This CL does same changes as
https://chromium-review.googlesource.com/c/540763/, but for async
compilation instead of for parallel compilation. The biggest difference
is that for async compilation I start background tasks again when half
of the memory is free again and not when all the memory is free again.
Original description:
It is possible that the foreground task is unable to clear the
scheduled unfinished work, eventually leading to an OOM.
We use either code_range on 64 bit, or the capacity of the code space,
as a heuristic for how much memory to use for compilation.
The change avoids blocking the background threads while we're over the
memory threshold. This is to avoid starving the GC.
R=mtrofin@chromium.org
Change-Id: I7399e2474f72f6727e6e50176dd7ba95cdcd3238
Reviewed-on: https://chromium-review.googlesource.com/543477
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Mircea Trofin <mtrofin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46120}
This will allow for embedders to easily implement their own Platform
without duplicating the tracing controller code.
BUG=v8:6511
R=fmeawad@chromium.org
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I7c64933d12b2cf53f0636fbc87f6ad5d22019f5c
Reviewed-on: https://chromium-review.googlesource.com/543015
Commit-Queue: Jochen Eisinger <jochen@chromium.org>
Reviewed-by: Fadi Meawad <fmeawad@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46118}
In edge cases such as the following, sloppy-mode block-scoped function
hoisting is expected to occur:
eval(`
with({a: 1}) {
function a() {}
}
`)
In this case, there should be the equivalent of a var declaration
outside of the eval, which gets set to the value of the local function
a when the body of the with is executed.
Previously, the way that var declarations are hoisted out of eval
meant that the assignment to that var was an ordinary DYNAMIC_GLOBAL
assignment. However, such a lookup mode meant that the object in the
with scope received the assignment!
This patch fixes that error by marking the assignments produced by
the sloppy mode block scoped function hoisting desugaring so as to
generate a different runtime call which skips with scopes.
Bug: chromium:720247, v8:5135
Change-Id: Ie36322ddc9ca848bf680163e8c016f50d4597748
Reviewed-on: https://chromium-review.googlesource.com/529230
Commit-Queue: Daniel Ehrenberg <littledan@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46116}
Collect code coverage by compiling for one or more target architectures
and then running tests, in the same directory. This way, gcov aggregates
results.
Bug:
Change-Id: I3bf05416c535c0c566e48d4e73adc4eb49ba2793
Reviewed-on: https://chromium-review.googlesource.com/527522
Commit-Queue: Mircea Trofin <mtrofin@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46110}
Port 24b7026d73
Original Commit Message:
For interpreted functions, use the optimized code slot in the feedback
vector to store an optimization marker (optimize/in optimization queue)
rather than changing the JSFunction's code object. Then, adapt the
self-healing mechanism to also dispatch based on this optimization
marker. Similarly, replace SFI marking with optimization marker checks
in CompileLazy.
This allows JSFunctions to share optimization information (replacing
shared function marking) without leaking this information across native
contexts. Non I+TF functions (asm.js or --no-turbo) use a
CheckOptimizationMarker shim which generalises the old
CompileOptimized/InOptimizationQueue builtins and also checks the same
optimization marker as CompileLazy and InterpreterEntryTrampoline.
This is a reland of https://chromium-review.googlesource.com/c/509716R=leszeks@chromium.org, joransiu@ca.ibm.com, bjaideep@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=
LOG=N
Review-Url: https://codereview.chromium.org/2947903002
Cr-Commit-Position: refs/heads/master@{#46109}
This was left over from the previous CL to change S128LoadMem/S128StoreMem to
use prefixed opcodes. Decoding prefixed opcodes already checks for the
prototype flag.
BUG=V8:6020
R=bbudge@chromium.org
Review-Url: https://codereview.chromium.org/2946303002
Cr-Commit-Position: refs/heads/master@{#46108}
This will make it easier if we want to split it into two intrinsics, one
for creating an object with `done == true` and one with `done == false`.
Also remove apparently-dead method FullCodegen::EmitCreateIteratorResult.
Bug: v8:6408, v8:6409
Change-Id: I3d6022a9eff517dd8b664d65950502c22447b364
Reviewed-on: https://chromium-review.googlesource.com/543567
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46107}
(Reland: NeedsManualRebaseline'd newly-fixed layout test in Chromium.)
This was never legal; the spec only allows '\0' in strict-mode strings
or templates when not followed by a decimal digit. Previously we were
only enforcing that it not be followed by an _octal_ digit.
This was already fixed for numeric literals, but not for escape
sequences in strings.
BUG=v8:6504
Review-Url: https://codereview.chromium.org/2948903002
Cr-Commit-Position: refs/heads/master@{#46106}
This method returns position of importing stmt in module source.
R=neis@chromium.org
Bug: chromium:721589
Change-Id: I8639796a001fdfec7cf5aa1bf1a27493f7a757a9
Reviewed-on: https://chromium-review.googlesource.com/541322
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Commit-Queue: Aleksey Kozyatinskiy <kozyatinskiy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46105}
UpdateMaxNumberKey calls are moved to clients, who do have the
dictionary-holder. ::Add should basically always UpdateMaxNumberKey. I'm
reducing the number of entry points before looking into how to guarantee this.
