6HA=.2ddafc47-fa50-4ec0-b2b7-511475ce994f@github.com>
References: <CLA94NkaNJi6ILR_4q1pDygTFp4cDPug3GV1jG_D6HA=.2ddafc47-fa50-4ec0-b2b7-511475ce994f@github.com>
Message-ID: <oYm4M2GkQMZPx47C5QyUuIBQzmfLcSvKCkHDSS6-iIs=.5cbe7335-259e-4f44-9a6f-ff14d63aae32@github.com>

On Fri, 5 Feb 2021 08:33:19 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> We assert when scalarizing an inline type in a safepoint because `debug_end()` (-> `_endoff`) of the attached `JVMState` is larger than `SafePointNode::_max`. I don't think JVMStates should be shared between safepoint nodes and I intend to fix this in mainline with [JDK-8261158](https://bugs.openjdk.java.net/browse/JDK-8261158).
> 
> Gory details:
> In `PhaseIdealLoop::is_counted_loop`, we `// Check for immediately preceding SafePoint and remove` by cloning _SafePointNode1_ to _SafePointNode2_ without cloning the attached _JVMState_. Loop unswitching then clones the loop and therefore also clones _SafePointNode2_ to _SafePointNode3_, again without cloning the _JVMState_. All three safepoint nodes now share the same _JVMState_. An inline type is scalarized in _SafePointNode2_, increasing _JVMState_`::_endoff` to account for the fields. Once another inline type is scalarized in _SafePointNode3_,  _JVMState_`::_endoff ` is out of bounds of _SafePointNode3_`::_max`.
> 
> This fix refactors the code (which also serves as workaround because we now access `debug_end()` only after we updated it to the correct value for the current safepoint) and adds a regression test.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: ce486c32
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/ce486c32
Stats:     35 lines in 3 files changed: 30 ins; 3 del; 2 mod

8261037: [lworld] Assert during C2 compilation due to inconsistent JVMState at safepoint

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

PR: https://git.openjdk.java.net/valhalla/pull/322

From tobias.hartmann at oracle.com  Fri Feb  5 09:17:17 2021
From: tobias.hartmann at oracle.com (Tobias Hartmann)
Date: Fri, 5 Feb 2021 10:17:17 +0100
Subject: Status of AArch64 lworld port
In-Reply-To: <85a6sidhyi.fsf@nicgas01-pc.shanghai.arm.com>
References: <85a6sidhyi.fsf@nicgas01-pc.shanghai.arm.com>
Message-ID: <1853542c-80f4-db40-97ae-f3b3b74f91d6@oracle.com>

Hi Nick,

yes, the AArch64 port misses significant changes and is currently not tested on our (Oracle) side.

Dmitriy Samersoff worked on this in the past but I'm not sure what his current plans are. In any
case, contributions are very welcome and I'm happy to answer questions about the JIT related changes.

Best regards,
Tobias

On 05.02.21 09:41, Nick Gasson wrote:
> Hi,
> 
> I tried building the lworld branch on AArch64 but it seems to have
> bit-rotted since the port was originally done. I also noticed
> JDK-8258038 was raised to track this but it's not assigned at the
> moment. Just wondering if anyone is planning to work on this?
> 
> --
> Thanks,
> Nick
> 

From thartmann at openjdk.java.net  Fri Feb  5 12:30:14 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 5 Feb 2021 12:30:14 GMT
Subject: [lworld] RFR: 8261226: [lworld] Array access profiling should be
 disabled at compilation level 1 and 2
Message-ID: <2eoFD8XkaCfXBIaqwbNSu2w_YZv_wCr8Qtbbdup65ps=.ba6dc036-7443-4b08-9e72-cca9c3fc9bb7@github.com>

C1 compilations at level 1 and 2 should not profile array accesses.

Best regards,
Tobias

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

Commit messages:
 - 8261226: [lworld] Array access profiling should be disabled at compilation level 1 and 2

Changes: https://git.openjdk.java.net/valhalla/pull/323/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=323&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261226
  Stats: 8 lines in 2 files changed: 7 ins; 1 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/323.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/323/head:pull/323

PR: https://git.openjdk.java.net/valhalla/pull/323

From dsimms at openjdk.java.net  Fri Feb  5 15:15:56 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Fri, 5 Feb 2021 15:15:56 GMT
Subject: [lworld] RFR: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
In-Reply-To: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
References: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
Message-ID: <nn88tR1Ei_Tm66C6m5WXrT_LMyG4OORJRIBlA6N1XNw=.3f61b124-ad65-4443-a247-1f647c2c025d@github.com>

On Fri, 5 Feb 2021 03:24:10 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> Can I have a review of this trivial fix?
> 
> The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.
> 
> Thanks,
> Yang Yi

Looks good, thanks for the find

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

Marked as reviewed by dsimms (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/321

From rriggs at openjdk.java.net  Fri Feb  5 15:15:56 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Fri, 5 Feb 2021 15:15:56 GMT
Subject: [lworld] RFR: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
In-Reply-To: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
References: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
Message-ID: <4A2n4Tj5T3QsLG6HPFcz5uwAbnc3FbNp75NpkxiipAA=.c55cef50-06a2-4de7-946d-ca28617acf5e@github.com>

On Fri, 5 Feb 2021 03:24:10 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> Can I have a review of this trivial fix?
> 
> The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.
> 
> Thanks,
> Yang Yi

Marked as reviewed by rriggs (Committer).

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

PR: https://git.openjdk.java.net/valhalla/pull/321

From thartmann at openjdk.java.net  Fri Feb  5 15:26:59 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 5 Feb 2021 15:26:59 GMT
Subject: [lworld] Integrated: 8261226: [lworld] Array access profiling should
 be disabled at compilation level 1 and 2
In-Reply-To: <2eoFD8XkaCfXBIaqwbNSu2w_YZv_wCr8Qtbbdup65ps=.ba6dc036-7443-4b08-9e72-cca9c3fc9bb7@github.com>
References: <2eoFD8XkaCfXBIaqwbNSu2w_YZv_wCr8Qtbbdup65ps=.ba6dc036-7443-4b08-9e72-cca9c3fc9bb7@github.com>
Message-ID: <gk3wMpdrJKMxNBKhSxfU6rChgRX1Lm1ehG0z5TLo2PY=.4762d4a6-dd7c-4ba1-b4f8-b3bf8fdc907e@github.com>

On Fri, 5 Feb 2021 12:26:42 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> C1 compilations at level 1 and 2 should not profile array accesses.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: ddfba612
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/ddfba612
Stats:     8 lines in 2 files changed: 7 ins; 1 del; 0 mod

8261226: [lworld] Array access profiling should be disabled at compilation level 1 and 2

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

PR: https://git.openjdk.java.net/valhalla/pull/323

From rriggs at openjdk.java.net  Fri Feb  5 17:27:54 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Fri, 5 Feb 2021 17:27:54 GMT
Subject: [lworld] Integrated: 8260606: Update Valhalla core-libs naming for
 methods related to primitive classes
In-Reply-To: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
References: <ONwmq8TFb0DsN6p0XNrIzoAkOWQcljTmRCMIHqZnIvM=.24f1691a-78cd-407b-b457-27d438c5b34d@github.com>
Message-ID: <TPIj1yeYZafVArCNb5pgTjUvMYuZD0D3oerM0AK1-qc=.8e3a24e8-46ec-427c-ad18-efa23cddb79d@github.com>

On Tue, 2 Feb 2021 20:55:07 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Update terminology used to refer to primitive classes (from inline classes).
> Drop the descriptions using "projection" and use "conversion" as needed.
> Updated dependencies on Class.isPrimitiveClass().
> 
> [Updating test names will be a separate task.]

This pull request has now been integrated.

Changeset: 5f1e8d97
Author:    Roger Riggs <rriggs at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/5f1e8d97
Stats:     209 lines in 28 files changed: 0 ins; 14 del; 195 mod

8260606: Update Valhalla core-libs naming for methods related to primitive classes

Reviewed-by: mchung

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

PR: https://git.openjdk.java.net/valhalla/pull/318

From mchung at openjdk.java.net  Fri Feb  5 17:53:55 2021
From: mchung at openjdk.java.net (Mandy Chung)
Date: Fri, 5 Feb 2021 17:53:55 GMT
Subject: [lworld] RFR: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
In-Reply-To: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
References: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
Message-ID: <PaNjW2Dgo9kcbVKvoF6PCO2DhqtbLIOl1EohmQ9tIOU=.b04e1796-0304-4660-89ff-274c583e3828@github.com>

On Fri, 5 Feb 2021 03:24:10 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> Can I have a review of this trivial fix?
> 
> The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.
> 
> Thanks,
> Yang Yi

Marked as reviewed by mchung (Committer).

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

PR: https://git.openjdk.java.net/valhalla/pull/321

From hseigel at openjdk.java.net  Fri Feb  5 20:35:00 2021
From: hseigel at openjdk.java.net (Harold Seigel)
Date: Fri, 5 Feb 2021 20:35:00 GMT
Subject: [lworld] RFR: 8261264: [lworld] runtime/valhalla/inlinetypes tests
 fail to compile
Message-ID: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>

Please review this small fix to have runtime/valhalla tests call method Class.isPrimitiveClass() instead of Class.isInlineClass().

Thanks, Harold

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

Commit messages:
 - 8261264: [lworld] runtime/valhalla/inlinetypes tests fail to compile

Changes: https://git.openjdk.java.net/valhalla/pull/324/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=324&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261264
  Stats: 4 lines in 2 files changed: 0 ins; 0 del; 4 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/324.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/324/head:pull/324

PR: https://git.openjdk.java.net/valhalla/pull/324

From rriggs at openjdk.java.net  Fri Feb  5 20:49:48 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Fri, 5 Feb 2021 20:49:48 GMT
Subject: [lworld] RFR: 8261264: [lworld] runtime/valhalla/inlinetypes
 tests fail to compile
In-Reply-To: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>
References: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>
Message-ID: <ljgNVSqZYk-Nxn2JQxvNw5f6Oph3JA6QpfXV98F94AA=.c3ace2e5-6069-48e1-be40-acd0694d7e96@github.com>

On Fri, 5 Feb 2021 20:30:45 GMT, Harold Seigel <hseigel at openjdk.org> wrote:

> Please review this small fix to have runtime/valhalla tests call method Class.isPrimitiveClass() instead of Class.isInlineClass().
> 
> Thanks, Harold

Sorry Harold I missed these.

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

Marked as reviewed by rriggs (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/324

From hseigel at openjdk.java.net  Fri Feb  5 20:57:59 2021
From: hseigel at openjdk.java.net (Harold Seigel)
Date: Fri, 5 Feb 2021 20:57:59 GMT
Subject: [lworld] RFR: 8261264: [lworld] runtime/valhalla/inlinetypes
 tests fail to compile
In-Reply-To: <ljgNVSqZYk-Nxn2JQxvNw5f6Oph3JA6QpfXV98F94AA=.c3ace2e5-6069-48e1-be40-acd0694d7e96@github.com>
References: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>
 <ljgNVSqZYk-Nxn2JQxvNw5f6Oph3JA6QpfXV98F94AA=.c3ace2e5-6069-48e1-be40-acd0694d7e96@github.com>
Message-ID: <4tkQoAGHn2DujoJ8u-jQIU48bb5y--sGrCZ5YaOzgVM=.39c719d2-f802-4f6b-8020-842bc94081e4@github.com>

On Fri, 5 Feb 2021 20:46:39 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

>> Please review this small fix to have runtime/valhalla tests call method Class.isPrimitiveClass() instead of Class.isInlineClass().
>> 
>> Thanks, Harold
>
> Sorry Harold I missed these.

Thanks Roger for the reivew!

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

PR: https://git.openjdk.java.net/valhalla/pull/324

From hseigel at openjdk.java.net  Fri Feb  5 20:58:00 2021
From: hseigel at openjdk.java.net (Harold Seigel)
Date: Fri, 5 Feb 2021 20:58:00 GMT
Subject: [lworld] Integrated: 8261264: [lworld] runtime/valhalla/inlinetypes
 tests fail to compile
In-Reply-To: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>
References: <Zn19fy7RH5Ex0mQfxJnoweCxBVVwxjQnvUOS3ooH5I8=.16834caf-8c29-4ac6-b571-d072fcb5fff8@github.com>
Message-ID: <tIRhyOEfI8HHdzb-cNEKTQ1z7S7ZR9kj9okPgGoVCTE=.e183b5a8-2981-4e29-ad11-251aebf36c78@github.com>

On Fri, 5 Feb 2021 20:30:45 GMT, Harold Seigel <hseigel at openjdk.org> wrote:

> Please review this small fix to have runtime/valhalla tests call method Class.isPrimitiveClass() instead of Class.isInlineClass().
> 
> Thanks, Harold

This pull request has now been integrated.

Changeset: c4f63d28
Author:    Harold Seigel <hseigel at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/c4f63d28
Stats:     4 lines in 2 files changed: 0 ins; 0 del; 4 mod

8261264: [lworld] runtime/valhalla/inlinetypes tests fail to compile

Reviewed-by: rriggs

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

PR: https://git.openjdk.java.net/valhalla/pull/324

From github.com+5010047+kelthuzadx at openjdk.java.net  Fri Feb  5 21:34:55 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Fri, 5 Feb 2021 21:34:55 GMT
Subject: [lworld] Integrated: 8261019: [lworld]
 valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;
 " in class file $Proxy5"
In-Reply-To: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
References: <pNkOJheX76_LFReCFIBoOfbL7TDneEZBMWZG_bS54n4=.f47c2eb8-d81d-4f3b-b4ca-a222664f1061@github.com>
Message-ID: <wSezeUao47jM19zxtAqZLFVBZEYn7ePSvMakt0i3FhA=.f5a0184b-20b9-4164-b7f5-9bd099ff875c@github.com>

On Thu, 4 Feb 2021 15:57:30 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> QTypeDescriptorTest.java[0] also fails on my machine, this is occurred after merging JDK[1], I did some investigation. The root cause of the problem is that after the JDK was merged[1], the value of the constant CONSTANT_CLASS_DESCRIPTORS was changed from 61 to 62, but the major version of the class bytes generated dynamically by java.lang.reflect.ProxyGenerator was not updated simultaneously, it remained at the value 60:
> private byte[] generateClassFile() {
>   // V16 == 60
>   visit(V16, accessFlags, dotToSlash(className), null,
>           JLR_PROXY, typeNames(interfaces));
> ...
> }
> ClassFileParser::verify_legal_class_name would go to an unexpected branch, which causes ClassFormatError.
> void ClassFileParser::verify_legal_class_name(...) const {
>   ...
>   if (length > 0) {
>     ...
>     // read _major_version(60) from dynamically generated class bytes,
>     // compare with updated CONSTANT_CLASS_DESCRIPTORS(61)
>     else if (_major_version >= CONSTANT_CLASS_DESCRIPTORS && bytes[length - 1] == ';' ) {
>       // EXPECTED HERE
>       legal = verify_unqualified_name(bytes + 1, length - 2, LegalClass);
>     } else {
>       // GO TO HERE
>       legal = verify_unqualified_name(bytes, length, LegalClass);
>     }
>   }
>   if (!legal) {
>     ResourceMark rm(THREAD);
>     assert(_class_name != NULL, "invariant");
>     Exceptions::fthrow(
>       THREAD_AND_LOCATION,
>       vmSymbols::java_lang_ClassFormatError(),
>       "Illegal class name "%.*s" in class file %s", length, bytes,
>       _class_name->as_C_string()
>     );
>     return;
>   }
> }
> JDK-826101 suspects that JDK-8261021[2]  has the same problem since their exception messages are relatively similar. But according to their stack traces, they might be caused by different but related problems.
> 
> [0] https://bugs.openjdk.java.net/browse/JDK-8261019
> [1] https://github.com/openjdk/valhalla/pull/316
> [2] https://bugs.openjdk.java.net/browse/JDK-8261021

This pull request has now been integrated.

Changeset: 4768ea20
Author:    Yang Yi <qingfeng.yy at alibaba-inc.com>
Committer: Harold Seigel <hseigel at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/4768ea20
Stats:     2 lines in 1 file changed: 0 ins; 0 del; 2 mod

8261019: [lworld] valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;" in class file $Proxy5"

Reviewed-by: mchung

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

PR: https://git.openjdk.java.net/valhalla/pull/320

From github.com+5010047+kelthuzadx at openjdk.java.net  Fri Feb  5 23:48:55 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Fri, 5 Feb 2021 23:48:55 GMT
Subject: [lworld] RFR: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
In-Reply-To: <nn88tR1Ei_Tm66C6m5WXrT_LMyG4OORJRIBlA6N1XNw=.3f61b124-ad65-4443-a247-1f647c2c025d@github.com>
References: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
 <nn88tR1Ei_Tm66C6m5WXrT_LMyG4OORJRIBlA6N1XNw=.3f61b124-ad65-4443-a247-1f647c2c025d@github.com>
Message-ID: <bWNRyPz4-NPGcXDVoyHuJQCO1xsXpH1sqFG8xOJNB1M=.163a4e53-e087-4238-aeb4-d97524d4c753@github.com>

On Fri, 5 Feb 2021 15:13:01 GMT, David Simms <dsimms at openjdk.org> wrote:

>> Hi all,
>> 
>> Can I have a review of this trivial fix?
>> 
>> The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.
>> 
>> Thanks,
>> Yang Yi
>
> Looks good, thanks for the find

Thank you all for your review? @MrSimms @RogerRiggs @mlchung

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

PR: https://git.openjdk.java.net/valhalla/pull/321

From github.com+5010047+kelthuzadx at openjdk.java.net  Mon Feb  8 08:35:55 2021
From: github.com+5010047+kelthuzadx at openjdk.java.net (Yang Yi)
Date: Mon, 8 Feb 2021 08:35:55 GMT
Subject: [lworld] Integrated: 8261021: [lworld]
 valhalla/valuetypes/LambdaConversion.java fails Illegal class name
 "QLambdaConversion$Pointer; "
In-Reply-To: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
References: <dFSc6YicLArJoL5HqMYzYCeworZD66z7KYZFxyPX2fU=.4af32192-5308-44ef-8b12-6824755b8cd3@github.com>
Message-ID: <CT5YpBvc1pyQOi6YC0NRsxQ_yM5ebcEtGL5B0QbAThc=.0fc5b617-e154-4818-a669-376dd131f359@github.com>

On Fri, 5 Feb 2021 03:24:10 GMT, Yang Yi <github.com+5010047+kelthuzadx at openjdk.org> wrote:

> Hi all,
> 
> Can I have a review of this trivial fix?
> 
> The root cause of this problem is similar to PR #320. As explained before, in short, the major version of the dynamically generated bytecode does not match the version required by ClassFileParser::verify_legal_class_name, resulting in a ClassFormatError.
> 
> Thanks,
> Yang Yi

This pull request has now been integrated.

