Change-Id: I31cf082abae120dc4f5aa7eadce6513e933d942b
Reviewed-on: https://chromium-review.googlesource.com/1178042
Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55191}
Use CheckBounds and reduce the number of checks required to sanitize the
indices for DataView accesses in optimized code. Also constant-fold the
[[ByteLength]] if the DataView is a known compile-time constant (similar
to what we do for TypedArrays already). This further improves performance
of DataViews by 2-7% depending on the exact test case.
With this change DataView and TypedArray accesses themselves are mostly
on par performance wise.
Bug: chromium:225811
Change-Id: I6838339108b8a4dcf9b13ddecab40f1c3632967c
Reviewed-on: https://chromium-review.googlesource.com/1179741
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55190}
This reduced the number of targets depending on assembler.h
from ~900 to ~350.
Bug: v8:8054
Change-Id: I74ae2ce7a4b27791d0ee25542ee0b2175bedf5f7
Reviewed-on: https://chromium-review.googlesource.com/1174534
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55188}
Now that we pass in assembler options during builtin setup, we can be
more precise in assembler configuration.
Isolate-independent generation is only requested for isolate-independent
builtins. And pc-relative jumps additionally need a valid code range.
Bug: v8:6666
Change-Id: I64dfb414549a2f1e87610244c48d9405e63a1b12
Reviewed-on: https://chromium-review.googlesource.com/1177707
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55187}
This way the snapshot will contain only old space objects.
BUG=v8:7308
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng
Change-Id: Ib19642795b1e8c7cbeda5162be86909c1a777767
Reviewed-on: https://chromium-review.googlesource.com/1177384
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Marja Hölttä <marja@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55186}
This moves write-barrier for writes into code to heap-write-barrier.h
and adds four new functions:
- WriteBarrierForCode(host, rinfo, object) - combined generational
and marking write barrier.
- WriteBarrierForCode(host) - combined write barrier that rescans
all pointers in the host (former RecordWritesIntoCode).
- GenerationalWriteBarrierForCode.
- MarkingWriteBarrierForCode.
Bug: v8:8054,v8:7490
Change-Id: Ib1e07cfa1d5998fca2fa44e2ad08c52305f1373f
Reviewed-on: https://chromium-review.googlesource.com/1174436
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55185}
When printing the number and size of the instance types in each
snapshot, break out each line into the number for each AllocationSpace.
(This removes the specific RO_SPACE path that just broke out its stats).
Change-Id: I9d5bcecc4f0e8e1ec5c83409f54164f9bf83c5be
Reviewed-on: https://chromium-review.googlesource.com/1175826
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55184}
JSArrays and JSArrayBuffers are very different animals. As such,
split the js-array.h header into two parts.
R=ulan@chromium.org,mstarzinger@chromium.org
Bug: v8:5402
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng
Change-Id: I82f987ecea3e2e1ceaf8f8962a2b88165558c57e
Reviewed-on: https://chromium-review.googlesource.com/1177760
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55183}
This CL refactors the CodeAssembler::Branch(condition, true_body, false_body)
which was introduced by https://crrev.com/c/1175488, and this reduces snapshot by 864 bytes.
Change-Id: Ifde7d6f39bd7f265e71fef5bdcc6e69d8ab5be85
Reviewed-on: https://chromium-review.googlesource.com/1175488
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Kanghua Yu <kanghua.yu@intel.com>
Cr-Commit-Position: refs/heads/master@{#55179}
Previously, FormatDate implemented two functions:
(a) "DateTime Format Functions" as in ecma402/#sec-datetime-format-functions
(b) "FormatDateTime( dateTimeFormat, x )" as in ecma402/#sec-formatdatetime
This patch split FormatDate into two separate functions exactly as they are specified.
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Change-Id: Iae47f4d992f1767968e5a6df7d1a45a95bc33192
Reviewed-on: https://chromium-review.googlesource.com/1178886
Commit-Queue: Frank Tang <ftang@chromium.org>
Reviewed-by: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55177}
Don't compute the ToNumber conversion of long strings as it is
expensive.
Bug: v8:7790
Change-Id: Ief104f9229f7f633e31d26c98ed24b08a161525b
Reviewed-on: https://chromium-review.googlesource.com/1177719
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55168}
Fix 0027c83440
FLAG_runtime_stats was declared as Int but it was accessed as
pointer. On 64-bit architectures pointer access needs to be
eight byte aligned. This created unaligned memory access
on 64-bit architectures for those architectures that do not
support unaligned memory access.
TEST=mjsunit/wasm/table,cctest/test-accessors/AccessorIC
Change-Id: I569b69583de98b4c5fb7266f23f927bbd006966d
Reviewed-on: https://chromium-review.googlesource.com/1177601
Commit-Queue: Ivica Bogosavljevic <ibogosavljevic@wavecomp.com>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55166}
Forcing it seems unnecessary and hinders concurrent compilation.
