We can use tag dispatching to distinguish between the synchronized and
non-synchronized accessors. Also eliminated the need of adding explicit
"synchronized" in the name when using the macros.
As a note, we currently have one case of using both relaxed and
synchronized accessors (Map::instance_descriptors).
Cleaned up:
* BytecodeArray::source_position_table
* Code::code_data_container
* Code::source_position_table
* FunctionTemplateInfo::call_code
* Map::instance_descriptors
* Map::layout_descriptor
* SharedFunctionInfo::function_data
Bug: v8:7790
Change-Id: I5a502f4b2df6addb6c45056e77061271012c7d90
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2424130
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70306}
This is a reland of c4a062a958
which was a reland of 28a30c578c
which was a reland of 5d7a29c90e
Fixes TSAN errors from non-atomic writes in the deserializer. Now all
writes are (relaxed) atomic.
Original change's description:
> Reland^2 "[serializer] Allocate during deserialization"
>
> This is a reland of 28a30c578c
> which was a reland of 5d7a29c90e
>
> The crashes were from calling RegisterDeserializerFinished on a null
> Isolate pointer, for a deserializer that was never initialised
> (specifically, ReadOnlyDeserializer when ROHeap is shared).
>
> Original change's description:
> > Reland "[serializer] Allocate during deserialization"
> >
> > This is a reland of 5d7a29c90e
> >
> > This reland shuffles around the order of checks in Heap::AllocateRawWith
> > to not check the new space addresses until it's known that this is a new
> > space allocation. This fixes an UBSan failure during read-only space
> > deserialization, which happens before the new space is initialized.
> >
> > It also fixes some issues discovered by --stress-snapshot, around
> > serializing ThinStrings (which are now elided as part of serialization),
> > handle counts (I bumped the maximum handle count in that check), and
> > clearing map transitions (the map backpointer field needed a Smi
> > uninitialized value check).
> >
> > Original change's description:
> > > [serializer] Allocate during deserialization
> > >
> > > This patch removes the concept of reservations and a specialized
> > > deserializer allocator, and instead makes the deserializer allocate
> > > directly with the Heap's Allocate method.
> > >
> > > The major consequence of this is that the GC can now run during
> > > deserialization, which means that:
> > >
> > > a) Deserialized objects are visible to the GC, and
> > > b) Objects that the deserializer/deserialized objects point to can
> > > move.
> > >
> > > Point a) is mostly not a problem due to previous work in making
> > > deserialized objects "GC valid", i.e. making sure that they have a valid
> > > size before any subsequent allocation/safepoint. We now additionally
> > > have to initialize the allocated space with a valid tagged value -- this
> > > is a magic Smi value to keep "uninitialized" checks simple.
> > >
> > > Point b) is solved by Handlifying the deserializer. This involves
> > > changing any vectors of objects into vectors of Handles, and any object
> > > keyed map into an IdentityMap (we can't use Handles as keys because
> > > the object's address is no longer a stable hash).
> > >
> > > Back-references can no longer be direct chunk offsets, so instead the
> > > deserializer stores a Handle to each deserialized object, and the
> > > backreference is an index into this handle array. This encoding could
> > > be optimized in the future with e.g. a second pass over the serialized
> > > array which emits a different bytecode for objects that are and aren't
> > > back-referenced.
> > >
> > > Additionally, the slot-walk over objects to initialize them can no
> > > longer use absolute slot offsets, as again an object may move and its
> > > slot address would become invalid. Now, slots are walked as relative
> > > offsets to a Handle to the object, or as absolute slots for the case of
> > > root pointers. A concept of "slot accessor" is introduced to share the
> > > code between these two modes, and writing the slot (including write
> > > barriers) is abstracted into this accessor.
> > >
> > > Finally, the Code body walk is modified to deserialize all objects
> > > referred to by RelocInfos before doing the RelocInfo walk itself. This
> > > is because RelocInfoIterator uses raw pointers, so we cannot allocate
> > > during a RelocInfo walk.
> > >
> > > As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> > > size rather than byte size -- the size is expected to be tagged-aligned
> > > anyway, so now we get an extra few bits in the size encoding.
> > >
> > > Bug: chromium:1075999
> > > Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> > > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > > Cr-Commit-Position: refs/heads/master@{#70229}
> >
> > Bug: chromium:1075999
> > Change-Id: Ibc77cc48b3440b4a28b09746cfc47e50c340ce54
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440828
> > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#70267}
>
> Tbr: jgruber@chromium.org,ulan@chromium.org
> Bug: chromium:1075999
> Change-Id: Iaa8dc54895866ada0e34a7c9e8fff9ae1cb13f2d
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2444991
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70279}
Tbr: jgruber@chromium.org,ulan@chromium.org
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng,v8_linux64_tsan_no_cm_rel_ng,v8_linux64_tsan_isolates_rel_ng
Bug: chromium:1075999
Change-Id: I0b9b11644aebc4cc8b07c62a0f765b24e4d73d89
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2445872
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70288}
This reverts commit c4a062a958.
Reason for revert: TSan issues: https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/33504
Original change's description:
> Reland^2 "[serializer] Allocate during deserialization"
>
> This is a reland of 28a30c578c
> which was a reland of 5d7a29c90e
>
> The crashes were from calling RegisterDeserializerFinished on a null
> Isolate pointer, for a deserializer that was never initialised
> (specifically, ReadOnlyDeserializer when ROHeap is shared).
>
> Original change's description:
> > Reland "[serializer] Allocate during deserialization"
> >
> > This is a reland of 5d7a29c90e
> >
> > This reland shuffles around the order of checks in Heap::AllocateRawWith
> > to not check the new space addresses until it's known that this is a new
> > space allocation. This fixes an UBSan failure during read-only space
> > deserialization, which happens before the new space is initialized.
