Commit Graph

25 Commits

Author SHA1 Message Date
Andreas Haas
aba951101c [arraybuffer][test] Test creating a BackingStore with nullptr
R= ulan@chromium.org

Bug: chromium:1104580
Change-Id: I5824de7aa4c71b1464fc8b2c35375bcb5d84f2e9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2320329
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69073}
2020-07-27 14:04:40 +00:00
Anna Henningsen
e29c62b748 [arraybuffer] Clean up BackingStore even if it pointer to nullptr
For a zero-length BackingStore allocation, it is valid for the
underlying memory to be a null pointer. However, some cleanup
is still necessary, since the BackingStore may hold a reference
to the allocator itself, which needs to be released when destroying
the `BackingStore` instance.

Change-Id: I1f168079d39e4592d2fde31fbe5f705586690e85
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2169646
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#67420}
2020-04-28 08:29:43 +00:00
Ulan Degenbaev
5cf02f0f20 [api] Support v8::BackingStores with empty deleters
This adjusts v8::[Shared]ArrayBuffer::NewBackingStore to allow passing
a known empty deleter -- v8::BackingStore::EmptyDeleter. Such API is
useful if the backing store memory is static or is manually managed.

We can skip adjusting the amount of external memory for ArrayBuffers
with empty deleters and thus avoid scheduling ineffective GCs.

Bug: chromium:1061960

Change-Id: I0ef5b2b0839098beb59d5cebbb28f9f81a73a042
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2105355
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#66740}
2020-03-17 08:10:24 +00:00
Ulan Degenbaev
b5c917ee80 [api] New function for reallocating ArrayBuffer backing store
This patch adds a new BackingStore::Reallocate function that internally
uses a new ArrayBuffer::Allocator::Reallocate provided by the embedder.

The default implementation of the function simply copies the backing
store. The embedder can override the function and provide a more
efficient implementation e.g. using realloc.

Bug: v8:9908, v8:9380

Change-Id: I2179c80ba199c045b6900c620a813916150e7098
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2007274
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#66044}
2020-01-30 12:05:17 +00:00
Andreas Haas
bfe3d6bce7 [api] Deprecate [Shared]ArrayBuffer::Externalize/GetContents and constructors
The new API with v8::BackingStore should be used instead as explained in
https://docs.google.com/document/d/1sTc_jRL87Fu175Holm5SV0kajkseGl2r8ifGY76G35k

This also relaxes the pre-condition for [Shared]ArrayBuffer::Detach to
not require externalization first.

Bug: v8:9380, v8:9908
Change-Id: Idd119fcd28be84a2fae74ae86f7381fd997766f5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1859628
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64625}
2019-10-29 22:01:54 +00:00
Anna Henningsen
6b0a9535e6 [api] Add possibility for BackingStore to keep Allocator alive
Add an `array_buffer_allocator_shared` field to the
`Isolate::CreateParams` struct that allows embedders to share
ownership of the ArrayBuffer::Allocator with V8, and which in
particular means that when this method is used that the
BackingStore deleter will not perform an use-after-free access to the
Allocator under certain circumstances.

For Background:

tl;dr: This is necessary for Node.js to perform the transition to
V8 7.9, because of the way that ArrayBuffer::Allocators and their
lifetimes currently work there.

In Node.js, each Worker thread has its own ArrayBuffer::Allocator.
Changing that would currently be impractical, as each allocator
depends on per-Isolate state. However, now that backing stores
are managed globally and keep a pointer to the original
ArrayBuffer::Allocator, this means that when transferring an
ArrayBuffer (e.g. from one Worker to another through postMessage()),
the original Allocator has to be kept alive until the ArrayBuffer
no longer exists in the receiving Isolate (or until that Isolate
is disposed). See [1] for an example Node.js test that fails with
V8 7.9.

This problem also existed for SharedArrayBuffers, where Node.js
was broken by V8 earlier for the same reasons (see [2] for the bug
report on that and [3] for the resolution in Node.js).
For SharedArrayBuffers, we already had extensive tracking logic,
so adding a shared_ptr to keep alive the ArrayBuffer::Allocator
was not a significant amount of work. However, the mechanism for
transferring non-shared ArrayBuffers is quite different, and
it seems both easier for us and better for V8 from an API standpoint
to keep the Allocator alive from where it is being referenced.

