Since the new space is always empty after a full GC, the old-to-new
remembered set is also always empty after a full GC. This means we can
get rid of the sweeping_slot_set_.
This slot set was used to allow the main thread to insert into the
old-to-new remembered set non-atomically. The sweeping slot set was
owned by the sweeper, which deletes slots in free memory from it. The
main thread would start with an empty old-to-new remembered set. After
sweeping both slot sets are merged again.
The sweeper now needs to behave differently during a GC. When sweeping
a page during full GC, the sweeper needs to delete old-to-new-slots in
free memory.
Outside of the GC the sweeper isn't allowed to remove from the
old-to-new slots anymore. This would race with the main thread that adds
slots to that remembered set while the sweeper is running. However,
there should be no recorded slots in free memory. DCHECKing this is
tricky though, because we would need to synchronize with the main
thread right-trimming objects and at least String::MakeThin only deletes
slots after the map release-store.
Bug: v8:12760
Change-Id: Ic0301851a714e894c3040595f456ab93b5875c81
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3560638
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79713}
- Always show the navigation buttons
- Format code with fixed-width font
- Limit the property-table height for more compact tooltips
Bug: v8:10644
Change-Id: I0a0f30056455371bad12b2c679d184948c5b52de
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3555772
Reviewed-by: Patrick Thier <pthier@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79652}
Search for all files with testing naming convention and run that on v8_presubmit.
Also modify all PRESUBMIT files in the tools directory to include any test file
with the appropriate naming convention.
Bug: chromium:1306474
Change-Id: I61c1b7c71badbbc3b99705289588aa8280824d66
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3532266
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Almothana Athamneh <almuthanna@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79651}
Bidirectional phase switching by shortcuts "n", "b".
Improved selection of nodes when they are splitting or raising to a common ancestor.
Fixed minor inconsistencies in some variable names with the project style.
Added name and email to the AUTHORS file for first-time contribution.
Change-Id: I0c903dbf81c3d1d75503004ce412a81aace06a61
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3537008
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79636}
The debugger maintains a stack of promises used for catch prediction
with promise builtins and async functions. Previously this stack would
hold on to the individual promises strongly, and subtle bugs that lead
to not properly cleaning up the stack in some corner cases would often
lead to significant memory issues (e.g. leaking whole iframes).
This refactors the PromiseOnStack to be
(a) on the V8 heap, rather than allocating C++ structs with global
handles pointing to the promises, and
(b) hold on to the promises only weakly.
While this will not guarantee proper promise stack management, it will
at least ensure that edge cases don't lead to catastrophic (debugger
only) leaks.
Bug: chromium:1292063
Change-Id: I9c293ca2032de3a59e1e9624f132d37187805567
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545176
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Simon Zünd <szuend@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79594}
Some edges are self looping because of incorrect `horizontalPos`. This is occuring because of an unexpected scenario caused due to incorrect calculation of `inputApproch` and `outputApproach`. And all of this is occuring because of insufficient distance between two nodes.
An example of the problem is shown in the image: https://imgur.com/aAmnzaK.
Change-Id: I056e1fbcc420ce65a3ae9201e187b22ad3fbaaba
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3535791
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79570}
- Avoid redrawing property-link tables if the contents don't change
- Don't update timeline legends if the selection doesn't change
- Use shorter class names for the flamechart for faster parsing
- Round positions in flamechart to avoid long strings that would be
created from raw double positions
- Don't redraw the tooltip if the content is the same
Change-Id: I925f1708400286c7c9f8db62f75c3b5fe8a16b12
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3521945
Reviewed-by: Patrick Thier <pthier@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79540}
JSObject::InitializeBody now checks whether the instance type of the
object being initialized can have embedder data slots around the
initialization logic for these slots. This fixes a performance
regression on certain benchmarks.
To perform this check efficiently, a new instance type,
JSObjectWithEmbedderSlots, is introduced so that the check becomes a
simple range check.
Bug: chromium:1304139
Change-Id: I00c892bc2276e950b59602257ca1c2435c10e517
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3507712
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Samuel Groß <saelo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79497}
It turned out that on arm and arm64 we over-estimated the code size of a
Wasm module quite a bit. This CL adds some more output for the
--trace-wasm-compilation-times flag, and adds a script to compute the
factors we use for code size estimates from that output.
I ran the script on a few benchmarks (an older Epic module, the current
Photoshop module, and the benchmark from the linked bug), and adjusted
the constants accordingly.
Also, simplify the API of {ReservationSize} to only return a single
number, and fail internally if we need to allocate more than the engine
supports (which would only fail for artificially large modules).
R=jkummerow@chromium.org
Bug: chromium:1302310
Change-Id: I5b2c27ff3e360fb6738cf5dd697bcee09e106b6d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3522067
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79482}
The fuzzers were passing the flag --es-staging which doesn't exist. This
CL updates them to pass the flag --harmony which does exist.
