This reverts commit 470a10015d.
Reason for revert: GC stress bot failures.
https://uberchromegw.corp.google.com/i/client.v8/builders/V8%20Mac%20GC%20Stress/builds/14522
Original change's description:
> Start migration of try/throw/catch to match proposal.
>
> This CL does the first baby steps on moving the current (experimental)
> exception handling to match that of the WebAssembly proposal.
>
> It does the following:
>
> 1) Use exception tags instead of integers.
>
> 2) Only handle empty exception signatures (i.e. no values associated
> with the exception tag.
>
> 3) Only handle one catch clause.
>
> 4) Be sure to rethrow the exception if the exception tag does not match.
>
> Note: There are many things that need to be fixed, and are too
> numerous to list here. However, the code should have TODO's on each
> missing parts of the implementation.
>
> Also note that the code currently doesn't handle nested catch blocks,
> nor does it change the throw value being an integer. Rather, the
> integer value is still being thrown, and currently is the exception
> tag. Therefore, we don't build an exception object. This is the reason
> why this CL doesn't handle exceptions that pass values.
>
> Also, the current implementation still can't handle multiple modules
> because tag resolution (between) modules has not be implemented yet.
>
> Bug: v8:6577
> Change-Id: Id6d08b641b3c42d1eec7d4db582f2dab35406114
> Reviewed-on: https://chromium-review.googlesource.com/591910
> Reviewed-by: Brad Nelson <bradnelson@chromium.org>
> Commit-Queue: Karl Schimpf <kschimpf@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#47087}
TBR=bradnelson@chromium.org,eholk@chromium.org,kschimpf@chromium.org
Change-Id: I01dc8c40cc1057333a988c1d275ce5f457b0cb64
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:6577
Reviewed-on: https://chromium-review.googlesource.com/598847
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#47090}
This CL does the first baby steps on moving the current (experimental)
exception handling to match that of the WebAssembly proposal.
It does the following:
1) Use exception tags instead of integers.
2) Only handle empty exception signatures (i.e. no values associated
with the exception tag.
3) Only handle one catch clause.
4) Be sure to rethrow the exception if the exception tag does not match.
Note: There are many things that need to be fixed, and are too
numerous to list here. However, the code should have TODO's on each
missing parts of the implementation.
Also note that the code currently doesn't handle nested catch blocks,
nor does it change the throw value being an integer. Rather, the
integer value is still being thrown, and currently is the exception
tag. Therefore, we don't build an exception object. This is the reason
why this CL doesn't handle exceptions that pass values.
Also, the current implementation still can't handle multiple modules
because tag resolution (between) modules has not be implemented yet.
Bug: v8:6577
Change-Id: Id6d08b641b3c42d1eec7d4db582f2dab35406114
Reviewed-on: https://chromium-review.googlesource.com/591910
Reviewed-by: Brad Nelson <bradnelson@chromium.org>
Commit-Queue: Karl Schimpf <kschimpf@chromium.org>
Cr-Commit-Position: refs/heads/master@{#47087}
Instead of auto-generating the Name() convenience accessor, use a macro to
avoid wasting code space.
BUILTIN_CODE(isolate, Name)
expands to
isolate->builtins()->builtin_handle(Builtins::kName);
This reduces the size of libv8.so by 134,752 bytes on a x64 release build.
Bug: v8:6624
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: Idff7ee5c45e344e73412c0f47e92553c7c7ff75f
Reviewed-on: https://chromium-review.googlesource.com/593607
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#47010}
Now that the maximum string length varies between platforms, the
correctness fuzzer is unhappy. It will ignore crashes, so when we know
we have reached platform-dependant behavior just crash if
--abort_on_stack_overflow is enabled.
Also rename abort_on_stack_overflow to
abort_on_stack_or_string_length_overflow.
Bug: chromium:748137
Change-Id: Ie4e96709b90029b5ce3c8408064d928f841b3b9f
Reviewed-on: https://chromium-review.googlesource.com/589269
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#47007}
Bytecode handlers are technically not builtins, but very similar to CSA
builtins in most respects (CSA-generated code, currently included in the
snapshot and deserialized for every isolate). This prints bytecode
handler sizes (in addition to standard CSA builtin sizes) when
--print-builtin-size is passed.
Bug:
Change-Id: Ibd78422c5138b77ccf298f97c7c1fc1b73a3a09b
Reviewed-on: https://chromium-review.googlesource.com/581191
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46831}
This reverts commit 20d5048a6f.
Revert "[heap] Ensure that concurrent marking tasks exit before heap tear down."
