With this change, on ia32 and x64, a load from memory into a register can be replaced by a memory operand for integer binops if it makes sense.
BUG=
Review-Url: https://codereview.chromium.org/2728533003
Cr-Commit-Position: refs/heads/master@{#43739}
Sometimes TurboFan is able to extract receiver maps from the surrounding
graph and thus is able to generate reasonable code for property accesses,
even if those haven't been executed in the baseline tier yet. So, only
stick in an SOFT deoptimization exit, if ExtractReceiverMaps failed to
infer proper receiver maps.
R=yangguo@chromium.org
BUG=v8:5267
Review-Url: https://codereview.chromium.org/2746013002
Cr-Commit-Position: refs/heads/master@{#43736}
These operations don't need the context, so no need to pass the context
to them. Also avoids the loading of context in the interpreter bytecode
handlers for StrictEqual and Typeof.
BUG=v8:5268,v8:5269
R=yangguo@chromium.org
Review-Url: https://codereview.chromium.org/2744173002
Cr-Commit-Position: refs/heads/master@{#43733}
To speed up Date.prototype.toString(), this patch adds a cache in
the DateCache for the string short name representing the time zone.
Because time zones in a particular location just have two short names
(for DST and standard time), and the DateCache already understands
whether a time is in DST or not, it is possible to keep the result
of OS::LocalTimezone around and select between the two based on
whether the time is DST or not.
In local microbenchmarks (calling Date.prototype.toString() in a
loop), I observed a 6-10% speedup with this patch. In the browser,
the speedup may be even greater as the system call needs to do
some extra work to break out of the sandbox. I don't think the
microbenchmark is extremely unrealistic; in any real program which
calls Date.prototype.toString() multiple times, the cache should
hit almost all of the time, as time zone changes are rare.
The proximate motivation for this patch was to enable ICU as a
backend for timezone information, which is drafted at
https://codereview.chromium.org/2724373002/
The ICU implementation of OS::LocalTimezone is even slower than
the system call one, but this patch makes their performance
indistinguishable on the microbenchmark.
In the tz database, many timezones actually do have a number of different
historical names. For example, America/Anchorage went through a number of
changes, from AST to AHST to YST to AKST. However, both ICU and the
Linux OS interfaces just report the modern timezone name in tests
for the appropriate timezone name, even for historical times. I can
see why this would be:
- For ICU, CLDR only has two short names in the data file: the one for
dst and non-dst
- For Linux, the timezone names do seem to make it into the
/etc/localtime file. However, glibc assumes there are only two relevant
names and selects between them, as you can see in its implementation
of localtime_r:
http://bazaar.launchpad.net/~vcs-imports/glibc/master/view/head:/time/tzset.c#L573
So, this cache should be valid until we switch to a more accurate source
of short timezone names.
BUG=v8:6031
Review-Url: https://codereview.chromium.org/2726253002
Cr-Commit-Position: refs/heads/master@{#43730}
There is no guarantee that Map::GetConstructor() returns a JSFunction.
Specifically, detached global proxies return the |null| sentinel. So
we have to check the object type before casting to JSFunction.
BUG=chromium:694141
Review-Url: https://codereview.chromium.org/2739303003
Cr-Commit-Position: refs/heads/master@{#43727}
Note that this changes the sampling interval from milliseconds to
microseconds -- this shouldn't cause issues except for tools that use
'profiler,"begin",<interval>' somehow.
Change-Id: I20222de91f7820e26eb3fc505a4752b0bc7e1642
Reviewed-on: https://chromium-review.googlesource.com/451658
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43726}
This fixes the catch predictions for the following builtins --
AsyncFunctionAwaitCaught
AsyncFunctionAwaitUncaught
PromiseResolveClosure
ResolvePromise
PromiseResolve
Added tests for each.
Added whitelist for builtins behind a flag.
BUG=chromium:691875
Change-Id: I816cafdb69f0c9f1eefc440a0a44c36713d0b7dc
Reviewed-on: https://chromium-review.googlesource.com/450894
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43725}
AllocateRawAligned called into AllocateRawUnaligned, which expected
the address of the pointer to the top of the stack, not the pointer
itself. Instead, the pointer itself was passed, causing segfaults
if this code is actually run.
Also do some drive-by clean up of the branching/labels and unused
vars etc. in AllocateRawAligned.
BUG=v8:6075
Change-Id: If71db4b61d777b6543e5246e92bb5b9e6c02c81f
Reviewed-on: https://chromium-review.googlesource.com/452374
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43722}
Removing it from default variants set which will substantially increase test speed for developers.
Pure ignition is intended to be tested only on bots unless opted in.
