This is necessary because the spec changed.
R=mstarzinger@chromium.org
Bug: v8:9865
Change-Id: Id8b4d85eafcf368d591666907036e6aa54664e63
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1921794
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65072}
Previously the fast path only asserted the correct instance types; but
when reading lastIndex we additionally rely on a specific object
shape. This is checked by HasInitialRegExpMap().
Bug: chromium:1024758
Change-Id: I0b401ffb246dd47153caf798446d8d41bc84bc8e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924354
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65071}
If a breakpoint is set on a non-breakable position, the wasm interpreter
just stores the value 0xFF (kInternalBreakpoint) in the function body
(actually, a copy of the function body). This might overwrite immediates
and cause subsequent failures in the wasm interpreter.
In JavaScript, breakpoints are just forwarded to the next breakable
position. This CL implements the same for WebAssembly.
A cctest tests this behavior, and the existing
wasm-stepping-byte-offsets.js inspector test is extended to also set the
breakpoint within an i32 constant immediate.
R=leese@chromium.org, mstarzinger@chromium.org
CC=bmeurer@chromium.org
Bug: chromium:1025184
Change-Id: Ia2706f8f1c3d686cbbe8e1e7339d9ee86247bb4a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1925152
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65070}
That's possible because JS builtins are JSFunctions that embed a
NativeContext.
Bug: v8:7793
Change-Id: Id2bf7844fcfb53df733100f1e3e554f25a78482a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1926150
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65068}
In {EmptyBackingStore}, the {free_on_destruct} flag was not set as an
optimization: Since there is no memory, it also does not have to be
freed. However, this flag has a side-effect: any backing store where
this flag is not set is considered {external}. The {external} flag is
mis-used by blink to indicate if ArrayBuffers need to be wrapped or not.
With this CL we set the {free_on_destruct} flag in {EmptyBackingStore},
but we change the ArrayBufferTracker to just ignore empty backing
stores.
R=ulan@chromium.org
Bug: chromium:1008840
Change-Id: I1552a6e013c8b23f39fba1c2d9d9c61dc30c0c74
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924263
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65067}
Bug: v8:9989
Change-Id: I6923f99398c0a1c8b447e18e0416a2630a09ee5d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924259
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Auto-Submit: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65065}
When importing a JS function, Wasm tries to guess the type of function
(parameters & strict/sloppy mode). This can sometimes fail which leads
to re-creation of the wrapper. With this change, the same wrapper can
be used for strict and sloppy mode requiring the re-creation only on
arity mismatch.
R=mstarzinger@chromium.org
Change-Id: I77ec2b853153dec0772873cfb60c064a74065732
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1921793
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Emanuel Ziegler <ecmziegler@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65064}
Don't overwrite WATCHLISTS each time with a checkout from the latest
release branch as that means it will never pick up changes from
master.
No-Try: true
Bug: chromium:832032
Change-Id: I3a9231369caa9a6591acb9b7f0c76dc031ab9178
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1926029
Commit-Queue: Dan Elphick <delphick@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65063}
Bug: v8:9972
Change-Id: Ia85520eea8d3bcadc2573c16bf2778b1c3ff0c5a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1926028
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65061}
"preparser" is a legacy test-suite written in Python. "cctest/test-parsing"
provides the same coverage and more for the preparser.
This CL removes "preparser" stand-alone test-suite
R=verwaest@chromium.org
CC=machenbach@chromium.org
Bug: v8:10001
Change-Id: I1823967e654e8d6d9e42eadfd667f90074d57ba9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1926027
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Auto-Submit: Tamer Tas <tmrts@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65059}
This reverts commit f2a74165bf.
Reason for revert: Clusterfuzz
Bug: chromium:1026479
Original change's description:
> [regexp] Re-execute regexp when '.indices' is accessed.
>
> Instead of storing a pointer to the last_match_info, which may
> change, this cl modifies JSRegExpResult to store a pointer to
> the original JSRegExp which generated it, as well as additional
> data needed to re-execute the match.