By sharing memory with the custom deleter function/data pair,
this comes at no memory overhead.

[1]: https://github.com/nodejs/node/pull/30044
[2]: https://github.com/nodejs/node-v8/issues/115
[3]: https://github.com/nodejs/node/pull/29637

Bug: v8:9380
Change-Id: Ibc2c4fb6341b53653cbd637bd8cb3d4ac43809c7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1874347
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64542}
2019-10-24 10:56:03 +00:00
Ulan Degenbaev
10b5e80637 [api] Add v8::BackingStore::IsShared
The new predicate indicates whether the backing store was created for
an ArrayBuffer or a SharedArrayBuffer. It is useful for some embedders.

Bug: v8:9380
Change-Id: I804063bb8c4c17815defd6538ce6a1b32f6a4531
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1873689
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64479}
2019-10-22 15:36:42 +00:00
Ulan Degenbaev
bba5f1f43d [api] Add API functions for constructing standalone BackingStores
These are useful for the cases when an embedder works with backing
stores without creating JS ArrayBuffer objects.

Bug: v8:9380
Change-Id: I452bd911e7b20fb38568f18f9d15ea1a7ffb5a57
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1825339
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64460}
2019-10-22 12:00:02 +00:00
Ulan Degenbaev
f7adf5f569 Avoid deduplication overhead for backing stores managed by embedder
If a BackingStore is marked as !free_on_destruct, then we don't have to
guarantee that there is only one such BackingStore pointing to the
underlying buffer. So we can skip costly registration in process-global
table of backing stores.

Bug: v8:9380,chromium:1002693
Change-Id: Iad1ec5c4811d6c52a9a9d78dd700acf69170db60
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1815136
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63941}
2019-09-24 10:39:25 +00:00
Ulan Degenbaev
c199f828a8 Fix initalization of embedder fields for ArrayBuffers
This moves the initialization to JSArrayBuffer::SetupEmpty, which is the
proper bottleneck for all paths constructing array buffers.

Bug: chromium:1006600,v8:9380
Change-Id: I1887cb867627d69ade20654e5bc372b1ba1ac4e3
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1815132
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63939}
2019-09-24 09:39:27 +00:00
Ulan Degenbaev
55c48820f8 [api] Add [Shared]ArrayBuffer::GetBackingStore()
This adds an additional V8 API to get the backing store of an array
buffer. Unlike the existing API, the backing store comes wrapped
in a std::shared_ptr, making lifetime management with the embedder
explicit. This obviates the need for the old GetContents() and
Externalize() APIs, which will be deprecated in a future CL.

Contributed by titzer@chromium.org


Bug: v8:9380
Change-Id: I8a87f5dc141dab684693fe536b636e33f6e45173
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807354
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63883}
2019-09-19 10:03:02 +00:00
Ulan Degenbaev
b6b7de0d60 Reland x6 [arraybuffer] Rearchitect backing store ownership
This reverts commit 9da3483136

Original change's description:
> "Reland x4 [arraybuffer] Rearchitect backing store ownership"
>
> This is a reland of bc33f5aeba
>
> Contributed by titzer@chromium.org
>
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> >
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> >
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> >
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
>
> TBR=yangguo@chromium.org
>
> BUG=v8:9380,v8:9221,chromium:986318
>
> Change-Id: If671a4a9ca0476e8f084efae46e0d2bf99ed99ef
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1731005
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>

> Cr-Commit-Position: refs/heads/master@{#63041}

TBR=yangguo@chromium.org

Change-Id: I3cc4bb80081c662b1751234bc16a821c20e744be
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1792166
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63617}
2019-09-09 13:07:42 +00:00
Ulan Degenbaev
9da3483136 Revert "Reland x5 [arraybuffer] Rearchitect backing store ownership"
This reverts commit 62e168308c.

