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}
It's extremely difficult to get right: there have been several bugs
related to this feature, especially when combined with
non-simple parameter lists in arrow functions.
BUG=chromium:727218
Change-Id: I97dfbc57a7650199964c5fe99de69143c8e537c2
Reviewed-on: https://chromium-review.googlesource.com/518145
Commit-Queue: Marja Hölttä <marja@chromium.org>
Reviewed-by: Daniel Vogelheim <vogelheim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45603}
The old logic made the cfi build fail on each clang update.
Bug: chromium:726584
Change-Id: Ia24181d3bc92eb18116c2ac2b42ac2c68f02ce25
Reviewed-on: https://chromium-review.googlesource.com/518185
Reviewed-by: Jochen Eisinger <jochen@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45601}
Retrieve length from the buffer instead of storing in as value in the
map.
Change-Id: Icdc87bb46c29fff779bf6e98139324662ced5bf6
Reviewed-on: https://chromium-review.googlesource.com/518171
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Jochen Eisinger <jochen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45600}
This CL improves break locations for expressions like 'var a = <expr>'. Without CL we use <expr> position as break location for initialization statement, with this CL we use position of first character after '=' as position.
Benefits (see test for details):
- only one break in expressions which includes mix of property lookup and calls, e.g. var p = Promise.resolve().then(x => x * 2),
- removed redundant break location for expressions like: let { x, y } = { x: 1, y: 2}.
TBR=dgozman@chromium.org,rmcilroy@chromium.org,machenbach@chromium.org,marja@chromium.org,kozyatinskiy@chromium.org,devtools-reviews@chromium.org,v8-reviews@googlegroups.com
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: v8:5909
Change-Id: Ie84fa79afeed09e28cf8478ba610a0cfbfdfc294
Reviewed-on: https://chromium-review.googlesource.com/518116
Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org>
Reviewed-by: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45598}
We add the bot via the post-upload step, that's good enough
R=machenbach@chromium.org,tandrii@chromium.org
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I8d582d101b4d2f308829da75df56e04a3dabadbf
Reviewed-on: https://chromium-review.googlesource.com/518114
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Jochen Eisinger <jochen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45594}
The problem with enums is that neither is_integral, nor is_signed or
is_unsigned is true for them. Thus, comparison with our CHECK/DCHECK
macros always just uses the default comparison, which fails if the
signedness of the underlying integer types does not match.
This CL fixes this by considering the underlying integer type of an
enum to choose the right comparison operator.
R=ishell@chromium.orgCC=ahaas@chromium.org
Change-Id: I5ef56d1b86228e879f5866967ab7e709f1e97f0b
Reviewed-on: https://chromium-review.googlesource.com/518123
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45593}
This may happen in KeyedStoreIC when the last store via setter deprecates the receiver map.
Bug: chromium:723366
Change-Id: Iff19e50c3761584401340d276ab4eead111883c3
Reviewed-on: https://chromium-review.googlesource.com/517952
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45592}
This CL replaces the last usages of std::vector in the AsmJsParser by
ZoneVector. This allows to also measure the memory consumption of these
vectors, since it is now contained in the zone memory.
ZoneVectors are reused to avoid accumulating lots of unused memory.
This also saves 2.6% performance (avg over 1000 runs) on my local
workstation.
R=mstarzinger@chromium.org
Change-Id: I04c96db558d9c362b1494ddd9e975edf2783403c
Reviewed-on: https://chromium-review.googlesource.com/516985
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45591}
Both lazy initialization and static initialization (via static
initializer) are bad. Fortunately, the arrays we are constructing are
constant anyway, so we can just compute them at compile time. This is
enforced by making them constexpr.
This also saves all code needed for the initialization, and makes
accesses to the tables faster, as they don't need any atomic operations
(via LazyInstance).
R=ahaas@chromium.org
Change-Id: I7d3ba9b0f2602f596a6c71c8c567e0d1bc306268
Reviewed-on: https://chromium-review.googlesource.com/517083
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45589}
This time for the current memory size. This call also used to use the
context object stored in the instance, hence it required the instance
to be set. This is no longer the case, so the DCHECKs can just be
removed.
R=ahaas@chromium.org
BUG=chromium:727222
Change-Id: I72a7e3e80c3beb15ecad00c5be068e803456797e
Reviewed-on: https://chromium-review.googlesource.com/517947
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45587}
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}
This reverts commit 2ba1fcda5f.