>
> Basically a straight copy and tidy off jgruber@'s prototype:
> https://chromium-review.googlesource.com/c/v8/v8/+/1876810
>
> Bug: v8:9548
> Change-Id: I11b7deae681b8287e41e8d0e342291ff484751fb
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1910129
> Commit-Queue: Joshua Litt <joshualitt@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#65053}
TBR=jgruber@chromium.org,joshualitt@chromium.org
Change-Id: I6294e3d7ac0b3e2bd9404697823b8d3cc2545c16
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9548
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1925651
Reviewed-by: Joshua Litt <joshualitt@chromium.org>
Commit-Queue: Joshua Litt <joshualitt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65057}
Minor cleanup: some classes in Torque don't yet have any definitions for
their fields, so it doesn't make sense to emit field layout macros for
those classes.
Bug: v8:7793
Change-Id: Iee38aa3cbe684f4a63329a676e2e94944dc05de1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1925010
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Seth Brenith <seth.brenith@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#65054}
Instead of storing a pointer to the last_match_info, which may
change, this cl modifies JSRegExpResult to store a pointer to
the original JSRegExp which generated it, as well as additional
data needed to re-execute the match.
Basically a straight copy and tidy off jgruber@'s prototype:
https://chromium-review.googlesource.com/c/v8/v8/+/1876810
Bug: v8:9548
Change-Id: I11b7deae681b8287e41e8d0e342291ff484751fb
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1910129
Commit-Queue: Joshua Litt <joshualitt@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65053}
These instructions should always treat inputs as signed, and saturate to
unsigned min/max values.
E.g. given -1, it should saturate to 0.
The spec text,
https://github.com/WebAssembly/simd/blob/master/proposals/simd/SIMD.md#integer-to-integer-narrowing,
has been updated to describe this.
The changes here include codegen changes to ia32, x64, arm, and arm64,
changes to arm simulator, assembler, and disassembler to handle the case
of treating input as signed and narrowing to unsigned. The vqmovn
instruction can handle this case, our assembler wasn't allowing callers
to specify this.
The interpreter and scalar lowering are also fixed with this change.
Bug: v8:9729
Change-Id: I6f72baa825f59037f7754485df6a2964af59fe31
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1879423
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Bill Budge <bbudge@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65051}
Spill/fill now take offsets instead of indices. We provide a
helper, GetStackOffsetFromIndex, for callers. This is currently only
useful while slot sizes are still fixed to 8 bytes.
StackTransferRecipe's RegisterLoad now works in terms of offset.
LiftoffStackSlots work in terms of offset as well.
TransferStackSlot currently still works in terms of indicies, but can be
converted to use offsets in a subsequent change.
Bug: v8:9909
Change-Id: If54fb844309bdfd641720d063135dd59551813e0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1922489
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65049}
This is part 3 of Torquifying DescriptorArray: making it possible to use
the "descriptors" indexed field from code written in Torque. A small
macro EnsureArrayLengthWritable is converted to demonstrate the new
functionality.
This CL also introduces the arrow token `->` and desugars a->b to (*a).b
so that the new builtin looks a little cleaner.
Bug: v8:7793
Change-Id: I84eaa97f664aa67273866760e6ede4346a3ee2f9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1900332
Commit-Queue: Seth Brenith <seth.brenith@microsoft.com>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65046}
This reduction relies on a known object layout of the regexp instance
in order to access the lastIndex field through a statically-determined
offset. Prior to this CL, we checked only for instance types, not for
the map, and thus it was possible to read garbage from either inside
or outside the current object.
Bug: chromium:1024758,v8:7779
Change-Id: I1eec8220797f443bdf3d05804e54f33b21fa2f00
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924353
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65039}
This CL implements torque builtins for BigInt subtraction and extends
the compilation pipeline to lower calls to the generic subtraction
to SpeculativeBigIntSubtract and later to BigIntSubtract with
necessary checks in case of BigInt feedback.
The CL also implements lowering of these operators to native machine
word operations on 64 bit architectures if they are used in a
truncating context (aka BigInt.asUintN).
