Fix two bugs with the runtime-profiler optimization heuristics for
interpreted code:
- Reset shared->tick_count for interpreted functions when optimizing
- Update ticks after checking whether to optimize functions, to be the
same as the FCG profiler checks (where updates are done to the code
ticks after deciding whether to optimize).
BUG=chromium:662071
Review-Url: https://codereview.chromium.org/2497933002
Cr-Commit-Position: refs/heads/master@{#40978}
Adds an IsInterpreted() function to both SharedFunctionInfo and JSFunction.
This is used to fix the test-heap code-aging tests since Ignition doesn't
age code.
BUG=v8:4680
Review-Url: https://codereview.chromium.org/2481433002
Cr-Commit-Position: refs/heads/master@{#40868}
It looks like waiting for 4 ticks before optimizing from interpreted
code is hurting performance in sunspider after turning on Ignition
for all TurboFan code. Set it back to 2 ticks.
BUG=chromium:661556
Review-Url: https://codereview.chromium.org/2488703002
Cr-Commit-Position: refs/heads/master@{#40845}
When checking for marking a function for optimization, we had a check if
the function is already optimized to return early. This works in non-OSR
cases. For Turbofan OSR even when the current execution of the function
has already been optimized, the function itself will not be replaced
with optimized code. Hence, we may end up checking a function that is
already marked for optimization again. A check for the frame being optimized
avoids these checks.
BUG=
Review-Url: https://codereview.chromium.org/2450233002
Cr-Commit-Position: refs/heads/master@{#40760}
All vector ICs use the TypeFeedbackVector::ComputeCounts method now,
while the remaining patching ICs still use the traditional way of
counting on the TypeFeedbackInfo hanging off the fullcodegen code
object. This fixes the problem that counts were sometimes off.
Drive-by-fix: Move FullCodeGenerator::CallIC to fullcodegen.cc.
R=yangguo@chromium.org
Review-Url: https://codereview.chromium.org/2472653002
Cr-Commit-Position: refs/heads/master@{#40690}
Turbofan requires a different tuning when compared to crankshaft. Crankshaft
typically has faster compilation times when compared to turbofan. Hence,
added a new parameter, so that crankshaft and turbofan can be tuned
independently.
OSRing too soon is not good for performance, especially for sunspider
benchmarks. Since they are really small functions and optimizing them is
more expensive than just executing unoptimized code. Tuning the code size
threshold of the functions that can be OSRed from ignition.
BUG=v8:4280,chromium:659111
Review-Url: https://codereview.chromium.org/2445203003
Cr-Commit-Position: refs/heads/master@{#40598}
For fullcodegen the RuntimeProfiler has a shortcut that allows it to
tier up small functions earlier, when enough type feedback is available.
Port the same optimization for the Ignition+TurboFan pipeline.
R=mstarzinger@chromium.org
Review-Url: https://chromiumcodereview.appspot.com/2427283004
Cr-Commit-Position: refs/heads/master@{#40435}
Previously we would not have a total count of ICs when interpreting and
thus the check for sufficient type info would always succeed.
Also use the optimization checks for OSR while waiting for baseline
compilation and refactor the check.
BUG=v8:4280
BUG=chromium:634884
Review-Url: https://codereview.chromium.org/2360913003
Cr-Commit-Position: refs/heads/master@{#39677}
Full code uses patching ICs for this feedback, and the interpreter uses
the type feedback vector. It's a good idea to code the vector slots
appropriately as ICs so that the runtime profiler can better gauge if
the function is ready for tiering up from Ignition to TurboFan.
As is, the feedback is stored in "general" slots which can't be
characterized by the runtime profiler into feedback states.
This CL addresses that problem. Note that it's also important to
carefully exclude these slots from the profiler's consideration when
determining if you want to optimize from Full code.
BUG=
Review-Url: https://codereview.chromium.org/2342853002
Cr-Commit-Position: refs/heads/master@{#39555}
This way, many files which only need CompilationInfo but not compiler.h
and its dependencies can include just compilation-info.h.