Reason for revert: it will be relanded after branch

Original change's description:
> Reland x5 [arraybuffer] Rearchitect backing store ownership
> 
> This reverts commit 8fdb23873b.
> 
> Original change's description:
> > "Reland x4 [arraybuffer] Rearchitect backing store ownership"
> >
> > This is a reland of bc33f5aeba
> >
> > Contributed by titzer@chromium.org
> >
> > Original change's description:
> > > [arraybuffer] Rearchitect backing store ownership
> > >
> > > This CL completely rearchitects the ownership of array buffer backing stores,
> > > consolidating ownership into a {BackingStore} C++ object that is tracked
> > > throughout V8 using unique_ptr and shared_ptr where appropriate.
> > >
> > > Overall, lifetime management is simpler and more explicit. The numerous
> > > ways that array buffers were initialized have been streamlined to one
> > > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > > and the construction and destruction of the BackingStore object itself
> > > handles the underlying page or embedder-allocated memory.
> > >
> > > The embedder API remains unchanged for now. We use the
> > > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > > keep the backing store alive properly, even in the case of aliases
> > > from live heap objects. Thus the embedder has a lower chance of making
> > > a mistake. Long-term, we should move the embedder to a model where they
> > > manage backing stores using shared_ptr to an opaque backing store object.
> >
> > TBR=yangguo@chromium.org
> >
> > BUG=v8:9380,v8:9221,chromium:986318
> >
> > Change-Id: If671a4a9ca0476e8f084efae46e0d2bf99ed99ef
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1731005
> > Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> > Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
> > Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#63041}
> 
> TBR=yangguo@chromium.org,clemensh@chromium.org,mstarzinger@chromium.org
> 
> Change-Id: Iba55c7ab71e5642b5cb6aeb699d6fc9cf9061486
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771795
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63461}

TBR=ulan@chromium.org,mlippautz@chromium.org

Change-Id: Id8f67a68ab398032eb2975b1b24ee125394d9c4b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1776095
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63471}
2019-08-30 08:23:40 +00:00
Ulan Degenbaev
62e168308c Reland x5 [arraybuffer] Rearchitect backing store ownership
This reverts commit 8fdb23873b.

Original change's description:
> "Reland x4 [arraybuffer] Rearchitect backing store ownership"
>
> This is a reland of bc33f5aeba
>
> Contributed by titzer@chromium.org
>
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> >
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> >
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> >
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
>
> TBR=yangguo@chromium.org
>
> BUG=v8:9380,v8:9221,chromium:986318
>
> Change-Id: If671a4a9ca0476e8f084efae46e0d2bf99ed99ef
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1731005
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63041}

TBR=yangguo@chromium.org,clemensh@chromium.org,mstarzinger@chromium.org

Change-Id: Iba55c7ab71e5642b5cb6aeb699d6fc9cf9061486
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771795
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63461}
2019-08-29 16:40:18 +00:00
Ulan Degenbaev
8fdb23873b Revert ""Reland x4 [arraybuffer] Rearchitect backing store ownership""
This reverts commit 5611f70b3d.

Reason for revert: flaky tests: v8:9588, v8:9587

Original change's description:
> "Reland x4 [arraybuffer] Rearchitect backing store ownership"
> 
> This is a reland of bc33f5aeba
> 
> Contributed by titzer@chromium.org
> 
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> >
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> >
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> >
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
> 
> TBR=yangguo@chromium.org
> 
> BUG=v8:9380,v8:9221,chromium:986318
> 
> Change-Id: If671a4a9ca0476e8f084efae46e0d2bf99ed99ef
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1731005
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63041}

TBR=ulan@chromium.org,yangguo@chromium.org,mstarzinger@chromium.org,clemensh@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug: v8:9380, v8:9221, chromium:986318
Change-Id: Ic7381239f4e90d0c437b7e47a5ac6e8bce60f882
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1736747
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63081}
2019-08-05 15:33:47 +00:00
Ulan Degenbaev
5611f70b3d "Reland x4 [arraybuffer] Rearchitect backing store ownership"
This is a reland of bc33f5aeba

Contributed by titzer@chromium.org

Original change's description:
> [arraybuffer] Rearchitect backing store ownership
>
> This CL completely rearchitects the ownership of array buffer backing stores,
> consolidating ownership into a {BackingStore} C++ object that is tracked
> throughout V8 using unique_ptr and shared_ptr where appropriate.
>
> Overall, lifetime management is simpler and more explicit. The numerous
> ways that array buffers were initialized have been streamlined to one
> Attach() method on JSArrayBuffer. The array buffer tracker in the
> GC implementation now manages std::shared_ptr<BackingStore> pointers,
> and the construction and destruction of the BackingStore object itself
> handles the underlying page or embedder-allocated memory.
>
> The embedder API remains unchanged for now. We use the
> v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> keep the backing store alive properly, even in the case of aliases
> from live heap objects. Thus the embedder has a lower chance of making
> a mistake. Long-term, we should move the embedder to a model where they
> manage backing stores using shared_ptr to an opaque backing store object.

