8311604: Simplify NOCOOPS requested addresses for archived heap objects #14792

iklam · 2023-07-06T23:39:34Z

This PR attempts to clean up some of the cruds in the existing code:

Simplified the calculation of "requested address" when UseCompressedOops is disabled -- the archived heap objects are always written starting from 0x10000000
Removed HeapShared::to_requested_address() so we don't have two kinds of "requested address"
Updated the comments about "source" vs "buffered" vs "requested" addresses in archiveHeapWriter.hpp
Removed SerializeClosure::oop() as the only oop we need to store into the archive header is HeapShared::roots(), which can be handled more easily with FileMapHeader::_heap_roots_offset
Removed some unnecessary dependencies on G1CollectedHeap::heap()->reserved()

Also:

Moved SerializeClosure to its own header file to improve build time.
Fixed DeterministicDump.java, which wasn't archiving Java objects when UseCompressedOops was disabled.

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-8311604: Simplify NOCOOPS requested addresses for archived heap objects (Enhancement - P4)

Reviewers

@ashu-mehra (no known openjdk.org user name / role) ⚠️ Review applies to 88d2b338
Calvin Cheung (@calvinccheung - Reviewer) ⚠️ Review applies to fb8a2165

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/14792/head:pull/14792
$ git checkout pull/14792

Update a local copy of the PR:
$ git checkout pull/14792
$ git pull https://git.openjdk.org/jdk.git pull/14792/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 14792

View PR using the GUI difftool:
$ git pr show -t 14792

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/14792.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2023-07-06T23:40:19Z

👋 Welcome back iklam! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2023-07-06T23:42:36Z

@iklam The following label will be automatically applied to this pull request:

hotspot

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

mlbridge · 2023-07-06T23:47:14Z

Webrevs

ashu-mehra · 2023-07-07T15:06:49Z

src/hotspot/share/cds/archiveHeapWriter.hpp

+  //   stored into the CDS archive. These are entered into HeapShared::archived_object_cache().
+  //
+  // - "buffered objects" are copies of the "source objects", and are stored in into
+  //   ArchiveHeapWriter::_buffer, which is a GrowableArray that sites outside of


s/sites/sits

ashu-mehra · 2023-07-07T15:10:20Z

src/hotspot/share/cds/filemap.cpp

@@ -2086,7 +2069,7 @@ address FileMapInfo::heap_region_requested_address() {
  } else {
    // We can avoid relocation if each region is mapped into the exact same address
    // where it was at dump time.
-    return /*dumptime*/header()->heap_begin() + r->mapping_offset();
+    return (address)ArchiveHeapWriter::NOCOOPS_REQUESTED_BASE;


Can you please update the comment above this line as well to remove each region as we only have single heap region now.

Thanks for the review. I updated the comment to clarify that this address was hard-coded at dump time, and explained the reason for that.

ashu-mehra · 2023-07-07T15:12:08Z

minor nitpicks, otherwise looks good!

calvinccheung

Just couple of nits.

calvinccheung · 2023-07-07T21:56:06Z

src/hotspot/share/cds/serializeClosure.hpp

+#ifndef SHARED_CDS_SERIALIZECLOSURE_HPP
+#define SHARED_CDS_SERIALIZECLOSURE_HPP


Since the directory name is share, I think the above should be SHARE_ instead of SHARED_.

calvinccheung · 2023-07-07T21:56:51Z

src/hotspot/share/cds/serializeClosure.hpp

+  template <typename T> void do_ptr(T** p) { do_ptr((void**)p); }
+};
+
+#endif // SHARED_CDS_SERIALIZECLOSURE_HPP


Should be SHARE_ instead of SHARED_.

I fixed the SHARED_ include guards in this file, plus a few other headers in the share/cds/directory. I also took this opportunity to fix some indent issues in cds_globals.hpp.

calvinccheung · 2023-07-07T22:27:27Z

src/hotspot/share/cds/filemap.cpp

-  st->print_cr("- heap_begin:                     " INTPTR_FORMAT, p2i(_heap_begin));
-  st->print_cr("- heap_end:                       " INTPTR_FORMAT, p2i(_heap_end));


Maybe print the _heap_roots_offset?

Hi Calvin, thanks for the review. I added the _heap_roots_offset into the debug output.

openjdk · 2023-07-07T22:32:55Z

@iklam This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

After integration, the commit message for the final commit will be:

8311604: Simplify NOCOOPS requested addresses for archived heap objects

Reviewed-by: ccheung

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been no new commits pushed to the master branch. If another commit should be pushed before you perform the /integrate command, your PR will be automatically rebased. If you prefer to avoid any potential automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

…ed line escapes in cds_globals.hpp

…r-archived-heap

iklam · 2023-07-09T15:17:52Z

Thanks @ashu-mehra and @calvinccheung for the review.
/integrate

openjdk · 2023-07-09T15:18:49Z

Going to push as commit 581f90e.

openjdk · 2023-07-09T15:18:58Z

@iklam Pushed as commit 581f90e.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

8311604: Simplify NOCOOPS requested addresses for archived heap objects

88d2b33

openjdk bot added the rfr Pull request is ready for review label Jul 6, 2023

openjdk bot added the hotspot hotspot-dev@openjdk.org label Jul 6, 2023

ashu-mehra approved these changes Jul 7, 2023

View reviewed changes

@ashu-mehra review; updated source code comments

fb8a216

calvinccheung approved these changes Jul 7, 2023

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jul 7, 2023

iklam added 3 commits July 7, 2023 17:17

@calvinccheung review: fixed include guard in headers; fixed misalign…

e1db066

…ed line escapes in cds_globals.hpp

@calvinccheung review: print heap_roots_offset

7f37fd2

Merge branch 'master' into 8311604-simplify-nocoops-requested-addr-fo…

d1b56a2

…r-archived-heap

openjdk bot added the integrated Pull request has been integrated label Jul 9, 2023

openjdk bot closed this Jul 9, 2023

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jul 9, 2023

		#ifndef SHARED_CDS_SERIALIZECLOSURE_HPP
		#define SHARED_CDS_SERIALIZECLOSURE_HPP

		st->print_cr("- heap_begin: " INTPTR_FORMAT, p2i(_heap_begin));
		st->print_cr("- heap_end: " INTPTR_FORMAT, p2i(_heap_end));

8311604: Simplify NOCOOPS requested addresses for archived heap objects #14792

8311604: Simplify NOCOOPS requested addresses for archived heap objects #14792

Uh oh!

Conversation

iklam commented Jul 6, 2023 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Webrev

Uh oh!

bridgekeeper bot commented Jul 6, 2023

Uh oh!

openjdk bot commented Jul 6, 2023

Uh oh!

mlbridge bot commented Jul 6, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ashu-mehra commented Jul 7, 2023

Uh oh!

calvinccheung left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Jul 7, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

iklam commented Jul 9, 2023

Uh oh!

openjdk bot commented Jul 9, 2023

Uh oh!

openjdk bot commented Jul 9, 2023

Uh oh!

Uh oh!

iklam commented Jul 6, 2023 •

edited by openjdk bot

Loading

mlbridge bot commented Jul 6, 2023 •

edited

Loading

openjdk bot commented Jul 7, 2023 •

edited

Loading