Change-Id: I02c83026e5b9bdf49e51e700f16702bf56cd49e8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3522064
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Marja Hölttä <marja@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79464}
Instead of implementing ExternalObjects as plain JSObjects with a single
EmbedderDataSlot pointing to a Foreign containing the actual raw
pointer, this CL now creates a new JSExternalObject type that directly
contains the external pointer.
As a side-effect of this refactoring, nullptr values are now no longer
valid for ExternalObjects.
Change-Id: Ic8ff334681c966e823ca70f34dd1efaaa21a0789
Cq-Include-Trybots: luci.v8.try:v8_linux64_heap_sandbox_dbg_ng,v8_linux_arm64_sim_heap_sandbox_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3513234
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Samuel Groß <saelo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79459}
Instead of using the high water mark for determining this metric, we use
a bitset for all active/used system pages on a V8 heap page. Each time
when allocating a LAB on a page, we add the pages of that memory range
to that bitset. During sweeping we rebuild that bitset from scratch and
replace it with the old one in case free pages are discarded by the GC.
We DCHECK here that the sweeper only ever removes pages. This has the
nice benefit of ensuring that we don't miss any allocations (like we
do now for concurrent allocations).
CommittedPhysicalMemory for a page is then calculated by counting the
set bits in the bitset and multiplying it with the system page size.
This should be simpler to verify and track the "real" effective size
more precisely.
One case where we are partially less precise than the current
implementation is for LABs. In order to reduce complexity we now treat
all pages of a LAB allocation as active immediately. In the current
implementation we tried to only account the actual used part of the LAB
when changing the LAB later. This is more complex to track correctly
but also doesn't account the currently used LAB in effective size.
Change-Id: Ia83df9ad5fbb852f0717c4c396b5074604bd21e9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3497363
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79428}
Clean up two flags which control features shipped in m99
harmony_intl_locale_info
harmony_intl_enumeration
Bug: v8:10743, v8:11638
Change-Id: I856afaceb0972052d1bcc70b162e8f9ee9071dd6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3508169
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Commit-Queue: Frank Tang <ftang@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79397}
Prepare gcmole.cc for the next update:
- Print possible GC locations when discovering stale/dead variables
- Make error messages less confusing for the modern V8 engineer
- Prepare gcmole to read suspects.allowlist instead of .whitelist
- Use more readable variable names
- Only log non-found types with --verbose
- Change the currently unusued gccauses format in gcmole.py and
support loading it back in gcmole.cc
- Implemented first basic gc call-chain printing (disabled by default)
GCmole packaging:
- Add debug mode to bootstrap.sh build script
- Update gcmole.py run instructions in bootstrap.sh and package.sh
Bug: v8:10009
Change-Id: I369d48baa2980455d2e8f57e7a803d0384fe83f1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3480095
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79357}
- Convert gcmole to python3-ish code
- use local Path implementation for future full migration
- Use optparse and explicit arguments for gcmole
- Add explicit directories flags
- Use backwards compatible env vars as fallbacks
- Add gn target v8_gcmole_files to avoid issues with missing or
incompatible generated files
Drive-by-fixes for running gcmole without ignored files:
- Disable gcmole in Isolate::UnwindAndFindHandle
- Partially disable gcmole in V8HeapExplorer::AddEntry
Bug: v8:10009
Change-Id: I5b4d1c6554db300190226361b6c518419109ff3d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3497320
Reviewed-by: Liviu Rau <liviurau@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79348}
Store the free registers as a RegList rather than stack of Register
values. This allows us to simplify some of the register freeing logic,
including passing the current free set to nodes for use as temporaries.
Drive-by: Replace ALWAYS_ALLOCATABLE_GENERAL_REGISTERS with
ALLOCATABLE_GENERAL_REGISTERS, which is the more general list (the former
is an implementation detail for optionally reserving a register for
the cage register).
Bug: v8:7700
Change-Id: I666e9a7547c2f4f4e578fbcbb4bd3fe3cb06dac5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3497767
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79344}
Profiler:
- Track profiler tick durations
- Various speedups due to low-level hacking
Improve code-panel:
- Better register highlighting
- Added address navigation and highlighting
- Removed obsolete inline source-view
Improve script-panel:
- Keep current source position focused when showing related entries
- Better tool-tip with buttons to focus on grouped entries per
source postion
- Focus by default on other views when showing related entries
Improve timeline-panel:
- Initialise event handlers late to avoid errors
- Lazy initialise chunks to avoid errors when zooming-in and trying to
create tooltips at the same time
Change-Id: I3f3c0fd51985aaa490d62f786ab52a4be1eed292
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3492521
Reviewed-by: Patrick Thier <pthier@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79329}
When --maglev is set, tier up to Maglev from unoptimized tiers based on
--interrupt-budget-for-maglev, initially set to 40KB (which should very
roughly by 1/10th of the time until the TF tierup decision is made).
On the first interrupt, a non-concurrent optimization to Maglev is
requested, which the next call to the marked function will perform.