BUG=
Review-Url: https://codereview.chromium.org/2284313003
Cr-Commit-Position: refs/heads/master@{#39038}
This stages the --ignition-preserve-bytecode flag which preserves the
bytecode even when switching to baseline code. It is now implied by the
combined --ignition-staging flag.
R=rmcilroy@chromium.org
Review-Url: https://codereview.chromium.org/2244303003
Cr-Commit-Position: refs/heads/master@{#38648}
This switches the interface of the runtime profiler to use frames as
opposed to functions for performing on-stack replacement. Requests for
such replacements need to target a specific frame. This will enable us
to activate bytecode as well as baseline code for the same function.
The existing %OptimizeOsr runtime function also had to adapted and now
takes an optional stack depth to target a specific stack frame.
R=bmeurer@chromium.org
BUG=v8:4764
Review-Url: https://codereview.chromium.org/2230783004
Cr-Commit-Position: refs/heads/master@{#38548}
This fixes the runtime profiler to no longer assume that seeing an
optimized frame on the stack implies the underlying function is not
being interpreted when entered normally. This no longer holds with code
generated for OSR directly from bytecode (not installed on function).
R=rmcilroy@chromium.org
TEST=mjsunit/regress/regress-crbug-632800
BUG=chromium:632800
Review-Url: https://codereview.chromium.org/2208603005
Cr-Commit-Position: refs/heads/master@{#38360}
This adds preliminary support for on-stack replacement from Ignition to
optimized code generated by TurboFan to the runtime profiler. Involved
heuristics (e.g. code size allowance) have been taken from existing code
without any re-evaluation in the new setting.
R=rmcilroy@chromium.org
BUG=v8:4764
Review-Url: https://codereview.chromium.org/2182183005
Cr-Commit-Position: refs/heads/master@{#38159}
This adds a new field to the header of every BytecodeArray which stores
the current nesting level up to which loop back edges are armed as OSR
points. The intention is to arm OSR points incrementally from outermost
to innermost until one fires (similar to OSR from FullCodegen).
R=rmcilroy@chromium.org
BUG=v8:4764
Review-Url: https://codereview.chromium.org/2172583002
Cr-Commit-Position: refs/heads/master@{#38017}
Always use the BytecodeGraphBuilder when the --turbo-from-bytecode
is enabled, assuming the function should be compiled for Ignition.
Adds a new MaybeOptimizeIgnition function to runtime-profiler
which is called if the function should be optimized from bytecode
rather than going via full-codegen.
BUG=v8:4280
Committed: https://crrev.com/9ca7db914be88e6792a88eab4a1988ee031d70c4
Review-Url: https://codereview.chromium.org/2156753002
Cr-Original-Commit-Position: refs/heads/master@{#37921}
Cr-Commit-Position: refs/heads/master@{#38002}
Reason for revert:
Breaks tsan:
https://build.chromium.org/p/client.v8/builders/V8%20Linux64%20TSAN/builds/10758
Original issue's description:
> [Intepreter] Always use BytecodeGraphBuilder when --turbo-from-bytecode
>
> Always use the BytecodeGraphBuilder when the --turbo-from-bytecode
> is enabled, assuming the function should be compiled for Ignition.
> Adds a new MaybeOptimizeIgnition function to runtime-profiler
> which is called if the function should be optimized from bytecode
> rather than going via full-codegen.
>
> BUG=v8:4280
>
> Committed: https://crrev.com/9ca7db914be88e6792a88eab4a1988ee031d70c4
> Cr-Commit-Position: refs/heads/master@{#37921}
TBR=mstarzinger@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/2165223002
Cr-Commit-Position: refs/heads/master@{#37925}
Always use the BytecodeGraphBuilder when the --turbo-from-bytecode
is enabled, assuming the function should be compiled for Ignition.
Adds a new MaybeOptimizeIgnition function to runtime-profiler
which is called if the function should be optimized from bytecode
rather than going via full-codegen.
BUG=v8:4280
Review-Url: https://codereview.chromium.org/2156753002
Cr-Commit-Position: refs/heads/master@{#37921}
Updates kProfilerTicksBeforeBaseline in runtime-profiler to allow
functions to switch from ignition to full-codgen earlier. This helps
on many benchmarks and does not impact the code size significantly.
