This CL optimizes the code in BytecodeArrayBuilder and
BytecodeArrayWriter by making the following main changes:
- Move operand scale calculation out of BytecodeArrayWriter to the
BytecodeNode constructor, where the decision on which operands are
scalable can generally be statically decided by the compiler.
- Move the maximum register calculation out of BytecodeArrayWriter
and into BytecodeRegisterOptimizer (which is the only place outside
BytecodeGenerator which updates which registers are used). This
avoids the BytecodeArrayWriter needing to know the operand types
of a node as it writes it.
- Modify EmitBytecodes to use individual push_backs rather than
building a buffer and calling insert, since this turns out to be faster.
- Initialize BytecodeArrayWriter's bytecode vector by reserving 512
bytes,
- Make common functions in Bytecodes constexpr so that they
can be statically calculated by the compiler.
- Move common functions and constructors in Bytecodes and
BytecodeNode to the header so that they can be inlined.
- Change large static switch statements in Bytecodes to const array
lookups, and move to the header to allow inlining.
I also took the opportunity to remove a number of unused helper
functions, and rework some others for consistency.
This reduces the percentage of time spent in making BytecodeArrays
in CodeLoad from ~15% to ~11% according to perf. The
CoadLoad score increase by around 2%.
BUG=v8:4280
Committed: https://crrev.com/b11a8b4d41bf09d6b3d6cf214fe3fb61faf01a64
Review-Url: https://codereview.chromium.org/2351763002
Cr-Original-Commit-Position: refs/heads/master@{#39599}
Cr-Commit-Position: refs/heads/master@{#39637}
WasmModule::Instantiate can return an empty handle if it is not possible
to instantiate the module. With this change the wasm-module-runner does
not assume anymore that WasmModule::Instantiate produces a valid handle.
BUG=chromium:648078
R=titzer@chromium.org
Review-Url: https://codereview.chromium.org/2365493003
Cr-Commit-Position: refs/heads/master@{#39629}
We can't assume that there's no entered context during microtask
execution. We also can't assume that we'll never enter an context during
microtask execution either.
We can however assume that we don't execute microtasks recursively, so
remembering the length of the entered context stack is enough to decide
which context to use.
BUG=chromium:635769
R=verwaest@chromium.org
Review-Url: https://codereview.chromium.org/2350933007
Cr-Commit-Position: refs/heads/master@{#39627}
This reverts commit b42ecda533.
That commit introduced a allocator field to hashmap, which indirectly
added a field to Scopes, where the field, effectively storing a Zone, is
unnecessary because the Zone can be accessed in other ways.
Review-Url: https://codereview.chromium.org/2351393003
Cr-Commit-Position: refs/heads/master@{#39623}
Generate TBZ/TBNZ for certain comparisons against zero. E.g. instead of:
cmp w0, 0x0
b.lt/ge <addr>
we can generate:
tbnz/tbz w0, 31, <addr>
BUG=
Review-Url: https://codereview.chromium.org/2359723004
Cr-Commit-Position: refs/heads/master@{#39620}
If a JSCallFunction node doesn't have any callee information, either
from feedback taken on input nodes, i.e. on property loads, or from
the CallIC, we insert a soft deoptimization exit instead.
R=jarin@chromium.org
BUG=v8:5267
Review-Url: https://codereview.chromium.org/2361773002
Cr-Commit-Position: refs/heads/master@{#39619}
Port for VisitFloat32Add, VisitFloat64Add, VisitFloat32Sub and
VisitFloat64Sub in InstructionSelector.
TEST=unittests/InstructionSelectorTest.Float32AddWithFloat32Mul,
unittests/InstructionSelectorTest.Float64AddWithFloat64Mul,
unittests/InstructionSelectorTest.Float32SubWithFloat32Mul,
unittests/InstructionSelectorTest.Float64SubWithFloat64Mul
BUG=
Review-Url: https://codereview.chromium.org/2341303002
Cr-Commit-Position: refs/heads/master@{#39616}
This makes sure cycles in the object states graph are detected early by
escape analysis instead of late in the scheduler. This is mainly done
for improved debuggability.
R=bmeurer@chromium.org
BUG=chromium:613923
Review-Url: https://codereview.chromium.org/2354263002
Cr-Commit-Position: refs/heads/master@{#39614}
Reason for revert:
Prime suspect for roll blocker: https://codereview.chromium.org/2362503002/
Original issue's description:
> [Interpreter] Optimize BytecodeArrayBuilder and BytecodeArrayWriter.