Changeset: b2c84c74
Author:    Yang Yi <qingfeng.yy at alibaba-inc.com>
Committer: David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/b2c84c74
Stats:     2 lines in 1 file changed: 0 ins; 0 del; 2 mod

8261021: [lworld] valhalla/valuetypes/LambdaConversion.java fails Illegal class name "QLambdaConversion$Pointer;"

Reviewed-by: dsimms, rriggs, mchung

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

PR: https://git.openjdk.java.net/valhalla/pull/321

From sadayapalam at openjdk.java.net  Mon Feb  8 12:37:40 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 8 Feb 2021 12:37:40 GMT
Subject: Integrated: 8260870: [type-restrictions] Generate RestrictedMethod
 attributes
Message-ID: <xe-Js30Tuil-MD7QyeCahBshd9chby0rl9hnrs0QMYM=.d9a28fef-9323-4aab-9f0e-07030205c475@github.com>

Support for generation of RestrictedMethod attribute both via command line option and type annotations.

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

Commit messages:
 - 8260870: [type-restrictions] Generate RestrictedMethod attributes

Changes: https://git.openjdk.java.net/valhalla/pull/325/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=325&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8260870
  Stats: 504 lines in 14 files changed: 502 ins; 0 del; 2 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/325.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/325/head:pull/325

PR: https://git.openjdk.java.net/valhalla/pull/325

From sadayapalam at openjdk.java.net  Mon Feb  8 12:37:41 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 8 Feb 2021 12:37:41 GMT
Subject: Integrated: 8260870: [type-restrictions] Generate RestrictedMethod
 attributes
In-Reply-To: <xe-Js30Tuil-MD7QyeCahBshd9chby0rl9hnrs0QMYM=.d9a28fef-9323-4aab-9f0e-07030205c475@github.com>
References: <xe-Js30Tuil-MD7QyeCahBshd9chby0rl9hnrs0QMYM=.d9a28fef-9323-4aab-9f0e-07030205c475@github.com>
Message-ID: <yvN417RzUfTebOFa0VghZ-ASm8rhC_EmXtEv0hx4bgA=.523119ea-3194-42c7-bfb4-8c80d327feda@github.com>

On Mon, 8 Feb 2021 12:30:45 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> Support for generation of RestrictedMethod attribute both via command line option and type annotations.

This pull request has now been integrated.

Changeset: d7af04e6
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/d7af04e6
Stats:     504 lines in 14 files changed: 502 ins; 0 del; 2 mod

8260870: [type-restrictions] Generate RestrictedMethod attributes

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

PR: https://git.openjdk.java.net/valhalla/pull/325

From dsimms at openjdk.java.net  Mon Feb  8 16:13:29 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Mon, 8 Feb 2021 16:13:29 GMT
Subject: RFR: Merge lworld
Message-ID: <6jrMay0kq9W8fSMbKvjTYrL_tlkjibLLXi9BLa2fN_Y=.9a91636f-e177-46fd-a2ad-46420678cf09@github.com>

Merge lworld at jdk-17+8

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

Commit messages:
 - Merge lworld
 - 8261021: [lworld] valhalla/valuetypes/LambdaConversion.java fails Illegal class name "QLambdaConversion$Pointer;"
 - 8261019: [lworld] valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;" in class file $Proxy5"
 - 8261264: [lworld] runtime/valhalla/inlinetypes tests fail to compile
 - 8260606: Update Valhalla core-libs naming for methods related to primitive classes
 - 8261226: [lworld] Array access profiling should be disabled at compilation level 1 and 2
 - 8261037: [lworld] Assert during C2 compilation due to inconsistent JVMState at safepoint
 - Merge jdk
 - 8260301: misc gc/g1/unloading tests fails with "RuntimeException: Method could not be enqueued for compilation at level N"
 - 8257858: [macOS]: Remove JNF dependency from libosxsecurity/KeystoreImpl.m
 - ... and 983 more: https://git.openjdk.java.net/valhalla/compare/d7af04e6...81297001

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - type-restrictions: https://webrevs.openjdk.java.net/?repo=valhalla&pr=326&range=00.0
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=326&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/326/files
  Stats: 208176 lines in 4372 files changed: 101136 ins; 64664 del; 42376 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/326.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/326/head:pull/326

PR: https://git.openjdk.java.net/valhalla/pull/326

From dsimms at openjdk.java.net  Mon Feb  8 16:32:51 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Mon, 8 Feb 2021 16:32:51 GMT
Subject: Integrated: Merge lworld
In-Reply-To: <6jrMay0kq9W8fSMbKvjTYrL_tlkjibLLXi9BLa2fN_Y=.9a91636f-e177-46fd-a2ad-46420678cf09@github.com>
References: <6jrMay0kq9W8fSMbKvjTYrL_tlkjibLLXi9BLa2fN_Y=.9a91636f-e177-46fd-a2ad-46420678cf09@github.com>
Message-ID: <PZBQknqPoMkGV_ZnhkJZF4YjhbZAiFhzejoKDj9mukc=.244cb442-0dd1-48ab-89b1-4a4aab0ecbc1@github.com>

On Mon, 8 Feb 2021 15:35:18 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge lworld at jdk-17+8

This pull request has now been integrated.

Changeset: 5f6ef0f4
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/5f6ef0f4
Stats:     208176 lines in 4372 files changed: 101136 ins; 64664 del; 42376 mod

Merge lworld

Merge lworld at jdk-17+8

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

PR: https://git.openjdk.java.net/valhalla/pull/326

From dsimms at openjdk.java.net  Tue Feb  9 10:37:47 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Tue, 9 Feb 2021 10:37:47 GMT
Subject: [lworld] RFR: 8241894: [lworld] crash in ZGC running
 InlineOops.java#id3
Message-ID: <ezQTR_hcOs5WRotx5ASm3wbk7QsuGdVtvd8eL0s_9IU=.437931c5-54d2-4f1a-9c4d-3d10f40dd12d@github.com>

Remove raw oop load from CollectOops::do_oop()

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

Commit messages:
 - 8241894: [lworld] crash in ZGC running InlineOops.java#id3

Changes: https://git.openjdk.java.net/valhalla/pull/327/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=327&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8241894
  Stats: 33 lines in 2 files changed: 2 ins; 7 del; 24 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/327.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/327/head:pull/327

PR: https://git.openjdk.java.net/valhalla/pull/327

From sadayapalam at openjdk.java.net  Tue Feb  9 11:47:22 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Tue, 9 Feb 2021 11:47:22 GMT
Subject: [lworld] Integrated: 8261393: [lworld] Adopt `primitive' as the
 modifier in declaration of identity free classes in lieu of `inline`
Message-ID: <C8X_3PqZYDkFiCtcdbSvmubgHxkId0pqc9Ydm5ihMBM=.eb76d4c1-0a4a-408f-a4ae-31f687055ff4@github.com>

Compiler changes to accept primitive as the modifier in class declaration plus test changes across components necessitated by said change.

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

Commit messages:
 - Fix white space problems
 - Fix white space problems
 - 8261393: [lworld] Adopt `primitive' as the modifier in declaration of identity free classes in lieu of `inline`

Changes: https://git.openjdk.java.net/valhalla/pull/328/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=328&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261393
  Stats: 678 lines in 275 files changed: 40 ins; 41 del; 597 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/328.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/328/head:pull/328

PR: https://git.openjdk.java.net/valhalla/pull/328

From sadayapalam at openjdk.java.net  Tue Feb  9 11:47:25 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Tue, 9 Feb 2021 11:47:25 GMT
Subject: [lworld] Integrated: 8261393: [lworld] Adopt `primitive' as the
 modifier in declaration of identity free classes in lieu of `inline`
In-Reply-To: <C8X_3PqZYDkFiCtcdbSvmubgHxkId0pqc9Ydm5ihMBM=.eb76d4c1-0a4a-408f-a4ae-31f687055ff4@github.com>
References: <C8X_3PqZYDkFiCtcdbSvmubgHxkId0pqc9Ydm5ihMBM=.eb76d4c1-0a4a-408f-a4ae-31f687055ff4@github.com>
Message-ID: <cTMAccMKwO7BSO077xN-Ce0eCkOFDTbcW49RMhxYTC8=.94ed4e8b-10fd-4938-9607-15e0e48e23ec@github.com>

On Tue, 9 Feb 2021 11:35:41 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> Compiler changes to accept primitive as the modifier in class declaration plus test changes across components necessitated by said change.

This pull request has now been integrated.

Changeset: 693282f3
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/693282f3
Stats:     678 lines in 275 files changed: 40 ins; 41 del; 597 mod

8261393: [lworld] Adopt `primitive' as the modifier in declaration of identity free classes in lieu of `inline`

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

PR: https://git.openjdk.java.net/valhalla/pull/328

From srikanth.adayapalam at oracle.com  Tue Feb  9 12:02:09 2021
From: srikanth.adayapalam at oracle.com (Srikanth Adayapalam)
Date: Tue, 9 Feb 2021 12:02:09 +0000
Subject: [lworld] Integrated: 8261393: [lworld] Adopt `primitive' as the
 modifier in declaration of identity free classes in lieu of `inline`
In-Reply-To: <cTMAccMKwO7BSO077xN-Ce0eCkOFDTbcW49RMhxYTC8=.94ed4e8b-10fd-4938-9607-15e0e48e23ec@github.com>
References: <C8X_3PqZYDkFiCtcdbSvmubgHxkId0pqc9Ydm5ihMBM=.eb76d4c1-0a4a-408f-a4ae-31f687055ff4@github.com>,
 <cTMAccMKwO7BSO077xN-Ce0eCkOFDTbcW49RMhxYTC8=.94ed4e8b-10fd-4938-9607-15e0e48e23ec@github.com>
Message-ID: <MN2PR10MB355243DC077BCCCE672BA5FE908E9@MN2PR10MB3552.namprd10.prod.outlook.com>

Hello!

Please take note of this push. It is not inline class anymore, but primitive class as in


primitive class Point implements Shape {
    private double x;
    private double y;

    public Point(double x, double y) {
        this.x = x;
        this.y = y;
    }

    public double x() { return x; }
    public double y() { return y; }

    public Point translate(double dx, double dy) {
        return new Point(x+dx, y+dy);
    }

    public boolean contains(Point p) {
        return equals(p);
    }
}


I have adjusted all the tests across components. This change merely adjusts the declaration sites.
It is likely that comments in source and tests, APIs internal and external, identifiers, documentation,
various experimental command line options etc still refer to inline classes. Respective component owners
should follow up.

Javac messages are not adjusted yet and will refer to inline types. I will follow up.

Srikanth
________________________________
From: valhalla-dev <valhalla-dev-retn at openjdk.java.net> on behalf of Srikanth Adayapalam <sadayapalam at openjdk.java.net>
Sent: 09 February 2021 17:17
To: valhalla-dev at openjdk.java.net <valhalla-dev at openjdk.java.net>
Subject: [lworld] Integrated: 8261393: [lworld] Adopt `primitive' as the modifier in declaration of identity free classes in lieu of `inline`

On Tue, 9 Feb 2021 11:35:41 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> Compiler changes to accept primitive as the modifier in class declaration plus test changes across components necessitated by said change.

This pull request has now been integrated.

Changeset: 693282f3
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/693282f3
Stats:     678 lines in 275 files changed: 40 ins; 41 del; 597 mod

8261393: [lworld] Adopt `primitive' as the modifier in declaration of identity free classes in lieu of `inline`

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

PR: https://git.openjdk.java.net/valhalla/pull/328

From dsimms at openjdk.java.net  Tue Feb  9 13:34:05 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Tue, 9 Feb 2021 13:34:05 GMT
Subject: [lworld] RFR: 8261439: [lworld] update use of 'inline class' to
 'primitive class' in java/lang/invoke and com/sun/jdi/ tests
Message-ID: <zGEhGEf9_6JqLTJ5rfWJ180D0r9Bm5DP6sUbXuCsQrU=.b3c53e5a-f426-42cf-8a94-897e640e042a@github.com>

Update 'inline' to 'primitive'

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

Commit messages:
 - Indentation with tabs
 - 8261439: [lworld] update use of 'inline class' to 'primitive class' in java/lang/invoke and com/sun/jdi/ tests

Changes: https://git.openjdk.java.net/valhalla/pull/329/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=329&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261439
  Stats: 62 lines in 6 files changed: 6 ins; 6 del; 50 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/329.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/329/head:pull/329

PR: https://git.openjdk.java.net/valhalla/pull/329

From dsimms at openjdk.java.net  Tue Feb  9 13:55:32 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Tue, 9 Feb 2021 13:55:32 GMT
Subject: git: openjdk/valhalla: lworld: 8261439: [lworld] update use of
 "inline class" to "primitive class" in java/lang/invoke and com/sun/jdi/
 tests
Message-ID: <df87240b-5ede-442c-bafe-df38543cbe71@openjdk.org>

Changeset: 3d47537a
Author:    David Simms <dsimms at openjdk.org>
Date:      2021-02-09 13:55:09 +0000
URL:       https://git.openjdk.java.net/valhalla/commit/3d47537a

8261439: [lworld] update use of "inline class" to "primitive class" in java/lang/invoke and com/sun/jdi/ tests

! test/hotspot/jtreg/runtime/cds/appcds/test-classes/HelloInlineClassApp.java
! test/hotspot/jtreg/runtime/cds/appcds/test-classes/Point.java
! test/hotspot/jtreg/vmTestbase/nsk/jvmti/GetObjectMonitorUsage/objmonusage007.java
! test/jdk/com/sun/jdi/JdbInlineTypesTest.java
! test/jdk/java/lang/invoke/VarHandles/Point.java
! test/jdk/java/lang/invoke/VarHandles/Value.java


From dsimms at openjdk.java.net  Tue Feb  9 13:57:47 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Tue, 9 Feb 2021 13:57:47 GMT
Subject: [lworld] Integrated: 8261439: [lworld] update use of "inline class"
 to "primitive class" in java/lang/invoke and com/sun/jdi/ tests
In-Reply-To: <zGEhGEf9_6JqLTJ5rfWJ180D0r9Bm5DP6sUbXuCsQrU=.b3c53e5a-f426-42cf-8a94-897e640e042a@github.com>
References: <zGEhGEf9_6JqLTJ5rfWJ180D0r9Bm5DP6sUbXuCsQrU=.b3c53e5a-f426-42cf-8a94-897e640e042a@github.com>
Message-ID: <3c9NfdAQzthzm4-3Z_KFjdWaBOcE-6EB30QnNiJV2MA=.9a1acad4-2ed8-40b7-ae65-45bd14e8f16d@github.com>

On Tue, 9 Feb 2021 13:30:11 GMT, David Simms <dsimms at openjdk.org> wrote:

> Update 'inline' to 'primitive'

This pull request has now been integrated.

Changeset: 3d47537a
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/3d47537a
Stats:     62 lines in 6 files changed: 6 ins; 6 del; 50 mod

8261439: [lworld] update use of "inline class" to "primitive class" in java/lang/invoke and com/sun/jdi/ tests

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

PR: https://git.openjdk.java.net/valhalla/pull/329

From fparain at openjdk.java.net  Tue Feb  9 15:06:56 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Tue, 9 Feb 2021 15:06:56 GMT
Subject: [lworld] RFR: 8241894: [lworld] crash in ZGC running
 InlineOops.java#id3
In-Reply-To: <ezQTR_hcOs5WRotx5ASm3wbk7QsuGdVtvd8eL0s_9IU=.437931c5-54d2-4f1a-9c4d-3d10f40dd12d@github.com>
References: <ezQTR_hcOs5WRotx5ASm3wbk7QsuGdVtvd8eL0s_9IU=.437931c5-54d2-4f1a-9c4d-3d10f40dd12d@github.com>
Message-ID: <76zQXm32PSgmYtoMcurvWZJv8KcFpqiYOciGEX1nd7k=.a9626817-dcfb-4270-a1b0-22b46a36b058@github.com>

On Tue, 9 Feb 2021 10:34:12 GMT, David Simms <dsimms at openjdk.org> wrote:

> Remove raw oop load from CollectOops::do_oop()

Looks good to me.
Thank you for fixing this long standing issue.

Fred

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

Marked as reviewed by fparain (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/327

From dsimms at openjdk.java.net  Tue Feb  9 15:06:57 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Tue, 9 Feb 2021 15:06:57 GMT
Subject: [lworld] Integrated: 8241894: [lworld] crash in ZGC running
 InlineOops.java#id3
In-Reply-To: <ezQTR_hcOs5WRotx5ASm3wbk7QsuGdVtvd8eL0s_9IU=.437931c5-54d2-4f1a-9c4d-3d10f40dd12d@github.com>
References: <ezQTR_hcOs5WRotx5ASm3wbk7QsuGdVtvd8eL0s_9IU=.437931c5-54d2-4f1a-9c4d-3d10f40dd12d@github.com>
Message-ID: <SAQhjjgJ1b0P5up9hikFBVcR374EuTj1R7y3TL3JbFI=.f30ff530-e616-4b7d-aabe-feb3f24ff217@github.com>

On Tue, 9 Feb 2021 10:34:12 GMT, David Simms <dsimms at openjdk.org> wrote:

> Remove raw oop load from CollectOops::do_oop()

This pull request has now been integrated.