BUG=v8:4280
LOG=N
Review-Url: https://codereview.chromium.org/2085153003
Cr-Commit-Position: refs/heads/master@{#37189}
We get less "pollution" of type feedback if we have one vector per native
context, rather than one for the whole system. This CL moves the vector
appropriately.
BUG=
Review-Url: https://codereview.chromium.org/1906823002
Cr-Commit-Position: refs/heads/master@{#36539}
Remove checks for IC hotness from Ignition tiering up decision since this is
not relevent for full-codegen compilation. Also make the decision about what
tier we are moving to more explicit and visible in --trace-opt.
BUG=v8:4280
LOG=N
Review-Url: https://codereview.chromium.org/1969773002
Cr-Commit-Position: refs/heads/master@{#36260}
This adds a baseline tier to the compilation pipeline. Currently this
tier is used to model a path from the interpreter to optimized code via
full-codegen code (to ensure sufficient type feedback). Switching from
the unoptimized tier to the baseline tier is limited to happen only when
there are no activations of the given function on the stack.
R=rmcilroy@chromium.org,bmeurer@chromium.org
Review URL: https://codereview.chromium.org/1903273004
Cr-Commit-Position: refs/heads/master@{#35757}
The JSFunction::PassesFilter predicate is not fine-grained enough to
actually distinguish different closures and hence can be changed into
SharedFunctionInfo::PassesFilter instead. This will allow the compiler
to use is more broadly.
R=jkummerow@chromium.org
Review URL: https://codereview.chromium.org/1823033002
Cr-Commit-Position: refs/heads/master@{#34981}
Adds a profiling counter to each BytecodeArray object, and adds
code to Jump and Return bytecode handlers to update this
counter by the size of the jump or the distance from the return
to the start of the function. This is more accurate than fullcodegen's
approach since it takes forward jumps into account as well as back-edges.
Modifies RuntimeProfiler to track ticks for interpreted frames.
Currently we use the SharedFunctionInfo::profiler_ticks() instead
of adding another to tick field to avoid adding another field to
BytecodeArray since SharedFunctionInfo::profiler_ticks() is only
used by Crankshaft otherwise so we shouldn't need both for
BUG=v8:4689
LOG=N
Review URL: https://codereview.chromium.org/1707693003
Cr-Commit-Position: refs/heads/master@{#34166}
Reason for revert:
Must revert for now due to chromium api natives issues.
Original issue's description:
> Type Feedback Vector lives in the closure
>
> (RELAND: the problem before was a missing write barrier for adding the code
> entry to the new closure. It's been addressed with a new macro instruction
> and test. The only change to this CL is the addition of two calls to
> __ RecordWriteCodeEntryField() in the platform CompileLazy builtin.)
>
> We get less "pollution" of type feedback if we have one vector per native
> context, rather than one for the whole system. This CL moves the vector
> appropriately.
>
> We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
> vector actually lives in the first slot of the literals array (indeed there is
> great commonality between those arrays, they can be thought of as the same
> thing). So we make greater effort to ensure there is a valid literals array
> after compilation.
>
> This meant, for performance reasons, that we needed to extend
> FastNewClosureStub to support creating closures with literals. And ultimately,
> it drove us to move the optimized code map lookup out of FastNewClosureStub
> and into the compile lazy builtin.
>
> The heap change is trivial so I TBR Hannes for it...
> Also, Yang has had a look at the debugger changes already and approved 'em. So he is TBR style too.
> And Benedikt reviewed it as well.
>
> TBR=hpayer@chromium.org, yangguo@chromium.org, bmeurer@chromium.org
>
> BUG=
>
> Committed: https://crrev.com/bb31db3ad6de16f86a61f6c7bbfd3274e3d957b5
> Cr-Commit-Position: refs/heads/master@{#33741}
TBR=bmeurer@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=
Review URL: https://codereview.chromium.org/1670813005
Cr-Commit-Position: refs/heads/master@{#33766}
(RELAND: the problem before was a missing write barrier for adding the code
entry to the new closure. It's been addressed with a new macro instruction
and test. The only change to this CL is the addition of two calls to
__ RecordWriteCodeEntryField() in the platform CompileLazy builtin.)
We get less "pollution" of type feedback if we have one vector per native
context, rather than one for the whole system. This CL moves the vector
appropriately.