TBR=yangguo@chromium.org

BUG=v8:9380,v8:9221,chromium:986318

Change-Id: If671a4a9ca0476e8f084efae46e0d2bf99ed99ef
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1731005
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63041}
2019-08-02 10:40:43 +00:00
Francis McCabe
195679de3f Revert ""Reland x3 [arraybuffer] Rearchitect backing store ownership""
This reverts commit df8e617772.

Reason for revert: Multiple flakes in apparently related areas:

https://logs.chromium.org/logs/v8/buildbucket/cr-buildbucket.appspot.com/8906409837768155568/+/steps/Check__flakes_/0/logs/BackingStoreTest.RacyGrowWasmMem.../0

Original change's description:
> "Reland x3 [arraybuffer] Rearchitect backing store ownership"
> 
> This is a reland of bc33f5aeba
> 
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> >
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> >
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> >
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
> 
> R=​mlippautz@chromium.org
> BUG=v8:9380,v8:9221,chromium:986318
> TBR=ulan@chromium.org
> 
> Change-Id: I6c49e2425029b5664ef1c68dab8b5146f4ed0ff2
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1719191
> Reviewed-by: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Commit-Queue: Ben Titzer <titzer@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63007}

TBR=mstarzinger@chromium.org,titzer@chromium.org,mlippautz@chromium.org

Change-Id: If0266e5893b1325a332d5986337fa7ece2cb6943
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9380, v8:9221, chromium:986318
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1729549
Reviewed-by: Francis McCabe <fgm@chromium.org>
Commit-Queue: Francis McCabe <fgm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63011}
2019-07-31 15:50:56 +00:00
Ben L. Titzer
df8e617772 "Reland x3 [arraybuffer] Rearchitect backing store ownership"
This is a reland of bc33f5aeba

Original change's description:
> [arraybuffer] Rearchitect backing store ownership
>
> This CL completely rearchitects the ownership of array buffer backing stores,
> consolidating ownership into a {BackingStore} C++ object that is tracked
> throughout V8 using unique_ptr and shared_ptr where appropriate.
>
> Overall, lifetime management is simpler and more explicit. The numerous
> ways that array buffers were initialized have been streamlined to one
> Attach() method on JSArrayBuffer. The array buffer tracker in the
> GC implementation now manages std::shared_ptr<BackingStore> pointers,
> and the construction and destruction of the BackingStore object itself
> handles the underlying page or embedder-allocated memory.
>
> The embedder API remains unchanged for now. We use the
> v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> keep the backing store alive properly, even in the case of aliases
> from live heap objects. Thus the embedder has a lower chance of making
> a mistake. Long-term, we should move the embedder to a model where they
> manage backing stores using shared_ptr to an opaque backing store object.

R=mlippautz@chromium.org
BUG=v8:9380,v8:9221,chromium:986318
TBR=ulan@chromium.org

Change-Id: I6c49e2425029b5664ef1c68dab8b5146f4ed0ff2
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1719191
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63007}
2019-07-31 14:13:59 +00:00
Ben L. Titzer
845b47a1d2 Revert "Reland "Reland "[arraybuffer] Rearchitect backing store ownership"""
This reverts commit 306cf40344.