This reverts commit 387f65d41a.
Reason: concurrent marking tasks waiting for a signal from the main thread
is susceptible to deadlocks. We should instead re-schedule concurrent marking
threads once they exit.
BUG=chromium:694255
Change-Id: I20db2f26b42e960f4cc04506d9598c1187b8a003
Reviewed-on: https://chromium-review.googlesource.com/571800
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46671}
The tail call implementation is hidden behind the --harmony-tailcalls
flag, which is off-by-default (and has been unstaged since February).
It is known to be broken in a variety of cases, including clusterfuzz
security issues (see sample Chromium issues below). To avoid letting
the implementation bitrot further on trunk, this patch removes it.
Bug: v8:4698, chromium:636914, chromium:724746
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng;master.tryserver.v8:v8_linux_noi18n_rel_ng
Change-Id: I9cb547101456a582374fdf7b1a3f044a9ef33e5c
Reviewed-on: https://chromium-review.googlesource.com/569069
Commit-Queue: Adam Klein <adamk@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46651}
This adds a convenience method for the common Smi to int conversion
pattern.
Bug:
Change-Id: I7d7b171c36cfec5f6d10c60f1d9c3e06e3aed0fa
Reviewed-on: https://chromium-review.googlesource.com/563205
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Andreas Rossberg <rossberg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46516}
Catch prediction for optimized frames had two issues:
Inlined frames were iterated from caller-to-callee (which could
result in incorrect predictions if one frame predicted CAUGHT and
another predicted PROMISE).
When encountering a builtin frame, we'd unconditionally return its
prediction (which is wrong if it predicted UNCAUGHT and another inlined
frame predicted either CAUGHT or PROMISE).
This CL fixes both issues and refactors the function to reduce nesting.
BUG=v8:6536
Change-Id: I764a4ec033e4476bd840134b5eacfe0e08b3c1a4
Reviewed-on: https://chromium-review.googlesource.com/555519
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46358}
V8's catch prediction mechanism tries to predict whether a thrown
exception will be caught, just by looking at the current call stack.
At the time when catch prediction was first introduced, only a few
builtins (mostly related to Promise and Generator) could end up being
fed into the catch prediction mechanism. This is no longer the case now
that builtins are used in new ways, e.g. Array.p.forEach's continuation
builtins.
This CL removes the need to explicitly mark all builtins visible to the
StackFrameIterator as CAUGHT/UNCAUGHT/PROMISE, and instead defaults to
treating unmarked builtins as UNCAUGHT.
BUG=v8:6536
Change-Id: Ibdc106a91b2b0ffb93099433077642cad02c71e2
Reviewed-on: https://chromium-review.googlesource.com/555518
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46357}
Previously V8 created a promise to return to userland,
but instead we let the embedder create and track the promise.
Bug: v8:5785
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I8903ffbabf3a256f1c8df844a656a873da304586
Reviewed-on: https://chromium-review.googlesource.com/492646
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Daniel Clifford <danno@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46333}
There are very few cases where OSR code can be re-used, and where the
function won't be non-concurrently optimized after OSR has happened.
Maintaining the OSR code cache is unnecessary complexity, and caching
OSR prevents us from e.g. seeding the optimizer with the actual OSR
values.
So, this patch removes it.
Change-Id: Ib9223de590f35ffc1dc2ab593b7cc9fe97dde4a6
Reviewed-on: https://chromium-review.googlesource.com/552637
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46306}
Removes from CL https://codereview.chromium.org/2929853003 code to fix
histogram timers in class WasmCompilationUnit. This was done because
the CL was reverted due to errors caused by background compiles that
updated UMA histogram timers.
The goal of this CL is to reland the remaining portion of the reverted
CL.
Bug:v8:6361
Change-Id: Ic03ceb118734bd55c463a843521bcd5b09342afe
Reviewed-on: https://chromium-review.googlesource.com/550196
Reviewed-by: Mircea Trofin <mtrofin@chromium.org>
Commit-Queue: Karl Schimpf <kschimpf@google.com>
Cr-Commit-Position: refs/heads/master@{#46268}
This removes support for code-stub to tail-call into the runtime via the
deoptimizer. The Hydrogen code-stubs would trigger a deopt in order to
materialize a trampoline frame, which would then continue execution in a
runtime function associated with each stub. This is no longer needed for
code-stubs built with the CSA.
R=jarin@chromium.org
BUG=v8:6408
Change-Id: I1ff8dc03ac716200b28e962259a3e233aeda1234
Reviewed-on: https://chromium-review.googlesource.com/548375
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46223}
This is a fix to https://codereview.chromium.org/2929853003 that got
reverted. The DCHECK checked to see that it was not in a background
thread. While this is a property we want for v8, it is also used
by blink, and blink violates this property.