Also turn a related condition into a CHECK.
Bug: v8:7790
Change-Id: Ie7d7e47e6bc2743d742cfe07e65436c1b8b994f3
Reviewed-on: https://chromium-review.googlesource.com/1177705
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55164}
Enforce both engine limitations and spec (http://asmjs.org/spec/latest/)
limitations on the size of asm.js heaps.
R=clemensh@chromium.org
CC=mstarzinger@chromium.org
Bug: chromium:873600
Change-Id: I104c23bbd0a9a7c494f97f8f9e83ac5a37496dfd
Reviewed-on: https://chromium-review.googlesource.com/1174411
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55163}
{AsyncCompileJob::FinishCompile} assumes that a context is set on the
isolate when it is called. This was not the case when it was called at
the end of streaming compilation.
R=mstarzinger@chromium.org
Bug: chromium:874525
Change-Id: Ia31cd9eb40598918262ff14606796027a6c80bc0
Reviewed-on: https://chromium-review.googlesource.com/1177392
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55161}
We really want to catch such errors. (std::optional throws an
exception in some of these cases.)
Change-Id: I7f8c3aff38aed1b3cbe52bcfb92062268fd3b201
Reviewed-on: https://chromium-review.googlesource.com/1177387
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55158}
This is a reland of 898f880aa7
Original change's description:
> [cpu-profiler] Turn on detailed line info for optimized code
>
> Cautiously turn on this flag by default to check the impact on
> performance bots. Could show minor regressions in old space and/or
> code_and_metadata memory buckets.
>
> Bug: v8:7983
> Change-Id: Ic4369cdb0231f4f88eada699da948e8bb48a25fd
> Reviewed-on: https://chromium-review.googlesource.com/1162234
> Commit-Queue: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#54908}
Bug: v8:7983
Change-Id: Iff53bc19c92f26f386948bd4a58e8dcf0493cf0d
Reviewed-on: https://chromium-review.googlesource.com/1177481
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55156}
Direct call to CloneFastJSArray is used to improve performance in that
case. Tests are also added.
Bug: v8:7980
Change-Id: Ifca34f3e182b776cd9862da8bf529fc13f6be9ed
Reviewed-on: https://chromium-review.googlesource.com/1172782
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Hai Dang <dhai@google.com>
Cr-Commit-Position: refs/heads/master@{#55154}
It looks like we are not putting doubles into the constant pool
on arm any time soon, so lets remove this feature.
A follow-up CL will remove the 64bit support from the constant pool.
Bug: v8:8054
Change-Id: I87865a9c252c8967b5ac71b4db5d8c3b93a0e705
Reviewed-on: https://chromium-review.googlesource.com/1174531
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55152}
This reverts commit cd28595110.
Reason for revert: No size improvements, but 5% regression on octane/NavierStrokes:
https://chromeperf.appspot.com/report?sid=ebcd213df609b4ff31b0b25b5c8aab9d0738ff46f623161bdcdbb6f594112761&start_rev=579365&end_rev=582538
Original change's description:
> [arm] Use constant pool for off-heap targets
>
> This CL puts off-heap targets (i.e. code addresses for embedded builtins)
> in the constant pool on ARM.
>
> We are landing this CL to evaluate impact on benchmarks and code size,
> and expect to revert it once we have gathered that data.
>
> Bug: v8:6666
>
> Change-Id: If4935a6fb162cd1ffb34489c6fa9630f10ca2c9f
> Reviewed-on: https://chromium-review.googlesource.com/1154924
> Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#54914}
TBR=sigurds@chromium.org,jgruber@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: v8:6666, chromium:872176
Change-Id: Icd0019c7f0c890d0eb09ddc2a74745c6b68db263
Reviewed-on: https://chromium-review.googlesource.com/1172482
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55151}
This fixes the bug where the reducer ignores a prototype that is not
initial. Tests are also added.
Bug: v8:8056
Change-Id: I428eed2d2790fffa22f67a051f7d1f1e4d3ce947
Reviewed-on: https://chromium-review.googlesource.com/1174542
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Hai Dang <dhai@google.com>
Cr-Commit-Position: refs/heads/master@{#55149}
This test would fail when running natively on arm64. If USE_SIMULATOR isn't
defined, then the assembler always generates a brk instruction, regarless of the
`enable_simulator_code` Assembler option.
Change-Id: Id56788ba60ab4a13857825ebf09bf55375250c9d
Reviewed-on: https://chromium-review.googlesource.com/1174712
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Pierre Langlois <pierre.langlois@arm.com>
Cr-Commit-Position: refs/heads/master@{#55148}