> >
> > It also fixes some issues discovered by --stress-snapshot, around
> > serializing ThinStrings (which are now elided as part of serialization),
> > handle counts (I bumped the maximum handle count in that check), and
> > clearing map transitions (the map backpointer field needed a Smi
> > uninitialized value check).
> >
> > Original change's description:
> > > [serializer] Allocate during deserialization
> > >
> > > This patch removes the concept of reservations and a specialized
> > > deserializer allocator, and instead makes the deserializer allocate
> > > directly with the Heap's Allocate method.
> > >
> > > The major consequence of this is that the GC can now run during
> > > deserialization, which means that:
> > >
> > > a) Deserialized objects are visible to the GC, and
> > > b) Objects that the deserializer/deserialized objects point to can
> > > move.
> > >
> > > Point a) is mostly not a problem due to previous work in making
> > > deserialized objects "GC valid", i.e. making sure that they have a valid
> > > size before any subsequent allocation/safepoint. We now additionally
> > > have to initialize the allocated space with a valid tagged value -- this
> > > is a magic Smi value to keep "uninitialized" checks simple.
> > >
> > > Point b) is solved by Handlifying the deserializer. This involves
> > > changing any vectors of objects into vectors of Handles, and any object
> > > keyed map into an IdentityMap (we can't use Handles as keys because
> > > the object's address is no longer a stable hash).
> > >
> > > Back-references can no longer be direct chunk offsets, so instead the
> > > deserializer stores a Handle to each deserialized object, and the
> > > backreference is an index into this handle array. This encoding could
> > > be optimized in the future with e.g. a second pass over the serialized
> > > array which emits a different bytecode for objects that are and aren't
> > > back-referenced.
> > >
> > > Additionally, the slot-walk over objects to initialize them can no
> > > longer use absolute slot offsets, as again an object may move and its
> > > slot address would become invalid. Now, slots are walked as relative
> > > offsets to a Handle to the object, or as absolute slots for the case of
> > > root pointers. A concept of "slot accessor" is introduced to share the
> > > code between these two modes, and writing the slot (including write
> > > barriers) is abstracted into this accessor.
> > >
> > > Finally, the Code body walk is modified to deserialize all objects
> > > referred to by RelocInfos before doing the RelocInfo walk itself. This
> > > is because RelocInfoIterator uses raw pointers, so we cannot allocate
> > > during a RelocInfo walk.
> > >
> > > As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> > > size rather than byte size -- the size is expected to be tagged-aligned
> > > anyway, so now we get an extra few bits in the size encoding.
> > >
> > > Bug: chromium:1075999
> > > Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> > > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > > Cr-Commit-Position: refs/heads/master@{#70229}
> >
> > Bug: chromium:1075999
> > Change-Id: Ibc77cc48b3440b4a28b09746cfc47e50c340ce54
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440828
> > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#70267}
>
> Tbr: jgruber@chromium.org,ulan@chromium.org
> Bug: chromium:1075999
> Change-Id: Iaa8dc54895866ada0e34a7c9e8fff9ae1cb13f2d
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2444991
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70279}
TBR=ulan@chromium.org,jgruber@chromium.org,leszeks@chromium.org
Change-Id: Ib2f01db4cd9b55639d6a4af971bda865edb45e84
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:1075999
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2445250
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70280}
This is a reland of 28a30c578c
which was a reland of 5d7a29c90e
The crashes were from calling RegisterDeserializerFinished on a null
Isolate pointer, for a deserializer that was never initialised
(specifically, ReadOnlyDeserializer when ROHeap is shared).
Original change's description:
> Reland "[serializer] Allocate during deserialization"
>
> This is a reland of 5d7a29c90e
>
> This reland shuffles around the order of checks in Heap::AllocateRawWith
> to not check the new space addresses until it's known that this is a new
> space allocation. This fixes an UBSan failure during read-only space
> deserialization, which happens before the new space is initialized.
>
> It also fixes some issues discovered by --stress-snapshot, around
> serializing ThinStrings (which are now elided as part of serialization),
> handle counts (I bumped the maximum handle count in that check), and
> clearing map transitions (the map backpointer field needed a Smi
> uninitialized value check).
>
> Original change's description:
> > [serializer] Allocate during deserialization
> >
> > This patch removes the concept of reservations and a specialized
> > deserializer allocator, and instead makes the deserializer allocate
> > directly with the Heap's Allocate method.
> >
> > The major consequence of this is that the GC can now run during
> > deserialization, which means that:
> >
> > a) Deserialized objects are visible to the GC, and
> > b) Objects that the deserializer/deserialized objects point to can
> > move.
> >
> > Point a) is mostly not a problem due to previous work in making
> > deserialized objects "GC valid", i.e. making sure that they have a valid
> > size before any subsequent allocation/safepoint. We now additionally
> > have to initialize the allocated space with a valid tagged value -- this
> > is a magic Smi value to keep "uninitialized" checks simple.
> >
> > Point b) is solved by Handlifying the deserializer. This involves
> > changing any vectors of objects into vectors of Handles, and any object
> > keyed map into an IdentityMap (we can't use Handles as keys because
> > the object's address is no longer a stable hash).
> >
> > Back-references can no longer be direct chunk offsets, so instead the
> > deserializer stores a Handle to each deserialized object, and the
> > backreference is an index into this handle array. This encoding could
> > be optimized in the future with e.g. a second pass over the serialized
> > array which emits a different bytecode for objects that are and aren't
> > back-referenced.
> >
> > Additionally, the slot-walk over objects to initialize them can no
> > longer use absolute slot offsets, as again an object may move and its
> > slot address would become invalid. Now, slots are walked as relative
> > offsets to a Handle to the object, or as absolute slots for the case of
> > root pointers. A concept of "slot accessor" is introduced to share the
> > code between these two modes, and writing the slot (including write
> > barriers) is abstracted into this accessor.