Bug:
Change-Id: Iefe8a7fdf7c1e0a6d731bfd948d22849714498a9
Reviewed-on: https://chromium-review.googlesource.com/542895
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46104}
Allows BitVector to resize, updating its own data and data length to
match the new length. We can fast-path resizes which fit into the same
data length (since high bits are already zero), and replace the pattern
where a BitVector is cloned using CopyFrom.
Change-Id: If79ca782c516e93b2a27c5e335e263554d522e88
Reviewed-on: https://chromium-review.googlesource.com/539522
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46101}
Currently, the number of ticks to wait before optimizing is a constant (if
sufficient feedback is available). This cl changes it so that, larger
functions would have to wait longer for optimizing. The number of ticks
required scales linearly with the function size.
Bug:
Change-Id: Id27bea715cf15960667cf63381b1cbe8dac94428
Reviewed-on: https://chromium-review.googlesource.com/538614
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46097}
I can't reproduce any issues with an lsan build, so we will remove
this for now and keep an eye out.
Bug: v8:6315
Change-Id: Iad2a1b23f3614ec9a09a83bb01e235969c3f9fcc
Reviewed-on: https://chromium-review.googlesource.com/542835
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46096}
This CL improves reported source range precision in a couple of ways:
Source ranges are now standardized to consist of an inclusive start
index and an exclusive end index (similar to what's reported for
functions). For example:
0123456789 // Offset.
{ f(); } // Block represented as range {0,8}.
Duplicate singleton ranges (i.e. same start and end offsets) are now
merged (this only becomes relevant once jump statement coverage is
added). For example:
for (.) break; // Break- and loop continuation have same positions.
SourceRangeScope incorrectly collected starting position
(unconditionally) and end position (when no semi-colon was present).
01234567890123 // Offset.
for (.) break // Loop body range is {8,13}, was {6,9}.
Bug: v8:6000
Change-Id: I62e7c70cc894a20f318330a2fbbcedc47da2b5db
Reviewed-on: https://chromium-review.googlesource.com/541358
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Marja Hölttä <marja@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46095}
Adds typed lowering of JSStringConcat to ConsString allocation if the
following conditions hold:
- All concatinations will result in a ConsString of >= ConString::kMinLength
- No concatinations will result in a empty string in the RHS unless there is
a sequential string in the LHS.
This also means JSStringConcat needs an eager checkpoint since it can
deopt if throwing a RangeError when the string length protector is valid.
BUG=v8:6243
Change-Id: I01ca79f884df467c10f2c032c72d51b5199c1a3c
Reviewed-on: https://chromium-review.googlesource.com/526636
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46093}
This switches all uses of the patching {ToBooleanICStub} over to the
existing and non-patching {ToBoolean} CSA-builtin, and removes some
supporting code.
R=verwaest@chromium.org
BUG=v8:6408
Change-Id: Iab60c95e6b54e426408390e056b679f6227e7ce0
Reviewed-on: https://chromium-review.googlesource.com/539576
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46089}
Add a new JSConstructWithArrayLike operator that is backed by the
ConstructWithArrayLike builtin (similar to what was done before
for the JSCallWithArrayLike operator), and use that operator to
optimize Reflect.construct inlining in TurboFan. This is handled
uniformly with JSConstructWithSpread in the JSCallReducer.
Also add missing test coverage for Reflect.construct in optimized
code, especially for some interesting corner cases.
R=petermarshall@chromium.org
BUG=v8:4587,v8:5269
Review-Url: https://codereview.chromium.org/2949813002
Cr-Commit-Position: refs/heads/master@{#46087}
This addresses a TODO about the correct location of the helper function
in question, it is now internal to TurboFan instead of being shared.
R=jarin@chromium.org
Change-Id: I7e6112e9bc9759255a416fa2e2a9f92a8e4248c8
Reviewed-on: https://chromium-review.googlesource.com/542840
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46086}
- Iterator advancing is kept mainly unchanged.
- The iterator stores the size of the object which is to be used by the
caller in follow ups. This way we might be able to avoid further out
of line loads.
- The iteartor follows the regular std conventions allowing range based
loops.
Bug: chromium:651354
Change-Id: I8928224a62d3a48a48145a2d00279a28608bc634
Reviewed-on: https://chromium-review.googlesource.com/543335
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46085}
Inlines some functions to improve reduce the stack requirements for
chains of binary operations in the bytecode generator, thereby
enabling support of deeper expression stacks.
BUG=chromium:731861
Change-Id: I5ca437d507e9b2a7eb74f33deaa708ecd646077b
Reviewed-on: https://chromium-review.googlesource.com/541356
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46080}
The fuzzer has already been removed from chromium. In addition I removed
code which was only used by this fuzzer.
BUG=chromium:734550
R=clemensh@chromium.orgCC=mstarzinger@chromium.org
Change-Id: I2ff4614e4d64131412ead759318e5c38e38f5d3d
Reviewed-on: https://chromium-review.googlesource.com/542816
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46078}
- Now that there are no boolean vector types, we can directly test the
results of relational ops.
Bug: v8:6020
Change-Id: Id2139133ae3a548a9985a26a3427cbeddc6272a6
Reviewed-on: https://chromium-review.googlesource.com/536176
Reviewed-by: Aseem Garg <aseemgarg@chromium.org>
Commit-Queue: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46075}