Therefore, this CL removes the DCHECK for now.
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2961443002
Cr-Commit-Position: refs/heads/master@{#46190}
Each frame within an optimized frame needs to be re-verified as being
subject to debugging. The specific crash from the linked bug was caused
by the fact that non-user JS frames don't necessarily have an associated
script. We'd ignore that and attempt to call
Script::GetPositionInfo() on the Undefined constant.
BUG=chromium:732408
Change-Id: Ib2982890529623d150092fa57e26fcb237fef59f
Reviewed-on: https://chromium-review.googlesource.com/545895
Reviewed-by: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46186}
HistoryTimer's can't run in the background because they use a timer
with a simple api of Start() and Stop(). This CL fixes this problem
by building a base class TimedHistogram that doesn't have a timer.
The class HistoryTimer is modified to use this base class so that
uses that run on the foreground thread do not need to be modified.
It also adds a new class TimedHistogramScope that defines the timer
in this class. This allows the corresopnding TimedHistogram class to
be type safe.
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2929853003
Cr-Commit-Position: refs/heads/master@{#46150}
For each Managed<T> (which is a Foreign), we create a weak global handle
with a finalizer which deletes the referenced C++ object once the
Foreign is dead.
Before calling this finalizer, the garbage collector needs to mark the
referenced object black (i.e. live), because the finalizer might
resurrect it.
Since this is never done for managed objects, we can use the more
lightweight phantom handle semantics, which allows the referenced
object to be garbage collected right away.
However, we can't access the global handle via the WeakCallbackInfo,
because the global handle will already be garbage collected. So we need
to store it explicitly. This is solved by storing the global handle
together with the finalizer.
In order to implement this, ownership of the ManagedObjectFinalizer
is moved from the isolate to the managed object.
R=ulan@chromium.org, mtrofin@chromium.org
BUG=v8:6505, chromium:734345
Change-Id: I94a245df601f70e19355d82439d30099e159231b
Reviewed-on: https://chromium-review.googlesource.com/539578
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46036}
This removes the heuristic from {JSStackFrame::IsConstructor} that tried
to infer whether a frame was called as a constructor or not from the
receiver value. We are now carrying along the appropriate bit derived
from the frame type instead.
R=jgruber@chromium.org
TEST=message/regress/regress-5727
BUG=v8:5727
Change-Id: I0e2f1d0f95485c84c4ebcd3cbfe0123c6afd2e01
Reviewed-on: https://chromium-review.googlesource.com/500313
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45972}
This CL is the first step in introducing a compilation manager for
asynchronous compile jobs in WebAssembly.
The compilation manager holds a list of currently active
AsyncCompileJobs. With the compilation manager these compile jobs get
deallocated when the isolate shuts down. Note that this CL is not enough
to provide a graceful isolate shutdown. For this we have to wait for all
compilation tasks to finish before we shut down, and we have to make the
tasks stateless. I plan to do these changes in separate CLs.
R=clemensh@chromium.org, mtrofin@chromium.org
BUG=v8:6436
Change-Id: I9a6e165dd2ef6d33944ca303fed49f7940eea7a2
Reviewed-on: https://chromium-review.googlesource.com/528079
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45858}
This CL takes advantage of the fact that StatsCounter is now local to
the Counters class. This includes:
1) Method StatsTable::SetCreateHistogramFunction() was only called in
one spot (in api.cc), which also called Counters::ResetHistograms()
and Counters::InitializeHistorgram(). InitializeHistogram can be
folded into Histogram.Reset().
2) Since Histogram::Reset() now regenerats the histogram, we no longer
need the field lookup_done_. Therefore there is no longer a race
between updating ptr_ and lookup_done_, making the Histogram class
thread safe.
3) Made the constructors of several classes private (except for class
Counters), minimizing the scope that they are used. When the couldn't
be moved, add comment that they were public only for test cases.
4) Removed the need for a mutex lock on StatsCounter::Reset(), since
it is now guaranteed to only be called when
StatsTable::SetCounterFunction() is called.
BUG=v8:6361
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_chromium_rel_ng
Review-Url: https://codereview.chromium.org/2918703002
Cr-Commit-Position: refs/heads/master@{#45791}
ThrowIfHole bytecodes were handled by introducing deopt points to check
for a hole. To avoid deopt loops a hole check protector was used to
generate control flow if there was a deopt due to a hole. However, the
normal control flow version should be as fast as the deopt version
in general. The deopt version could potentially consume less compile time
but it may not be worth the complexity added. Hence simplifying it to
only construct the control flow.