> >
> > Finally, the Code body walk is modified to deserialize all objects
> > referred to by RelocInfos before doing the RelocInfo walk itself. This
> > is because RelocInfoIterator uses raw pointers, so we cannot allocate
> > during a RelocInfo walk.
> >
> > As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> > size rather than byte size -- the size is expected to be tagged-aligned
> > anyway, so now we get an extra few bits in the size encoding.
> >
> > Bug: chromium:1075999
> > Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#70229}
>
> Bug: chromium:1075999
> Change-Id: Ibc77cc48b3440b4a28b09746cfc47e50c340ce54
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440828
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70267}
Tbr: jgruber@chromium.org,ulan@chromium.org
Bug: chromium:1075999
Change-Id: Iaa8dc54895866ada0e34a7c9e8fff9ae1cb13f2d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2444991
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70279}
This reverts commit 28a30c578c.
Reason for revert: Broke Test262 https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20shared/38638?
Original change's description:
> Reland "[serializer] Allocate during deserialization"
>
> This is a reland of 5d7a29c90e
>
> This reland shuffles around the order of checks in Heap::AllocateRawWith
> to not check the new space addresses until it's known that this is a new
> space allocation. This fixes an UBSan failure during read-only space
> deserialization, which happens before the new space is initialized.
>
> It also fixes some issues discovered by --stress-snapshot, around
> serializing ThinStrings (which are now elided as part of serialization),
> handle counts (I bumped the maximum handle count in that check), and
> clearing map transitions (the map backpointer field needed a Smi
> uninitialized value check).
>
> Original change's description:
> > [serializer] Allocate during deserialization
> >
> > This patch removes the concept of reservations and a specialized
> > deserializer allocator, and instead makes the deserializer allocate
> > directly with the Heap's Allocate method.
> >
> > The major consequence of this is that the GC can now run during
> > deserialization, which means that:
> >
> > a) Deserialized objects are visible to the GC, and
> > b) Objects that the deserializer/deserialized objects point to can
> > move.
> >
> > Point a) is mostly not a problem due to previous work in making
> > deserialized objects "GC valid", i.e. making sure that they have a valid
> > size before any subsequent allocation/safepoint. We now additionally
> > have to initialize the allocated space with a valid tagged value -- this
> > is a magic Smi value to keep "uninitialized" checks simple.
> >
> > Point b) is solved by Handlifying the deserializer. This involves
> > changing any vectors of objects into vectors of Handles, and any object
> > keyed map into an IdentityMap (we can't use Handles as keys because
> > the object's address is no longer a stable hash).
> >
> > Back-references can no longer be direct chunk offsets, so instead the
> > deserializer stores a Handle to each deserialized object, and the
> > backreference is an index into this handle array. This encoding could
> > be optimized in the future with e.g. a second pass over the serialized
> > array which emits a different bytecode for objects that are and aren't
> > back-referenced.
> >
> > Additionally, the slot-walk over objects to initialize them can no
> > longer use absolute slot offsets, as again an object may move and its
> > slot address would become invalid. Now, slots are walked as relative
> > offsets to a Handle to the object, or as absolute slots for the case of
> > root pointers. A concept of "slot accessor" is introduced to share the
> > code between these two modes, and writing the slot (including write
> > barriers) is abstracted into this accessor.
> >
> > Finally, the Code body walk is modified to deserialize all objects
> > referred to by RelocInfos before doing the RelocInfo walk itself. This
> > is because RelocInfoIterator uses raw pointers, so we cannot allocate
> > during a RelocInfo walk.
> >
> > As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> > size rather than byte size -- the size is expected to be tagged-aligned
> > anyway, so now we get an extra few bits in the size encoding.
> >
> > Bug: chromium:1075999
> > Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> > Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> > Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#70229}
>
> Bug: chromium:1075999
> Change-Id: Ibc77cc48b3440b4a28b09746cfc47e50c340ce54
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440828
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70267}
TBR=ulan@chromium.org,jgruber@chromium.org,leszeks@chromium.org
Change-Id: Ieed68332ef6a7ad36db061e3f48be0f28673d7a2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:1075999
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2441608
Reviewed-by: Zhi An Ng <zhin@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70268}
This is a reland of 5d7a29c90e
This reland shuffles around the order of checks in Heap::AllocateRawWith
to not check the new space addresses until it's known that this is a new
space allocation. This fixes an UBSan failure during read-only space
deserialization, which happens before the new space is initialized.
It also fixes some issues discovered by --stress-snapshot, around
serializing ThinStrings (which are now elided as part of serialization),
handle counts (I bumped the maximum handle count in that check), and
clearing map transitions (the map backpointer field needed a Smi
uninitialized value check).
Original change's description:
> [serializer] Allocate during deserialization
>
> This patch removes the concept of reservations and a specialized
> deserializer allocator, and instead makes the deserializer allocate
> directly with the Heap's Allocate method.
>
> The major consequence of this is that the GC can now run during
> deserialization, which means that:
>
> a) Deserialized objects are visible to the GC, and
> b) Objects that the deserializer/deserialized objects point to can
> move.
>
> Point a) is mostly not a problem due to previous work in making
> deserialized objects "GC valid", i.e. making sure that they have a valid
> size before any subsequent allocation/safepoint. We now additionally
> have to initialize the allocated space with a valid tagged value -- this
> is a magic Smi value to keep "uninitialized" checks simple.
>
> Point b) is solved by Handlifying the deserializer. This involves
> changing any vectors of objects into vectors of Handles, and any object
> keyed map into an IdentityMap (we can't use Handles as keys because
> the object's address is no longer a stable hash).
>
> Back-references can no longer be direct chunk offsets, so instead the
> deserializer stores a Handle to each deserialized object, and the
> backreference is an index into this handle array. This encoding could
> be optimized in the future with e.g. a second pass over the serialized
> array which emits a different bytecode for objects that are and aren't
> back-referenced.
