Store segment size in SHSegmentInfo #1506
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D61807366 |
lavenzg
added a commit
to lavenzg/hermes
that referenced
this pull request
Oct 21, 2024
Summary: We need the segment size in several places, such as CardTable, heap segment itself and getting sizes for large object GCCell. Add this size into `SHSegmentInfo`. In addition, in CardTableNCTest, when search dirty bits, we should start from kFirstUsedIndex instead of 0, since the value of `SHSegmentInfo` may write some bytes to non-zero. Differential Revision: D61807366
lavenzg
force-pushed
the
export-D61807366-to-static_h
branch
from
90bc2f2 to
0820b2b
Compare
lavenzg
added a commit
to lavenzg/hermes
that referenced
this pull request
Oct 21, 2024
Summary: We need the segment size in several places, such as CardTable, heap segment itself and getting sizes for large object GCCell. Add this size into `SHSegmentInfo`. In addition, in CardTableNCTest, when search dirty bits, we should start from kFirstUsedIndex instead of 0, since the value of `SHSegmentInfo` may write some bytes to non-zero. Differential Revision: D61807366
|
This pull request was exported from Phabricator. Differential Revision: D61807366 |
lavenzg
added a commit
to lavenzg/hermes
that referenced
this pull request
Oct 22, 2024
Summary: We need the segment size in several places, such as CardTable, heap segment itself and getting sizes for large object GCCell. Add this size into `SHSegmentInfo`. In addition, in CardTableNCTest, when search dirty bits, we should start from kFirstUsedIndex instead of 0, since the value of `SHSegmentInfo` may write some bytes to non-zero. Differential Revision: D61807366
…dHeapSegmentBase (facebook#1510) Summary: The large heap segment type should have the same storage layout as current AlignedHeapSegment, and share a few common methods. Abstract these to a base class, and make both AlignedHeapSegment and JumboHeapSegment inherit from the base type. Differential Revision: D61675022
…er (facebook#1504) Summary: Large segment needs to be backed by a large storage size. StorageProvider currently always allocate fixed size of storage determined by HERMESVM_LOG_HEAP_SEGMENT_SIZE. This diffs adds support of allocating larger storage with below changes: 1. `newStorage()` and `deleteStorage()` takes additional `sz` parameter. 2. For `MallocStorageProvider` and `VMAllocateStorageProvider`, simply change the previous fixed storage size to passed in `sz`. 3. For `ContiguousVAStorageProvider`, use a BitVector to manage allocations and deallocations. This can be improved later if we observe fragmentations. The support of enabling different sizes of heap segment will be added later. Differential Revision: D61676721
Summary: Currently `SHSegmentInfo` lives in the prefix of CardTable inline storage (to be specific, prefix of the `cards_` array). But this is only defined in one comment. Add it into a union with the `cards_` array to make it clear. It also simplifies the reasoning of following diffs, in which we need to add more fields to `SHSegmentInfo`. In addition, `kFirstUsedIndex` should take into account of the size of `SHSegmentInfo`, since the size of `SHSegmentInfo` could be larger than `(2 * kCardTableSize) >> kLogCardSize)` for small segment size. Differential Revision: D61747499
Summary: We need the segment size in several places, such as CardTable, heap segment itself and getting sizes for large object GCCell. Add this size into `SHSegmentInfo`. In addition, in CardTableNCTest, when search dirty bits, we should start from kFirstUsedIndex, since the value o `SHSegmentInfo` may write some bytes to non-zero. Differential Revision: D61807366
lavenzg
force-pushed
the
export-D61807366-to-static_h
branch
from
0820b2b to
08045e6
Compare
|
This pull request was exported from Phabricator. Differential Revision: D61807366 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
We need the segment size in several places, such as CardTable, heap
segment itself and getting sizes for large object GCCell. Add this size
into
SHSegmentInfo.In addition, in CardTableNCTest, when search dirty bits, we should
start from kFirstUsedIndex instead of 0, since the value of
SHSegmentInfomay write some bytes to non-zero.Differential Revision: D61807366