Bug: v8:6383
Change-Id: Icace11f7a6e21e64e1cebd104496e3f559bc85f7
Reviewed-on: https://chromium-review.googlesource.com/525573
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45783}
This CL contains a few pieces:
- A new mechanism to create "BuiltinContinuation" checkpoints in TurboFan
graphs, which--when triggered--swizzle the values in the the FrameState to be
parameters to a typically TF-generated builtin that resumes execution to finish
the slow-case functionality.
- Continuation builtins that have special handling in the deoptimizer and their own
new frame type to ensure that the values they need to begin executing can be stashed
away and restored immediately before the builtin is called via a trampoline that runs
when the continuation builtin's frame execution resumes.
- An implementation of Array.prototype.forEach in TurboFan that can be used to
inline it. The inlined forEach implementation uses the checkpoints mechanism
described above to deopt in the middle of the forEach in the cases that optimization
invariants are violated. There is a slightly different continuation stub for each
deopt point in the forEach implementation to ensure the correct side-effects, i.e.
that the deopt of the builtin isn't programmatically observable.
Review-Url: https://codereview.chromium.org/2803853005
Cr-Commit-Position: refs/heads/master@{#45764}
Fixes issues raised in CL https://codereview.chromium.org/2887193002.
That is:
1) Remove using mutex in Isolate::InitializeCounters().
2) Use counters_shared_.get() instead of counters_ (and hence, also
remove field counters_).
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2919953003
Cr-Commit-Position: refs/heads/master@{#45743}
Introduces ThrowReferenceErrorIfHole / ThrowSuperNotCalledIfHole
/ ThrowSuperAlreadyCalledIfNotHole bytecodes to handle hole checks.
In the bytecode-graph builder they are handled by introducing a deopt point
instead of adding explicit control flow. JumpIfNotHole / JumpIfNotHoleConstant
bytecodes are removed since they are no longer required.
Bug: v8:4280, v8:6383
Change-Id: I58b70c556b0ffa30e41a0cd44016874c3e9c5fe1
Reviewed-on: https://chromium-review.googlesource.com/509613
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45720}
Most prototype implementations are not fully supported in the
interpreter. This is the case at least for exception handling, simd, and
atomics. Any function can be redirected to the interpreter though,
either by passing --wasm-interpret-all, or by dynamically redirecting to
the interpreter for debugging.
Making the flags experimental keeps the fuzzer from playing around with
these flags.
Drive-by: Refactor tests which explicitly set the prototype flag to use
a new scope for that.
R=ahaas@chromium.org
BUG=chromium:727584
Change-Id: I67da79f579f1ac93c67189afef40c6524bdd4430
Reviewed-on: https://chromium-review.googlesource.com/519402
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45639}
Passing --print-builtin-size will print the size of all builtins on
isolate creation.
BUG=v8:5737
Review-Url: https://codereview.chromium.org/2895163002
Cr-Commit-Position: refs/heads/master@{#45605}
This is consistent with C++ memory model and avoids confusion with GC
write barrier.
BUG=
Review-Url: https://codereview.chromium.org/2912773002
Cr-Commit-Position: refs/heads/master@{#45584}
By moving StatsTable from class Isolate to class Counters, it make the
class StatsTable thead safe. This is needed because these two classes
call each other, and for background compilation, instances of the
Counters class can persist longer that the corresponding Isolate it
came from.
It also removes unnecessary hops to the the Isolate, and checks if the
StatsTable has been created, for these communications.
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2906063002
Cr-Commit-Position: refs/heads/master@{#45576}
All APIs that can throw exceptions should return Maybe<> values
BUG=none
R=neis@chromium.org,gsathya@chromium.org
Change-Id: I6a6e5888cd71257bb02bdcfcc587c909d0c1d8f4
Reviewed-on: https://chromium-review.googlesource.com/517785
Commit-Queue: Jochen Eisinger <jochen@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45557}
Creates a new class StatsCounterThreadSafe to be used by counters that
can be updated when compiling/decoding etc. are done using workers.
Does this by using a mutex on all opreations.
Also updates the StatsCounterThreadSafe constructor to force counter
initialization, as well as method Reset(). In addition, whenever the
method StatsTable::SetCounterFunction() is called (from the main
thread), it forces counter initialization for all thread safe stats
counters.
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2887193002
Cr-Commit-Position: refs/heads/master@{#45526}