>
> Additionally, the slot-walk over objects to initialize them can no
> longer use absolute slot offsets, as again an object may move and its
> slot address would become invalid. Now, slots are walked as relative
> offsets to a Handle to the object, or as absolute slots for the case of
> root pointers. A concept of "slot accessor" is introduced to share the
> code between these two modes, and writing the slot (including write
> barriers) is abstracted into this accessor.
>
> Finally, the Code body walk is modified to deserialize all objects
> referred to by RelocInfos before doing the RelocInfo walk itself. This
> is because RelocInfoIterator uses raw pointers, so we cannot allocate
> during a RelocInfo walk.
>
> As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> size rather than byte size -- the size is expected to be tagged-aligned
> anyway, so now we get an extra few bits in the size encoding.
>
> Bug: chromium:1075999
> Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70229}
Bug: chromium:1075999
Change-Id: Ibc77cc48b3440b4a28b09746cfc47e50c340ce54
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440828
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70267}
CodeKind::OPTIMIZED_CODE -> TURBOFAN
Kinds are now more fine-grained and distinguish between TF, TP, NCI.
CodeKind::STUB -> DEOPT_ENTRIES_OR_FOR_TESTING
Code stubs (like builtins, but generated at runtime) were removed from
the codebase years ago, this is the last remnant. This kind is used
only for deopt entries (which should be converted into builtins) and
for tests.
Change-Id: I67beb15377cb60f395e9b051b25f3e5764982e93
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440335
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70234}
This reverts commit 5d7a29c90e.
Reason for revert: UBSan -- https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20UBSan/13100
Original change's description:
> [serializer] Allocate during deserialization
>
> This patch removes the concept of reservations and a specialized
> deserializer allocator, and instead makes the deserializer allocate
> directly with the Heap's Allocate method.
>
> The major consequence of this is that the GC can now run during
> deserialization, which means that:
>
> a) Deserialized objects are visible to the GC, and
> b) Objects that the deserializer/deserialized objects point to can
> move.
>
> Point a) is mostly not a problem due to previous work in making
> deserialized objects "GC valid", i.e. making sure that they have a valid
> size before any subsequent allocation/safepoint. We now additionally
> have to initialize the allocated space with a valid tagged value -- this
> is a magic Smi value to keep "uninitialized" checks simple.
>
> Point b) is solved by Handlifying the deserializer. This involves
> changing any vectors of objects into vectors of Handles, and any object
> keyed map into an IdentityMap (we can't use Handles as keys because
> the object's address is no longer a stable hash).
>
> Back-references can no longer be direct chunk offsets, so instead the
> deserializer stores a Handle to each deserialized object, and the
> backreference is an index into this handle array. This encoding could
> be optimized in the future with e.g. a second pass over the serialized
> array which emits a different bytecode for objects that are and aren't
> back-referenced.
>
> Additionally, the slot-walk over objects to initialize them can no
> longer use absolute slot offsets, as again an object may move and its
> slot address would become invalid. Now, slots are walked as relative
> offsets to a Handle to the object, or as absolute slots for the case of
> root pointers. A concept of "slot accessor" is introduced to share the
> code between these two modes, and writing the slot (including write
> barriers) is abstracted into this accessor.
>
> Finally, the Code body walk is modified to deserialize all objects
> referred to by RelocInfos before doing the RelocInfo walk itself. This
> is because RelocInfoIterator uses raw pointers, so we cannot allocate
> during a RelocInfo walk.
>
> As a drive-by, the VariableRawData bytecode is tweaked to use tagged
> size rather than byte size -- the size is expected to be tagged-aligned
> anyway, so now we get an extra few bits in the size encoding.
>
> Bug: chromium:1075999
> Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70229}
TBR=ulan@chromium.org,jgruber@chromium.org,leszeks@chromium.org
Change-Id: I2bd792a24861e8f54897e51522769b50f8f814e2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:1075999
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2440827
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70231}
This patch removes the concept of reservations and a specialized
deserializer allocator, and instead makes the deserializer allocate
directly with the Heap's Allocate method.
The major consequence of this is that the GC can now run during
deserialization, which means that:
a) Deserialized objects are visible to the GC, and
b) Objects that the deserializer/deserialized objects point to can
move.
Point a) is mostly not a problem due to previous work in making
deserialized objects "GC valid", i.e. making sure that they have a valid
size before any subsequent allocation/safepoint. We now additionally
have to initialize the allocated space with a valid tagged value -- this
is a magic Smi value to keep "uninitialized" checks simple.
Point b) is solved by Handlifying the deserializer. This involves
changing any vectors of objects into vectors of Handles, and any object
keyed map into an IdentityMap (we can't use Handles as keys because
the object's address is no longer a stable hash).
Back-references can no longer be direct chunk offsets, so instead the
deserializer stores a Handle to each deserialized object, and the
backreference is an index into this handle array. This encoding could
be optimized in the future with e.g. a second pass over the serialized
array which emits a different bytecode for objects that are and aren't
back-referenced.
Additionally, the slot-walk over objects to initialize them can no
longer use absolute slot offsets, as again an object may move and its
slot address would become invalid. Now, slots are walked as relative
offsets to a Handle to the object, or as absolute slots for the case of
root pointers. A concept of "slot accessor" is introduced to share the
code between these two modes, and writing the slot (including write
barriers) is abstracted into this accessor.
Finally, the Code body walk is modified to deserialize all objects
referred to by RelocInfos before doing the RelocInfo walk itself. This
is because RelocInfoIterator uses raw pointers, so we cannot allocate
during a RelocInfo walk.
As a drive-by, the VariableRawData bytecode is tweaked to use tagged
size rather than byte size -- the size is expected to be tagged-aligned
anyway, so now we get an extra few bits in the size encoding.
Bug: chromium:1075999
Change-Id: I672c42f553f2669888cc5e35d692c1b8ece1845e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404451
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70229}
This test checks SizeOfObjects after GC, but there might be concurrent
allocations in-between.