Reason for revert: performance regressions / too near branch point

TBR=mslekova@chromium.org
BUG=v8:9380

Change-Id: If77630b73eafbf1190c823199fe2a34361da303f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1714867
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62867}
2019-07-23 12:25:16 +00:00
Ben L. Titzer
306cf40344 Reland "Reland "[arraybuffer] Rearchitect backing store ownership""
This is a reland of bc33f5aeba

Original change's description:
> Reland "[arraybuffer] Rearchitect backing store ownership"
> 
> This is a reland of 31cd5d83d3
> 
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> > 
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> > 
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> > 
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
> > 
> > R=mlippautz@chromium.org
> > BUG=v8:9380,v8:9221
> > 
> > Change-Id: I48fae5ac85dcf6172a83f252439e77e7c1a16ccd
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1584323
> > Commit-Queue: Ben Titzer <titzer@chromium.org>
> > Reviewed-by: Ben Titzer <titzer@chromium.org>
> > Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> > Reviewed-by: Yang Guo <yangguo@chromium.org>
> > Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#62572}
> 
> Bug: v8:9380, v8:9221
> Change-Id: If3f72967a8ebeb067c0edcfc16ed631e36829dbc
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1691906
> Commit-Queue: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#62809}

Bug: v8:9380, v8:9221
Change-Id: I9a2525753ae2424108d074fa81df5f25d945c824
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1709409
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62847}
2019-07-22 10:33:26 +00:00
Clemens Hammacher
6e0473f343 Revert "Reland "[arraybuffer] Rearchitect backing store ownership""
This reverts commit bc33f5aeba.

Reason for revert: Still failing (OOM on win32): https://ci.chromium.org/p/v8/builders/ci/V8%20Win32/22210

Original change's description:
> Reland "[arraybuffer] Rearchitect backing store ownership"
> 
> This is a reland of 31cd5d83d3
> 
> Original change's description:
> > [arraybuffer] Rearchitect backing store ownership
> > 
> > This CL completely rearchitects the ownership of array buffer backing stores,
> > consolidating ownership into a {BackingStore} C++ object that is tracked
> > throughout V8 using unique_ptr and shared_ptr where appropriate.
> > 
> > Overall, lifetime management is simpler and more explicit. The numerous
> > ways that array buffers were initialized have been streamlined to one
> > Attach() method on JSArrayBuffer. The array buffer tracker in the
> > GC implementation now manages std::shared_ptr<BackingStore> pointers,
> > and the construction and destruction of the BackingStore object itself
> > handles the underlying page or embedder-allocated memory.
> > 
> > The embedder API remains unchanged for now. We use the
> > v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> > keep the backing store alive properly, even in the case of aliases
> > from live heap objects. Thus the embedder has a lower chance of making
> > a mistake. Long-term, we should move the embedder to a model where they
> > manage backing stores using shared_ptr to an opaque backing store object.
> > 
> > R=mlippautz@chromium.org
> > BUG=v8:9380,v8:9221
> > 
> > Change-Id: I48fae5ac85dcf6172a83f252439e77e7c1a16ccd
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1584323
> > Commit-Queue: Ben Titzer <titzer@chromium.org>
> > Reviewed-by: Ben Titzer <titzer@chromium.org>
> > Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> > Reviewed-by: Yang Guo <yangguo@chromium.org>
> > Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#62572}
> 
> Bug: v8:9380, v8:9221
> Change-Id: If3f72967a8ebeb067c0edcfc16ed631e36829dbc
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1691906
> Commit-Queue: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#62809}

TBR=ulan@chromium.org,yangguo@chromium.org,mstarzinger@chromium.org,titzer@chromium.org,gdeepti@chromium.org,mlippautz@chromium.org

Change-Id: Iea755df9aaa1e95d284135bd0a6681b1340b6832
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9380, v8:9221
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1708487
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62811}
2019-07-18 15:56:45 +00:00
Ben L. Titzer
bc33f5aeba Reland "[arraybuffer] Rearchitect backing store ownership"
This is a reland of 31cd5d83d3

Original change's description:
> [arraybuffer] Rearchitect backing store ownership
> 
> This CL completely rearchitects the ownership of array buffer backing stores,
> consolidating ownership into a {BackingStore} C++ object that is tracked
> throughout V8 using unique_ptr and shared_ptr where appropriate.
> 
> Overall, lifetime management is simpler and more explicit. The numerous
> ways that array buffers were initialized have been streamlined to one
> Attach() method on JSArrayBuffer. The array buffer tracker in the
> GC implementation now manages std::shared_ptr<BackingStore> pointers,
> and the construction and destruction of the BackingStore object itself
> handles the underlying page or embedder-allocated memory.
> 
> The embedder API remains unchanged for now. We use the
> v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> keep the backing store alive properly, even in the case of aliases
> from live heap objects. Thus the embedder has a lower chance of making
> a mistake. Long-term, we should move the embedder to a model where they
> manage backing stores using shared_ptr to an opaque backing store object.
> 
> R=mlippautz@chromium.org
> BUG=v8:9380,v8:9221
> 
> Change-Id: I48fae5ac85dcf6172a83f252439e77e7c1a16ccd
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1584323
> Commit-Queue: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#62572}