Bug: v8:9407
Change-Id: Idf5da14c380bc7c12375e7f084a3e1c455303f5f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1895566
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Nico Hartmann <nicohartmann@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65037}
Bytecode flushing bit me again.
Bug: v8:9945, v8:9983
Change-Id: I9e4f9dd5e1793d60b24def447a8374e550fa248a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924352
Auto-Submit: Georg Neis <neis@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65036}
For arm64 decompressing a TaggedSigned results in a 64-bit value, with
the top 32 bits zeroed and the lower 32 bits being the SMI. This patch,
with the --debug-code flag enabled, will corrupt the top 32 bits to try
and catch places that are using 64-bit operations instead of 32-bit operations.
Change-Id: I3d3048c4f4b87f3bce26d0c7eb41789ec6b99f5b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1917099
Reviewed-by: Santiago Aboy Solanes <solanes@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Joey Gouly <joey.gouly@arm.com>
Cr-Commit-Position: refs/heads/master@{#65035}
The optimization behavior in these configurations is strange, I'm
still trying to understand what exactly is going on.
Bug: v8:9945, v8:9983
Change-Id: I52782b9e73decb9f3b2439cddd5e23068faebdf4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924349
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65032}
Former CL 352bbb1https://crrev.com/c/1893192 made an optimization that move non-JS
linkage builtins code objects into RO_SPACE, but caused v8 crash on mips platform,
so temporarily disable the change on mips platform.
Change-Id: Iffba1a35985ccf6688e29e35db20602d69a496e6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1919442
Auto-Submit: Zhao Jiazhong <zhaojiazhong-hf@loongson.cn>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Commit-Queue: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65031}
Drive-by: For more flexibility (e.g. for future IsNull methods), remove
'Constant' from names in the list of constants.
Bug: v8:9972
Change-Id: I66ec64c30cb397641d77cd26b514e8ac52763e16
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1924348
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65030}
In most cases where we bail out to TurboFan, we actually have a fallback
that just calls a C function instead. This fallback did not exist
initially, but was added later for other platforms (see e.g.
https://crrev.com/c/1044187).
By reusing that fallback on intel, we remove another portion of function
that could not be compiled in Liftoff on specific CPUs.
R=ahaas@chromium.org
Bug: v8:9919
Change-Id: I151000c004dc330362337bf7ddeaec68390c3467
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1921986
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65029}
... by disallowing checkpoint elimination across function boundaries.
See the comment in checkpoint-elimination.cc and the tests for details.
Bug: v8:9945
Change-Id: Ibf4ab6f0e4e709e26d3c4428a082ef45dcbeb8b0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1906208
Auto-Submit: Georg Neis <neis@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65027}
Individual frames of a stack frame in the frame cache might point
to the JSFunction of that corresponding stack frame. It is illegal to
serialize JSFunction objects in the isolate snapshot, so the attempt
to serialize the stack frame cache results in a crash. This can happen
when a warmup script is run, before a snapshot is created.
This CL fixes the crash by not utilizing the stack frame cache in case
the serializer is enabled.
Change-Id: I8b79a06b8cff36e1f54b54d3d8e5397b07ba52e7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1923068
Auto-Submit: Simon Zünd <szuend@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65026}
We recently extended function-entry stack checks by an offset
representing the difference in optimized and unoptimized frame sizes,
with the intent of avoiding stack overflows during deopts. Although
the generated code is very efficient (just a single additional
register subtraction, executed exactly once per call), perf impact
is measurable.
To avoid the overhead in most cases, this CL adds a stack slack,
currently set to 256 bytes, by which deopts are allowed to exceed the
real V8 stack limit. For function-entry stack checks with an offset
less than stack slack, the offset is not applied and the more
efficient version of the stack check is emitted.
The V8 limit is chosen to be smaller than OS stack size (assumed to
be at least 1 MB). This guarantee is upheld even with slack.
Bug: chromium:1020989,v8:9534
Change-Id: Idee2e7ad1fa7810bf086a9f72ce00a9717010310
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1910099
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#65025}