BUG=chromium:682617
R=mstarzinger@chromium.org,mvstanton@chromium.org,rmcilroy@chromium.org
Change-Id: I98adb81951a0a2d16f52e6b34b98d3a90e178d8a
Reviewed-on: https://chromium-review.googlesource.com/452580
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43721}
Now that the --turbo flag is on by default, this implication makes it
impossible to disable the runtime profiler to prevent functions from
being considered hot. Essentially the --nocrankshaft flag was borked.
R=machenbach@chromium.org
Change-Id: I698008b4b69038e8bfab6d8ec3ccf57e2fd71701
Reviewed-on: https://chromium-review.googlesource.com/452619
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43720}
Also format flags in single lines to make configs easier to read and modify.
BUG=chromium:673246
NOTRY=true
R=mstarzinger@chromium.org,titzer@chromium.org,rmcilroy@chromium.org
Change-Id: If37486c98af161467b639271b035207c3ae1077a
Reviewed-on: https://chromium-review.googlesource.com/452579
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43719}
The callers of this function already do the flushing (as its comment demands).
BUG=
Change-Id: Iff6720f4ddecb1a4569b6cac66eba4b857c46b17
Reviewed-on: https://chromium-review.googlesource.com/452504
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43718}
- lower the cutoff-threshold for bottom up graphs to 1%
- show all builtins/stubs/ics by default
R=jkummerow@chromium.org
Change-Id: I522173155b817ae20a37c40ecf411f2b55d82105
Reviewed-on: https://chromium-review.googlesource.com/452482
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43717}
Add log message when RuntimeProfiler checks whether to optimize a
function, but that function is already in the optimization queue
(with --trace-opt-verbose).
R=yangguo@chromium.org
Review-Url: https://codereview.chromium.org/2740103003
Cr-Commit-Position: refs/heads/master@{#43711}
ICU's UChar was uint16_t (non-Win) or wchar_t (Windows). It's switching
to char16_t in both C/C++ API. It needs some changes. Fortunately,
v8 needs only a couple of changes because v8 has been using
reinterpret_cast in many places calling ICU API.
This change was confirmed to work fine with ICU-59-to-be.
BUG=v8:6062
TEST=trybot
Review-Url: https://codereview.chromium.org/2738503008
Cr-Commit-Position: refs/heads/master@{#43707}
Chrome no longer calls v8::Isolate::Dispose on shutdown, essentially preventing
the use of V8 stats within chrome/content_shell. This CL adds a basic hook to
the api that is then used to only print the stats.
Chrome change: https://codereview.chromium.org/2693353002
Change-Id: I1481c14afe611e9c08ae67c815201a45940daa57
Reviewed-on: https://chromium-review.googlesource.com/452338
Reviewed-by: Jochen Eisinger <jochen@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43706}
- lower the cutoff-threshold for bottom up graphs to 1%
- show all builtins/stubs/ics by default
R=jkummerow@chromium.org
Change-Id: I85f8c4f00c5f36407bc5d9f81bead8f39a5ef632
NOTRY=true
Change-Id: I85f8c4f00c5f36407bc5d9f81bead8f39a5ef632
Reviewed-on: https://chromium-review.googlesource.com/452300
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43704}
Specifically, add bytecodes for Call0, Call1, Call2, CallProperty0, CallProperty1,
and CallProperty2. Also share the bytecode handler code between between
equivalent CallX and CallPropertyX handlers.
Review-Url: https://codereview.chromium.org/2684993002
Cr-Original-Commit-Position: refs/heads/master@{#43290}
Committed: 00d6f1f80a
Review-Url: https://codereview.chromium.org/2684993002
Cr-Commit-Position: refs/heads/master@{#43700}
The optimization
NumberFloor(NumberDivide(lhs, rhs))
to
NumberToInt32(NumberDivide(lhs, rhs))
for potentially negative lhs is not valid, since Math.floor rounds
towards -infinity, whereas ToInt32 truncates.
BUG=chromium:699282
R=jarin@chromium.org
Review-Url: https://codereview.chromium.org/2743673002
Cr-Commit-Position: refs/heads/master@{#43699}
We used to embed a string address as description right after
a stop instruction, which the simulator would read and print.
We removed that a while ago to make the snapshot predictable.
R=petermarshall@chromium.org
BUG=v8:6071
Review-Url: https://codereview.chromium.org/2744503003
Cr-Commit-Position: refs/heads/master@{#43698}
We need to check whether advancing the iterator moved us beyong the end
of the bitmap. This has not been flushed out as our inlined bitmap is
still in valid memory.
In practice this is not a problem because the value is never used as we
are at the end of the bitmap. Asan rightfully complains when using an
external bitmap though.
BUG=chromium:651354
Change-Id: I8b141a467e9552f8ac2287dd62a725a14a289a37
Reviewed-on: https://chromium-review.googlesource.com/452497
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43696}