Bug: v8:9380, v8:9221
Change-Id: If3f72967a8ebeb067c0edcfc16ed631e36829dbc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1691906
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62809}
2019-07-18 15:30:52 +00:00
Clemens Hammacher
bf92fbf4c8 Revert "[arraybuffer] Rearchitect backing store ownership"
This reverts commit 31cd5d83d3.

Reason for revert: It breaks my heart to revert this, but it fails differently on several bots, e.g. https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20debug/26671.

Original change's description:
> [arraybuffer] Rearchitect backing store ownership
> 
> This CL completely rearchitects the ownership of array buffer backing stores,
> consolidating ownership into a {BackingStore} C++ object that is tracked
> throughout V8 using unique_ptr and shared_ptr where appropriate.
> 
> Overall, lifetime management is simpler and more explicit. The numerous
> ways that array buffers were initialized have been streamlined to one
> Attach() method on JSArrayBuffer. The array buffer tracker in the
> GC implementation now manages std::shared_ptr<BackingStore> pointers,
> and the construction and destruction of the BackingStore object itself
> handles the underlying page or embedder-allocated memory.
> 
> The embedder API remains unchanged for now. We use the
> v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
> keep the backing store alive properly, even in the case of aliases
> from live heap objects. Thus the embedder has a lower chance of making
> a mistake. Long-term, we should move the embedder to a model where they
> manage backing stores using shared_ptr to an opaque backing store object.
> 
> R=​mlippautz@chromium.org
> BUG=v8:9380,v8:9221
> 
> Change-Id: I48fae5ac85dcf6172a83f252439e77e7c1a16ccd
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1584323
> Commit-Queue: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Ben Titzer <titzer@chromium.org>
> Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#62572}

TBR=ulan@chromium.org,yangguo@chromium.org,mstarzinger@chromium.org,titzer@chromium.org,gdeepti@chromium.org,mlippautz@chromium.org

Change-Id: Ib35788ba8c31192d90cbc72df3dbc41030f109de
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9380, v8:9221
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1691034
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62578}
2019-07-08 17:20:51 +00:00
Ben L. Titzer
31cd5d83d3 [arraybuffer] Rearchitect backing store ownership
This CL completely rearchitects the ownership of array buffer backing stores,
consolidating ownership into a {BackingStore} C++ object that is tracked
throughout V8 using unique_ptr and shared_ptr where appropriate.

Overall, lifetime management is simpler and more explicit. The numerous
ways that array buffers were initialized have been streamlined to one
Attach() method on JSArrayBuffer. The array buffer tracker in the
GC implementation now manages std::shared_ptr<BackingStore> pointers,
and the construction and destruction of the BackingStore object itself
handles the underlying page or embedder-allocated memory.

The embedder API remains unchanged for now. We use the
v8::ArrayBuffer::Contents struct to hide an additional shared_ptr to
keep the backing store alive properly, even in the case of aliases
from live heap objects. Thus the embedder has a lower chance of making
a mistake. Long-term, we should move the embedder to a model where they
manage backing stores using shared_ptr to an opaque backing store object.

R=mlippautz@chromium.org
BUG=v8:9380,v8:9221

Change-Id: I48fae5ac85dcf6172a83f252439e77e7c1a16ccd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1584323
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62572}
2019-07-08 16:06:48 +00:00
Ben L. Titzer
d05871bf08 [test-api] Extract arraybufs and typed arrays
Extract tests related to array buffers and typed arrays to their own
.cc files.

R=mstarzinger@chromium.org

Change-Id: Ic80205d02b62db1565670ecf2bb4c0dbe52fab49
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1662301
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62219}
2019-06-17 15:27:01 +00:00