Changeset: 84c3e324
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/84c3e324
Stats:     33 lines in 2 files changed: 2 ins; 7 del; 24 mod

8241894: [lworld] crash in ZGC running InlineOops.java#id3

Reviewed-by: fparain

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

PR: https://git.openjdk.java.net/valhalla/pull/327

From dsimms at openjdk.java.net  Thu Feb 11 20:15:28 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 11 Feb 2021 20:15:28 GMT
Subject: [lworld] RFR: Merge jdk
Message-ID: <wFzabRc5wvtBT1UZRqI6JIfjMv6LpHRMRwJpBDQKmjk=.3ec39120-4920-436e-8971-99ff40ba4137@github.com>

Merge tag 'jdk-17+9' into lworld_merge_jdk_17_9

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

Commit messages:
 - Logical merge compile fixes
 - Merge tag 'jdk-17+9' into lworld_merge_jdk_17_9
 - 8260934: java/lang/StringBuilder/HugeCapacity.java fails without Compact Strings
 - 6211242: AreaAveragingScaleFilter(int, int): IAE is not specified
 - 6211257: BasicStroke.createStrokedShape(Shape): NPE is not specified
 - 8261472: BasicConstraintsExtension::toString shows "PathLen:2147483647" if there is no pathLenConstraint
 - 8223188: Removed unnecessary #ifdef __cplusplus from .cpp sources
 - 8259656: fixpath.sh changes broke _NT_SYMBOL_PATH in RunTests.gmk
 - 8225081: Remove Telia Company CA certificate expiring in April 2021
 - 8261209: isStandalone property: remove dependency on pretty-print
 - ... and 91 more: https://git.openjdk.java.net/valhalla/compare/84c3e324...c17673a8

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=330&range=00.0
 - jdk: https://webrevs.openjdk.java.net/?repo=valhalla&pr=330&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/330/files
  Stats: 21441 lines in 680 files changed: 13674 ins; 5372 del; 2395 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/330.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/330/head:pull/330

PR: https://git.openjdk.java.net/valhalla/pull/330

From dsimms at openjdk.java.net  Thu Feb 11 21:05:32 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 11 Feb 2021 21:05:32 GMT
Subject: [lworld] RFR: Merge jdk [v2]
In-Reply-To: <wFzabRc5wvtBT1UZRqI6JIfjMv6LpHRMRwJpBDQKmjk=.3ec39120-4920-436e-8971-99ff40ba4137@github.com>
References: <wFzabRc5wvtBT1UZRqI6JIfjMv6LpHRMRwJpBDQKmjk=.3ec39120-4920-436e-8971-99ff40ba4137@github.com>
Message-ID: <zto5AOaqhdGnTEh4gNq2X1z5bV2Rf1R55M4F78BceAQ=.1c60c3a3-0730-4526-a20e-355542ea23b9@github.com>

> Merge tag 'jdk-17+9' into lworld_merge_jdk_17_9

David Simms has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 1093 commits:

 - Logical merge compile fixes
 - Merge tag 'jdk-17+9' into lworld_merge_jdk_17_9
   
   Added tag jdk-17+9 for changeset ad54d8dd
   
   # Conflicts:
   #	src/hotspot/share/runtime/jniHandles.cpp
   #	src/hotspot/share/runtime/reflection.cpp
   #	test/hotspot/jtreg/runtime/cds/appcds/ArchiveRelocationTest.java
 - 8241894: [lworld] crash in ZGC running InlineOops.java#id3
   
   Reviewed-by: fparain
 - 8261439: [lworld] update use of "inline class" to "primitive class" in java/lang/invoke and com/sun/jdi/ tests
 - 8261393: [lworld] Adopt `primitive' as the modifier in declaration of identity free classes in lieu of `inline`
 - 8261021: [lworld] valhalla/valuetypes/LambdaConversion.java fails Illegal class name "QLambdaConversion$Pointer;"
   
   Reviewed-by: dsimms, rriggs, mchung
 - 8261019: [lworld] valhalla/valuetypes/QTypeDescriptorTest.java "Illegal class name "QLine;" in class file $Proxy5"
   
   Reviewed-by: mchung
 - 8261264: [lworld] runtime/valhalla/inlinetypes tests fail to compile
   
   Reviewed-by: rriggs
 - 8260606: Update Valhalla core-libs naming for methods related to primitive classes
   
   Reviewed-by: mchung
 - 8261226: [lworld] Array access profiling should be disabled at compilation level 1 and 2
 - ... and 1083 more: https://git.openjdk.java.net/valhalla/compare/ad54d8dd...c17673a8

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

Changes: https://git.openjdk.java.net/valhalla/pull/330/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=330&range=01
  Stats: 144143 lines in 1307 files changed: 137514 ins; 2040 del; 4589 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/330.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/330/head:pull/330

PR: https://git.openjdk.java.net/valhalla/pull/330

From dsimms at openjdk.java.net  Thu Feb 11 21:05:34 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 11 Feb 2021 21:05:34 GMT
Subject: [lworld] Integrated: Merge jdk
In-Reply-To: <wFzabRc5wvtBT1UZRqI6JIfjMv6LpHRMRwJpBDQKmjk=.3ec39120-4920-436e-8971-99ff40ba4137@github.com>
References: <wFzabRc5wvtBT1UZRqI6JIfjMv6LpHRMRwJpBDQKmjk=.3ec39120-4920-436e-8971-99ff40ba4137@github.com>
Message-ID: <0CTSaIy1U-bxn8JSlv2doYwalrEBw9qTIDJmsVYvGEY=.0cb107f2-3ced-4f74-8366-609e30e9e45b@github.com>

On Thu, 11 Feb 2021 20:09:48 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge tag 'jdk-17+9' into lworld_merge_jdk_17_9

This pull request has now been integrated.

Changeset: ac814eac
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/ac814eac
Stats:     21441 lines in 680 files changed: 13674 ins; 5372 del; 2395 mod

Merge jdk

Merge tag 'jdk-17+9'

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

PR: https://git.openjdk.java.net/valhalla/pull/330

From thartmann at openjdk.java.net  Fri Feb 12 08:49:58 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 08:49:58 GMT
Subject: [lworld] RFR: 8261631: [lworld] Fatal error in C1 compiled code due
 to unexpected klass
Message-ID: <SIKJC181B9phFDcG4ixiMQ_R6I45J_gBI5ysoQExWwc=.1d8eecad-4390-40bf-a46c-0776ca5ecc3d@github.com>

An assertion in C1 compiled profiling code triggers because the actual (exact) klass of a field is not as expected. The problem should have been solved by [JDK-8260225](https://bugs.openjdk.java.net/browse/JDK-8260225) but was only partially fixed due to a typo (`!=` vs. `==`). I've also noticed that we are profiling inline types for null, which should be avoided.

Best regards,
Tobias

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

Commit messages:
 - 8261631: [lworld] Fatal error in C1 compiled code due to unexpected klass

Changes: https://git.openjdk.java.net/valhalla/pull/331/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=331&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261631
  Stats: 11 lines in 3 files changed: 10 ins; 0 del; 1 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/331.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/331/head:pull/331

PR: https://git.openjdk.java.net/valhalla/pull/331

From thartmann at openjdk.java.net  Fri Feb 12 09:16:08 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 09:16:08 GMT
Subject: [lworld] RFR: 8261633: [lworld] TestLWorld::test10 fails IR
 verification
Message-ID: <B5PDfOdAn0GnntVTNcAA7lazU30kaaK5Rbr-TZsGC7Y=.3508158f-667e-41e4-a00e-6905bbb56edd@github.com>

IR verification fails because there is an unexpected inline type allocation in the code. This is a test bug. The root cause is a missing `@ForceInline` annotation that makes sure the `test10_helper` method is always inlined such that the allocation can be avoided.

Best regards,
Tobias

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

Commit messages:
 - 8261633: [lworld] TestLWorld::test10 fails IR verification

Changes: https://git.openjdk.java.net/valhalla/pull/332/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=332&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261633
  Stats: 1 line in 1 file changed: 1 ins; 0 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/332.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/332/head:pull/332

PR: https://git.openjdk.java.net/valhalla/pull/332

From thartmann at openjdk.java.net  Fri Feb 12 09:23:49 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 09:23:49 GMT
Subject: [lworld] Integrated: 8261633: [lworld] TestLWorld::test10 fails IR
 verification
In-Reply-To: <B5PDfOdAn0GnntVTNcAA7lazU30kaaK5Rbr-TZsGC7Y=.3508158f-667e-41e4-a00e-6905bbb56edd@github.com>
References: <B5PDfOdAn0GnntVTNcAA7lazU30kaaK5Rbr-TZsGC7Y=.3508158f-667e-41e4-a00e-6905bbb56edd@github.com>
Message-ID: <GdkTKm9qtG-L2TUMB0QbIWSniccnNrYgCE9WH5Wy-n0=.b6149975-a1fc-4a28-add5-0cc36afaf1f4@github.com>

On Fri, 12 Feb 2021 09:12:00 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> IR verification fails because there is an unexpected inline type allocation in the code. This is a test bug. The root cause is a missing `@ForceInline` annotation that makes sure the `test10_helper` method is always inlined such that the allocation can be avoided.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: f1d50a1a
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/f1d50a1a
Stats:     1 line in 1 file changed: 1 ins; 0 del; 0 mod

8261633: [lworld] TestLWorld::test10 fails IR verification

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

PR: https://git.openjdk.java.net/valhalla/pull/332

From thartmann at openjdk.java.net  Fri Feb 12 10:20:57 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 10:20:57 GMT
Subject: [lworld] Integrated: 8261631: [lworld] Fatal error in C1 compiled
 code due to unexpected klass
In-Reply-To: <SIKJC181B9phFDcG4ixiMQ_R6I45J_gBI5ysoQExWwc=.1d8eecad-4390-40bf-a46c-0776ca5ecc3d@github.com>
References: <SIKJC181B9phFDcG4ixiMQ_R6I45J_gBI5ysoQExWwc=.1d8eecad-4390-40bf-a46c-0776ca5ecc3d@github.com>
Message-ID: <LAKgQwi9obdv03PDKBTwY21rO63yO_t-6JlWNOI1JzQ=.1d60f7c4-8dbd-42da-8e07-f40ea0dd33c0@github.com>

On Fri, 12 Feb 2021 08:46:17 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> An assertion in C1 compiled profiling code triggers because the actual (exact) klass of a field is not as expected. The problem should have been solved by [JDK-8260225](https://bugs.openjdk.java.net/browse/JDK-8260225) but was only partially fixed due to a typo (`!=` vs. `==`). I've also noticed that we are profiling inline types for null, which should be avoided.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 4d953ffa
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/4d953ffa
Stats:     11 lines in 3 files changed: 10 ins; 0 del; 1 mod

8261631: [lworld] Fatal error in C1 compiled code due to unexpected klass

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

PR: https://git.openjdk.java.net/valhalla/pull/331

From thartmann at openjdk.java.net  Fri Feb 12 10:38:00 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 10:38:00 GMT
Subject: [lworld] RFR: 8261156: [lworld] Re-enable C1 testing after mainline
 issues are fixed
Message-ID: <yOl7ZSifnw3A5-8eKxWQTAfpfP4iC-W3xPTxC40-GOo=.82310b2f-f7aa-4862-bbde-7d7884368543@github.com>

Explicitly set `TieredStopAtLevel=4` in `TestLWorldProfiling` to avoid hitting mainline bug [JDK-8261225](https://bugs.openjdk.java.net/browse/JDK-8261225) and allow re-enabling C1 testing in the CI. I've also removed an unused flag from the framework.

Best regards,
Tobias

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

Commit messages:
 - Workaround
 - 8261156: [lworld] Re-enable C1 testing after mainline issues are fixed

Changes: https://git.openjdk.java.net/valhalla/pull/333/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=333&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261156
  Stats: 13 lines in 2 files changed: 3 ins; 8 del; 2 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/333.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/333/head:pull/333

PR: https://git.openjdk.java.net/valhalla/pull/333

From thartmann at openjdk.java.net  Fri Feb 12 12:14:55 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 12 Feb 2021 12:14:55 GMT
Subject: [lworld] Integrated: 8261156: [lworld] Re-enable C1 testing after
 mainline issues are fixed
In-Reply-To: <yOl7ZSifnw3A5-8eKxWQTAfpfP4iC-W3xPTxC40-GOo=.82310b2f-f7aa-4862-bbde-7d7884368543@github.com>
References: <yOl7ZSifnw3A5-8eKxWQTAfpfP4iC-W3xPTxC40-GOo=.82310b2f-f7aa-4862-bbde-7d7884368543@github.com>
Message-ID: <SyseKGMDeEcwAErVKhdGojjsS2kEzwQD-v_VW13OP-o=.e46793bc-7b2d-40d2-8922-75f8f43489fa@github.com>

On Fri, 12 Feb 2021 10:33:13 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> Explicitly set `TieredStopAtLevel=4` in `TestLWorldProfiling` to avoid hitting mainline bug [JDK-8261225](https://bugs.openjdk.java.net/browse/JDK-8261225) and allow re-enabling C1 testing in the CI. I've also removed an unused flag from the framework.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 9e577c0e
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/9e577c0e
Stats:     13 lines in 2 files changed: 3 ins; 8 del; 2 mod

8261156: [lworld] Re-enable C1 testing after mainline issues are fixed

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

PR: https://git.openjdk.java.net/valhalla/pull/333

From sadayapalam at openjdk.java.net  Mon Feb 15 07:40:29 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 15 Feb 2021 07:40:29 GMT
Subject: [lworld] Integrated: 8261721: [lworld] Javac messages still refer to
 inline types
Message-ID: <Pyk3deh4WcJWzYdNd958DQtOxSBtvgBqt_gOl1d2_VA=.84c340f0-712a-44cb-bb34-869c9d328fda@github.com>

Fix message keys and text; adjust sources and tests accordingly.

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

Commit messages:
 - 8261721: [lworld] Javac messages still refer to inline types

Changes: https://git.openjdk.java.net/valhalla/pull/334/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=334&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261721
  Stats: 135 lines in 27 files changed: 10 ins; 10 del; 115 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/334.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/334/head:pull/334

PR: https://git.openjdk.java.net/valhalla/pull/334

From sadayapalam at openjdk.java.net  Mon Feb 15 07:40:30 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 15 Feb 2021 07:40:30 GMT
Subject: [lworld] Integrated: 8261721: [lworld] Javac messages still refer to
 inline types
In-Reply-To: <Pyk3deh4WcJWzYdNd958DQtOxSBtvgBqt_gOl1d2_VA=.84c340f0-712a-44cb-bb34-869c9d328fda@github.com>
References: <Pyk3deh4WcJWzYdNd958DQtOxSBtvgBqt_gOl1d2_VA=.84c340f0-712a-44cb-bb34-869c9d328fda@github.com>
Message-ID: <dh80sPiCYzo4zLvqE1ULKKrHVQMsNNjkZjsv6v5Gt6A=.133f6998-4232-49af-bda7-700d8281bc12@github.com>

On Mon, 15 Feb 2021 07:33:09 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> Fix message keys and text; adjust sources and tests accordingly.

This pull request has now been integrated.

Changeset: dad7d191
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/dad7d191
Stats:     135 lines in 27 files changed: 10 ins; 10 del; 115 mod

8261721: [lworld] Javac messages still refer to inline types

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

PR: https://git.openjdk.java.net/valhalla/pull/334

From thartmann at openjdk.java.net  Mon Feb 15 12:13:14 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 15 Feb 2021 12:13:14 GMT
Subject: [lworld] RFR: 8261647: [lworld] Missing default initialization of
 non-flattened field of empty inline type
Message-ID: <Hco3kpIBkzZ1XHSgeYmf5ijTxfjKKi5kD9xiWCwyX0o=.0e48a54d-643c-4dc4-8d0e-2d45f90819c0@github.com>

After [JDK-8253893](https://bugs.openjdk.java.net/browse/JDK-8253893) has been fixed, I've re-enabled some compiler tests for empty inline types and noticed that they are failing. The problem is that both during slow path array access through `OptoRuntime::load_unknown_inline()` and when re-allocating during deoptimization, non-flattened fields of an empty inline type are not properly initialized with the default value.

Best regards,
Tobias

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

Commit messages:
 - 8261647: [lworld] Missing default initialization of non-flattened field of empty inline type

Changes: https://git.openjdk.java.net/valhalla/pull/335/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=335&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261647
  Stats: 65 lines in 4 files changed: 24 ins; 28 del; 13 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/335.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/335/head:pull/335

PR: https://git.openjdk.java.net/valhalla/pull/335

From thartmann at openjdk.java.net  Mon Feb 15 12:26:48 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 15 Feb 2021 12:26:48 GMT
Subject: [lworld] Integrated: 8261647: [lworld] Missing default initialization
 of non-flattened field of empty inline type
In-Reply-To: <Hco3kpIBkzZ1XHSgeYmf5ijTxfjKKi5kD9xiWCwyX0o=.0e48a54d-643c-4dc4-8d0e-2d45f90819c0@github.com>
References: <Hco3kpIBkzZ1XHSgeYmf5ijTxfjKKi5kD9xiWCwyX0o=.0e48a54d-643c-4dc4-8d0e-2d45f90819c0@github.com>
Message-ID: <S-1jCBaLd4f2IpS1FtlOjxA6jR9AztTQBoBFO6J9c4w=.f1070499-2187-4e2e-9fce-2632c26efc02@github.com>

On Mon, 15 Feb 2021 12:09:21 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> After [JDK-8253893](https://bugs.openjdk.java.net/browse/JDK-8253893) has been fixed, I've re-enabled some compiler tests for empty inline types and noticed that they are failing. The problem is that both during slow path array access through `OptoRuntime::load_unknown_inline()` and when re-allocating during deoptimization, non-flattened fields of an empty inline type are not properly initialized with the default value.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: d4ce0d4e
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/d4ce0d4e
Stats:     65 lines in 4 files changed: 24 ins; 28 del; 13 mod

8261647: [lworld] Missing default initialization of non-flattened field of empty inline type

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

PR: https://git.openjdk.java.net/valhalla/pull/335

From sadayapalam at openjdk.java.net  Mon Feb 15 14:10:07 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 15 Feb 2021 14:10:07 GMT
Subject: [lworld] Integrated: 8261751: [lworld] Fix lingering references to
 inline/value types
Message-ID: <SF2o1kNLaE7RTlbAwGjAo6ljV0b0nqORC97y_zuGm4o=.eb2271e6-9eba-49e7-9890-555c79562c1b@github.com>

Adopt new nomenclature from Primitive Objects JEP

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

Commit messages:
 - 8261751: [lworld] Fix lingering references to inline/value types

Changes: https://git.openjdk.java.net/valhalla/pull/336/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=336&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261751
  Stats: 158 lines in 36 files changed: 0 ins; 0 del; 158 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/336.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/336/head:pull/336

PR: https://git.openjdk.java.net/valhalla/pull/336

From sadayapalam at openjdk.java.net  Mon Feb 15 14:10:07 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 15 Feb 2021 14:10:07 GMT
Subject: [lworld] Integrated: 8261751: [lworld] Fix lingering references to
 inline/value types
In-Reply-To: <SF2o1kNLaE7RTlbAwGjAo6ljV0b0nqORC97y_zuGm4o=.eb2271e6-9eba-49e7-9890-555c79562c1b@github.com>
References: <SF2o1kNLaE7RTlbAwGjAo6ljV0b0nqORC97y_zuGm4o=.eb2271e6-9eba-49e7-9890-555c79562c1b@github.com>
Message-ID: <h8HqdkKF0Di-c9KDIVIpYCvDx62OXoAvZXUagYVVJM8=.269780ae-5403-4632-a3a0-bb39675bec69@github.com>

On Mon, 15 Feb 2021 14:05:01 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> Adopt new nomenclature from Primitive Objects JEP

This pull request has now been integrated.