Bug: v8:10315
Change-Id: Id904c8865e44ac5c3b486ff6f1316e536cf20e9f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2428864
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70122}
Make sure that tests grow the new space in a safepoint. This fixes
races with concurrent allocation.
Bug: v8:10315
Change-Id: I6fce6740bc3c9385f18bbbcde4b06ba881a03635
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2428946
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70120}
Avoid data race by only setting FLAG_local_heaps to true if not
already enabled.
Bug: v8:10315
Change-Id: Ib562b6d525448f5c088da39bf60928debd97db43
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2426610
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70115}
Otherwise concurrent allocation might start incremental marking, which
would then mark the global handle.
Bug: v8:10315
Change-Id: Ibc681b001847a7c52e9fd8a0420e42a0d0ecfbda
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2424004
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70054}
This reverts commit 4a2b2b2e56.
Reason for revert: Speculative revert due to https://ci.chromium.org/p/chromium/builders/try/linux-rel/495075?
Original change's description:
> [Heap]: Marking use Jobs.
>
> StopRequest is removed in favor of:
> COMPLETE_TASKS_FOR_TESTING -> JoinForTesting()
> PREEMPT_TASKS -> Pause()
> COMPLETE_ONGOING_TASKS now has the same behavior as PREEMPT_TASKS
> - we should avoid waiting on the main thread as much as possible.
>
> Change-Id: Icceeb4f0c0fda2ed234b2f26fe308b11410fcfb7
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2376166
> Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#70037}
TBR=ulan@chromium.org,etiennep@chromium.org
Change-Id: I63f24bffa0f56c6ffa1d1977fc4fb8a76b6f3ba2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2423722
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Nico Hartmann <nicohartmann@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70049}
Building and running tests with v8_enabled_concurrent_marking=false
currently produces two failures:
1) Segmentation fault on attempt to mark a read-only object.
This is fixed by changing MarkBit::Set to be a no-op if the object
is already marked (which is the case for the readonly space).
2) Missing write-barrier due to bogus condition in the bailout.
The barrier can be skipped only if the host object is not marked yet.
This also disables two concurrent allocation tests that rely on
concurrent marking write-barrier.
Bug: v8:10875
Change-Id: Ib3a238fc34c8f20c697470e0bd4ac427fb4bdc0e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2421816
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70041}
StopRequest is removed in favor of:
COMPLETE_TASKS_FOR_TESTING -> JoinForTesting()
PREEMPT_TASKS -> Pause()
COMPLETE_ONGOING_TASKS now has the same behavior as PREEMPT_TASKS
- we should avoid waiting on the main thread as much as possible.
Change-Id: Icceeb4f0c0fda2ed234b2f26fe308b11410fcfb7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2376166
Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70037}
Test was asserting heap size before and after GC. With background
thread allocation those assertions might not hold.
Bug: v8:10315
Change-Id: I4f8c0f6d0b80040b3c89f85e801416abb29ed30e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2421999
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70034}
When a compaction space allocates a new code page, that pages needs to
be added to the Isolate::code_pages_ array used for stack unwinding.
Since the array is owned by the main thread, compaction thread cannot
directly modify it. Because of that code pages are added upon merging
of the compaction space to the main space in MergeLocalSpace.
The bug was that all code pages coming from the compaction space
were added to the code_pages_ array. However, some of the pages are
not newly allocated but merely borrowed from the main space.
This CL keeps track of all newly allocated paged by a compaction space.
Bug: v8:10900
Change-Id: Iff3ff5d608df60fb752d2e0ffc29e51f2d967936
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2418718
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#70023}
This reverts commit af5f437cd9.
Reason for revert: Seems to break TSAN - https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/33286?
Original change's description:
> [heap] Fix tracking of code pages for V8 stack unwinder
>
> When a compaction space allocates a new code page, that pages needs to
> be added to the Isolate::code_pages_ array used for stack unwinding.
> Since the array is owned by the main thread, compaction thread cannot
> directly modify it. Because of that code pages are added upon merging
> of the compaction space to the main spage in MergeLocalSpace.
>
> The bug was that all code pages coming from the compaction space
> were added to the code_pages_ array. However, some of the pages are
> not newly allocated but merely borrowed from the main space.
>
> This CL introduces a new page flag for marking pages that are borrowed
> during compaction and skips them in MergeLocalSpace.
>
> Bug: v8:10900
> Change-Id: I786dc5747bd7c785ae58dfd8b841c00774efb15e
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2416500
> Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69992}
TBR=ulan@chromium.org,jkummerow@chromium.org,dinfuehr@chromium.org
Change-Id: I13f8b64014750af95423166152dc9bee8cec12d0
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:10900
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2418395
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69993}
When a compaction space allocates a new code page, that pages needs to
be added to the Isolate::code_pages_ array used for stack unwinding.
Since the array is owned by the main thread, compaction thread cannot
directly modify it. Because of that code pages are added upon merging
of the compaction space to the main spage in MergeLocalSpace.
The bug was that all code pages coming from the compaction space
were added to the code_pages_ array. However, some of the pages are
not newly allocated but merely borrowed from the main space.
This CL introduces a new page flag for marking pages that are borrowed
during compaction and skips them in MergeLocalSpace.
Bug: v8:10900
Change-Id: I786dc5747bd7c785ae58dfd8b841c00774efb15e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2416500
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69992}
This reverts commit 1aa9ab7384.
The reverted CL chain had an issue where ThinStrings could accidentally
end up in compilation artifacts, causing issues down the line with ICs
that expected direct internalized strings.
The reason for this bug was that forward references to internalized
strings were resolved before PostProcessNewObject. When this happened,
the internalized string A would be written to the field where it was
previously deferred, then PostProcessNewObject would change string A to
string A', and update string A to a ThinString. This means any _future_
back references to A would see the ThinString and follow it to receive
A', but any _past_ forward references would keep pointing to the
ThinString A.
