port 9e7af9efc5 (r29280).
original commit message:
It's useful for the megamorphic keyed store case to not require a
vector and slot as input. Analogous to the load case, we have a dummy
one-ic-slot vector to aid. Since the only kind of MISS is for
megamorphic cache stub failures, we don't need the real vector.
The reason is that megamorphic cache stub failures don't result in any
change to the type feedback vector state.
BUG=
Review URL: https://codereview.chromium.org/1212653011
Cr-Commit-Position: refs/heads/master@{#29410}
port c1a4f7477f (r29281).
original commit message:
The idea is that TurboFan can use this information for more intelligent
inlining.
BUG=
Review URL: https://codereview.chromium.org/1213723005
Cr-Commit-Position: refs/heads/master@{#29409}
port a58ba8d801 (r29279).
original commit message:
This introduces some initial building blocks for calling out to
C/C++ functions directly from TurboFan generated code objects.
BUG=
Review URL: https://codereview.chromium.org/1214903007
Cr-Commit-Position: refs/heads/master@{#29408}
This CL fixes the following situations:
1, Operands are -0.0 or 0.0. it always return right value.
1, Operands contains NaN. it always return right value.
BUG=
Review URL: https://codereview.chromium.org/1214893004
Cr-Commit-Position: refs/heads/master@{#29407}
port c019d7f498 (r29248).
original commit message:
Use big-boy Types to annotate interface descriptor parameters
- Thread Type::FunctionType through stubs and the TF pipeline.
- Augment Typer to decorate parameter nodes with types from
a Type::FunctionType associated with interface descriptors.
- Factor interface descriptors into platform-specific and
platform-independent components so that all descriptors share
a common Type::FunctionType for all platforms.
BUG=
Review URL: https://codereview.chromium.org/1210463002
Cr-Commit-Position: refs/heads/master@{#29406}
Up until now the context register was listed explicitly in each stub's
CallInterfaceDescriptor. This was problematic, because it was listed
first in the list of register parameters--which is fine for Crankshaft,
which is more or less built to handle the context as the first
parameter-- but not ideal for TurboFan, which adds the context at
the end of all function parameters. Now the context register is no
longer in the register list and can be handled appropriately by both
compilers. Specifically, this allows the FunctionType specified for
each CallInterfaceDescriptor to exactly match the parameter register
list.
Review URL: https://codereview.chromium.org/1211333003
Cr-Commit-Position: refs/heads/master@{#29402}
This operator< overload is rather dangerous for handles. And we don't
actually need a std::set in the chunk builder, since duplicates are
automatically eliminated later when the code dependencies are committed.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1214573004
Cr-Commit-Position: refs/heads/master@{#29401}
We now consistently ignore native and extension scripts for debugging. Only native scripts and extension scripts can disallow lazy compilation. So we can make sure that break points are only set to code that are compiled with debug break slots.
R=mvstanton@chromium.org
Review URL: https://codereview.chromium.org/1216193002
Cr-Commit-Position: refs/heads/master@{#29400}
port 8196c28a94 (r29238).
original commit message:
JSConstructStub for subclass constructors instead locates new.target in
a known location on the stack.
BUG=
Review URL: https://codereview.chromium.org/1207613003
Cr-Commit-Position: refs/heads/master@{#29398}
port 882055ff6a (r29082).
The original CL covers part of X87 port. This CL addes the missing changes.
original commit message:
- fix truthfulness of comments
- use InitializeFieldsWithFiller more consistently
- use unsigned comparisons for pointers
No change in functionality intended.
Bonus: improve JavaScriptFrame::Print() for an enhanced debugging experience:
BUG=
Review URL: https://codereview.chromium.org/1210153007
Cr-Commit-Position: refs/heads/master@{#29397}
The enumerable expression in a for-in/of loop is supposed to have a TDZ for any
lexically bound names in that loop (there can be more than one with destructuring).
This patch accomplishes this with an almost-correct desugaring. The only thing missing
is proper debugger support (the let declarations added by the desugaring, while invisible
to code due to shadowing, are visible to the debugger).
