Remove all IsHeapObject/IsSmi checks from assembler and also from
the macro-assembler functions that Turbofan code generation uses.
Note for porters: In case it's unclear which macro-assembler
functions need to be modified, it may be best to wait until I
split MacroAssembler in a followup-CL, which will make that clear.
Bug: v8:6048
Change-Id: Ife0735cc6f48713c9ec493faf2dac5e553d1c06b
Reviewed-on: https://chromium-review.googlesource.com/561015
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46436}
Introduce a stripped down version of UseScratchRegisterScope for ARM and use it
inside the assembler and macro-assembler. At the exception of the Call
instructions, we now use this scope instead of using the ip register
directly. This is inspired from how the ARM64 backend works.
In general, the benefit of doing this is we can catch cases where ip is being
used both by the caller and by the assembler. But more specifically, TurboFan
reserves r9 as an extra scratch register because ip can already be used by the
assembler. With this utility, we can isolate the cases in the code generator
which need an extra register and potentially fix them, allowing us to give r9
back to the register allocator.
This patch uncovered places in the assembler where we were using ip
unconditionally when we could have re-used the destination register instead.
Bug: v8:6553
Change-Id: Ib7134e3ed64dd1f90baf209ae831ed8f644cac78
Reviewed-on: https://chromium-review.googlesource.com/544956
Commit-Queue: Pierre Langlois <pierre.langlois@arm.com>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46425}
On x86, signalling NaNs get converted to quiet NaNs when they get push
on the stack and popped again. This happens in the code generation for
arm, specifically for the vmov instruction with the immediate parameter.
This CL replaces the vmov function in assembler-arm to take the
immediate as a uint64_t instead of a double, to guarantee that the bit
pattern does not change even if the parameter is a signalling NaN.
BUG=v8:6564
Change-Id: I062559f9a7ba8b0f560628e5c39621ca578c3e7d
Reviewed-on: https://chromium-review.googlesource.com/558964
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Martyn Capewell <martyn.capewell@arm.com>
Cr-Commit-Position: refs/heads/master@{#46418}
Instead of counting profiler ticks on the shared function info (which is
shared between native contexts), count them on the feedback vector
(which is not). This allows us to continue pushing optimization
decisions off the SFI, onto the feedback vector.
Note that a side-effect of this is that ICs don't have to walk the stack
to reset profiler ticks, as they can access the feedback vector directly
from their feedback nexus.
Change-Id: I232ae9e759fca75cd89d393148a4ff42caa2646f
Reviewed-on: https://chromium-review.googlesource.com/544888
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46411}
On Loongson 3A, MADD/MSUB instructions are actually fused MADD/MSUB and
they can cause failure in some of the tests. Since this optimization is
rarely used, and not used at all on MIPS64R6, MADD/MSUB instructions
are removed from the source base.
TEST=
BUG=
Change-Id: Ifbb5508a62731bb061f332864ffd1e210e97f963
Reviewed-on: https://chromium-review.googlesource.com/558066
Reviewed-by: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Commit-Queue: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Cr-Commit-Position: refs/heads/master@{#46387}
Instructions ins and ext didn't work properly when size = 32 because of
incorrect mask initialization, this CL fixes this. A test for Ins is
also added.
BUG=
Change-Id: I95cc8e13aaa2341b34ae59dae1eefb64c551b8b4
Reviewed-on: https://chromium-review.googlesource.com/558872
Commit-Queue: Miran Karić <Miran.Karic@imgtec.com>
Reviewed-by: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Cr-Commit-Position: refs/heads/master@{#46378}
These were originally written as part of
https://chromium-review.googlesource.com/c/550396/. I've separated them
out into a separate CL with the intent of landing it first, so that it's
easier to see the difference these CLs will have on generated bytecode.
BUG=v8:5855
TBR=tebbi@chromium.org, rmcilroy@chromium.org
Change-Id: Ib84e65847d7396e31b0e38d28f59454cf7c58fc1
Reviewed-on: https://chromium-review.googlesource.com/558221
Reviewed-by: Caitlin Potter <caitp@igalia.com>
Commit-Queue: Caitlin Potter <caitp@igalia.com>
Cr-Commit-Position: refs/heads/master@{#46370}
The `FAST_` prefix doesn’t make much sense — they’re all just different cases
with their own optimizations. Packedness being implicit (e.g. `FAST_ELEMENTS`
vs. `FAST_HOLEY_ELEMENTS`) is not ideal, either.