Reason for revert: Speculative for:
https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064%20%28dbg%29/builds/8413
Original change's description:
> Remove ENTER_V8_DO_NOT_USE usage from other macros
>
> Eventually I want to delete that macro, so just inline it at the places
> where we'll need to keep it.
>
> BUG=v8:5830
> R=marja@chromium.org
>
> Change-Id: I904a1dd3555c23c69e457e078faaaa86a9514932
> Reviewed-on: https://chromium-review.googlesource.com/518043
> Reviewed-by: Marja Hölttä <marja@chromium.org>
> Commit-Queue: Jochen Eisinger <jochen@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#45561}
NOTRY=true
NOTREECHECKS=true
NOPRESUBMIT=true
TBR=marja@chromium.org,jochen@chromium.org
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
BUG=v8:5830
Change-Id: I7097597d2695980f71fc93a79999d59956dd76c9
Reviewed-on: https://chromium-review.googlesource.com/518107
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45578}
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}
Furthermore avoid lock-step between pointer updating phases as they
should execute in parallel without synchronization restrictions.
Bug: chromium:726040
Change-Id: I26ce0d1f2a4637ff5610cae556113e3d736788e2
Reviewed-on: https://chromium-review.googlesource.com/518103
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45574}
This widens the range of value output counts to 32 bit on the {Operator}
class. Note that the limit imposed by the parser is 65535 parameters for
each function, but the {Start} node has additional value outputs.
R=jarin@chromium.org
TEST=mjsunit/regress/regress-crbug-724153
BUG=chromium:724153
Change-Id: I21b5d947cc2305b255ddbbff6ec1dfa5c02784c7
Reviewed-on: https://chromium-review.googlesource.com/517489
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45573}
This removes special support in the compilation pipeline to keep alive
unoptimized code for inlined functions (i.e. preventing the code flusher
from clearing it). Now that the code flusher is gone, this is obsolete.
R=jarin@chromium.org
Change-Id: I00361d12f2d17556056d06c10ad19f09a4772cd8
Reviewed-on: https://chromium-review.googlesource.com/517948
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45568}
For non-simple param lists, the parser first declares a TEMPORARY for each
param, and then the named variables as locals. The TEMPORARY variables determine
the parameter count.
This CL makes the PreParser produce the same parameter count as the Parser.
BUG=v8:5516
Change-Id: I8a794d6a8342145ab7934d922e2d69450d67b199
Reviewed-on: https://chromium-review.googlesource.com/517944
Commit-Queue: Marja Hölttä <marja@chromium.org>
Reviewed-by: Daniel Vogelheim <vogelheim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45566}
Eventually I want to delete that macro, so just inline it at the places
where we'll need to keep it.
BUG=v8:5830
R=marja@chromium.org
Change-Id: I904a1dd3555c23c69e457e078faaaa86a9514932
Reviewed-on: https://chromium-review.googlesource.com/518043
Reviewed-by: Marja Hölttä <marja@chromium.org>
Commit-Queue: Jochen Eisinger <jochen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45561}
Replace the macro with more generic macros expanding to the same
code.
BUG=v8:5830
R=marja@chromium.org
Change-Id: Ibf31fa4d89960a025af859198e190910c5608a20
Reviewed-on: https://chromium-review.googlesource.com/518006
Commit-Queue: Jochen Eisinger <jochen@chromium.org>
Reviewed-by: Marja Hölttä <marja@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45560}
For lazy compilation, we encode information about table exports in the
deoptimization data. This information is rebuilt on each instantiation,
so we need to reset it when reusing code objects from another instance.
R=ahaas@chromium.org
BUG=chromium:727219
Change-Id: I90557ef06e692d0a8323223cac26679efcfa408b
Reviewed-on: https://chromium-review.googlesource.com/517945
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45559}
It happens frequently that unrelated compiler errors lead to the
instantiation of {read_leb_tail} with an invalid {byte_index}, which
again triggers compiler errors.
This change fixes this, such that illegal instantiation just triggers a
static_assert and nothing more. This helps to spot the actual error.
Drive-by: Mark {Min} and {Max} constexpr, such that they can produce
constexpr results. Note that the result will only be constexpr if both
arguments are constexpr.
R=ahaas@chromium.org
Change-Id: I35d6865a7f569d72ace0debb045ac615cfd87d92
Reviewed-on: https://chromium-review.googlesource.com/518005
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#45558}