Changeset: 9a601e0d
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/9a601e0d
Stats:     158 lines in 36 files changed: 0 ins; 0 del; 158 mod

8261751: [lworld] Fix lingering references to inline/value types

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

PR: https://git.openjdk.java.net/valhalla/pull/336

From thartmann at openjdk.java.net  Tue Feb 16 10:34:21 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 16 Feb 2021 10:34:21 GMT
Subject: [lworld] RFR: 8255045: [lworld] LoadNode::Value() use of
 markWord::prototype() may strip type bits
Message-ID: <AObryKldMNj5ZyngO0j2oid8wYNX3tiRlv7eQ7vdUrE=.313458ee-9f93-4669-8795-cd6448cfec81@github.com>

`LoadNode::Value` detects mark word loads and replaces them by `markWord::prototype()` which would strip inline/flat/null-free property bits. We need to use the prototype header from the klass. When writing tests, I've found an independent problem (due to a typo) in `Parse::do_monitor_enter()`.

Best regards,
Tobias

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

Commit messages:
 - 8255045: [lworld] LoadNode::Value() use of markWord::prototype() may strip type bits

Changes: https://git.openjdk.java.net/valhalla/pull/337/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=337&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8255045
  Stats: 100 lines in 6 files changed: 96 ins; 0 del; 4 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/337.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/337/head:pull/337

PR: https://git.openjdk.java.net/valhalla/pull/337

From thartmann at openjdk.java.net  Tue Feb 16 11:55:00 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 16 Feb 2021 11:55:00 GMT
Subject: [lworld] Integrated: 8255045: [lworld] LoadNode::Value() use of
 markWord::prototype() may strip type bits
In-Reply-To: <AObryKldMNj5ZyngO0j2oid8wYNX3tiRlv7eQ7vdUrE=.313458ee-9f93-4669-8795-cd6448cfec81@github.com>
References: <AObryKldMNj5ZyngO0j2oid8wYNX3tiRlv7eQ7vdUrE=.313458ee-9f93-4669-8795-cd6448cfec81@github.com>
Message-ID: <dDNZrsAbgduJT_Fmpn68tQOYSEVeeN1mIgk0ILMvUG0=.ff02b4cc-b7cf-4e9c-9c30-f45052297e6c@github.com>

On Tue, 16 Feb 2021 10:29:29 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> `LoadNode::Value` detects mark word loads and replaces them by `markWord::prototype()` which would strip inline/flat/null-free property bits. We need to use the prototype header from the klass. When writing tests, I've found an independent problem (due to a typo) in `Parse::do_monitor_enter()`.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: b9d9ca46
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/b9d9ca46
Stats:     100 lines in 6 files changed: 96 ins; 0 del; 4 mod

8255045: [lworld] LoadNode::Value() use of markWord::prototype() may strip type bits

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

PR: https://git.openjdk.java.net/valhalla/pull/337

From rriggs at openjdk.java.net  Tue Feb 16 16:49:04 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Tue, 16 Feb 2021 16:49:04 GMT
Subject: [lworld] RFR: 8261830: [lworld] [test] ArchiveRelocationTest
 compilation failure
Message-ID: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>

Restore a missed merge to fix a compilation failure in ArchiveRelationTest.java.

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

Commit messages:
 - 8261830: [lworld] [test] ArchiveRelocationTest compilation failure

Changes: https://git.openjdk.java.net/valhalla/pull/338/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=338&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261830
  Stats: 8 lines in 1 file changed: 0 ins; 2 del; 6 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/338.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/338/head:pull/338

PR: https://git.openjdk.java.net/valhalla/pull/338

From iklam at openjdk.java.net  Tue Feb 16 17:14:45 2021
From: iklam at openjdk.java.net (Ioi Lam)
Date: Tue, 16 Feb 2021 17:14:45 GMT
Subject: [lworld] RFR: 8261830: [lworld] [test] ArchiveRelocationTest
 compilation failure
In-Reply-To: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
References: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
Message-ID: <LYmB5lDBELUMgx1ZlPkSkDpimaBBwM7trMiwTiRYXcM=.b3b1a5f2-1e11-45ba-b120-5de9d7160b25@github.com>

On Tue, 16 Feb 2021 16:44:45 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Restore a missed merge to fix a compilation failure in ArchiveRelationTest.java.

LGTM

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

Marked as reviewed by iklam (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/338

From rriggs at openjdk.java.net  Tue Feb 16 17:23:18 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Tue, 16 Feb 2021 17:23:18 GMT
Subject: [lworld] RFR: 8261830: [lworld] [test] ArchiveRelocationTest
 compilation failure [v2]
In-Reply-To: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
References: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
Message-ID: <kWkJnWONE7oUMZQgcIyLCVy6frPrhYbtVuB9cVDNdqc=.7be28e3b-5b77-4b37-9c99-45520c6d82fd@github.com>

> Restore a missed merge to fix a compilation failure in ArchiveRelationTest.java.

Roger Riggs has updated the pull request incrementally with one additional commit since the last revision:

  Removed unnecessary check in HelloDyanmicInlineClass

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/338/files
  - new: https://git.openjdk.java.net/valhalla/pull/338/files/582e4e2a..47a28c70

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=338&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=338&range=00-01

  Stats: 2 lines in 1 file changed: 0 ins; 1 del; 1 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/338.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/338/head:pull/338

PR: https://git.openjdk.java.net/valhalla/pull/338

From iklam at openjdk.java.net  Tue Feb 16 17:26:58 2021
From: iklam at openjdk.java.net (Ioi Lam)
Date: Tue, 16 Feb 2021 17:26:58 GMT
Subject: [lworld] RFR: 8261830: [lworld] [test] ArchiveRelocationTest
 compilation failure [v2]
In-Reply-To: <kWkJnWONE7oUMZQgcIyLCVy6frPrhYbtVuB9cVDNdqc=.7be28e3b-5b77-4b37-9c99-45520c6d82fd@github.com>
References: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
 <kWkJnWONE7oUMZQgcIyLCVy6frPrhYbtVuB9cVDNdqc=.7be28e3b-5b77-4b37-9c99-45520c6d82fd@github.com>
Message-ID: <IN7qi5vcQrhcLqKukcpTDqP42FK8SX6DuWXPlD5FcrI=.366dd59a-7b9f-4039-98eb-a0a9e073eec0@github.com>

On Tue, 16 Feb 2021 17:23:18 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

>> Restore a missed merge to fix a compilation failure in ArchiveRelationTest.java.
>
> Roger Riggs has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Removed unnecessary check in HelloDyanmicInlineClass

LGTM

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

Marked as reviewed by iklam (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/338

From rriggs at openjdk.java.net  Tue Feb 16 17:32:51 2021
From: rriggs at openjdk.java.net (Roger Riggs)
Date: Tue, 16 Feb 2021 17:32:51 GMT
Subject: [lworld] Integrated: 8261830: [lworld] [test] ArchiveRelocationTest
 compilation failure
In-Reply-To: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
References: <8eQgVPZZysZnZQS_pYTLkSigEfQRm1iFQKTT06IMVfE=.16887a25-a29a-4ac0-a25e-05713f6eb492@github.com>
Message-ID: <XV6nUU7t2lZu1_wKhHBd-Hx35-Nv_mLqwPcgHMMwnmg=.70581fb7-553f-4bdb-8bd5-775cd6f013d1@github.com>

On Tue, 16 Feb 2021 16:44:45 GMT, Roger Riggs <rriggs at openjdk.org> wrote:

> Restore a missed merge to fix a compilation failure in ArchiveRelationTest.java.

This pull request has now been integrated.

Changeset: 3899e114
Author:    Roger Riggs <rriggs at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/3899e114
Stats:     10 lines in 2 files changed: 0 ins; 3 del; 7 mod

8261830: [lworld] [test] ArchiveRelocationTest compilation failure

Reviewed-by: iklam

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

PR: https://git.openjdk.java.net/valhalla/pull/338

From nlisker at gmail.com  Tue Feb 16 20:48:38 2021
From: nlisker at gmail.com (Nir Lisker)
Date: Tue, 16 Feb 2021 22:48:38 +0200
Subject: Primitive values and void
Message-ID: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>

Hi,

After reading the JEPs about primitive objects (.ref and .val) and the
conversion of primitives, I was wondering what happens with void and Void.
They are not interchangeable like primitives - a method returning void must
have 'return;` while a method returning Void must 'return null;' - but they
are the same semantically. Since Void mainly (only?) exists for generic
types purposes, will void be valid as a generic type and make Void more or
less useless? Will they have reference-value conversions like primitives,
if that makes sense?

- Nir

From brian.goetz at oracle.com  Tue Feb 16 21:15:40 2021
From: brian.goetz at oracle.com (Brian Goetz)
Date: Tue, 16 Feb 2021 16:15:40 -0500
Subject: Primitive values and void
In-Reply-To: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>
References: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>
Message-ID: <ee695b43-656b-1734-7522-fa755f0c31aa@oracle.com>

It's an open question, but one we want to address.? Adjusting the 
language to understand that `void` and `Void` are related is not hard, 
allowing us to ignore the need to return a value.

Turning `void` into a proper (unit) type is harder, but ultimately 
valuable; if you look at the explosion of functional interfaces we added 
in 8, one contributor was all the primitive specializations 
(IntPredicate could be Predicate<int> which in turn could be 
Function<int,boolean>), but another was the fact that `void` is not a 
type (Consumer<T> could be Function<T, void>).? Valhalla will surely 
collapse the first (hurrah), but it would be a shame not to collapse the 
second as well.

There's a certain amount of accidental complexity that will have to be 
compounded to get there; for example, there are "statement lambdas" and 
"expression lambdas" but if void is a type, then a statement lambda 
could also be an expression lambda returning void.? I doubt any of these 
are blockers, but there's a certain amount of un-fun grinding here to 
clear out the underbrush.



On 2/16/2021 3:48 PM, Nir Lisker wrote:
> Hi,
>
> After reading the JEPs about primitive objects (.ref and .val) and the
> conversion of primitives, I was wondering what happens with void and Void.
> They are not interchangeable like primitives - a method returning void must
> have 'return;` while a method returning Void must 'return null;' - but they
> are the same semantically. Since Void mainly (only?) exists for generic
> types purposes, will void be valid as a generic type and make Void more or
> less useless? Will they have reference-value conversions like primitives,
> if that makes sense?
>
> - Nir


From nlisker at gmail.com  Wed Feb 17 02:08:00 2021
From: nlisker at gmail.com (Nir Lisker)
Date: Wed, 17 Feb 2021 04:08:00 +0200
Subject: Primitive values and void
In-Reply-To: <ee695b43-656b-1734-7522-fa755f0c31aa@oracle.com>
References: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>
 <ee695b43-656b-1734-7522-fa755f0c31aa@oracle.com>
Message-ID: <CA+0ynh-Oq_RsGvsgFaGCgJ1yH-HMLsJy7atr2nXc8yO0iw2Gww@mail.gmail.com>

>
> There's a certain amount of accidental complexity that will have to be
> compounded to get there; for example, there are "statement lambdas" and
> "expression lambdas" but if void is a type, then a statement lambda could
> also be an expression lambda returning void.  I doubt any of these are
> blockers, but there's a certain amount of un-fun grinding here to clear out
> the underbrush.


This made me think of cases of the so-called "ternary expression" that
doesn't work with `void`:

boolean b = i == 3 ? boolean1() : boolean2();
??? = i ==3 ? void1() : void2();

where the methods return `boolean` and `void`. While I understand that the
point is assignment, and that assigning to `void` is meaningless, we can
still assign to technically `Void`. Probably, the only reason to deal with
this is if we accept the "ternary expression" as a full shortcut to
if-else, and not only to if-else assignments. This is probably too far off
the mark though.

On Tue, Feb 16, 2021 at 11:15 PM Brian Goetz <brian.goetz at oracle.com> wrote:

> It's an open question, but one we want to address.  Adjusting the language
> to understand that `void` and `Void` are related is not hard, allowing us
> to ignore the need to return a value.
>
> Turning `void` into a proper (unit) type is harder, but ultimately
> valuable; if you look at the explosion of functional interfaces we added in
> 8, one contributor was all the primitive specializations (IntPredicate
> could be Predicate<int> which in turn could be Function<int,boolean>), but
> another was the fact that `void` is not a type (Consumer<T> could be
> Function<T, void>).  Valhalla will surely collapse the first (hurrah), but
> it would be a shame not to collapse the second as well.
>
> There's a certain amount of accidental complexity that will have to be
> compounded to get there; for example, there are "statement lambdas" and
> "expression lambdas" but if void is a type, then a statement lambda could
> also be an expression lambda returning void.  I doubt any of these are
> blockers, but there's a certain amount of un-fun grinding here to clear out
> the underbrush.
>
>
>
> On 2/16/2021 3:48 PM, Nir Lisker wrote:
>
> Hi,
>
> After reading the JEPs about primitive objects (.ref and .val) and the
> conversion of primitives, I was wondering what happens with void and Void.
> They are not interchangeable like primitives - a method returning void must
> have 'return;` while a method returning Void must 'return null;' - but they
> are the same semantically. Since Void mainly (only?) exists for generic
> types purposes, will void be valid as a generic type and make Void more or
> less useless? Will they have reference-value conversions like primitives,
> if that makes sense?
>
> - Nir
>
>
>

From brian.goetz at oracle.com  Wed Feb 17 16:59:44 2021
From: brian.goetz at oracle.com (Brian Goetz)
Date: Wed, 17 Feb 2021 11:59:44 -0500
Subject: [External] : Re: Primitive values and void
In-Reply-To: <CA+0ynh-Oq_RsGvsgFaGCgJ1yH-HMLsJy7atr2nXc8yO0iw2Gww@mail.gmail.com>
References: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>
 <ee695b43-656b-1734-7522-fa755f0c31aa@oracle.com>
 <CA+0ynh-Oq_RsGvsgFaGCgJ1yH-HMLsJy7atr2nXc8yO0iw2Gww@mail.gmail.com>
Message-ID: <ca53abd3-d7a3-d439-3e6a-fdabfd0222bf@oracle.com>



On 2/16/2021 9:08 PM, Nir Lisker wrote:
>
>     There's a certain amount of accidental complexity that will have
>     to be compounded to get there; for example, there are "statement
>     lambdas" and "expression lambdas" but if void is a type, then a
>     statement lambda could also be an expression lambda returning
>     void.? I doubt any of these are blockers, but there's a certain
>     amount of un-fun grinding here to clear out the underbrush.
>
>
> This made me think of cases of the so-called "ternary expression" that 
> doesn't work with `void`:
>
> boolean b = i == 3 ? boolean1() : boolean2();
> ??? = i ==3 ? void1() : void2();

Indeed, this is the sort of accidental complexity that would have to 
either deal with `void` as a new type, or be adjusted to explicitly 
disallow void expressions in either of the arms.? None of this is likely 
to be impossible, just tedious.

> where the methods return?`boolean` and `void`. While I understand that 
> the point is assignment, and that assigning to `void` is meaningless, 
> we can still assign to technically `Void`. Probably, the only reason 
> to deal with this is if we accept the "ternary expression" as a full 
> shortcut?to if-else, and not only to if-else assignments. This is 
> probably too far off the mark though.

I think the logical way to look at this is to let the construct define 
how it handles void or not, and then treat Void as the box for void, 
just as we treat Integer as the box for int.? (Boxing conversions will 
get adjusted as part of Valhalla, but the concept will still be around.)

