Separated core greedy allocator concepts, exposing the APIs we would want to continue working with. In particular, this change completely reworks CoalescedLiveRanges to reflect the fact that we expect more than one possible conflict, scrapping the initial design of the structure. Since this is a critical part of the design, this change may be thought of as a full rewrite of the algorithm.
Reduced all heuristics to just 2 essential ones: split "somewhere", which we'll still need when all other heuristics fail; and spill.
Introduced a simple primitive for splitting - at GapPosition::START. The goal is to use such primitives to quickly and reliably author heuristics.
I expected this primitive to "just work" for any arbitrary instruction index within a live range - e.g. its middle. That's not the case, it seems to upset execution in certain scenarios. Restricting to either before/after use positions seems to work. I'm still investigating what the source of failures is in the case of "arbitrary instruction in the range" case.
I intended to document the rationale and prove the soundness of always using START for splits, but I will postpone to after this last remaining issue is resolved.
Review URL: https://codereview.chromium.org/1205173002
Cr-Commit-Position: refs/heads/master@{#29352}
Dumb typo introduced in refs/heads/master@{#29306}. I thought I was turning on
report_exceptions in Shell::ExecuteString, but instead I turned on print_result
(which assumes an interactive debugger and a HandleScope for the
utility_context_).
BUG=chromium:504727,chromium:504728
R=mstarzinger@chromium.org
LOG=n
Review URL: https://codereview.chromium.org/1219563002
Cr-Commit-Position: refs/heads/master@{#29350}
The android_* configs were never real v8 target
architectures, only make targets. This doesn't make sense
with ninja anymore.
BUG=chromium:502176
LOG=n
NOTRY=true
Review URL: https://codereview.chromium.org/1220743002
Cr-Commit-Position: refs/heads/master@{#29349}
RegExpCompileEvent acquieres mutex from Log class during MessageBuilder creation. LogRegExpSource, called from RegExpCompileEvent creates another MessageBuilder object which also acquires the same mutex. This mutex is not recursive, so during second acquirement, assertion fail is happening. Solution: LogRegExpSource should use the same MessageBuilder object as RegExpCompileEvent.
Review URL: https://codereview.chromium.org/1207433002
Cr-Commit-Position: refs/heads/master@{#29347}
Make clang dir absolute to avoid differences between ninja
and make gyp generator.
BUG=chromium:502176
LOG=n
Review URL: https://codereview.chromium.org/1217783002
Cr-Commit-Position: refs/heads/master@{#29341}
Report builtins by name (e.g. "Builtin:ArgumentsAdaptorTrampoline")
instead of labeling everything "Builtin:A builtin from the snapshot".
BUG=
Review URL: https://codereview.chromium.org/1216833002
Cr-Commit-Position: refs/heads/master@{#29339}
The test language/asi/S7.9_A5.7_T1 is failing intermittently.
BUG=v8:4253
LOG=N
TBR=adamk
Review URL: https://codereview.chromium.org/1215813002.
Cr-Commit-Position: refs/heads/master@{#29332}
The use of jalr ra is unpredictable if instruction in branch delay slot
is in next page.
This finally fixes random failures in JS debugger and InteruptRequest tests.
TEST=mjsunit/debug-*,
cctest/test-api/RequestInterruptTestWithNativeAccessor
BUG=
Review URL: https://codereview.chromium.org/1220443002
Cr-Commit-Position: refs/heads/master@{#29331}
Now that we keep tabs on shared function infos from a script, we can speed up finding shared function infos for debugging. However, in case we have to compile a function that cannot be lazily compiled without context, we fall back to the slow heap iteration.
R=mstarzinger@chromium.org
BUG=v8:4132,v8:4052
LOG=N
Committed: https://crrev.com/cfe89a71a332ef9ed481c8210bc3ad6d2822034b
Cr-Commit-Position: refs/heads/master@{#29296}
Review URL: https://codereview.chromium.org/1206573004
Cr-Commit-Position: refs/heads/master@{#29327}
Note that prior to having canonical shared function infos, this has
been a source of duplicate shared function infos.
R=bmeurer@chromium.org
BUG=chromium:504787
LOG=N
Review URL: https://codereview.chromium.org/1209383002
Cr-Commit-Position: refs/heads/master@{#29326}
This optimization is already implemented in fullcodegen, and
basically makes sure that we do not unecessarily blow up the
code with duplicated return sequences everywhere.
R=danno@chromium.org
Review URL: https://codereview.chromium.org/1211373002
Cr-Commit-Position: refs/heads/master@{#29315}
This allows context-independent code generated by TurboFan to be cached
in the optimized code map and reused across native contexts. Note that
currently this cache is still flushed at GC time.
R=bmeurer@chromium.org,mvstanton@chromium.org
TEST=cctest/test-compiler/OptimizedCodeSharing
Review URL: https://codereview.chromium.org/1208013002
Cr-Commit-Position: refs/heads/master@{#29313}
This will enable tail call optimization even across inlining. Plus it
might enable some other interesting optimizations as well. In order to
avoid blowing up the generated code, we can still canonicalize the
epilogue in the CodeGenerator, similar to what fullcodegen does.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1215623002
Cr-Commit-Position: refs/heads/master@{#29311}
Reserving space for deserialization can cause GC, which
can evict entries from the string table. Having more deleted
entries now, StringTable::EnsureCapacity could cause a GC
later during deserialization even when we actually still
have enough capacity.
Instead, we now keep new internalized strings in a separate list
and commit them to the string table at the end.
R=ulan@chromium.org
BUG=chromium:502085
LOG=N
Review URL: https://codereview.chromium.org/1204863006
Cr-Commit-Position: refs/heads/master@{#29308}
The issue is that Worker.prototype.terminate was deleting the C++ Worker
object, and then Worker.prototype.getMessage was trying to read messages from
the queue.
The simplest solution is to keep workers in a zombie state when they have been
terminated. They won't be reaped until Shell::CleanupWorkers is called.
I've also fixed some threading issues with Workers:
* Workers can be created by another Worker, so the Shell::workers_ variable
must be protected by a mutex.
* An individual Worker can typically only be accessed by the isolate that
created it, but the main thread can always terminate it, so the Worker::state_
must be accessed in a thread-safe way.
BUG=chromium:504136
R=jochen@chromium.org
LOG=n
Review URL: https://codereview.chromium.org/1208733002
Cr-Commit-Position: refs/heads/master@{#29306}