This patch renames the FAST elements kinds as follows:
- e.g. FAST_ELEMENTS => PACKED_ELEMENTS
- e.g. FAST_HOLEY_ELEMENTS => HOLEY_ELEMENTS
The following exceptions are left intact, for lack of a better name:
- FAST_SLOPPY_ARGUMENTS_ELEMENTS
- SLOW_SLOPPY_ARGUMENTS_ELEMENTS
- FAST_STRING_WRAPPER_ELEMENTS
- SLOW_STRING_WRAPPER_ELEMENTS
This makes it easier to reason about elements kinds, and less confusing to
explain how they’re used.
R=jkummerow@chromium.org, cbruni@chromium.org
BUG=v8:6548
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: Ie7c6bee85583c3d84b730f7aebbd70c1efa38af9
Reviewed-on: https://chromium-review.googlesource.com/556032
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Mathias Bynens <mathias@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46361}
V8's catch prediction mechanism tries to predict whether a thrown
exception will be caught, just by looking at the current call stack.
At the time when catch prediction was first introduced, only a few
builtins (mostly related to Promise and Generator) could end up being
fed into the catch prediction mechanism. This is no longer the case now
that builtins are used in new ways, e.g. Array.p.forEach's continuation
builtins.
This CL removes the need to explicitly mark all builtins visible to the
StackFrameIterator as CAUGHT/UNCAUGHT/PROMISE, and instead defaults to
treating unmarked builtins as UNCAUGHT.
BUG=v8:6536
Change-Id: Ibdc106a91b2b0ffb93099433077642cad02c71e2
Reviewed-on: https://chromium-review.googlesource.com/555518
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46357}
This reverts commit 0d2ed6c328.
The CL introduced perf regressions: crbug.com/735649.
We are going to reland the CL in an isolated V8 roll to ensure
that perf regressions are attributed correctly.
Original commit message:
> [heap] Allow a minimum semi-space size of 512K.
> This CL also reduces the minimum semi-space size to 512K.
> BUG=chromium:716032
BUG=chromium:735649
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I1f1b08ca6853347c00070f000c309d839ff8a4bb
Reviewed-on: https://chromium-review.googlesource.com/552541
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46349}
This way, each lazy function needs to handle only the data relevant to
itself. This reduced data handling overheads.
Other changes:
1) Don't deserialize the data; once it's on the heap, it can stay there. Lazy
function compilation is only done in the main thread.
2) Separate ProducedPreParsedScopeData and ConsumedPreParsedScopeData. It's clearer, because:
- The data looks fundamentally different when we're producing it and when we're
consuming it.
- Cleanly separates the operations we can do in the "producing phase" and in the
"consuming phase".
Bug: v8:5516
Change-Id: I6985a6621f71b348a55155724765624b5d5f7c33
Reviewed-on: https://chromium-review.googlesource.com/528094
Commit-Queue: Marja Hölttä <marja@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46347}
Previously V8 created a promise to return to userland,
but instead we let the embedder create and track the promise.
Bug: v8:5785
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I8903ffbabf3a256f1c8df844a656a873da304586
Reviewed-on: https://chromium-review.googlesource.com/492646
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Daniel Clifford <danno@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46333}
The naming convention in v8 has trivial getters named like the field,
no 'get_' prefix, and dropping the '_' suffix of the field.
BUG=
Review-Url: https://codereview.chromium.org/2958283003
Cr-Commit-Position: refs/heads/master@{#46332}
When internalization of the key fails because the string does not
exist in the StringTable yet, then no regular object can possibly
have a property with that name, so just returning "false" is safe.
However, for objects with interceptors this is not true, as there
may well be intercepted properties whose keys have not been
internalized. So "special API objects" must take the slow path to
query any interceptors.
Bug: chromium:735990
Change-Id: Ibe6c4f8b14fef65738115f12167d3602bec3d9b7
Reviewed-on: https://chromium-review.googlesource.com/552550
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46323}
There are very few cases where OSR code can be re-used, and where the
function won't be non-concurrently optimized after OSR has happened.
Maintaining the OSR code cache is unnecessary complexity, and caching
OSR prevents us from e.g. seeding the optimizer with the actual OSR
values.
So, this patch removes it.
Change-Id: Ib9223de590f35ffc1dc2ab593b7cc9fe97dde4a6
Reviewed-on: https://chromium-review.googlesource.com/552637
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46306}
icu-case-mapping was shipped a few months ago. By dropping
the flag, unibrow's case conversion code won't be included
by default because V8_INTL_SUPPORT is on by default.