From dsimms at openjdk.java.net  Thu Feb 18 08:25:27 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 08:25:27 GMT
Subject: [lworld] RFR: Merge jdk
Message-ID: <xxy51pxZ_heykSSYWFcU1NoPKtzhTJACGhXk-BxxC-w=.4e4fafcb-845d-4d5b-9672-dba04de96dc4@github.com>

Merge tag 'jdk-17+10'
# Conflicts:
#	src/hotspot/cpu/aarch64/interpreterRT_aarch64.cpp
#	src/hotspot/cpu/x86/gc/shared/barrierSetAssembler_x86.cpp
#	src/hotspot/share/memory/archiveBuilder.cpp
#	src/hotspot/share/oops/instanceKlass.hpp
#	src/hotspot/share/oops/typeArrayKlass.hpp
#	src/hotspot/share/runtime/sharedRuntime.cpp
#	test/hotspot/jtreg/ProblemList.txt

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

Commit messages:
 - Merge tag 'jdk-17+10' into lworld_merge_jdk_17_10
 - 8261711: Clhsdb "versioncheck true" throws NPE every time
 - 8261607: SA attach is exceeding JNI Local Refs capacity
 - 8241372: Several test failures due to javax.net.ssl.SSLException: Connection reset
 - 8260415: Remove unused class ReferenceProcessorMTProcMutator
 - 8248223: KeyAgreement spec update on multi-party key exchange support
 - 8261753: Test java/lang/System/OsVersionTest.java still failing on BigSur patch versions after JDK-8253702
 - 8261336: IGV: enhance default filters
 - 8261501: Shenandoah: reconsider heap statistics memory ordering
 - 8259668: Make SubTasksDone use-once
 - ... and 69 more: https://git.openjdk.java.net/valhalla/compare/3899e114...d2da4e1f

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=339&range=00.0
 - jdk: https://webrevs.openjdk.java.net/?repo=valhalla&pr=339&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/339/files
  Stats: 9363 lines in 376 files changed: 4632 ins; 2645 del; 2086 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/339.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/339/head:pull/339

PR: https://git.openjdk.java.net/valhalla/pull/339

From dsimms at openjdk.java.net  Thu Feb 18 09:52:35 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 09:52:35 GMT
Subject: [lworld] RFR: Merge jdk [v2]
In-Reply-To: <xxy51pxZ_heykSSYWFcU1NoPKtzhTJACGhXk-BxxC-w=.4e4fafcb-845d-4d5b-9672-dba04de96dc4@github.com>
References: <xxy51pxZ_heykSSYWFcU1NoPKtzhTJACGhXk-BxxC-w=.4e4fafcb-845d-4d5b-9672-dba04de96dc4@github.com>
Message-ID: <o1b_K3lxEIc4lfhQYTeaBOs4nr9q7lNjF1uWVxCKatk=.08da9206-d0ae-469d-9fc7-c32312ec1062@github.com>

> Merge tag 'jdk-17+10'
> # Conflicts:
> #	src/hotspot/cpu/aarch64/interpreterRT_aarch64.cpp
> #	src/hotspot/cpu/x86/gc/shared/barrierSetAssembler_x86.cpp
> #	src/hotspot/share/memory/archiveBuilder.cpp
> #	src/hotspot/share/oops/instanceKlass.hpp
> #	src/hotspot/share/oops/typeArrayKlass.hpp
> #	src/hotspot/share/runtime/sharedRuntime.cpp
> #	test/hotspot/jtreg/ProblemList.txt

David Simms has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 1101 commits:

 - Merge tag 'jdk-17+10' into lworld_merge_jdk_17_10
   
   Added tag jdk-17+10 for changeset e7e20d4e
   
   # Conflicts:
   #	src/hotspot/cpu/aarch64/interpreterRT_aarch64.cpp
   #	src/hotspot/cpu/x86/gc/shared/barrierSetAssembler_x86.cpp
   #	src/hotspot/share/memory/archiveBuilder.cpp
   #	src/hotspot/share/oops/instanceKlass.hpp
   #	src/hotspot/share/oops/typeArrayKlass.hpp
   #	src/hotspot/share/runtime/sharedRuntime.cpp
   #	test/hotspot/jtreg/ProblemList.txt
 - 8261830: [lworld] [test] ArchiveRelocationTest compilation failure
   
   Reviewed-by: iklam
 - 8255045: [lworld] LoadNode::Value() use of markWord::prototype() may strip type bits
 - 8261751: [lworld] Fix lingering references to inline/value types
 - 8261647: [lworld] Missing default initialization of non-flattened field of empty inline type
 - 8261721: [lworld] Javac messages still refer to inline types
 - 8261156: [lworld] Re-enable C1 testing after mainline issues are fixed
 - 8261631: [lworld] Fatal error in C1 compiled code due to unexpected klass
 - 8261633: [lworld] TestLWorld::test10 fails IR verification
 - Merge jdk
   
   Merge tag 'jdk-17+9'
 - ... and 1091 more: https://git.openjdk.java.net/valhalla/compare/e7e20d4e...d2da4e1f

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

Changes: https://git.openjdk.java.net/valhalla/pull/339/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=339&range=01
  Stats: 144237 lines in 1307 files changed: 137613 ins; 2042 del; 4582 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/339.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/339/head:pull/339

PR: https://git.openjdk.java.net/valhalla/pull/339

From dsimms at openjdk.java.net  Thu Feb 18 09:52:37 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 09:52:37 GMT
Subject: [lworld] Integrated: Merge jdk
In-Reply-To: <xxy51pxZ_heykSSYWFcU1NoPKtzhTJACGhXk-BxxC-w=.4e4fafcb-845d-4d5b-9672-dba04de96dc4@github.com>
References: <xxy51pxZ_heykSSYWFcU1NoPKtzhTJACGhXk-BxxC-w=.4e4fafcb-845d-4d5b-9672-dba04de96dc4@github.com>
Message-ID: <rjb7tjJu4p9IeokvBr1LGpYB0qCmIpHCD7c83DKD_rU=.5e072ae3-1755-494c-878b-e90ad05b6a7b@github.com>

On Thu, 18 Feb 2021 08:20:33 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge tag 'jdk-17+10'
> # Conflicts:
> #	src/hotspot/cpu/aarch64/interpreterRT_aarch64.cpp
> #	src/hotspot/cpu/x86/gc/shared/barrierSetAssembler_x86.cpp
> #	src/hotspot/share/memory/archiveBuilder.cpp
> #	src/hotspot/share/oops/instanceKlass.hpp
> #	src/hotspot/share/oops/typeArrayKlass.hpp
> #	src/hotspot/share/runtime/sharedRuntime.cpp
> #	test/hotspot/jtreg/ProblemList.txt

This pull request has now been integrated.

Changeset: e31335fa
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/e31335fa
Stats:     9363 lines in 376 files changed: 4632 ins; 2645 del; 2086 mod

Merge jdk

Merge tag 'jdk-17+10'

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

PR: https://git.openjdk.java.net/valhalla/pull/339

From dsimms at openjdk.java.net  Thu Feb 18 11:30:21 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 11:30:21 GMT
Subject: RFR: Merge lworld
Message-ID: <sNDxgHQIEBjSJxhH1p-46CiidRK0zuC0ajinhLZlS8c=.c1e2ad29-c9e0-4f05-890f-17386f627c85@github.com>

Merge 'jdk-17+10'

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

Commit messages:
 - More inline to primitive keyword changes
 - inline to primitive keyword changes
 - Merge branch 'lworld' into type-restrictions_merge_lworld_17_10
 - Merge jdk
 - 8261711: Clhsdb "versioncheck true" throws NPE every time
 - 8261607: SA attach is exceeding JNI Local Refs capacity
 - 8241372: Several test failures due to javax.net.ssl.SSLException: Connection reset
 - 8260415: Remove unused class ReferenceProcessorMTProcMutator
 - 8248223: KeyAgreement spec update on multi-party key exchange support
 - 8261753: Test java/lang/System/OsVersionTest.java still failing on BigSur patch versions after JDK-8253702
 - ... and 183 more: https://git.openjdk.java.net/valhalla/compare/5f6ef0f4...ccfcad5d

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - type-restrictions: https://webrevs.openjdk.java.net/?repo=valhalla&pr=340&range=00.0
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=340&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/340/files
  Stats: 33321 lines in 1324 files changed: 19153 ins; 8776 del; 5392 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/340.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/340/head:pull/340

PR: https://git.openjdk.java.net/valhalla/pull/340

From dsimms at openjdk.java.net  Thu Feb 18 12:14:22 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 12:14:22 GMT
Subject: Integrated: Merge lworld
In-Reply-To: <sNDxgHQIEBjSJxhH1p-46CiidRK0zuC0ajinhLZlS8c=.c1e2ad29-c9e0-4f05-890f-17386f627c85@github.com>
References: <sNDxgHQIEBjSJxhH1p-46CiidRK0zuC0ajinhLZlS8c=.c1e2ad29-c9e0-4f05-890f-17386f627c85@github.com>
Message-ID: <IWLOCmRa5e7lRrO8VJOYS6BQbkkozCX-931ee_9kT7s=.72683612-05d6-4dc6-b2f9-82699ffdcf2a@github.com>

On Thu, 18 Feb 2021 11:24:11 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge 'jdk-17+10'

This pull request has now been integrated.

Changeset: 803ebc43
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/803ebc43
Stats:     33321 lines in 1324 files changed: 19153 ins; 8776 del; 5392 mod

Merge lworld

Merge 'jdk-17+10'

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

PR: https://git.openjdk.java.net/valhalla/pull/340

From dsimms at openjdk.java.net  Thu Feb 18 12:14:18 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 12:14:18 GMT
Subject: RFR: Merge lworld [v2]
In-Reply-To: <sNDxgHQIEBjSJxhH1p-46CiidRK0zuC0ajinhLZlS8c=.c1e2ad29-c9e0-4f05-890f-17386f627c85@github.com>
References: <sNDxgHQIEBjSJxhH1p-46CiidRK0zuC0ajinhLZlS8c=.c1e2ad29-c9e0-4f05-890f-17386f627c85@github.com>
Message-ID: <g9OFBs1obbl02DvAZU-dxosv9SRhSRPTtV1FmHscPdU=.7ea9762f-066b-4b67-b505-9aa42ba78499@github.com>

> Merge 'jdk-17+10'

David Simms has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 20 commits:

 - More inline to primitive keyword changes
 - inline to primitive keyword changes
 - Merge branch 'lworld' into type-restrictions_merge_lworld_17_10
   
   # Conflicts:
   #	src/jdk.compiler/share/classes/com/sun/tools/javac/code/Symbol.java
 - Merge lworld
   
   Merge lworld at jdk-17+8
 - 8260870: [type-restrictions] Generate RestrictedMethod attributes
 - 8257600: [type-restrictions] Implement RestrictedField in C1
   
   Reviewed-by: lfoltan, thartmann
 - Merge lworld
   
   Merge tag 'jdk-16+26'
 - 8257028: [type-restrictions] Assorted issues with generation of RestrictedField attributes from annotations
 - 8255856: Generate RestrictedField attributes from annotations
 - Merge lworld
   
   Merge jdk-16+25
 - ... and 10 more: https://git.openjdk.java.net/valhalla/compare/e31335fa...ccfcad5d

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

Changes: https://git.openjdk.java.net/valhalla/pull/340/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=340&range=01
  Stats: 1919 lines in 56 files changed: 1851 ins; 11 del; 57 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/340.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/340/head:pull/340

PR: https://git.openjdk.java.net/valhalla/pull/340

From fparain at openjdk.java.net  Thu Feb 18 14:31:09 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Thu, 18 Feb 2021 14:31:09 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter
Message-ID: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>

Please review these changes adding support for RestrictedMethod in the interpreter.
Types restrictions are enforced in the following bytecodes:
  - invokevirtual
  - invokestatic
  - invokespecial
  - invokeinterface
  
The changeset also includes a small fix to prevent previous type restrictions related tests to fail.

Thank you,

Fred

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

Commit messages:
 - Add test for restricted method, fix other tests
 - Fix type restriction check on return value
 - Code simplification and add invokeinterface support
 - Simple type restriction support for invokevirtual/invokestatic

Changes: https://git.openjdk.java.net/valhalla/pull/341/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=341&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261956
  Stats: 474 lines in 14 files changed: 451 ins; 1 del; 22 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/341.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/341/head:pull/341

PR: https://git.openjdk.java.net/valhalla/pull/341

From thartmann at openjdk.java.net  Thu Feb 18 15:19:05 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Thu, 18 Feb 2021 15:19:05 GMT
Subject: [lworld] RFR: 8261874: [lworld] Non-flattened array blocks
 scalarization of inline type
Message-ID: <DgPw1IeWlKJztuOKcv4goyvHjsbecvy_ORHiZmGY91A=.83e539f9-d8ff-4571-b0c0-092aa7e2bb59@github.com>

We hit an assert because an inline type is not scalarized in the debug info of a safepoint. The problem is that the inline type is stored in a non-flattened array which blocks scalarization. Once the non-flattened array is scalarized itself, the inline type could be scalarized as well but we don't keep track of its field values long enough (we remove the `InlineTypePtrNode` before macro expansion). The solution is to keep `InlineTypePtrNodes` until after macro expansion. I've also converted scalarization in safepoints into an ideal transformation, refactored related code and added the corresponding IR verification tests.

Best regards,
Tobias

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

Commit messages:
 - 8261874: [lworld] Non-flattened array blocks scalarization of inline type

Changes: https://git.openjdk.java.net/valhalla/pull/342/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=342&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261874
  Stats: 176 lines in 7 files changed: 127 ins; 25 del; 24 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/342.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/342/head:pull/342

PR: https://git.openjdk.java.net/valhalla/pull/342

From dsimms at openjdk.java.net  Thu Feb 18 17:13:48 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 18 Feb 2021 17:13:48 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter
In-Reply-To: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
Message-ID: <zbzu9jSJ0u0IOmvCvYlg_x74HcBJnUYsadBiRssotLg=.a16b508b-d299-4e28-acbe-62518bb2a854@github.com>

On Thu, 18 Feb 2021 14:26:29 GMT, Frederic Parain <fparain at openjdk.org> wrote:

> Please review these changes adding support for RestrictedMethod in the interpreter.
> Types restrictions are enforced in the following bytecodes:
>   - invokevirtual
>   - invokestatic
>   - invokespecial
>   - invokeinterface
>   
> The changeset also includes a small fix to prevent previous type restrictions related tests to fail.
> 
> Thank you,
> 
> Fred

Looks good

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

Marked as reviewed by dsimms (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/341

From hseigel at openjdk.java.net  Thu Feb 18 20:02:02 2021
From: hseigel at openjdk.java.net (Harold Seigel)
Date: Thu, 18 Feb 2021 20:02:02 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter
In-Reply-To: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
Message-ID: <0ZgfAoR6lQtb0IJma3pN00iRUjYG8hImySpfnD8SNH8=.684a568f-a6d4-4323-a939-bd6f73e5c0ea@github.com>

On Thu, 18 Feb 2021 14:26:29 GMT, Frederic Parain <fparain at openjdk.org> wrote:

> Please review these changes adding support for RestrictedMethod in the interpreter.
> Types restrictions are enforced in the following bytecodes:
>   - invokevirtual
>   - invokestatic
>   - invokespecial
>   - invokeinterface
>   
> The changeset also includes a small fix to prevent previous type restrictions related tests to fail.
> 
> Thank you,
> 
> Fred

Hi Fred,
Maybe move the classFileParser.cpp code, that checks for the restricted_method attribute, inside of an "if (_major_version >= JAVA_17_VERSION)" conditional?

Thanks, Harold

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

PR: https://git.openjdk.java.net/valhalla/pull/341

From fparain at openjdk.java.net  Thu Feb 18 22:13:28 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Thu, 18 Feb 2021 22:13:28 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter [v2]
In-Reply-To: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
Message-ID: <AdKRlSe52JS2JD__E4RQsluRbpkTLPpa5A79TnLcO-0=.d6db2400-de50-4654-958e-5a4988f24436@github.com>

> Please review these changes adding support for RestrictedMethod in the interpreter.
> Types restrictions are enforced in the following bytecodes:
>   - invokevirtual
>   - invokestatic
>   - invokespecial
>   - invokeinterface
>   
> The changeset also includes a small fix to prevent previous type restrictions related tests to fail.
> 
> Thank you,
> 
> Fred

Frederic Parain has updated the pull request incrementally with one additional commit since the last revision:

  Check class file version for RestrictedMethod attribute

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/341/files
  - new: https://git.openjdk.java.net/valhalla/pull/341/files/6b0face9..912eee2c

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=341&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=341&range=00-01

  Stats: 28 lines in 1 file changed: 16 ins; 12 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/341.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/341/head:pull/341

PR: https://git.openjdk.java.net/valhalla/pull/341

From fparain at openjdk.java.net  Thu Feb 18 22:13:30 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Thu, 18 Feb 2021 22:13:30 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter [v2]
In-Reply-To: <0ZgfAoR6lQtb0IJma3pN00iRUjYG8hImySpfnD8SNH8=.684a568f-a6d4-4323-a939-bd6f73e5c0ea@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
 <0ZgfAoR6lQtb0IJma3pN00iRUjYG8hImySpfnD8SNH8=.684a568f-a6d4-4323-a939-bd6f73e5c0ea@github.com>
Message-ID: <HzWd47TdG8FUnqni89N_2Acdzh9RLzebnPv5pfT-cW4=.8ec02636-3c0f-4efa-843c-6d90ec53ee83@github.com>

On Thu, 18 Feb 2021 19:58:58 GMT, Harold Seigel <hseigel at openjdk.org> wrote:

>> Frederic Parain has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Check class file version for RestrictedMethod attribute
>
> Hi Fred,
> Maybe move the classFileParser.cpp code, that checks for the restricted_method attribute, inside of an "if (_major_version >= JAVA_17_VERSION)" conditional?
> 
> Thanks, Harold

Harold,

Following your suggestion, a class file version check has been added for the RestrictedMethod attribute.

Fred

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

PR: https://git.openjdk.java.net/valhalla/pull/341

From thartmann at openjdk.java.net  Fri Feb 19 08:29:06 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 19 Feb 2021 08:29:06 GMT
Subject: [lworld] Integrated: 8261874: [lworld] Non-flattened array blocks
 scalarization of inline type
In-Reply-To: <DgPw1IeWlKJztuOKcv4goyvHjsbecvy_ORHiZmGY91A=.83e539f9-d8ff-4571-b0c0-092aa7e2bb59@github.com>
References: <DgPw1IeWlKJztuOKcv4goyvHjsbecvy_ORHiZmGY91A=.83e539f9-d8ff-4571-b0c0-092aa7e2bb59@github.com>
Message-ID: <pPPTDFzfLyOU7lO3-w1_BPB9_QqpdVFZ7nkUnkR5QtU=.3fe9d257-1ec8-4087-8047-6519e23ee429@github.com>

On Thu, 18 Feb 2021 15:14:26 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> We hit an assert because an inline type is not scalarized in the debug info of a safepoint. The problem is that the inline type is stored in a non-flattened array which blocks scalarization. Once the non-flattened array is scalarized itself, the inline type could be scalarized as well but we don't keep track of its field values long enough (we remove the `InlineTypePtrNode` before macro expansion). The solution is to keep `InlineTypePtrNodes` until after macro expansion. I've also converted scalarization in safepoints into an ideal transformation, refactored related code and added the corresponding IR verification tests.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: e88ff83c
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/e88ff83c
Stats:     176 lines in 7 files changed: 127 ins; 25 del; 24 mod

8261874: [lworld] Non-flattened array blocks scalarization of inline type

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

PR: https://git.openjdk.java.net/valhalla/pull/342

From thartmann at openjdk.java.net  Fri Feb 19 13:02:26 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 19 Feb 2021 13:02:26 GMT
Subject: [lworld] RFR: 8262025: [lworld] C2 should optimize acmp of the same
 inline type
Message-ID: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>

I've noticed that a simple `val == val` where `val` is an inline type is not optimized by C2. The reason is that we are buffering the inline type operands before applying optimizations. The fix is to simply check for equality before that.

