8251330: Reorder CDS archived heap to speed up relocation #17350

matias9927 · 2024-01-10T17:19:39Z

We should reorder the archived heap to segregate the objects that don't need marking. This will save space in the archive and improve start-up time

This patch reorders the archived heap to segregate the objects that don't need marking. The leading zeros present in the bitmaps thanks to the reordering can be easily removed, and this the size of the archive is reduced.

Given that the bitmaps are word aligned, some leading zeroes will remain. Verified with tier 1-5 tests.

Metrics, courtesy of @iklam :

Oopmap = 15872 bytes

calculate_oopmap: objects = 4590 (335872 bytes, 178120 bytes don't need marking), embedded oops = 46487, nulls = 29019
Oopmap = 10496 bytes

(332752 + 178120) / (507904 + 335872.0) = 0.6054592688106796 
More than 60% of the space used by the archived heap doesn't need to be marked by the oopmap. 

$ java -Xshare:dump -Xlog:cds+map | grep lead
[3.777s][info][cds,map] - heap_oopmap_leading_zeros: 143286
[3.777s][info][cds,map] - heap_ptrmap_leading_zeros: 50713

So we can reduce the "bm" region by (143286 + 50713) / 8 = 24249 bytes.

Current output:
$ java -XX:+UseNewCode -Xshare:dump -Xlog:cds+map | grep lead 
[5.339s][info][cds,map] - heap_oopmap_leading_zeros:      26
[5.339s][info][cds,map] - heap_ptrmap_leading_zeros:      8

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-8251330: Reorder CDS archived heap to speed up relocation (Enhancement - P4)

Reviewers

Ioi Lam (@iklam - Reviewer) ⚠️ Review applies to bdd74ec9
Calvin Cheung (@calvinccheung - Reviewer) ⚠️ Review applies to c06c4d40

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/17350/head:pull/17350
$ git checkout pull/17350

Update a local copy of the PR:
$ git checkout pull/17350
$ git pull https://git.openjdk.org/jdk.git pull/17350/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 17350

View PR using the GUI difftool:
$ git pr show -t 17350

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/17350.diff

Webrev

Link to Webrev Comment

bridgekeeper · 2024-01-10T17:21:10Z

👋 Welcome back matsaave! 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 · 2024-01-10T17:23:02Z

@matias9927 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 · 2024-02-22T17:12:04Z

Webrevs

iklam

Looks good. I have some minor remarks for refactoring.

iklam · 2024-02-27T19:37:56Z

src/hotspot/share/cds/archiveHeapLoader.cpp

@@ -164,18 +164,19 @@ void ArchiveHeapLoader::patch_compressed_embedded_pointers(BitMapView bm,

  // Optimization: if dumptime shift is the same as runtime shift, we can perform a
  // quick conversion from "dumptime narrowOop" -> "runtime narrowOop".
+  narrowOop* patching_start = (narrowOop*)region.start() + MetaspaceShared::oopmap_start_pos();


I think there is no need to introduce a new MetaspaceShared::oopmap_start_pos() API. You can get this info from

FileMapInfo::current_info()->header()->heap_oopmap_start_pos()

The same thing can be done for MetaspaceShared::ptrmap_start_pos()

iklam · 2024-02-27T19:39:58Z

src/hotspot/share/cds/filemap.cpp

@@ -289,6 +289,8 @@ void FileMapHeader::print(outputStream* st) {
  st->print_cr("- requested_base_address:         " INTPTR_FORMAT, p2i(_requested_base_address));
  st->print_cr("- mapped_base_address:            " INTPTR_FORMAT, p2i(_mapped_base_address));
  st->print_cr("- heap_roots_offset:              " SIZE_FORMAT, _heap_roots_offset);
+  st->print_cr("- _heap_oopmap_start_pos:      " SIZE_FORMAT, _heap_oopmap_start_pos);
+  st->print_cr("- _heap_ptrmap_start_pos:      " SIZE_FORMAT, _heap_ptrmap_start_pos);


Need two more spaces before SIZE_FORMAT for alignment?

iklam · 2024-02-27T20:58:31Z

src/hotspot/share/cds/filemap.cpp

+  size_t new_zeros = map->find_first_set_bit(0);
+  size_t removed_zeros = old_zeros - new_zeros;
+
+  assert(new_zeros <= old_zeros, "Should have removed leading zeros");


Maybe add this comment above:

// The start of the archived heap has many primitive arrays (String // bodies) that are not marked by the oop/ptr maps. So we must have // lots of leading zeros.

iklam · 2024-02-27T21:08:00Z

src/hotspot/share/cds/archiveHeapWriter.cpp

@@ -565,6 +634,12 @@ bool ArchiveHeapWriter::is_marked_as_native_pointer(ArchiveHeapInfo* heap_info,
  assert((Metadata**)_requested_bottom <= requested_field_addr && requested_field_addr < (Metadata**) _requested_top, "range check");