BUG=v8:4477, v8:4476
TEST=test262/{built-ins,intl402}/Strings/*, webkit/fast/js/*,
mjsunit/string-case, intl/general/case*
Cq-Include-Trybots: master.tryserver.v8:v8_linux_noi18n_rel_ng
Change-Id: I78be9cc64b4588bc5af79ecbbadf93af6e84a1df
Reviewed-on: https://chromium-review.googlesource.com/534541
Commit-Queue: Jungshik Shin <jshin@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Daniel Ehrenberg <littledan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46304}
It appears we actually get a compile time boost, and sometimes a
runtime boost, at the cost of some reloc info growth.
Bug:
Change-Id: I1d1dc48f364e6611f895ebd00f86451199dd8626
Reviewed-on: https://chromium-review.googlesource.com/544713
Commit-Queue: Mircea Trofin <mtrofin@chromium.org>
Reviewed-by: Brad Nelson <bradnelson@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46302}
V8 now takes care of recording a module's status, as proposed
in https://github.com/tc39/ecma262/pull/916.
R=adamk@chromium.org
Bug: v8:1569, chromium:594639
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: Id884f1c817e1dc3eea79a5d5a7f5cd996db1dbb0
Reviewed-on: https://chromium-review.googlesource.com/548500
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46293}
This implements the changes proposed at
https://github.com/tc39/ecma262/pull/916.
The API will be extended in a follow-up CL.
R=adamk@chromium.org
Bug: v8:1569
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I79476b5b674c924fea390dff1b9bee7f86a111c6
Reviewed-on: https://chromium-review.googlesource.com/544970
Reviewed-by: Adam Klein <adamk@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46289}
Removes from CL https://codereview.chromium.org/2929853003 code to fix
histogram timers in class WasmCompilationUnit. This was done because
the CL was reverted due to errors caused by background compiles that
updated UMA histogram timers.
The goal of this CL is to reland the remaining portion of the reverted
CL.
Bug:v8:6361
Change-Id: Ic03ceb118734bd55c463a843521bcd5b09342afe
Reviewed-on: https://chromium-review.googlesource.com/550196
Reviewed-by: Mircea Trofin <mtrofin@chromium.org>
Commit-Queue: Karl Schimpf <kschimpf@google.com>
Cr-Commit-Position: refs/heads/master@{#46268}
This reverts commit 3c04ee6d4e.
Reason for revert: Speculative revert for OOMs in:
https://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20shared/builds/18563
Original change's description:
> [heap] Make aborting compaction a fatal
>
> Last attempt failed because of too agressive growing and no proper GC
> scheduling.
>
> Also refactor live object iteration on the way to avoid a branch when we
> know that we will succeed.
>
> Bug: chromium:651354
> Change-Id: I8f52cbc79ac293989fb7c29c492d4ae4afe5ebe4
> Reviewed-on: https://chromium-review.googlesource.com/544829
> Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#46245}
TBR=ulan@chromium.org,mlippautz@chromium.org
Change-Id: Ifb9fd0c873c4ec6d4bd895f2978849cc2a223a05
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:651354
Reviewed-on: https://chromium-review.googlesource.com/549324
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46247}
Last attempt failed because of too agressive growing and no proper GC
scheduling.
Also refactor live object iteration on the way to avoid a branch when we
know that we will succeed.
Bug: chromium:651354
Change-Id: I8f52cbc79ac293989fb7c29c492d4ae4afe5ebe4
Reviewed-on: https://chromium-review.googlesource.com/544829
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46245}
Uninitialized property accesses are replaced with SOFT deopts in
TurboFan, but uninitialized JSCall nodes are not, and instead they
just stick around and are also not being inlined because the heurstic
in TurboFan doesn't consider those candidates since their call frequency
is below the threshold. This unifies the behavior and also replaces
uninitialized calls with SOFT deopts, addressing some inconsistency in
optimization behavior as discovered by Brian White of Node for example
here: https://twitter.com/mscdexdotexe/status/879005026202640385R=jarin@chromium.org
BUG=v8:4551, v8:5267
Review-Url: https://codereview.chromium.org/2956843002
Cr-Commit-Position: refs/heads/master@{#46231}
This is a fix to https://codereview.chromium.org/2929853003 that got
reverted. The DCHECK checked to see that it was not in a background
thread. While this is a property we want for v8, it is also used
by blink, and blink violates this property.
Therefore, this CL removes the DCHECK for now.
BUG=v8:6361
Review-Url: https://codereview.chromium.org/2961443002
Cr-Commit-Position: refs/heads/master@{#46190}