Best regards,
Tobias

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

Commit messages:
 - 8262025: [lworld] C2 should optimize acmp of the same inline type

Changes: https://git.openjdk.java.net/valhalla/pull/343/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=343&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262025
  Stats: 72 lines in 2 files changed: 72 ins; 0 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/343.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/343/head:pull/343

PR: https://git.openjdk.java.net/valhalla/pull/343

From thartmann at openjdk.java.net  Fri Feb 19 14:11:08 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 19 Feb 2021 14:11:08 GMT
Subject: [lworld] RFR: 8262025: [lworld] C2 should optimize acmp of the
 same inline type [v2]
In-Reply-To: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
References: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
Message-ID: <ls5M9rSMw2zna5eoH8NscyhqJu2ePsfcOGz76OI8cKw=.1da1739a-363d-4b64-a25c-0baf9b40fc71@github.com>

> I've noticed that a simple `val == val` where `val` is an inline type is not optimized by C2. The reason is that we are buffering the inline type operands before applying optimizations. The fix is to simply check for equality before that.
> 
> Best regards,
> Tobias

Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:

  Fixed test bug

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/343/files
  - new: https://git.openjdk.java.net/valhalla/pull/343/files/6b9dc63d..e0ccb503

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=343&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=343&range=00-01

  Stats: 1 line in 1 file changed: 1 ins; 0 del; 0 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/343.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/343/head:pull/343

PR: https://git.openjdk.java.net/valhalla/pull/343

From thartmann at openjdk.java.net  Fri Feb 19 14:13:16 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Fri, 19 Feb 2021 14:13:16 GMT
Subject: [lworld] RFR: 8262025: [lworld] C2 should optimize acmp of the
 same inline type
In-Reply-To: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
References: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
Message-ID: <VFyfuEiaM6RraFU7ElfOaFo9EOa6mXA1ueH-Acw1SQ8=.75a039ed-bb58-4e7a-8e20-7bd339e3be5c@github.com>

On Fri, 19 Feb 2021 12:57:21 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> I've noticed that a simple `val == val` where `val` is an inline type is not optimized by C2. The reason is that we are buffering the inline type operands before applying optimizations. The fix is to simply check for equality before that.
> 
> Best regards,
> Tobias

I've noticed that with the fix, the `Ifacmp.java` test started to time out due to an endless loop in the `testAlot` method. It seems that this is due to C2 now being able to fully remove the loop (because it can prove that `a == b`) and therefore also the safepoint that would trigger a GC. I've added a call to `System.gc`. @MrSimms does that seem reasonable to you?

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

PR: https://git.openjdk.java.net/valhalla/pull/343

From sadayapalam at openjdk.java.net  Mon Feb 22 06:46:33 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 22 Feb 2021 06:46:33 GMT
Subject: [lworld] Integrated: 8261782: [lworld] Add support for
 PrimitiveObject interface
Message-ID: <M1ptuu9Vmih0p2Bovs3s2ngbOrWLZBqLd1c3zLuXfcA=.ebb904f6-34e8-4277-84ac-722b155862e4@github.com>

- Compiler support for PrimitiveObject top type

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

Commit messages:
 - 8261782: [lworld] Add support for PrimitiveObject interface

Changes: https://git.openjdk.java.net/valhalla/pull/344/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=344&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8261782
  Stats: 162 lines in 8 files changed: 150 ins; 1 del; 11 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/344.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/344/head:pull/344

PR: https://git.openjdk.java.net/valhalla/pull/344

From sadayapalam at openjdk.java.net  Mon Feb 22 06:46:36 2021
From: sadayapalam at openjdk.java.net (Srikanth Adayapalam)
Date: Mon, 22 Feb 2021 06:46:36 GMT
Subject: [lworld] Integrated: 8261782: [lworld] Add support for
 PrimitiveObject interface
In-Reply-To: <M1ptuu9Vmih0p2Bovs3s2ngbOrWLZBqLd1c3zLuXfcA=.ebb904f6-34e8-4277-84ac-722b155862e4@github.com>
References: <M1ptuu9Vmih0p2Bovs3s2ngbOrWLZBqLd1c3zLuXfcA=.ebb904f6-34e8-4277-84ac-722b155862e4@github.com>
Message-ID: <BW5EpZ9M5n8aIr-h4DuON9orHdKcGKUu7AFSrDR8g2o=.5cdadc78-fb83-4bee-9378-c6aec729cf5d@github.com>

On Mon, 22 Feb 2021 06:39:48 GMT, Srikanth Adayapalam <sadayapalam at openjdk.org> wrote:

> - Compiler support for PrimitiveObject top type

This pull request has now been integrated.

Changeset: f421b939
Author:    Srikanth Adayapalam <sadayapalam at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/f421b939
Stats:     162 lines in 8 files changed: 150 ins; 1 del; 11 mod

8261782: [lworld] Add support for PrimitiveObject interface

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

PR: https://git.openjdk.java.net/valhalla/pull/344

From dsimms at openjdk.java.net  Mon Feb 22 09:25:00 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Mon, 22 Feb 2021 09:25:00 GMT
Subject: [lworld] RFR: 8262025: [lworld] C2 should optimize acmp of the
 same inline type [v2]
In-Reply-To: <ls5M9rSMw2zna5eoH8NscyhqJu2ePsfcOGz76OI8cKw=.1da1739a-363d-4b64-a25c-0baf9b40fc71@github.com>
References: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
 <ls5M9rSMw2zna5eoH8NscyhqJu2ePsfcOGz76OI8cKw=.1da1739a-363d-4b64-a25c-0baf9b40fc71@github.com>
Message-ID: <mLOzLPzghGPYAmDs0xTNazdFevvYNYx5BuIwxh8Pqgc=.3584f1e6-d0bf-4c5b-b912-3415eb521e01@github.com>

On Fri, 19 Feb 2021 14:11:08 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

>> I've noticed that a simple `val == val` where `val` is an inline type is not optimized by C2. The reason is that we are buffering the inline type operands before applying optimizations. The fix is to simply check for equality before that.
>> 
>> Best regards,
>> Tobias
>
> Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Fixed test bug

Runtime bits look good.

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

Marked as reviewed by dsimms (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/343

From thartmann at openjdk.java.net  Mon Feb 22 09:29:01 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 09:29:01 GMT
Subject: [lworld] RFR: 8262025: [lworld] C2 should optimize acmp of the
 same inline type [v2]
In-Reply-To: <mLOzLPzghGPYAmDs0xTNazdFevvYNYx5BuIwxh8Pqgc=.3584f1e6-d0bf-4c5b-b912-3415eb521e01@github.com>
References: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
 <ls5M9rSMw2zna5eoH8NscyhqJu2ePsfcOGz76OI8cKw=.1da1739a-363d-4b64-a25c-0baf9b40fc71@github.com>
 <mLOzLPzghGPYAmDs0xTNazdFevvYNYx5BuIwxh8Pqgc=.3584f1e6-d0bf-4c5b-b912-3415eb521e01@github.com>
Message-ID: <5deiSgXkU8kYzM2zuvQSOh76F5vD36E75zj2X8b2PaY=.d39f93c6-34d9-420d-8925-af6c90bb7816@github.com>

On Mon, 22 Feb 2021 09:22:08 GMT, David Simms <dsimms at openjdk.org> wrote:

>> Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Fixed test bug
>
> Runtime bits look good.

Thanks for the review!

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

PR: https://git.openjdk.java.net/valhalla/pull/343

From thartmann at openjdk.java.net  Mon Feb 22 09:29:02 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 09:29:02 GMT
Subject: [lworld] Integrated: 8262025: [lworld] C2 should optimize acmp of the
 same inline type
In-Reply-To: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
References: <eXdyHJdY5YtubO6qfyPn4IrFigrMikiJspkrWX7KN8w=.3607bf24-5815-42fe-9262-e3696f394309@github.com>
Message-ID: <wG2F3CoOaFGbBDe4jDz1uPimbwA9bIg-KwzpCITdxsQ=.834b1e3d-02b8-4c4f-a0c4-73345960c94a@github.com>

On Fri, 19 Feb 2021 12:57:21 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> I've noticed that a simple `val == val` where `val` is an inline type is not optimized by C2. The reason is that we are buffering the inline type operands before applying optimizations. The fix is to simply check for equality before that.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 11cda281
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/11cda281
Stats:     73 lines in 3 files changed: 73 ins; 0 del; 0 mod

8262025: [lworld] C2 should optimize acmp of the same inline type

Reviewed-by: dsimms

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

PR: https://git.openjdk.java.net/valhalla/pull/343

From thartmann at openjdk.java.net  Mon Feb 22 14:11:24 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 14:11:24 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization does
 not correctly handle delayed accesses
Message-ID: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>

C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.

Best regards,
Tobias

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

Commit messages:
 - 8262128: [lworld] C1's ValueNumbering optimization does not correctly handle delayed accesses

Changes: https://git.openjdk.java.net/valhalla/pull/345/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=345&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262128
  Stats: 44 lines in 4 files changed: 35 ins; 2 del; 7 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/345.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/345/head:pull/345

PR: https://git.openjdk.java.net/valhalla/pull/345

From thartmann at openjdk.java.net  Mon Feb 22 15:04:16 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 15:04:16 GMT
Subject: [lworld] RFR: 8262129: [lworld] C2 compilation fails with assert 'can
 not use dead node'
Message-ID: <oIBGLfxcNBb3JgGQGzwBvHHVGDgBK64aGwe8T2CVPAU=.45a11a15-b2b3-44d8-bfb4-0b5323c67a01@github.com>

We hit an assert with `-XX:++VerifyIterativeGVN` because the hook nodes created by `PhaseMacroExpand::expand_mh_intrinsic_return` are dead. Similar to the mainline fix for [https://bugs.openjdk.java.net/browse/JDK-8238756](JDK-8238756), we have to add an input to these nodes.

Best regards,
Tobias

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

Commit messages:
 - 8262129: [lworld] C2 compilation fails with assert 'can not use dead node'

Changes: https://git.openjdk.java.net/valhalla/pull/346/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=346&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262129
  Stats: 9 lines in 1 file changed: 2 ins; 0 del; 7 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/346.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/346/head:pull/346

PR: https://git.openjdk.java.net/valhalla/pull/346

From thartmann at openjdk.java.net  Mon Feb 22 15:09:00 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 15:09:00 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses
In-Reply-To: <3V0_UgdjCYm5UKM4t49sdXBxzQRDhmkRSlYFbb_1Sfk=.fe3df927-2a8d-492e-8486-89babc955fec@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
 <3V0_UgdjCYm5UKM4t49sdXBxzQRDhmkRSlYFbb_1Sfk=.fe3df927-2a8d-492e-8486-89babc955fec@github.com>
Message-ID: <0xdFqSKW5tNe6PbPnsdDYRKVxnH1xMIegTbSd37V-OU=.7bc68071-dbac-42ca-8e13-3ef2f69d46dc@github.com>

On Mon, 22 Feb 2021 15:03:53 GMT, Frederic Parain <fparain at openjdk.org> wrote:

>> C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.
>> 
>> Best regards,
>> Tobias
>
> Hi Tobias,
> 
> Thank you for fixing this.
> However, the fix seems to have an issue. The hash_inline_access() method encapsulates the additional offset from the delayed access into a hash value, this encapsulation prevents the is_equal(Value v) method (from the the HASHING macro) to see all individual arguments of the hash function and could cause the method return an incorrect result.
> 
> Fred

Hi Fred,

thanks for looking at this, you are right.

Best regards,
Tobias

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

PR: https://git.openjdk.java.net/valhalla/pull/345

From fparain at openjdk.java.net  Mon Feb 22 15:08:59 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Mon, 22 Feb 2021 15:08:59 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses
In-Reply-To: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
Message-ID: <3V0_UgdjCYm5UKM4t49sdXBxzQRDhmkRSlYFbb_1Sfk=.fe3df927-2a8d-492e-8486-89babc955fec@github.com>

On Mon, 22 Feb 2021 14:06:35 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.
> 
> Best regards,
> Tobias

Hi Tobias,

Thank you for fixing this.
However, the fix seems to have an issue. The hash_inline_access() method encapsulates the additional offset from the delayed access into a hash value, this encapsulation prevents the is_equal(Value v) method (from the the HASHING macro) to see all individual arguments of the hash function and could cause the method return an incorrect result.

Fred

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

PR: https://git.openjdk.java.net/valhalla/pull/345

From thartmann at openjdk.java.net  Mon Feb 22 15:41:12 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 15:41:12 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses [v2]
In-Reply-To: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
Message-ID: <2U7hpJsiRl7ghdFwPEMu31yRPCb0vjS5rUYdJvRTKSQ=.87444a50-f547-455f-8a05-30f81e626ce0@github.com>

> C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.
> 
> Best regards,
> Tobias

Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:

  Disable ValueNumbering for now

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/345/files
  - new: https://git.openjdk.java.net/valhalla/pull/345/files/3363401b..f17af09f

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=345&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=345&range=00-01

  Stats: 11 lines in 2 files changed: 0 ins; 10 del; 1 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/345.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/345/head:pull/345

PR: https://git.openjdk.java.net/valhalla/pull/345

From thartmann at openjdk.java.net  Mon Feb 22 15:41:12 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Mon, 22 Feb 2021 15:41:12 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses
In-Reply-To: <0xdFqSKW5tNe6PbPnsdDYRKVxnH1xMIegTbSd37V-OU=.7bc68071-dbac-42ca-8e13-3ef2f69d46dc@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
 <3V0_UgdjCYm5UKM4t49sdXBxzQRDhmkRSlYFbb_1Sfk=.fe3df927-2a8d-492e-8486-89babc955fec@github.com>
 <0xdFqSKW5tNe6PbPnsdDYRKVxnH1xMIegTbSd37V-OU=.7bc68071-dbac-42ca-8e13-3ef2f69d46dc@github.com>
Message-ID: <n7fMsn4_h_m9BaSLVzW57zM8_ATUiNZ80hiQT0Oznn8=.412cbdbc-6865-4199-864d-7bbc57ab5dcb@github.com>

On Mon, 22 Feb 2021 15:05:43 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

>> Hi Tobias,
>> 
>> Thank you for fixing this.
>> However, the fix seems to have an issue. The hash_inline_access() method encapsulates the additional offset from the delayed access into a hash value, this encapsulation prevents the is_equal(Value v) method (from the the HASHING macro) to see all individual arguments of the hash function and could cause the method return an incorrect result.
>> 
>> Fred
>
> Hi Fred,
> 
> thanks for looking at this, you are right.
> 
> Best regards,
> Tobias

As we've discussed offline, I've disabled ValueNumbering for delayed accesses for now and filed [https://bugs.openjdk.java.net/browse/JDK-8262136](JDK-8262136]) to re-enable it later.

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

PR: https://git.openjdk.java.net/valhalla/pull/345

From fparain at openjdk.java.net  Mon Feb 22 16:07:08 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Mon, 22 Feb 2021 16:07:08 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses [v2]
In-Reply-To: <2U7hpJsiRl7ghdFwPEMu31yRPCb0vjS5rUYdJvRTKSQ=.87444a50-f547-455f-8a05-30f81e626ce0@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
 <2U7hpJsiRl7ghdFwPEMu31yRPCb0vjS5rUYdJvRTKSQ=.87444a50-f547-455f-8a05-30f81e626ce0@github.com>
Message-ID: <Us5Yv9Q_IqD3_BJVgo1saDOMzh2q1VN8tq0XobpGRFM=.7147304d-6b7d-4406-98e3-7883d05fc8a6@github.com>

On Mon, 22 Feb 2021 15:41:12 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

>> C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.
>> 
>> Best regards,
>> Tobias
>
> Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Disable ValueNumbering for now

Looks good to me.

Fred

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

Marked as reviewed by fparain (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/345

From hseigel at openjdk.java.net  Mon Feb 22 16:27:56 2021
From: hseigel at openjdk.java.net (Harold Seigel)
Date: Mon, 22 Feb 2021 16:27:56 GMT
Subject: RFR: 8261956: [type-restrictions] Support RestrictedMethod in the
 interpreter [v2]
In-Reply-To: <AdKRlSe52JS2JD__E4RQsluRbpkTLPpa5A79TnLcO-0=.d6db2400-de50-4654-958e-5a4988f24436@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
 <AdKRlSe52JS2JD__E4RQsluRbpkTLPpa5A79TnLcO-0=.d6db2400-de50-4654-958e-5a4988f24436@github.com>
Message-ID: <QGRs-tYw0qoj7wT82L4ZB9v8SSiabK5ch_jFRHtRvfo=.4db7b04f-8a4e-44c4-8b23-f3c88c518ca6@github.com>

On Thu, 18 Feb 2021 22:13:28 GMT, Frederic Parain <fparain at openjdk.org> wrote:

>> Please review these changes adding support for RestrictedMethod in the interpreter.
>> Types restrictions are enforced in the following bytecodes:
>>   - invokevirtual
>>   - invokestatic
>>   - invokespecial
>>   - invokeinterface
>>   
>> The changeset also includes a small fix to prevent previous type restrictions related tests to fail.
>> 
>> Thank you,
>> 
>> Fred
>
> Frederic Parain has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Check class file version for RestrictedMethod attribute

The classFileParser.cpp change looks good.  Thanks for changing it.