>
> This CL optimizes the code in BytecodeArrayBuilder and
> BytecodeArrayWriter by making the following main changes:
>
> - Move operand scale calculation out of BytecodeArrayWriter to the
> BytecodeNode constructor, where the decision on which operands are
> scalable can generally be statically decided by the compiler.
> - Move the maximum register calculation out of BytecodeArrayWriter
> and into BytecodeRegisterOptimizer (which is the only place outside
> BytecodeGenerator which updates which registers are used). This
> avoids the BytecodeArrayWriter needing to know the operand types
> of a node as it writes it.
> - Modify EmitBytecodes to use individual push_backs rather than
> building a buffer and calling insert, since this turns out to be faster.
> - Initialize BytecodeArrayWriter's bytecode vector by reserving 512
> bytes,
> - Make common functions in Bytecodes constexpr so that they
> can be statically calculated by the compiler.
> - Move common functions and constructors in Bytecodes and
> BytecodeNode to the header so that they can be inlined.
> - Change large static switch statements in Bytecodes to const array
> lookups, and move to the header to allow inlining.
>
> I also took the opportunity to remove a number of unused helper
> functions, and rework some others for consistency.
>
> This reduces the percentage of time spent in making BytecodeArrays
> in CodeLoad from ~15% to ~11% according to perf. The
> CoadLoad score increase by around 2%.
>
> BUG=v8:4280
>
> Committed: https://crrev.com/b11a8b4d41bf09d6b3d6cf214fe3fb61faf01a64
> Cr-Commit-Position: refs/heads/master@{#39599}
TBR=mythria@chromium.org,leszeks@chromium.org,rmcilroy@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:4280
Review-Url: https://codereview.chromium.org/2360193003
Cr-Commit-Position: refs/heads/master@{#39612}
Rolling v8/base/trace_event/common to 6232c13e4edb36c84c61653fdae5a4afb5af9745
Rolling v8/build to 8438b8f51c0b125b298ad497c1bc1aa016c908ae
Rolling v8/buildtools to 57649e5e2001ba1f5e5d45f5a838c616ea0e9cb9
Rolling v8/tools/clang to c1f39cb405e8af68a20e08e9485275a42260a88f
TBR=machenbach@chromium.org,vogelheim@chromium.org,hablich@chromium.org
Review-Url: https://codereview.chromium.org/2353363007
Cr-Commit-Position: refs/heads/master@{#39611}
This CL optimizes the code in BytecodeArrayBuilder and
BytecodeArrayWriter by making the following main changes:
- Move operand scale calculation out of BytecodeArrayWriter to the
BytecodeNode constructor, where the decision on which operands are
scalable can generally be statically decided by the compiler.
- Move the maximum register calculation out of BytecodeArrayWriter
and into BytecodeRegisterOptimizer (which is the only place outside
BytecodeGenerator which updates which registers are used). This
avoids the BytecodeArrayWriter needing to know the operand types
of a node as it writes it.
- Modify EmitBytecodes to use individual push_backs rather than
building a buffer and calling insert, since this turns out to be faster.
- Initialize BytecodeArrayWriter's bytecode vector by reserving 512
bytes,
- Make common functions in Bytecodes constexpr so that they
can be statically calculated by the compiler.
- Move common functions and constructors in Bytecodes and
BytecodeNode to the header so that they can be inlined.
- Change large static switch statements in Bytecodes to const array
lookups, and move to the header to allow inlining.
I also took the opportunity to remove a number of unused helper
functions, and rework some others for consistency.
This reduces the percentage of time spent in making BytecodeArrays
in CodeLoad from ~15% to ~11% according to perf. The
CoadLoad score increase by around 2%.
BUG=v8:4280
Review-Url: https://codereview.chromium.org/2351763002
Cr-Commit-Position: refs/heads/master@{#39599}
This add support for ConsString objects allocated inline to the escape
analysis pass. The raw hash field in such strings needs special handling
similar to existing raw fields. This also contains materialization code
within the deoptimizer as usual.
R=bmeurer@chromium.org
TEST=mjsunit/regress/regress-crbug-648737
BUG=chromium:648737
Review-Url: https://codereview.chromium.org/2357153002
Cr-Commit-Position: refs/heads/master@{#39594}