- There is no support for tiering from Maglev to TF yet.
- Maglev's language support is minimal and tests are not expected to
pass with --maglev.
- Disable --maglev by default for now.
Drive-by: fixes related to Maglev flag definitions.
Bug: v8:7700
Change-Id: I121bb3f4f3830fdd20e1d4a12d3e04f08a99be38
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3500302
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79328}
Maglev is currently enabled only on x64; BUILD.gn knows about this
through v8_enable_maglev, but v8_check_header_includes doesn't. It
thus tries to compile maglev files on platforms that don't have maglev
support yet.
Add an explicit maglev exclude rule until we support other platforms.
Bug: v8:7700
Change-Id: Iac991741c7d630dc4ed9f9fbf6df7656853cc743
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3494536
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79305}
The per-Isolate Symbol tables are implemented using NameDictionary
before, which has additional property details overhead
And NameDictionary is limited to 2^23, which limits the Symbol
tables to be a maximum of 2^23.
- replace NameDictionary with SymbolTable in isolate
Bug: v8:12575
Change-Id: Ica4f05aac3494f7dfa3a074c240d4ba25df814e9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3476897
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79285}
- List the current v8 version
- Minor code cleanup
Change-Id: Ic7a89e42d27465cc5df8e2249eaeacf8ca1eb6a7
No-Try: true
No-Presubmit: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3477034
Auto-Submit: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79222}
This is a reland of 1025bf26e3
Changes since revert:
- TSAN issue fixed by https://crrev.com/c/3475084
- Skip the shared-struct-workers test until shared GC deadlock is fixed,
being tracked in v8:12645
Original change's description:
> [shared-struct] Prototype JS shared structs
>
> Unlike the Stage 1 proposal, for simplicity the prototype does not add
> any new syntax, instead opting for exposing a SharedStructType
> constructor which takes an array of field names. This type constructor
> returns constructors for shared structs.
>
> Shared structs can be shared across Isolates, are fixed layout, have no
> prototype, have no .constructor, and can only store primitives and
> other shared structs.
>
> The initial prototype does not have TurboFan support.
>
> Bug: v8:12547
> Change-Id: I23bdd819940b42139692bcdb53d372099b0d4426
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3390643
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Marja Hölttä <marja@chromium.org>
> Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Commit-Queue: Shu-yu Guo <syg@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79156}
Bug: v8:12547
Change-Id: Ic1f5cf9fa9791ae2d5d5dc7c110614ca10b5d98e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3475078
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Marja Hölttä <marja@chromium.org>
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79215}
Migrate predictable_wrapper to py3. Run test in v8_presubmit.
R=liviurau@chromium.org, machenbach@chromium.org
Bug: chromium:1245634
Change-Id: I941e248ffcf12ce26a55a5f5889dab06ee74e66e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3448379
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Liviu Rau <liviurau@chromium.org>
Commit-Queue: Alexander Schulze <alexschulze@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79200}
The tool has been unmaintained for a while and doesn't work.
We do have either the system-analyzer or profview as valid web-based
replacements. For all other use-cases we recommend using the
command-line versions.
Change-Id: I3a07e80aebfb1f8d6ba16d6bffe16d9da7b9eac4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3474677
Reviewed-by: Patrick Thier <pthier@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79178}
This reverts commit 1025bf26e3.
Reason for revert: https://crbug.com/v8/12645
Original change's description:
> [shared-struct] Prototype JS shared structs
>
> Unlike the Stage 1 proposal, for simplicity the prototype does not add
> any new syntax, instead opting for exposing a SharedStructType
> constructor which takes an array of field names. This type constructor
> returns constructors for shared structs.
>
> Shared structs can be shared across Isolates, are fixed layout, have no
> prototype, have no .constructor, and can only store primitives and
> other shared structs.
>
> The initial prototype does not have TurboFan support.
>
> Bug: v8:12547
> Change-Id: I23bdd819940b42139692bcdb53d372099b0d4426
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3390643
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Marja Hölttä <marja@chromium.org>
> Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Commit-Queue: Shu-yu Guo <syg@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79156}
Bug: v8:12547
Change-Id: I44f2b8bb7487b4d39ba1282585e0b2282501230f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3474676
Auto-Submit: Michael Achenbach <machenbach@chromium.org>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Owners-Override: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79170}
Now that we are able to compact map space, we can also get rid of the
map space and allocate maps in the old space instead. This CL introduces
a FLAG_map_space for enabling/disabling the map space but the map space
remains enabled by default for now.
Without a separate space for maps, the GC can't prevent relocation of
maps anymore. Therefore this CL always allows compaction of maps when
running without a map space. Rename flag to --compact-maps to better fit
this scenario.
mkgrokdump and debug_helper also need to be updated to look for maps
also in the old space. The map space is now optional.
Bug: v8:12578
Change-Id: Ic4e4abd0b58bee26e64329b1c92dbccb07d8105a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3424483
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Seth Brenith <seth.brenith@microsoft.com>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79165}