We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
vector actually lives in the first slot of the literals array (indeed there is
great commonality between those arrays, they can be thought of as the same
thing). So we make greater effort to ensure there is a valid literals array
after compilation.
This meant, for performance reasons, that we needed to extend
FastNewClosureStub to support creating closures with literals. And ultimately,
it drove us to move the optimized code map lookup out of FastNewClosureStub
and into the compile lazy builtin.
The heap change is trivial so I TBR Hannes for it...
Also, Yang has had a look at the debugger changes already and approved 'em. So he is TBR style too.
And Benedikt reviewed it as well.
TBR=hpayer@chromium.org, yangguo@chromium.org, bmeurer@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1668103002
Cr-Commit-Position: refs/heads/master@{#33741}
Reason for revert:
Bug: failing to use write barrier when writing code entry into closure.
Original issue's description:
> Reland of Type Feedback Vector lives in the closure
>
> (Fixed a bug found by nosnap builds.)
>
> We get less "pollution" of type feedback if we have one vector per native
> context, rather than one for the whole system. This CL moves the vector
> appropriately.
>
> We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
> vector actually lives in the first slot of the literals array (indeed there is
> great commonality between those arrays, they can be thought of as the same
> thing). So we make greater effort to ensure there is a valid literals array
> after compilation.
>
> This meant, for performance reasons, that we needed to extend
> FastNewClosureStub to support creating closures with literals. And ultimately,
> it drove us to move the optimized code map lookup out of FastNewClosureStub
> and into the compile lazy builtin.
>
> The heap change is trivial so I TBR Hannes for it...
>
> TBR=hpayer@chromium.org
> BUG=
>
> Committed: https://crrev.com/d984b3b0ce91e55800f5323b4bb32a06f8a5aab1
> Cr-Commit-Position: refs/heads/master@{#33548}
TBR=bmeurer@chromium.org,yangguo@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=
Review URL: https://codereview.chromium.org/1643533003
Cr-Commit-Position: refs/heads/master@{#33556}
(Fixed a bug found by nosnap builds.)
We get less "pollution" of type feedback if we have one vector per native
context, rather than one for the whole system. This CL moves the vector
appropriately.
We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
vector actually lives in the first slot of the literals array (indeed there is
great commonality between those arrays, they can be thought of as the same
thing). So we make greater effort to ensure there is a valid literals array
after compilation.
This meant, for performance reasons, that we needed to extend
FastNewClosureStub to support creating closures with literals. And ultimately,
it drove us to move the optimized code map lookup out of FastNewClosureStub
and into the compile lazy builtin.
The heap change is trivial so I TBR Hannes for it...
TBR=hpayer@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1642613002
Cr-Commit-Position: refs/heads/master@{#33548}
Reason for revert:
FAilure on win32 bot, need to investigate webkit failures.
Original issue's description:
> Type Feedback Vector lives in the closure
>
> We get less "pollution" of type feedback if we have one vector per native
> context, rather than one for the whole system. This CL moves the vector
> appropriately.
>
> We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
> vector actually lives in the first slot of the literals array (indeed there is
> great commonality between those arrays, they can be thought of as the same
> thing). So we make greater effort to ensure there is a valid literals array
> after compilation.
>
> This meant, for performance reasons, that we needed to extend
> FastNewClosureStub to support creating closures with literals. And ultimately,
> it drove us to move the optimized code map lookup out of FastNewClosureStub
> and into the compile lazy builtin.
>
> The heap change is trivial so I TBR Hannes for it...
>
> TBR=hpayer@chromium.org
>
> BUG=
>
> Committed: https://crrev.com/a5200f7ed4d11c6b882fa667da7a1864226544b4
> Cr-Commit-Position: refs/heads/master@{#33518}
TBR=bmeurer@chromium.org,akos.palfi@imgtec.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=
Review URL: https://codereview.chromium.org/1632993003
Cr-Commit-Position: refs/heads/master@{#33520}
We get less "pollution" of type feedback if we have one vector per native
context, rather than one for the whole system. This CL moves the vector
appropriately.