BUG=v8:4210
LOG=n
Review URL: https://codereview.chromium.org/1218543003
Cr-Commit-Position: refs/heads/master@{#29396}
v8::Internal::List will DCHECK when indexing out of the array, even if just to
get the address, and the value is never used. So this construct will fail:
memcpy(p, &data[0], length);
When data is empty and length is 0.
BUG=chromium:505778
R=mstarzinger@chromium.org
LOG=n
Review URL: https://codereview.chromium.org/1216853003
Cr-Commit-Position: refs/heads/master@{#29388}
This makes sure that both, the shared function info and it's optimized
code map get revisited when the code map is evicted from the flusher.
R=hpayer@chromium.org
TEST=mjsunit/debug-set-variable-value
Review URL: https://codereview.chromium.org/14703009
Cr-Commit-Position: refs/heads/master@{#29387}
Revert "Revert relanded strong property access CL"
Regression issues should be solved. Initial patchset is the original, subsequent patchsets are the fixing modifications.
This reverts commit 4ac7be5656.
BUG=v8:3956
LOG=N
Review URL: https://codereview.chromium.org/1199983002
Cr-Commit-Position: refs/heads/master@{#29384}
Also fixes a crankshaft bug with strong implicit conversions.
It turns out that the implicit conversion of oddball values
is smushed into so many places in crankshaft that it would
have been pretty invasive surgery to make everything fall
out naturally.
BUG=v8:3956
LOG=N
Review URL: https://codereview.chromium.org/1216463003
Cr-Commit-Position: refs/heads/master@{#29381}
Move the arithmetic shift from Int32MulHigh to a following Int32Add on ARM64.
This graph is commonly generated on reduction of signed integer division.
Review URL: https://codereview.chromium.org/1209413008
Cr-Commit-Position: refs/heads/master@{#29380}
This change makes possible to save and restore the FP registers
in the Prologue and Return parts for the CallAddress kind functions.
TEST=test-simplified-lowering/RunNumberDivide_2_TruncatingToUint32,
test-simplified-lowering/RunNumberMultiply_TruncatingToUint32
Review URL: https://codereview.chromium.org/1191513003
Cr-Commit-Position: refs/heads/master@{#29378}
The deoptimizer (and probably various other places) cannot deal properly
with recursive function inlining, so we disallow it in TurboFan as well.
We might want to reconsider that decision at some point in the future.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1211243007
Cr-Commit-Position: refs/heads/master@{#29374}
This has become necessary as we have more experimental natives
that run after deserializing from the snapshot.
R=hpayer@chromium.org
BUG=v8:4200
LOG=N
Review URL: https://codereview.chromium.org/1218973002
Cr-Commit-Position: refs/heads/master@{#29367}
Ideally inliner itself should not deal with context specialization at
all, since this is all handled in the pipeline instead (actually
inlining already runs together with context specialization), and the
inlining logic should not care about the specialization mode.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1217973003
Cr-Commit-Position: refs/heads/master@{#29366}
This optimization never triggers currently, and is inherently native
context dependent for no real reason (for example it will not properly
detect those constructors in the case of cross native context inlining),
plus it is slow and awkward. In case we really need this functionality
at some point, we should find a way to make it work with the builtin
function id mechanism that is already in place to match other builtins.
R=jarin@chromium.org,rossberg@chromium.org
Review URL: https://codereview.chromium.org/1221683006
Cr-Commit-Position: refs/heads/master@{#29365}
This makes sure that the exit control flow that merges the try-block
with the catch-block after a try-catch-statement creates a new merge
node in cases where it has to. Otherwise dangling phi nodes might have
the wrong number of value inputs.
R=bmeurer@chromium.org
TEST=mjsunit/regress/regress-crbug-505354
BUG=chromium:505354
LOG=N
Review URL: https://codereview.chromium.org/1213183003
Cr-Commit-Position: refs/heads/master@{#29362}