This reland fixes this by preventing InternalizedString deferral, so
that all references to InternalizedStrings are back references. It also
adds some additional verification to the heap verifier that constant
pools and object boilerplate descriptors aren't allowed to hold thin
strings.
This patch also fixes an additional bug in the original CL, where weak
forward refs weren't being serialized with a weak prefix.
Original change's description:
> Revert recent de/serializer related changes
>
> They are suspected to be causing Canary crashes, confirmed through
> local reverts and repro attempts.
>
> This reverts:
> - "Reland "[serializer] Change deferring to use forward refs""
> commit 76d684cc82.
> - "Reland "[serializer] Remove new space""
> commit 81231c23a9.
> - "[serializer] Clean-up and de-macro ReadDataCase"
> commit c06d24b915.
> - "[serializer] DCHECK deserializer allocations are initialized"
> commit fbc1f32d8e.
>
> Bug: chromium:1128872
> Change-Id: Id2bb3b8fac526fdf9ffb033222ae08cd423f8238
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2414220
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Reviewed-by: Adam Klein <adamk@chromium.org>
> Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69955}
Tbr: jgruber@chromium.org,dinfuehr@chromium.org
Bug: chromium:1075999
Bug: chromium:1127610
Bug: chromium:1128848
Bug: chromium:1128872
Bug: chromium:1128957
Change-Id: I8b7bbabf77eb8cb942a28316afbfaa5f9a0aa4cb
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2418101
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69988}
They are suspected to be causing Canary crashes, confirmed through
local reverts and repro attempts.
This reverts:
- "Reland "[serializer] Change deferring to use forward refs""
commit 76d684cc82.
- "Reland "[serializer] Remove new space""
commit 81231c23a9.
- "[serializer] Clean-up and de-macro ReadDataCase"
commit c06d24b915.
- "[serializer] DCHECK deserializer allocations are initialized"
commit fbc1f32d8e.
Bug: chromium:1128872
Change-Id: Id2bb3b8fac526fdf9ffb033222ae08cd423f8238
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2414220
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69955}
With the --always_promote_young_mc flag the mark-compact collector
cannot gracefully handle allocation failures when evacuating the young
generation. In some scenarios this causes OOM crashes without invoking
NearHeapLimitCallback.
This CL ensures that the young generation is evacuated before the old
generation because old generation evacuation can be aborted if needed.
Additionally, the CL cleans up usages of CanExpandOldGeneration.
Bug: v8:10843
Change-Id: I50d83912137afa3d3dac797dd4c6bddb51612334
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2404829
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69937}
This commit adds a check in Heap::AllocateRaw when setting the
large_object variable, when the AllocationType is of type kCode, to
take into account the size of the CodeSpace's area size.
The motivation for this change is that without this check it is
possible that size_in_bytes is less than 128, and hence not considered
a large object, but it might be larger than the available space
in code_space->AreaSize(), which will cause the object to be created
in the CodeLargeObjectSpace. This will later cause a segmentation fault
when calling the following chain of functions:
if (!large_object) {
MemoryChunk::FromHeapObject(heap_object)
->GetCodeObjectRegistry()
->RegisterNewlyAllocatedCodeObject(heap_object.address());
}
We (Red Hat) ran into this issue when running Node.js v12.16.1 in
combination with yarn on aarch64 (this was the only architecture that
this happed on).
Bug: v8:10808
Change-Id: I0c396b0eb64bc4cc91d9a3be521254f3130eac7b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390665
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69876}
Tests failed from time-to-time with --stress-concurrent-allocation. So
run those tests with that flag disabled.
Bug: v8:10315
Change-Id: I8a2b9f03d7bcd8a797134510f608dffb78dd1cdf
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2403257
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69826}
This reverts https://chromium-review.googlesource.com/c/v8/v8/+/2372545
in favour of different solution. In order to simulate filling up a page,
it's not suitable to look at the limit() since there might be observers
that have lowered it, so the page will not actually be full.
Instead, let's relax the CHECK() in CreatePadding() to not look at the
limit() but all available space.
For instance, the test-heap/Regress978156 cctest uses FillCurrentPage()
to fill the current page. However if there's an observer on the current
page, it will not be filled entirely and the test will fail. This works
because by default, when the new space is empty, the scavenger observer
happens to be on the second page of the space. However if one changes
the V8 page size to 512k, then it fails.
This can be reproduced as such:
# Make sure the scavenge trigger is on the first page.
./cctest test-heap/Regress978156 --scavenge-task-trigger=10
# Stress marking adds random observers to trigger incremental
# marking.
./cctest test-heap/Regress978156 --stress-marking=100
This issue also causes crashes when using the %SimulateNewspaceFull()
runtime test function, as found by fuzzing and you can find more details
in the bug.
Bug: v8:10808, v8:9906, chromium:1122848
Change-Id: Ie043ae0a1d3754d2423cb5d97f2b3e1ee860e5c8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2401427
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Pierre Langlois <pierre.langlois@arm.com>
Cr-Commit-Position: refs/heads/master@{#69805}
This is a reland of 1c7618abad
The revert was due to an missing dependency in the incremental build,
fixed in https://crrev.com/c/2400987.
Original change's description:
> [serializer] Remove new space
>
> The new space is unused in the snapshot, as we convert all new objects
> to old space objects when serializing. This means we can get rid of
> the snapshot new space entirely, and as a result get rid of the write
> barrier checks.
>
> This also rejiggles the order of the general spaces enum so that the new
> spaces are at the end, and can be truncated off for the SnapshotSpace
> enum.
>
> As a drive by, fix a bug in an unrelated test-api test which this patch
> exposed.
>
> Change-Id: If67ff8be5bf03104a3ffae7df707c22460bba3a1
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390762
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69761}
Tbr: jgruber@chromium.org,dinfuehr@chromium.org
Change-Id: I9fbc61a124fae09d12d6281baaca60eb6c39a6e5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2401420
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69785}
This reverts commit 1c7618abad.
