Rather than pushing zero for the initial bytecode offset, we should push the
offset of the first bytecode handler, Smi tagged. This fixes the line number
for the top stack frame on overflow errors.
BUG=v8:4981
LOG=N
Review-Url: https://codereview.chromium.org/1950913004
Cr-Commit-Position: refs/heads/master@{#36137}
Trying to re-land http://crrev.com/1948453002 after fixing assembler-x64.cc in http://crrev.com/1962563003.
Before this patch, we would emit a cmp or test with a memory operand only if both of the operands in the IR were loads. Now if either of them is a load and the other one is an immediate, we can use a memory operand if the load representation machine size is wide enough to represent the latter.
Review-Url: https://codereview.chromium.org/1968453002
Cr-Commit-Position: refs/heads/master@{#36136}
Reason for revert:
Causes regressions.
Original issue's description:
> [build] Use -O2 on linux by default
>
> This switches O2 -> O3 to match chromium's configuration.
>
> This makes the difference between gn and gyp a bit smaller
> before switching.
>
> BUG=chromium:474921
> LOG=n
>
> Committed: https://crrev.com/a37ab71e343b8cc24df05868b941a49e0c271376
> Cr-Commit-Position: refs/heads/master@{#36124}
TBR=jochen@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:474921
Review-Url: https://codereview.chromium.org/1958253004
Cr-Commit-Position: refs/heads/master@{#36135}
In the bytecode graphbuilder, translate the two generator-specific
bytecodes as a couple of runtime calls for now.
BUG=v8:4907
LOG=n
Review-Url: https://codereview.chromium.org/1957393004
Cr-Commit-Position: refs/heads/master@{#36134}
The new allocation folding implementation avoids fragmentation between folded allocation. As a consequence, our heap will always be iterable i.e. we do not have to perform a garbage collection before iterating the heap.
BUG=chromium:580959
LOG=n
Review-Url: https://codereview.chromium.org/1899813003
Cr-Commit-Position: refs/heads/master@{#36133}
This adds a new pass MemoryOptimizer that walks over the effect chain
from Start and lowers all Allocate, LoadField, StoreField, LoadElement,
and StoreElement nodes, trying to fold allocations into allocation
groups and eliminate write barriers on StoreField and StoreElement if
possible (i.e. if the object belongs to the current allocation group and
that group allocates in new space).
R=hpayer@chromium.org, jarin@chromium.org
BUG=v8:4931, chromium:580959
LOG=n
Review-Url: https://codereview.chromium.org/1963583004
Cr-Commit-Position: refs/heads/master@{#36128}
This implements declaration of lookup slots for variables and functions
within optimized code. Such a declaration only appears with top-level
eval code, which we only recently started handling in TurboFan.
R=jarin@chromium.org
Review-Url: https://codereview.chromium.org/1962723002
Cr-Commit-Position: refs/heads/master@{#36125}
This switches O2 -> O3 to match chromium's configuration.
This makes the difference between gn and gyp a bit smaller
before switching.
BUG=chromium:474921
LOG=n
Review-Url: https://codereview.chromium.org/1963693004
Cr-Commit-Position: refs/heads/master@{#36124}
Treat --trace-turbo and --trace-turbo-graph similar to --trace-hydrogen
and --trace-hydrogen-stubs, in that we disable concurrent recompilation
in that case (otherwise we cannot print heap objects and stuff).
R=ishell@chromium.org
BUG=v8:4969
LOG=n
Review-Url: https://codereview.chromium.org/1965653003
Cr-Commit-Position: refs/heads/master@{#36123}
- Inline reads to the species protector
- Put V8_LIKELY/V8_UNLIKELY to guide Array.prototype.{slice,splice,concat}
to the fast paths
- Put the Array species good path checks directly in
Array.prototype.concat to avoid a couple reads of the array constructor
from the native context
These changes together bring a ~4-6% win on
kraken-orig/stanford-crypto-pbkdf2-orig which more than makes up from
the performance degradation from a recent other species-related change.
BUG=chromium:609739
R=cbruni@chromium.org
LOG=Y
Review-Url: https://codereview.chromium.org/1958713003
Cr-Commit-Position: refs/heads/master@{#36121}
Rolling v8/build to 143dcc2b1b07c16858f16f25fefe04311f663279
Rolling v8/tools/clang to 18b63c680a59a7125514b1e05ca42cdfb89a19c7
TBR=machenbach@chromium.org,vogelheim@chromium.org,hablich@chromium.org
Review-Url: https://codereview.chromium.org/1959293002
Cr-Commit-Position: refs/heads/master@{#36120}
Changes some platform specific class names and uses typedefs
to define FloatRegister, DoubleRegister, etc.
This will be needed to do register allocation on ARM/MIPS where
registers combine to form larger registers.