We rely more heavily on the Optimized Code Map in the SharedFunctionInfo. The
vector actually lives in the first slot of the literals array (indeed there is
great commonality between those arrays, they can be thought of as the same
thing). So we make greater effort to ensure there is a valid literals array
after compilation.
This meant, for performance reasons, that we needed to extend
FastNewClosureStub to support creating closures with literals. And ultimately,
it drove us to move the optimized code map lookup out of FastNewClosureStub
and into the compile lazy builtin.
The heap change is trivial so I TBR Hannes for it...
TBR=hpayer@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1563213002
Cr-Commit-Position: refs/heads/master@{#33518}
It's cumbersome to maintain IC profiler statistics all the time.
Let's just do it as needed.
BUG=
Review URL: https://codereview.chromium.org/1507903004
Cr-Commit-Position: refs/heads/master@{#32693}
Moves all files related to AST and scopes into ast/,
and all files related to scanner & parser to parsing/.
Also eliminates a couple of spurious dependencies.
R=mstarzinger@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1481613002
Cr-Commit-Position: refs/heads/master@{#32351}
This removes several methods from JSFunction that just delegate to
SharedFunctionInfo. These methods are especially dangerous when they
hide the fact that they potentially affect all function instances
deriving from the same underlying SharedFunctionInfo.
R=bmeurer@chromium.org
Review URL: https://codereview.chromium.org/1417213005
Cr-Commit-Position: refs/heads/master@{#31792}
This is a first step towards constraining down the heap interface to
just the heap.h file. Note that many includes still leak through that
file to the global "src" directory, but there now is a single place
controlling which declarations leak that way. Especially inclusion of
inline header files within "heap" has been limited drastically.
R=hpayer@chromium.org,mlippautz@chromium.org
Review URL: https://codereview.chromium.org/1281233003
Cr-Commit-Position: refs/heads/master@{#30092}
Up until now we were unable to have profiler ticks beyong 255, which
basically disabled OSR for moderately large functions.
BUG=chromium:508741
LOG=n
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1224173003
Cr-Commit-Position: refs/heads/master@{#29597}
When compiling on a laptop I like to concatenate the small test files.
This makes a big difference to compile times. These changes make that
easier.
R=ulan@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1163803002
Cr-Commit-Position: refs/heads/master@{#28742}
This just delegates to SharedFunctionInfo::optimization_disabled and
was primarily used for assertions. Removing it due to misleading name
because already optimized functions reported being "non-optimizable".
This relands commit 181d7b8597.
R=titzer@chromium.org
Review URL: https://codereview.chromium.org/1146423002
Cr-Commit-Position: refs/heads/master@{#28577}
Reason for revert:
Causes assertions to fire when serializing optimized code.
Original issue's description:
> Remove obsolete JSFunction::IsOptimizable predicate.
>
> This just delegates to SharedFunctionInfo::optimization_disabled and
> was primarily used for assertions. Removing it due to misleading name
> because already optimized functions reported being "non-optimizable".
>
> R=titzer@chromium.org
>
> Committed: https://crrev.com/181d7b85977eb752b19e1de902093783e31330ef
> Cr-Commit-Position: refs/heads/master@{#28551}
TBR=titzer@chromium.org,bmeurer@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/1148973005
Cr-Commit-Position: refs/heads/master@{#28554}
This just delegates to SharedFunctionInfo::optimization_disabled and
was primarily used for assertions. Removing it due to misleading name
because already optimized functions reported being "non-optimizable".
R=titzer@chromium.org
Review URL: https://codereview.chromium.org/1150683002
Cr-Commit-Position: refs/heads/master@{#28551}
This flag mostly duplicates SharedFunctionInfo::optimization_disabled
and is only queried in places where the original is available. Remove
the brittle and error-prone duplication.
R=bmeurer@chromium.org
Review URL: https://codereview.chromium.org/1148043002
Cr-Commit-Position: refs/heads/master@{#28520}