Reason for revert: https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20-%20shared/10544
Original change's description:
> [serializer] Remove new space
>
> The new space is unused in the snapshot, as we convert all new objects
> to old space objects when serializing. This means we can get rid of
> the snapshot new space entirely, and as a result get rid of the write
> barrier checks.
>
> This also rejiggles the order of the general spaces enum so that the new
> spaces are at the end, and can be truncated off for the SnapshotSpace
> enum.
>
> As a drive by, fix a bug in an unrelated test-api test which this patch
> exposed.
>
> Change-Id: If67ff8be5bf03104a3ffae7df707c22460bba3a1
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390762
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69761}
TBR=jgruber@chromium.org,leszeks@chromium.org,dinfuehr@chromium.org
Change-Id: Iaf2362d8cd3a17d8410030aca0dd2250c5a0a7af
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2398533
Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org>
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69762}
The new space is unused in the snapshot, as we convert all new objects
to old space objects when serializing. This means we can get rid of
the snapshot new space entirely, and as a result get rid of the write
barrier checks.
This also rejiggles the order of the general spaces enum so that the new
spaces are at the end, and can be truncated off for the SnapshotSpace
enum.
As a drive by, fix a bug in an unrelated test-api test which this patch
exposed.
Change-Id: If67ff8be5bf03104a3ffae7df707c22460bba3a1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390762
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69761}
Tests do not expect concurrent allocation and are simply skipped in that
configuration.
Bug: v8:10315
Change-Id: Ia371efa3c27e1f8b76fab47abcce2d7c218224bc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390774
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69704}
This is a reland of 9eb090d261
The android-pie-arm64-dbg compiler error was fixed in:
https://chromium-review.googlesource.com/c/v8/v8/+/2381450
Original change's description:
> [heap] Add concurrent typed slot recording
>
> Since the typed slot set is not thread-safe, each concurrent marking
> barrier collects typed slots locally and publishes them to the main
> typed slot set in safepoints.
> Bug: v8:10315
>
> Change-Id: If1f5c5df786df88aac7bc27088afe91a4173c826
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2370302
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69576}
Bug: v8:10315
Change-Id: Iae2882bad1cd0ffcae28c96318ba5fd7937f2215
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390763
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69694}
Tests that use SimulateFullSpace and SealCurrentObjects do not work
if there is a background thread allocating concurrently.
Bug: v8:10315
Change-Id: I73a4c9db8eb32fdf3e07fcb8f5dda309de797709
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2390765
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69693}
With conservative stack scanning enabled, a snapshot of the call stack
upon entry to GC will be used to determine part of the root-set. When
the collector walks the stack, it looks at each value and determines
whether it could be a potential on-heap object pointer. However, unlike
with Handles, these on-stack pointers aren't guaranteed to point to the
start of the object: the compiler may decide hide these pointers, and
create interior pointers in C++ frames which the GC doesn't know about.
The solution to this is to include an object start bitmap in the header
of each page. Each bit in the bitmap represents a word in the page
payload which is set when an object is allocated. This means that when
the collector finds an arbitrary potential pointer into the page, it can
walk backwards through the bitmap until it finds the relevant object's
base pointer. To prevent the bitmap becoming stale after compaction, it
is rebuilt during object sweeping.
This is experimental, and currently only works with inline allocation
disabled, and single generational collection.
Bug: v8:10614
Change-Id: I28ebd9562f58f335f8b3c2d1189cdf39feaa1f52
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2375195
Commit-Queue: Anton Bikineev <bikineev@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Anton Bikineev <bikineev@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69615}
This reverts commit 9eb090d261.
Reason for revert: breaks trybot android-pie-arm64-dbg, repro steps: build cctest with target_cpu="arm64" in the args.
See thread:
https://chromium.slack.com/archives/CGJ5WKRUH/p1598563610118900
Original change's description:
> [heap] Add concurrent typed slot recording
>
> Since the typed slot set is not thread-safe, each concurrent marking
> barrier collects typed slots locally and publishes them to the main
> typed slot set in safepoints.
> Bug: v8:10315
>
> Change-Id: If1f5c5df786df88aac7bc27088afe91a4173c826
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2370302
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69576}
TBR=ulan@chromium.org,dinfuehr@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: v8:10315
Change-Id: Iade0443e5eccef06e3ea77913e18fd1f563995f5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2380613
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69597}
Since the typed slot set is not thread-safe, each concurrent marking
barrier collects typed slots locally and publishes them to the main
typed slot set in safepoints.
Bug: v8:10315
Change-Id: If1f5c5df786df88aac7bc27088afe91a4173c826
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2370302
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69576}
The cctests need to simulate full pages, however the calculation of the
remainging available space is wrong causing an assertion when we change
the V8 page size to 512K:
$ cctest test-array-buffer-tracker/ArrayBuffer_PagePromotion
#
# Fatal error in ../../test/cctest/heap/heap-utils.cc, line 94
# Check failed: padding_size <= overall_free_memory || overall_free_memory == 0.
The reason is:
- On startup, we register a scavenger observer on the new space. The
observer is set to trigger when the new space capacity is at 80% by
default.
- On linux, the initial capacity of the new space is 512K, so the
scavenger observer will be placed at 80% of 512K, which will either
be in the second page of the space if the page size is 256K, or in
the first page if the page size is 512K.
- When placing the observer, if the observer hits the first page, we
lower the allocation limit (see `ComputeLimit()`). This makes sure
the observer isn't skipped by allocations inlined in generated code.
However, when we simulate filling the current page, we compute the space
left in the current page by comparing the top with the `page_high()`
rather than `limit()`. This was done so the tests would also work when
inlined allocations are disabled. If we don't look at the `limit()`, we
don't take the observer into account and fill more space than is
available, triggering the assertion.