Harold

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

Marked as reviewed by hseigel (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/341

From fparain at openjdk.java.net  Mon Feb 22 16:31:56 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Mon, 22 Feb 2021 16:31:56 GMT
Subject: Integrated: 8261956: [type-restrictions] Support RestrictedMethod in
 the interpreter
In-Reply-To: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
References: <rfuew3AsAHl7sAU5SjhMpJczbV1Vi6Jzb-wpKhSSNx0=.504341f0-c33b-49fe-97e9-63ac16e81973@github.com>
Message-ID: <YNJ6H4PKaonk3nldOhl-ZohJ7NIBCgEwnecMnIjtYoY=.1da1743f-3e11-4436-bb51-73950ced0fd1@github.com>

On Thu, 18 Feb 2021 14:26:29 GMT, Frederic Parain <fparain at openjdk.org> wrote:

> Please review these changes adding support for RestrictedMethod in the interpreter.
> Types restrictions are enforced in the following bytecodes:
>   - invokevirtual
>   - invokestatic
>   - invokespecial
>   - invokeinterface
>   
> The changeset also includes a small fix to prevent previous type restrictions related tests to fail.
> 
> Thank you,
> 
> Fred

This pull request has now been integrated.

Changeset: 9c991d9b
Author:    Frederic Parain <fparain at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/9c991d9b
Stats:     502 lines in 14 files changed: 467 ins; 13 del; 22 mod

8261956: [type-restrictions] Support RestrictedMethod in the interpreter

Reviewed-by: dsimms, hseigel

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

PR: https://git.openjdk.java.net/valhalla/pull/341

From thartmann at openjdk.java.net  Tue Feb 23 06:58:52 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 06:58:52 GMT
Subject: [lworld] RFR: 8262128: [lworld] C1's ValueNumbering optimization
 does not correctly handle delayed accesses [v2]
In-Reply-To: <Us5Yv9Q_IqD3_BJVgo1saDOMzh2q1VN8tq0XobpGRFM=.7147304d-6b7d-4406-98e3-7883d05fc8a6@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
 <2U7hpJsiRl7ghdFwPEMu31yRPCb0vjS5rUYdJvRTKSQ=.87444a50-f547-455f-8a05-30f81e626ce0@github.com>
 <Us5Yv9Q_IqD3_BJVgo1saDOMzh2q1VN8tq0XobpGRFM=.7147304d-6b7d-4406-98e3-7883d05fc8a6@github.com>
Message-ID: <EHIDOO8eLtaB7ASyPWQ2unuO9ZU_OBO9dahrrZuMXys=.dc7edb62-4bda-45d9-b861-0629b0e1b83c@github.com>

On Mon, 22 Feb 2021 16:04:01 GMT, Frederic Parain <fparain at openjdk.org> wrote:

>> Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Disable ValueNumbering for now
>
> Looks good to me.
> 
> Fred

Thanks, Fred!

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

PR: https://git.openjdk.java.net/valhalla/pull/345

From thartmann at openjdk.java.net  Tue Feb 23 06:58:53 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 06:58:53 GMT
Subject: [lworld] Integrated: 8262128: [lworld] C1's ValueNumbering
 optimization does not correctly handle delayed accesses
In-Reply-To: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
References: <QFSmkonkS0fYYjfX6kAJ4P7pgYMZND7KFOlDave1PBo=.db01a74d-a202-48a4-8c38-1afa22967930@github.com>
Message-ID: <0kXLkn82Sy7AGHeajtzWvDGtd1JBawKmW92SoAVIP2k=.4a41bdf8-fe40-473e-800e-e2e3a6c25110@github.com>

On Mon, 22 Feb 2021 14:06:35 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> C1's ValueNumbering optimization replaces an array+field load `array[0].intArray` by another non-equivalent array+field load `array[0].floatArray` because it does not handle delayed accesses. The hash value of `LoadIndexed` needs to take delayed accesses into account.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 1c9a7a9a
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/1c9a7a9a
Stats:     34 lines in 3 files changed: 25 ins; 2 del; 7 mod

8262128: [lworld] C1's ValueNumbering optimization does not correctly handle delayed accesses

Reviewed-by: fparain

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

PR: https://git.openjdk.java.net/valhalla/pull/345

From thartmann at openjdk.java.net  Tue Feb 23 07:01:01 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 07:01:01 GMT
Subject: [lworld] Integrated: 8262129: [lworld] C2 compilation fails with
 assert "can not use dead node"
In-Reply-To: <oIBGLfxcNBb3JgGQGzwBvHHVGDgBK64aGwe8T2CVPAU=.45a11a15-b2b3-44d8-bfb4-0b5323c67a01@github.com>
References: <oIBGLfxcNBb3JgGQGzwBvHHVGDgBK64aGwe8T2CVPAU=.45a11a15-b2b3-44d8-bfb4-0b5323c67a01@github.com>
Message-ID: <dVbRD802PF-Seypo8pjyr_8KWHq1hDRxjHewGaMRXx4=.dba586b9-e487-4230-8086-263e2b2d351d@github.com>

On Mon, 22 Feb 2021 15:00:03 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> We hit an assert with `-XX:++VerifyIterativeGVN` because the hook nodes created by `PhaseMacroExpand::expand_mh_intrinsic_return` are dead. Similar to the mainline fix for [https://bugs.openjdk.java.net/browse/JDK-8238756](JDK-8238756), we have to add an input to these nodes.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 5ac8ba0f
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/5ac8ba0f
Stats:     9 lines in 1 file changed: 2 ins; 0 del; 7 mod

8262129: [lworld] C2 compilation fails with assert "can not use dead node"

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

PR: https://git.openjdk.java.net/valhalla/pull/346

From fparain at openjdk.java.net  Tue Feb 23 13:44:09 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Tue, 23 Feb 2021 13:44:09 GMT
Subject: Integrated: 8262202: [type-restrictions] RestrictedMethodTest failed
 to compile
Message-ID: <dPbwylBq-C-uer2msu_OUHNzDVbeMWO-GU25udXe9kc=.f8be20c5-56db-413e-977a-e53cf926ecb6@github.com>

Fix syntax in test

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

Commit messages:
 - Fix test with new syntax

Changes: https://git.openjdk.java.net/valhalla/pull/347/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=347&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262202
  Stats: 1 line in 1 file changed: 0 ins; 0 del; 1 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/347.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/347/head:pull/347

PR: https://git.openjdk.java.net/valhalla/pull/347

From fparain at openjdk.java.net  Tue Feb 23 13:44:10 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Tue, 23 Feb 2021 13:44:10 GMT
Subject: Integrated: 8262202: [type-restrictions] RestrictedMethodTest failed
 to compile
In-Reply-To: <dPbwylBq-C-uer2msu_OUHNzDVbeMWO-GU25udXe9kc=.f8be20c5-56db-413e-977a-e53cf926ecb6@github.com>
References: <dPbwylBq-C-uer2msu_OUHNzDVbeMWO-GU25udXe9kc=.f8be20c5-56db-413e-977a-e53cf926ecb6@github.com>
Message-ID: <xMTxfgz1K0rwKHCi-7YRcdXxirVjcXpV1Z234zOk6fg=.4c8ccf6b-3aee-4ee1-a402-1c258694548b@github.com>

On Tue, 23 Feb 2021 13:37:07 GMT, Frederic Parain <fparain at openjdk.org> wrote:

> Fix syntax in test

This pull request has now been integrated.

Changeset: 75d27907
Author:    Frederic Parain <fparain at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/75d27907
Stats:     1 line in 1 file changed: 0 ins; 0 del; 1 mod

8262202: [type-restrictions] RestrictedMethodTest failed to compile

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

PR: https://git.openjdk.java.net/valhalla/pull/347

From thartmann at openjdk.java.net  Tue Feb 23 17:15:19 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 17:15:19 GMT
Subject: [lworld] RFR: 8262231: [lworld] C2 compilation fails with assert
 "user must call transfer_exceptions_into_jvms"
Message-ID: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>

When running with `-XX:PerMethodSpecTrapLimit=0 -XX:PerMethodTrapLimit=0`, we hit an assert because the unsafe access intrinsic adds an exception to the `GraphKit` state but then bails out and therefore does not transfer that exception into the `JVMState`. In general, we should only modify the graph once we are sure we won't bail out from intrinsification. I've added an assert that catches this even without any VM flags and fixed all affected intrinsics.

Best regards,
Tobias

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

Commit messages:
 - Simplified assert
 - 8262231: [lworld] C2 compilation fails with assert "user must call transfer_exceptions_into_jvms"

Changes: https://git.openjdk.java.net/valhalla/pull/348/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=348&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262231
  Stats: 59 lines in 1 file changed: 21 ins; 27 del; 11 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/348.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/348/head:pull/348

PR: https://git.openjdk.java.net/valhalla/pull/348

From thartmann at openjdk.java.net  Tue Feb 23 18:02:28 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 18:02:28 GMT
Subject: [lworld] RFR: 8262231: [lworld] C2 compilation fails with assert
 "user must call transfer_exceptions_into_jvms" [v2]
In-Reply-To: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
References: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
Message-ID: <M8R8mkR6rHBombf2Yoh7h2tRtxiKtgA9eZhrP-0ONh4=.8e3deeb0-15a4-4590-a2fe-6a7b04fca92a@github.com>

> When running with `-XX:PerMethodSpecTrapLimit=0 -XX:PerMethodTrapLimit=0`, we hit an assert because the unsafe access intrinsic adds an exception to the `GraphKit` state but then bails out and therefore does not transfer that exception into the `JVMState`. In general, we should only modify the graph once we are sure we won't bail out from intrinsification. I've added an assert that catches this even without any VM flags and fixed all affected intrinsics.
> 
> Best regards,
> Tobias

Tobias Hartmann has updated the pull request incrementally with one additional commit since the last revision:

  Completely removed assert because it triggers bugs in mainline code

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

Changes:
  - all: https://git.openjdk.java.net/valhalla/pull/348/files
  - new: https://git.openjdk.java.net/valhalla/pull/348/files/2a0facae..bca88c7a

Webrevs:
 - full: https://webrevs.openjdk.java.net/?repo=valhalla&pr=348&range=01
 - incr: https://webrevs.openjdk.java.net/?repo=valhalla&pr=348&range=00-01

  Stats: 5 lines in 1 file changed: 1 ins; 3 del; 1 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/348.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/348/head:pull/348

PR: https://git.openjdk.java.net/valhalla/pull/348

From thartmann at openjdk.java.net  Tue Feb 23 18:02:29 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Tue, 23 Feb 2021 18:02:29 GMT
Subject: [lworld] RFR: 8262231: [lworld] C2 compilation fails with assert
 "user must call transfer_exceptions_into_jvms"
In-Reply-To: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
References: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
Message-ID: <-PQ-9T2sWuHJLfeXRhkzG0jFdiJX6EAHrVAea0aXu7g=.7e881783-4458-4f40-8230-cf3e9845a33c@github.com>

On Tue, 23 Feb 2021 17:11:05 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> When running with `-XX:PerMethodSpecTrapLimit=0 -XX:PerMethodTrapLimit=0`, we hit an assert because the unsafe access intrinsic adds an exception to the `GraphKit` state but then bails out and therefore does not transfer that exception into the `JVMState`. In general, we should only modify the graph once we are sure we won't bail out from intrinsification. I've added an assert that catches this even without any VM flags and fixed all affected intrinsics.
> 
> Best regards,
> Tobias

I've removed the assert because it triggers issues in mainline code. Will add the assert back into mainline after fixing issues there.

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

PR: https://git.openjdk.java.net/valhalla/pull/348

From john.r.rose at oracle.com  Tue Feb 23 18:20:28 2021
From: john.r.rose at oracle.com (John Rose)
Date: Tue, 23 Feb 2021 10:20:28 -0800
Subject: [External] : Re: Primitive values and void
In-Reply-To: <ca53abd3-d7a3-d439-3e6a-fdabfd0222bf@oracle.com>
References: <CA+0ynh8PuWaVp2hj75KV4y01x4oSHr643V+Gw-envnDqpnd0jg@mail.gmail.com>
 <ee695b43-656b-1734-7522-fa755f0c31aa@oracle.com>
 <CA+0ynh-Oq_RsGvsgFaGCgJ1yH-HMLsJy7atr2nXc8yO0iw2Gww@mail.gmail.com>
 <ca53abd3-d7a3-d439-3e6a-fdabfd0222bf@oracle.com>
Message-ID: <9B602613-A54D-46BB-A30E-9E88FB16994C@oracle.com>

On Feb 17, 2021, at 8:59 AM, Brian Goetz <brian.goetz at oracle.com> wrote:
> On 2/16/2021 9:08 PM, Nir Lisker wrote:
>> 
>>    There's a certain amount of accidental complexity that will have
>>    to be compounded to get there; for example, there are "statement
>>    lambdas" and "expression lambdas" but if void is a type, then a
>>    statement lambda could also be an expression lambda returning
>>    void.  I doubt any of these are blockers, but there's a certain
>>    amount of un-fun grinding here to clear out the underbrush.
>> 
>> 
>> This made me think of cases of the so-called "ternary expression" that doesn't work with `void`:
>> 
>> boolean b = i == 3 ? boolean1() : boolean2();
>> ??? = i ==3 ? void1() : void2();
> 
> Indeed, this is the sort of accidental complexity that would have to either deal with `void` as a new type, or be adjusted to explicitly disallow void expressions in either of the arms.  None of this is likely to be impossible, just tedious.
> 
>> where the methods return `boolean` and `void`. While I understand that the point is assignment, and that assigning to `void` is meaningless, we can still assign to technically `Void`. Probably, the only reason to deal with this is if we accept the "ternary expression" as a full shortcut to if-else, and not only to if-else assignments. This is probably too far off the mark though.
> 
> I think the logical way to look at this is to let the construct define how it handles void or not, and then treat Void as the box for void, just as we treat Integer as the box for int.  (Boxing conversions will get adjusted as part of Valhalla, but the concept will still be around.)

If Void were a regular reference-box for a primitive value void,
then the range of the type Void contains two values, reference
null and the single value of the (unit type) void.  But, unlike,
other wrappers, there is nothing *other than* the remainder
value null in Void.  This is an irregularity when trying to make
Void be a wrapper for (unit type) void.  It?s not a deal-killer,
but more annoying un-fun underbrush to cut away.  Do we
keep the irregularity, or add a second primitive void value?

Or we could say (by ad hoc fiat) that (Void)null interconverts
to the void value (not otherwise denotable as a reference).
That?s the approach that method handle conversions takes
(in explicitCastArguments).

Just saying?

? John

P.S. One more thing:  Void also *acts like* a nullable reference-box
type for a primitive type with *zero* value points, also known
as an ?empty type? or ?impossible type?.  That hypothetical
primitive type (call it ?Unreached??) denotes:

 - the return type of a method which must never return normally
 - parameter type of a method which must never be called
 - the field type of an class/species which must never be instantiated
 - a local variable in a declaration which must never complete normally

Of course if you adjoin null to that empty type you get a one-point
reference type containing only null, just like Void.

(I?m *not* saying that Void is a candidate name for ?Unreached?.)


From thartmann at openjdk.java.net  Wed Feb 24 06:46:03 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Wed, 24 Feb 2021 06:46:03 GMT
Subject: [lworld] Integrated: 8262231: [lworld] C2 compilation fails with
 assert "user must call transfer_exceptions_into_jvms"
In-Reply-To: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
References: <69_7K0rIYSIJaH0SKg96kRIxR9yU5znOdR9qO_nUQAE=.33ef3822-a598-44f5-871f-73c28ef24497@github.com>
Message-ID: <t2O1A_qMZZT_ij7JXgy8o12zFKJ7DhY4qSifAMgziTk=.8a14bcda-3110-4de4-ab53-fd5db8a8c5d5@github.com>

On Tue, 23 Feb 2021 17:11:05 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> When running with `-XX:PerMethodSpecTrapLimit=0 -XX:PerMethodTrapLimit=0`, we hit an assert because the unsafe access intrinsic adds an exception to the `GraphKit` state but then bails out and therefore does not transfer that exception into the `JVMState`. In general, we should only modify the graph once we are sure we won't bail out from intrinsification. I've added an assert that catches this even without any VM flags and fixed all affected intrinsics.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 1816db37
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/1816db37
Stats:     54 lines in 1 file changed: 18 ins; 26 del; 10 mod

8262231: [lworld] C2 compilation fails with assert "user must call transfer_exceptions_into_jvms"

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

PR: https://git.openjdk.java.net/valhalla/pull/348

From dsimms at openjdk.java.net  Wed Feb 24 15:23:32 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Wed, 24 Feb 2021 15:23:32 GMT
Subject: [lworld] RFR: 8262104: [lworld] Runtime should automagically have
 primitive classes implement PrimitiveObject
Message-ID: <Vr86oiGEeRYUnXHNXlNWFc1_lNwtPtD_YjHT5QsT-ag=.ad2d0e64-e2db-4e4f-a076-a7520f300339@github.com>

Follows in the footsteps of JDK-8245216: [lworld] The JVM should inject the IdentityObject interface to types which need it
Also includes a similar footprint tweak to JDK-8245646: [lworld] LW3 Reduce impact of IdentityObject on metaspace

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

Commit messages:
 - 8262104: [lworld] Runtime should automagically have primitive classes implement PrimitiveObject

Changes: https://git.openjdk.java.net/valhalla/pull/349/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=349&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262104
  Stats: 406 lines in 20 files changed: 382 ins; 10 del; 14 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/349.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/349/head:pull/349

PR: https://git.openjdk.java.net/valhalla/pull/349

From fparain at openjdk.java.net  Wed Feb 24 19:37:56 2021
From: fparain at openjdk.java.net (Frederic Parain)
Date: Wed, 24 Feb 2021 19:37:56 GMT
Subject: [lworld] RFR: 8262104: [lworld] Runtime should automagically have
 primitive classes implement PrimitiveObject
In-Reply-To: <Vr86oiGEeRYUnXHNXlNWFc1_lNwtPtD_YjHT5QsT-ag=.ad2d0e64-e2db-4e4f-a076-a7520f300339@github.com>
References: <Vr86oiGEeRYUnXHNXlNWFc1_lNwtPtD_YjHT5QsT-ag=.ad2d0e64-e2db-4e4f-a076-a7520f300339@github.com>
Message-ID: <hPdfzGOxeP1isUuV79lZCgsG-wAKeTvRPZKR-XjIa18=.ffc55082-14d2-41c3-9efc-81612bce3517@github.com>

On Wed, 24 Feb 2021 15:18:24 GMT, David Simms <dsimms at openjdk.org> wrote:

> Follows in the footsteps of JDK-8245216: [lworld] The JVM should inject the IdentityObject interface to types which need it
> Also includes a similar footprint tweak to JDK-8245646: [lworld] LW3 Reduce impact of IdentityObject on metaspace

Looks good to me.