  BitMap::idx_t idx = requested_field_addr - (Metadata**) _requested_bottom;
+  // Leading zeros have been removed so some addresses may not be in the ptrmap
+  if (idx < MetaspaceShared::ptrmap_start_pos()) {


This is now the only place you need to know ptrmap_start_pos during dump time (at runtime you can get the info from FileMapInfo::current_info()). I think it's better to store ptrmap_start_pos as a static field in the ArchiveHeapWriter class instead.

iklam

LGTM. Just some small nits.

iklam · 2024-03-12T05:14:33Z

src/hotspot/share/cds/filemap.cpp

+  size_t new_zeros = map->find_first_set_bit(0);
+
+  assert(new_zeros == 0, "Should have removed leading zeros");
+  assert(map->size_in_bytes() <= old_size, "Map size should have decreased");


The comment doesn't match the code, since you are comparing with <=. I think it's safe to change to < as we should have removed more than 64 leading bits, due to the abundance of primitive arrays.

iklam · 2024-03-12T05:18:34Z

src/hotspot/share/cds/heapShared.cpp

@@ -1138,10 +1156,26 @@ class WalkOopAndArchiveClosure: public BasicOopIterateClosure {

 WalkOopAndArchiveClosure* WalkOopAndArchiveClosure::_current = nullptr;

-HeapShared::CachedOopInfo HeapShared::make_cached_oop_info() {
+class PointsToOopsChecker : public BasicOopIterateClosure {


This class needs a short comment. How about:

// Checks if an oop has any non-null oop fields

openjdk · 2024-03-12T05:21:35Z

@matias9927 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:

8251330: Reorder CDS archived heap to speed up relocation

Reviewed-by: iklam, 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 27 new commits pushed to the master branch:

22f10e0: 8327856: Convert applet test SpanishDiacriticsTest.java to a main program
7283c8b: 8327972: Convert java/awt/FileDialog/SaveFileNameOverrideTest/SaveFileNameOverrideTest.html applet test to main
30249c4: 8327838: Convert java/awt/FileDialog/MultipleMode/MultipleMode.html applet test to main
94b4ed5: 8327096: (fc) java/nio/channels/FileChannel/Size.java fails on partition incapable of creating large files
b9c3dc3: 8327738: Remove unused internal method sun.n.w.p.h.HttpURLConnection.setDefaultAuthenticator
5b41466: 8327729: Remove deprecated xxxObject methods from jdk.internal.misc.Unsafe
313e814: 8324682: Remove redefinition of NULL for XLC compiler
8a3bdd5: 8327995: Remove unused Unused_Variable
201042f: 8327487: Further augment WorstCaseTests with more cases
379ad1f: 8312444: Delete unused parameters and variables in SocketPermission
... and 17 more: https://git.openjdk.org/jdk/compare/c65d308973f1e2c41b6910fd844223597e70972f...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this 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.

calvinccheung

Looks good. I just have a minor comment below.
Also, the copyright year of some files need to be updated.

calvinccheung · 2024-03-12T17:44:49Z

src/hotspot/share/cds/filemap.cpp

+  map->truncate(old_zeros, map->size());
+
+  // We want to keep track of how many zeros were removed
+  size_t new_zeros = map->find_first_set_bit(0);


I think this statement could be enclosed with DEBUG_ONLY since the new_zeros is only used in the assert statement following it.

matias9927 · 2024-03-13T14:00:06Z

Thanks for the reviews @calvinccheung and @iklam!
/integrate

openjdk · 2024-03-13T14:01:00Z

Going to push as commit 7e05a70.
Since your change was applied there have been 40 commits pushed to the master branch:

7d8561d: 8327109: Refactor data graph cloning used in create_new_if_for_predicate() into separate class
a4a5196: 8327872: Convert javax/swing/JToolTip/4644444/bug4644444.java applet test to main
da4dd7c: 8327989: java/net/httpclient/ManyRequest.java should not use "localhost" in URIs
49d8008: 8327452: G1: Improve scalability of Merge Log Buffers
0ae4fa7: 8327210: AIX: Delete obsolete parameter Use64KPagesThreshold
107cb53: 8327701: Remove the xlc toolchain
07acc0b: 8326385: [aarch64] C2: lightweight locking nodes kill the box register without specifying this effect
cc9a8ab: 8327460: Compile tests with the same visibility rules as product code
3b18c5d: 8323605: Java source launcher should not require --source ... to enable preview
5d4bfad: 8327693: C1: LIRGenerator::_instruction_for_operand is only read by assertion code
... and 30 more: https://git.openjdk.org/jdk/compare/c65d308973f1e2c41b6910fd844223597e70972f...master

Your commit was automatically rebased without conflicts.

openjdk · 2024-03-13T14:01:07Z

@matias9927 Pushed as commit 7e05a70.