This can also be reproduced by reducing the % at which the scavenger is
triggered so that it hits the first page instead of the second when the
page size is 256K, for example passing --scavenge-task-trigger=10.
Bug v8:10808, v8:9906
Change-Id: Iad50bb68995de5ee017dcbe069d1fb229c9f5985
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2372545
Commit-Queue: Pierre Langlois <pierre.langlois@arm.com>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69550}
This is a reland of 1dd7f3a953
Original change's description:
> [heap] Add concurrent marking write barrier
>
> A LocalHeap creates and owns an instance of MarkingBarrier. A pointer to
> the marking barrier is set to a thread_local variable for a quick access.
>
> WriteBarrier::MarkingSlow fetches the thread_local variable and invokes
> the write barrier if it is set. Otherwise, it invokes the main thread
> heap()->marking_barrier().
>
> Each marking barrier has its own local marking worklist that is
> published during scavenge (for updating pointers) and at finalization
> of incremental marking.
>
> Typed-slot recording does not work yet because it is not thread-safe.
> It will be fixed in a subsequent CL.
>
> Bug: v8:10315
> Change-Id: I221a906436cd91e7405a253ce0eb06cf68046f2c
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2354809
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69448}
Bug: v8:10315
Change-Id: I155bb0aadd53a5333672fb085b33d8da86f3f336
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2364509
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69517}
This reverts commit 1dd7f3a953.
Reason for revert: Breaks TSAN - https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/32846?
Original change's description:
> [heap] Add concurrent marking write barrier
>
> A LocalHeap creates and owns an instance of MarkingBarrier. A pointer to
> the marking barrier is set to a thread_local variable for a quick access.
>
> WriteBarrier::MarkingSlow fetches the thread_local variable and invokes
> the write barrier if it is set. Otherwise, it invokes the main thread
> heap()->marking_barrier().
>
> Each marking barrier has its own local marking worklist that is
> published during scavenge (for updating pointers) and at finalization
> of incremental marking.
>
> Typed-slot recording does not work yet because it is not thread-safe.
> It will be fixed in a subsequent CL.
>
> Bug: v8:10315
> Change-Id: I221a906436cd91e7405a253ce0eb06cf68046f2c
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2354809
> Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#69448}
TBR=ulan@chromium.org,dinfuehr@chromium.org
Change-Id: I9719d565aaa313cd23f5e759dcef1246f475eb46
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:10315
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2362689
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69451}
A LocalHeap creates and owns an instance of MarkingBarrier. A pointer to
the marking barrier is set to a thread_local variable for a quick access.
WriteBarrier::MarkingSlow fetches the thread_local variable and invokes
the write barrier if it is set. Otherwise, it invokes the main thread
heap()->marking_barrier().
Each marking barrier has its own local marking worklist that is
published during scavenge (for updating pointers) and at finalization
of incremental marking.
Typed-slot recording does not work yet because it is not thread-safe.
It will be fixed in a subsequent CL.
Bug: v8:10315
Change-Id: I221a906436cd91e7405a253ce0eb06cf68046f2c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2354809
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69448}
Starting the task right on NotifyDeserializationComplete() turned out
to be too early for some tests. So let an AllocationObserver start
the stress allocation after deserialization is finished.
Bug: v8:10315
Change-Id: I00a4294add5ec87074f9e775c602b3031cc16d58
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2358735
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69441}
This patch introduces a new LocalIsolate and LocalFactory, which use
LocalHeap and replace OffThreadIsolate and OffThreadFactory. This allows
us to remove those classes, as well as the related OffThreadSpace,
OffThreadLargeObjectSpace, OffThreadHeap, and OffThreadTransferHandle.
OffThreadLogger becomes LocalLogger.
LocalHeap behaves more like Heap than OffThreadHeap did, so this allows
us to additionally remove the concept of "Finish" and "Publish" that the
OffThreadIsolate had, and allows us to internalize strings directly with
the newly-concurrent string table (where the implementation can now move
to FactoryBase).
This patch also removes the off-thread support from the deserializer
entirely, as well as removing the LocalIsolateWrapper which allowed
run-time distinction between Isolate and OffThreadIsolate. LocalHeap
doesn't support the reservation model used by the deserializer, and we
will likely move the deserializer to use LocalIsolate unconditionally
once we figure out the details of how to do this.
Bug: chromium:1011762
Change-Id: I1a1a0a72952b19a8a4c167c11a863c153a1252fc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2315990
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69397}
Run SimulateFullSpace in safepoint. Needed for FreeLinearAllocationArea
which would race with concurrent allocation.
Bug: v8:10315
Change-Id: If0c3c4e20edea1b05cc1a3e1a3e37fc29254a196
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2354812
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69395}
Two tests already test concurrent allocation. Another one measures
memory, so non-deterministic allocation would lead to higher memory
usage than anticipated.
Bug: v8:10315
Change-Id: If994233c87ea04076b850ffdf024f7783653e9db
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2352781
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69383}
ArrayBufferTracker was superseded by ArrayBufferList and
ArrayBufferSweeper. Now that ArrayBufferSweeper is used in production,
we can remove the unused ArrayBufferTracker mechanism.
Bug: v8:10064
Change-Id: I479169c76b6c5c634672024f77e689bb64a36504
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2339105
Reviewed-by: Simon Zünd <szuend@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69351}
We shouldn't spill weak pointers onto the stack when calling functions
that can trigger GC. DynamicMapChecks operator was using feedback loaded
from the feedback vector across the TryMigrateInstance function call.
The feedback can be a weak pointer to receiver map for monomorphic cases
and TryMigrateInstance can trigger a GC. This cl fixes it by holding
a holding a strong reference to the feedback.
Bug: v8:10774,v8:10582,v8:9684
Change-Id: Ia36f4d8ad46421ae570f41439bc1f0875081deee
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2336804
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69338}