LOG=N
BUG=v8:4124
Review-Url: https://codereview.chromium.org/1954953002
Cr-Commit-Position: refs/heads/master@{#36117}
This patch installs %TypedArray% and its prototype on the native
context, and wires them up to each TypedArray subclass. This is later
used to check the holder of length, byteLength and byteOffset is
%Typedarray% and apply the appropriate optimizations.
BUG=chromium:593634
LOG=Y
Review-Url: https://codereview.chromium.org/1949863002
Cr-Commit-Position: refs/heads/master@{#36116}
Fixes 4 compile errors on Mac due to specifiers not matching type casts.
LOG=N
BUG=
Review-Url: https://codereview.chromium.org/1960113002
Cr-Commit-Position: refs/heads/master@{#36115}
atomic-utils.h only uses functionality from base/, and also by moving it into
base/, code outside of v8_base can benefit from it.
BUG=v8:4991
LOG=n
Review-Url: https://codereview.chromium.org/1954603002
Cr-Commit-Position: refs/heads/master@{#36114}
Assembler::testw(Register, Immediate) and Assembler::testw(const Operand&, Immediate) were emitting only the first 8 bits of a 16-bit immediate, causing unexpected crashes.
This went unnoticed because before http://crrev.com/1948453002 no compiler was using them.
Review-Url: https://codereview.chromium.org/1962563003
Cr-Commit-Position: refs/heads/master@{#36110}
The bailout points for named and keyed property loads when doing super
property calls are not being prepared by full-codegen, even though we
are using them in TurboFan for deopts and stack traces.
R=jarin@chromium.org
TEST=mjsunit/regress/regress-4971
BUG=v8:4971
LOG=n
Review-Url: https://codereview.chromium.org/1960083002
Cr-Commit-Position: refs/heads/master@{#36109}
This catches malformed code like the following example:
<expr>
<block begin>
<expr>
<binop>
<end>
Which is illegal because the inputs to the binop cross the block boundary.
R=rossberg@chromium.org,ahaas@chromium.org
BUG=
Review-Url: https://codereview.chromium.org/1961853002
Cr-Commit-Position: refs/heads/master@{#36103}
The flag in question used to fall-back to Crankshaft whenever an OSR
request couldn't be handled by TurboFan. By now OSR in TurboFan is
sufficiently stabilized that one single --use-osr flag should do it.
R=titzer@chromium.org
Review-Url: https://codereview.chromium.org/1960043002
Cr-Commit-Position: refs/heads/master@{#36102}
mjsunit/es6/mirror-collections fails with ignition because dead registers
may hold references to objects. This prevents GC from collecting
otherwise dead objects. Dead registers are not cleared because the cost
of clearing them outweighs its benefits. Hence, modifying this test to
work around this problem.
BUG=v8:4280,v8:4853
LOG=N
Review-Url: https://codereview.chromium.org/1945223002
Cr-Commit-Position: refs/heads/master@{#36098}
Handles of this type are automatically reset by the garbage collector
when their objects are not longer reachable.
The motivation is to reduce pause time of external.weak_global_handles
phase of the garbage collector by not maintaing the list of pending
callbacks and not calling the callbacks.
Local testing on discourse page of the v8.inifinite_scroll benchmark
shows 7x improvement for this GC phase.
Before:
external.weak_global_handles
len: 21
min: 0.0
max: 4.5
avg: 0.757142857143
After:
external.weak_global_handles
len: 21
min: 0.0
max: 0.5
avg: 0.109523809524
A follow-up patch will enable the new phantom handles in Chromium.
BUG=chromium:608333
LOG=NO
Review-Url: https://codereview.chromium.org/1950963002
Cr-Commit-Position: refs/heads/master@{#36095}
port 3b7ff999f3 (r36066)
original commit message:
Reason for revert:
Breaks WASM; please also add tests when relanding.
Original issue's description:
> [turbofan] Take the immediate size in account when narrowing ia32/x64 word comparison operators.
>
> Before this patch, we would emit a cmp or test with a memory operand only if both of the operands in the IR were loads. Now if either of them is a load and the other one is an immediate, we can use a
>
> Committed: https://crrev.com/a0543313dbd46b0c2e72c91ee3488a7dc6db73e4
> Cr-Commit-Position: refs/heads/master@{#36009}
BUG=
Review-Url: https://codereview.chromium.org/1958863003
Cr-Commit-Position: refs/heads/master@{#36094}
Rolling v8/build to e3d87b142123f2da73c94c276ee915c099afe909
Rolling v8/tools/clang to 15dd77e3ea10e43596ec6ac07b73431135915b30
TBR=machenbach@chromium.org,vogelheim@chromium.org,hablich@chromium.org
Review-Url: https://codereview.chromium.org/1961713002
Cr-Commit-Position: refs/heads/master@{#36093}