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

matias9927 added 3 commits November 21, 2023 16:22

8251330: Reorder CDS archived heap to speed up relocation

be4c573

Added bitmap slicing

a9de204

Improved slicing

b7c4daf

openjdk bot added the hotspot label Jan 10, 2024

matias9927 added 24 commits January 10, 2024 17:31

Added asserts

18598ad

Merge branch 'master' into reorder_archive_heap_8251330

8d58577

Removed UseNewCode

634964d

Merge branch 'master' into reorder_archive_heap_8251330

dbbf2e2

Changed buffer size and cleanup

1e20080

Bitmap size reduced

fb6c748

OSX fix and cleanup

1a479a1

Added oopmap offset

837d6cc

Removed UseNewCode

2f12930

Cleanup and fix

790889b

Removed debug print

43b6315

Merge branch 'master' into reorder_archive_heap_8251330

6326b2e

Relocated code to fix test failures

b9cb60c

cleanup

40da429

Merge branch 'master' into reorder_archive_heap_8251330

76b3193

Initialized leading zeros

8bf2b74

Merge branch 'master' into reorder_archive_heap_8251330

7b76b73

Improved setting leading zeros

733c7cf

Only set zeros once

c865503

Fixed sorting of equal rank objs

14369fe

Merge branch 'master' into reorder_archive_heap_8251330

14029c9

Merge branch 'master' into reorder_archive_heap_8251330

817a37e

Adjusted word alignment in slice method

ca03e00

Removed commented code

bf759be

matias9927 marked this pull request as ready for review February 22, 2024 17:07

openjdk bot added the rfr label Feb 22, 2024

matias9927 added 3 commits February 22, 2024 18:14

Merge branch 'master' into reorder_archive_heap_8251330

021fe98

Added helper function

bb1409d

Renamed variables and moved pointer patching shift

411bc1a

iklam suggested changes Feb 27, 2024

View reviewed changes

matias9927 added 8 commits February 28, 2024 20:26

Slice method now copies and Ioi Comments

3fb3c60

Merge branch 'master' into reorder_archive_heap_8251330

939536e

Removed unused methods and reordered code

029e602

Updated slice and truncate to be bit aligned

f665662

Moved slice method to new PR

f8a141e

Merge branch 'master' into reorder_archive_heap_8251330

95c4083

Merge branch 'master' into reorder_archive_heap_8251330

a09cd64

Removed metaspaceshared changes

bdd74ec

iklam approved these changes Mar 12, 2024

View reviewed changes

openjdk bot added the ready label Mar 12, 2024

Nits

c06c4d4

calvinccheung approved these changes Mar 12, 2024

View reviewed changes

Fixed copyright headers and made assert debug only

c45d58c

openjdk bot added the integrated label Mar 13, 2024

openjdk bot closed this Mar 13, 2024

openjdk bot removed ready rfr labels Mar 13, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

8251330: Reorder CDS archived heap to speed up relocation #17350

8251330: Reorder CDS archived heap to speed up relocation #17350

matias9927 commented Jan 10, 2024 •

edited by openjdk bot

Loading

bridgekeeper bot commented Jan 10, 2024

openjdk bot commented Jan 10, 2024

mlbridge bot commented Feb 22, 2024 •

edited

Loading

iklam left a comment

iklam Feb 27, 2024

iklam Feb 27, 2024

iklam Feb 27, 2024

iklam Feb 27, 2024

iklam left a comment

iklam Mar 12, 2024

iklam Mar 12, 2024

openjdk bot commented Mar 12, 2024 •

edited

Loading

calvinccheung left a comment

calvinccheung Mar 12, 2024

matias9927 commented Mar 13, 2024

openjdk bot commented Mar 13, 2024

openjdk bot commented Mar 13, 2024

8251330: Reorder CDS archived heap to speed up relocation #17350

8251330: Reorder CDS archived heap to speed up relocation #17350

Conversation

matias9927 commented Jan 10, 2024 • edited by openjdk bot Loading

Progress

Issue

Reviewers

Reviewing

Webrev

bridgekeeper bot commented Jan 10, 2024

openjdk bot commented Jan 10, 2024

mlbridge bot commented Feb 22, 2024 • edited Loading

Webrevs

iklam left a comment

Choose a reason for hiding this comment

iklam Feb 27, 2024

Choose a reason for hiding this comment

iklam Feb 27, 2024

Choose a reason for hiding this comment

iklam Feb 27, 2024

Choose a reason for hiding this comment

iklam Feb 27, 2024

Choose a reason for hiding this comment

iklam left a comment

Choose a reason for hiding this comment

iklam Mar 12, 2024

Choose a reason for hiding this comment

iklam Mar 12, 2024

Choose a reason for hiding this comment

openjdk bot commented Mar 12, 2024 • edited Loading

calvinccheung left a comment

Choose a reason for hiding this comment

calvinccheung Mar 12, 2024

Choose a reason for hiding this comment

matias9927 commented Mar 13, 2024

openjdk bot commented Mar 13, 2024

openjdk bot commented Mar 13, 2024

matias9927 commented Jan 10, 2024 •

edited by openjdk bot

Loading

mlbridge bot commented Feb 22, 2024 •

edited

Loading

openjdk bot commented Mar 12, 2024 •

edited

Loading