Fred

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

Marked as reviewed by fparain (Committer).

PR: https://git.openjdk.java.net/valhalla/pull/349

From dsimms at openjdk.java.net  Thu Feb 25 05:37:55 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 05:37:55 GMT
Subject: [lworld] Integrated: 8262104: [lworld] Runtime should automagically
 have primitive classes implement PrimitiveObject
In-Reply-To: <Vr86oiGEeRYUnXHNXlNWFc1_lNwtPtD_YjHT5QsT-ag=.ad2d0e64-e2db-4e4f-a076-a7520f300339@github.com>
References: <Vr86oiGEeRYUnXHNXlNWFc1_lNwtPtD_YjHT5QsT-ag=.ad2d0e64-e2db-4e4f-a076-a7520f300339@github.com>
Message-ID: <a8glbDmfu2yjMMlX9339Wm9dtzF9HG2XSl1ngzTGksA=.249ef289-f887-4cf7-bed4-44be9ef051ac@github.com>

On Wed, 24 Feb 2021 15:18:24 GMT, David Simms <dsimms at openjdk.org> wrote:

> Follows in the footsteps of JDK-8245216: [lworld] The JVM should inject the IdentityObject interface to types which need it
> Also includes a similar footprint tweak to JDK-8245646: [lworld] LW3 Reduce impact of IdentityObject on metaspace

This pull request has now been integrated.

Changeset: b5bf7a90
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/b5bf7a90
Stats:     406 lines in 20 files changed: 382 ins; 10 del; 14 mod

8262104: [lworld] Runtime should automagically have primitive classes implement PrimitiveObject

Reviewed-by: fparain

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

PR: https://git.openjdk.java.net/valhalla/pull/349

From dsimms at openjdk.java.net  Thu Feb 25 06:49:15 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 06:49:15 GMT
Subject: [lworld] RFR: Merge jdk
Message-ID: <TKJvO6XFrTa_wolHyHYp90ZFYFJT6dbwMQ37y4qqWgY=.81ca3074-c5f0-4ccc-8e13-939689de8032@github.com>

Merge tag 'jdk-17+11'

# Conflicts:
#	src/hotspot/share/runtime/vmOperations.hpp
#	src/java.base/share/classes/java/lang/Object.java

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

Commit messages:
 - Merge tag 'jdk-17+11' into lworld_merge_jdk_17_11
 - 8254001: [Metrics] Enhance parsing of cgroup interface files for version detection
 - 8262315: missing ';' in generated entities
 - 8258855: Two tests sun/security/krb5/auto/ReplayCacheTestProc.java and ReplayCacheTestProcWithMD5.java failed on OL8.3
 - 8261919: java/util/Locale/LocaleProvidersRun.java failed with "RuntimeException: Expected log was not emitted. LogRecord: null"
 - 8261920: [AIX] jshell command throws java.io.IOError on non English locales
 - 8262259: Remove unused variable in MethodLiveness::BasicBlock::compute_gen_kill_single
 - 8262027: Improve how HttpConnection detects a closed channel when taking/returning a connection to the pool
 - 8256438: AArch64: Implement match rules with ROR shift register value
 - 8262269: javadoc test TestGeneratedClasses.java fails on Windows
 - ... and 99 more: https://git.openjdk.java.net/valhalla/compare/b5bf7a90...f96697b7

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=350&range=00.0
 - jdk: https://webrevs.openjdk.java.net/?repo=valhalla&pr=350&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/350/files
  Stats: 15675 lines in 391 files changed: 10831 ins; 2560 del; 2284 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/350.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/350/head:pull/350

PR: https://git.openjdk.java.net/valhalla/pull/350

From dsimms at openjdk.java.net  Thu Feb 25 07:20:55 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 07:20:55 GMT
Subject: [lworld] Integrated: Merge jdk
In-Reply-To: <TKJvO6XFrTa_wolHyHYp90ZFYFJT6dbwMQ37y4qqWgY=.81ca3074-c5f0-4ccc-8e13-939689de8032@github.com>
References: <TKJvO6XFrTa_wolHyHYp90ZFYFJT6dbwMQ37y4qqWgY=.81ca3074-c5f0-4ccc-8e13-939689de8032@github.com>
Message-ID: <-t4PwLw7UVy-3A7_7oqCG7zil2S5-tbmgr6ZhOm6oPY=.84f6503b-080b-4b53-9b98-fc1da7488805@github.com>

On Thu, 25 Feb 2021 06:44:31 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge tag 'jdk-17+11'
> 
> # Conflicts:
> #	src/hotspot/share/runtime/vmOperations.hpp
> #	src/java.base/share/classes/java/lang/Object.java

This pull request has now been integrated.

Changeset: ab51b601
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/ab51b601
Stats:     15675 lines in 391 files changed: 10831 ins; 2560 del; 2284 mod

Merge jdk

Merge tag 'jdk-17+11'

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

PR: https://git.openjdk.java.net/valhalla/pull/350

From thartmann at openjdk.java.net  Thu Feb 25 07:36:07 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Thu, 25 Feb 2021 07:36:07 GMT
Subject: [lworld] RFR: 8262308: [lworld] Various small C2 fixes for bugs found
 by stress testing
Message-ID: <_PzETwlIuvPzRlBimJJMW5JQaMemHcDhMwVwwGh50cY=.26eec163-07f6-4316-822e-e4ade2da2ad7@github.com>

More testing with stress flags enabled revealed the following bugs:
- We should not call `remove_dead_node` from `InlineTypeBaseNode::Ideal`. Instead, we should just make sure the dead node is on the IGVN worklist to be removed. I've also removed the `VerifyVM` option from the framework. Such flags can be added manually or via the CI job definitions.
- `PhaseMacroExpand::process_users_of_allocation` needs to handle `InlineTypePtrNodes` because we now keep them until after macro expansion.
- `GraphKit::type_check_receiver` should create a `InlineTypePtrNode` instead of a `InlineTypeNode` because it operates on pointers. 

I have a job definition file with all these stress flags enabled that I will integrate separately soon.

Best regards,
Tobias

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

Commit messages:
 - More fixes
 - 8262308: [lworld] VerifyIterativeGVN fails with "can not use dead node"

Changes: https://git.openjdk.java.net/valhalla/pull/351/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=351&range=00
  Issue: https://bugs.openjdk.java.net/browse/JDK-8262308
  Stats: 21 lines in 4 files changed: 11 ins; 7 del; 3 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/351.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/351/head:pull/351

PR: https://git.openjdk.java.net/valhalla/pull/351

From dsimms at openjdk.java.net  Thu Feb 25 08:23:17 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 08:23:17 GMT
Subject: RFR: Merge lworld
Message-ID: <peIcQEuVtXnpaBqMfCKoU4b2EeDvcxxuy8SLVG2ARC0=.1c778d23-c927-4ea7-ae8f-522e6b301ada@github.com>

Merge tag 'jdk-17+11'

# Conflicts:
#	src/hotspot/share/classfile/classFileParser.cpp
#	src/hotspot/share/classfile/classFileParser.hpp
#	src/hotspot/share/oops/instanceKlass.hpp
#	src/jdk.compiler/share/classes/com/sun/tools/javac/code/Symtab.java

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

Commit messages:
 - Merge branch 'lworld' into type-restrictions_merge_lworld_17_11
 - Merge jdk
 - 8254001: [Metrics] Enhance parsing of cgroup interface files for version detection
 - 8262315: missing ';' in generated entities
 - 8258855: Two tests sun/security/krb5/auto/ReplayCacheTestProc.java and ReplayCacheTestProcWithMD5.java failed on OL8.3
 - 8261919: java/util/Locale/LocaleProvidersRun.java failed with "RuntimeException: Expected log was not emitted. LogRecord: null"
 - 8261920: [AIX] jshell command throws java.io.IOError on non English locales
 - 8262259: Remove unused variable in MethodLiveness::BasicBlock::compute_gen_kill_single
 - 8262027: Improve how HttpConnection detects a closed channel when taking/returning a connection to the pool
 - 8256438: AArch64: Implement match rules with ROR shift register value
 - ... and 107 more: https://git.openjdk.java.net/valhalla/compare/75d27907...94ce3695

The webrevs contain the adjustments done while merging with regards to each parent branch:
 - type-restrictions: https://webrevs.openjdk.java.net/?repo=valhalla&pr=352&range=00.0
 - lworld: https://webrevs.openjdk.java.net/?repo=valhalla&pr=352&range=00.1

Changes: https://git.openjdk.java.net/valhalla/pull/352/files
  Stats: 16586 lines in 429 files changed: 11608 ins; 2624 del; 2354 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/352.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/352/head:pull/352

PR: https://git.openjdk.java.net/valhalla/pull/352

From dsimms at openjdk.java.net  Thu Feb 25 09:31:28 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 09:31:28 GMT
Subject: RFR: Merge lworld [v2]
In-Reply-To: <peIcQEuVtXnpaBqMfCKoU4b2EeDvcxxuy8SLVG2ARC0=.1c778d23-c927-4ea7-ae8f-522e6b301ada@github.com>
References: <peIcQEuVtXnpaBqMfCKoU4b2EeDvcxxuy8SLVG2ARC0=.1c778d23-c927-4ea7-ae8f-522e6b301ada@github.com>
Message-ID: <Zw-RbE3JfP_xfHYmuY0u88k9si6-nq0tTPHiTQesICY=.fa07ca4b-9079-45ce-b234-204368eefa30@github.com>

> Merge tag 'jdk-17+11'
> 
> # Conflicts:
> #	src/hotspot/share/classfile/classFileParser.cpp
> #	src/hotspot/share/classfile/classFileParser.hpp
> #	src/hotspot/share/oops/instanceKlass.hpp
> #	src/jdk.compiler/share/classes/com/sun/tools/javac/code/Symtab.java

David Simms has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains 21 commits:

 - Merge branch 'lworld' into type-restrictions_merge_lworld_17_11
   
   # Conflicts:
   #	src/hotspot/share/classfile/classFileParser.cpp
   #	src/hotspot/share/classfile/classFileParser.hpp
   #	src/hotspot/share/oops/instanceKlass.hpp
   #	src/jdk.compiler/share/classes/com/sun/tools/javac/code/Symtab.java
 - 8262202: [type-restrictions] RestrictedMethodTest failed to compile
 - 8261956: [type-restrictions] Support RestrictedMethod in the interpreter
   
   Reviewed-by: dsimms, hseigel
 - Merge lworld
   
   Merge 'jdk-17+10'
 - Merge lworld
   
   Merge lworld at jdk-17+8
 - 8260870: [type-restrictions] Generate RestrictedMethod attributes
 - 8257600: [type-restrictions] Implement RestrictedField in C1
   
   Reviewed-by: lfoltan, thartmann
 - Merge lworld
   
   Merge tag 'jdk-16+26'
 - 8257028: [type-restrictions] Assorted issues with generation of RestrictedField attributes from annotations
 - 8255856: Generate RestrictedField attributes from annotations
 - ... and 11 more: https://git.openjdk.java.net/valhalla/compare/ab51b601...94ce3695

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

Changes: https://git.openjdk.java.net/valhalla/pull/352/files
 Webrev: https://webrevs.openjdk.java.net/?repo=valhalla&pr=352&range=01
  Stats: 2395 lines in 62 files changed: 2306 ins; 12 del; 77 mod
  Patch: https://git.openjdk.java.net/valhalla/pull/352.diff
  Fetch: git fetch https://git.openjdk.java.net/valhalla pull/352/head:pull/352

PR: https://git.openjdk.java.net/valhalla/pull/352

From dsimms at openjdk.java.net  Thu Feb 25 09:31:29 2021
From: dsimms at openjdk.java.net (David Simms)
Date: Thu, 25 Feb 2021 09:31:29 GMT
Subject: Integrated: Merge lworld
In-Reply-To: <peIcQEuVtXnpaBqMfCKoU4b2EeDvcxxuy8SLVG2ARC0=.1c778d23-c927-4ea7-ae8f-522e6b301ada@github.com>
References: <peIcQEuVtXnpaBqMfCKoU4b2EeDvcxxuy8SLVG2ARC0=.1c778d23-c927-4ea7-ae8f-522e6b301ada@github.com>
Message-ID: <C8Mi3Pcrw1T2cVJJ72st8aagbAgfE7T8ao_B5eUoOhg=.5ded63e9-cbbf-49f7-8cae-526927e4907f@github.com>

On Thu, 25 Feb 2021 08:18:14 GMT, David Simms <dsimms at openjdk.org> wrote:

> Merge tag 'jdk-17+11'
> 
> # Conflicts:
> #	src/hotspot/share/classfile/classFileParser.cpp
> #	src/hotspot/share/classfile/classFileParser.hpp
> #	src/hotspot/share/oops/instanceKlass.hpp
> #	src/jdk.compiler/share/classes/com/sun/tools/javac/code/Symtab.java

This pull request has now been integrated.

Changeset: 0a26489c
Author:    David Simms <dsimms at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/0a26489c
Stats:     16586 lines in 429 files changed: 11608 ins; 2624 del; 2354 mod

Merge lworld

Merge tag 'jdk-17+11'

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

PR: https://git.openjdk.java.net/valhalla/pull/352

From nick.gasson at arm.com  Thu Feb 25 10:03:51 2021
From: nick.gasson at arm.com (Nick Gasson)
Date: Thu, 25 Feb 2021 18:03:51 +0800
Subject: Status of AArch64 lworld port
In-Reply-To: <1853542c-80f4-db40-97ae-f3b3b74f91d6@oracle.com>
References: <85a6sidhyi.fsf@nicgas01-pc.shanghai.arm.com>
 <1853542c-80f4-db40-97ae-f3b3b74f91d6@oracle.com>
Message-ID: <85a6rslb08.fsf@arm.com>

On 02/05/21 17:17 pm, Tobias Hartmann wrote:
>
> yes, the AArch64 port misses significant changes and is currently not tested on our (Oracle) side.
>
> Dmitriy Samersoff worked on this in the past but I'm not sure what his current plans are. In any
> case, contributions are very welcome and I'm happy to answer questions about the JIT related changes.
>

Hi Tobias,

OK I started working on this. I've put my in-progress changes as draft
pull request here:

https://github.com/openjdk/valhalla/pull/353

It's not ready for review yet but simple test cases are passing. For the
time being I'm building without C2 and focusing on getting
C1/interpreter to work.

I had to make one change/hack to the shared code in
LIRGenerator::get_and_load_element_address():

https://github.com/openjdk/valhalla/pull/353/files#diff-cc27fc51caa691478ce875140a75940aea610cb6b3a67976852867b05672e56c

The problem is LIR_Address on AArch64 cannot have both an index and a
displacement. I thought it should be possible to call
LIRGenerator::generate_address() instead of explicitly constructing a
new LIR_Address() because generate_address() handles the necessary
adjustments in the AArch64-specific code. However that fails with the
error below and I haven't worked out why yet.

Error: live_in set of first block must be empty (when this fails, virtual registers are used before they are defined)
affected registers:
344 353 360 416 440 456
* vreg 344 (HIR instruction  0)
  used in block B0
  ...

--
Thanks,
Nick

From thartmann at openjdk.java.net  Thu Feb 25 11:04:56 2021
From: thartmann at openjdk.java.net (Tobias Hartmann)
Date: Thu, 25 Feb 2021 11:04:56 GMT
Subject: [lworld] Integrated: 8262308: [lworld] Various small C2 fixes for
 bugs found by stress testing
In-Reply-To: <_PzETwlIuvPzRlBimJJMW5JQaMemHcDhMwVwwGh50cY=.26eec163-07f6-4316-822e-e4ade2da2ad7@github.com>
References: <_PzETwlIuvPzRlBimJJMW5JQaMemHcDhMwVwwGh50cY=.26eec163-07f6-4316-822e-e4ade2da2ad7@github.com>
Message-ID: <-1EQ90G8UytrW1o6K2xDzXYVH05GlkUrXwk5DyAjHFQ=.cc2cb8b8-91e7-4909-8f91-43d024016663@github.com>

On Thu, 25 Feb 2021 07:31:27 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> More testing with stress flags enabled revealed the following bugs:
> - We should not call `remove_dead_node` from `InlineTypeBaseNode::Ideal`. Instead, we should just make sure the dead node is on the IGVN worklist to be removed. I've also removed the `VerifyVM` option from the framework. Such flags can be added manually or via the CI job definitions.
> - `PhaseMacroExpand::process_users_of_allocation` needs to handle `InlineTypePtrNodes` because we now keep them until after macro expansion.
> - `GraphKit::type_check_receiver` should create a `InlineTypePtrNode` instead of a `InlineTypeNode` because it operates on pointers. 
> 
> I have a job definition file with all these stress flags enabled that I will integrate separately soon.
> 
> Best regards,
> Tobias

This pull request has now been integrated.

Changeset: 0dfcc948
Author:    Tobias Hartmann <thartmann at openjdk.org>
URL:       https://git.openjdk.java.net/valhalla/commit/0dfcc948
Stats:     21 lines in 4 files changed: 11 ins; 7 del; 3 mod

8262308: [lworld] Various small C2 fixes for bugs found by stress testing

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

PR: https://git.openjdk.java.net/valhalla/pull/351