Quasi Fused Multiply-Add and Quasi Fused Multiply-Subtract performs, on floats, a + b * c and a - b * c respectively.
When there is only a single rounding, it is a fused operation. Quasi in this case means that the result can either be fused or not fused (two roundings), depending on hardware support.
It is tricky to write the test because we need to calculate the expected value, and there is no easy way to express fused or unfused operation in C++, i.e.
we cannot confirm that float expected = a + b * c will perform a fused or unfused operation (unless we use intrinsics).
Thus in the test we have a list of simple checks, plus interesting values that we know will produce different results depending on whether it was fused or not.
The difference between 32x4 and 64x2 qfma/qfms is the type, and also the values of b and c that will cause an overflow, and thus the intermediate rounding will affect the final result.
The same array can be copy pasted for both types, but with a bit of templating we can avoid that duplication.
Change-Id: I0973a3d28468d25f310b593c72f21bff54d809a7
Bug: v8:9415
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1779325
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63878}
Disabling the RegExp compilation cache comes with performance implications,
and it doesn't seem to be necessary for debugging.
Bug: chromium:992277
Change-Id: I24841f4814bcacb18a3968c37490f201c0c1ccac
Cq-Include-Trybots: luci.chromium.try:linux-rel,win7-rel
Cq-Include-Trybots: luci.v8.try:v8_linux_gc_stress_dbg,v8_linux64_gc_stress_custom_snapshot_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1805637
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63876}
WASM only supports Little-endian byte ordering and we need a mechanism to
reverse the ordering efficiently on Big-endian machines.
Up until now this was done using TF graphs within wasm-compiler.
The new approach allows for having more machine level optimizations
by introducing the new "kSimd128ReverseBytes" opcode which gets executed
only on Big-endian machines.
Change-Id: I63c6c3c42ca9ff9d9b2af2d45070a70cf1b3cefc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1803494
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Milad Farazmand <miladfar@ca.ibm.com>
Cr-Commit-Position: refs/heads/master@{#63875}
On-heap typed arrays contain HeapObject value in |base_pointer| field
and an offset in |external_pointer| field. When pointer compression is
enabled we want to combine decompression with the offset addition.
In order to do that we add an isolate root to the external_pointer value
and therefore the data pointer computation can is a simple addition of
a (potentially sign-extended) |base_pointer| loaded as Tagged_t value
and an |external_pointer| value.
Bug: v8:9706
Change-Id: Id5c546c353c81fb25e3598921bc78165d10a9c44
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807369
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63874}
We see crashes in the wild, however the minidump does not contain enough
information for properly investigating this. This is temporary code
that should be deleted as soon as possible.
Bug: chromium:1004365
Change-Id: I25e119d891e03f980d63f5d559fcb4429b574cb1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809373
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63873}
The serialization step in the pipeline gets an initial zone, and thus
far, it's allocated all of it's hinting information in that zone.
However, much of this comes from stepping into calls and
walking the bytecode of the called function. Once we finished
recursing into a call, we should be able to throw all those
hints away -- they've served their purpose, and the "output"
of their work is a set of new objects made visible to the broker.
Therefore, we should create and destroy a child zone.
On a run of typescript, this reduces absolute max (high water mark)
allocation of the serialization phase from 10 MB to 5 MB.
Bug: v8:7790
Change-Id: Icbb35abed28b1a924328541df82be23594152a8f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1800570
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Michael Stanton <mvstanton@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63872}
This introduces {GetContainingWasmFunction} to replace the above method,
since calculating offsets into the wire bytes is independent of the
concrete module object and hence only needs the shared decoded module.
R=clemensh@chromium.org
BUG=v8:6847
Change-Id: I145d527506289686653979dbb135480cc42ea4c8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809369
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63870}
... and do not artificially add at least one dummy.js file.
Until this CL we used to ensure the existence of at least one 'extra
native' by appending a short dummy.js file if v8_extra_library_files
was empty. This file is uselessly compiled and run at startup.
This CL removes that mechanism along with a minor tweak to handle an
empty natives blob.
Bug: v8:9736
Change-Id: I05bf7c54380e77a9105d5c1a1a76983c692faf60
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809372
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63868}
Separate creating Wasm translations from reporting them to an agent.
This is done in order to support multiple connected sessions.
Previously connecting more than one agent would fail assertion in debug
mode and overwrite translation objects over and over
(and potentially do something worse) in release mode.
Bug: v8:9725
Change-Id: I13fde5ebf6e64e7268eb6870f9c21ac9a5bed81e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807273
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Ingvar Stepanyan <rreverser@google.com>
Cr-Commit-Position: refs/heads/master@{#63867}
During deletion in the write-back phase, we do not necessarly need
to check with HasOwnProperty before calling DeleteProperty. Since this
is observable behavior when using proxies, we remove the HasOwnProperty
check to mirror more closely what Spidermonkey does.
R=jgruber@chromium.org
Change-Id: I67768b1d8b13e695b956b2fa0624dcb28fd0664d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809366
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Simon Zünd <szuend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63866}
In the mjsunit/wasm/asm-wasm-f32 test, {IsValid} caused 10% of the
overall runtime. In absolute numbers, {IsValid} wasm called more
than 6.000.000.000 times. I moved the calls to {IsValid} from
before-use to after-modification. I removed the calls in {Combine}
and {Remove}, because these operations cannot create an invalid
CPURegList.
These changes reduce the number of calls to {IsValid} to about
100.000.000 in the test mentioned above, and saves for that test
about 10 seconds out of before 2 minutes absolute runtime.
R=v8-arm-port@googlegroups.com
Bug: v8:9396
Change-Id: If3059f471e423405ec6f34ddef89e314dee1cbaf
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1801851
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Martyn Capewell <martyn.capewell@arm.com>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63864}
This is an unmodified reland of
60624b5692. Nosnap bots do not block
LKGR any more: https://crbug.com/v8/9737#c10.
Original change's description:
> Disallow nullptr arguments for {CopyChars}
>
> This allows to remove special casing for the {count == 0} case, which
> was needed because {memmove} does not accept {nullptr} arguments even
> if the {count} is zero.
>
> R=leszeks@chromium.org
>
> Bug: v8:9396
> Change-Id: Iaef3cdbbffa74c2ba1c4e4501dafd943282cbcd9
> Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807366
> Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63838}
TBR=leszeks@chromium.org
Bug: v8:9396
Change-Id: I6ab13575f13df060b450ff105e4b9db516671dcf
Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809365
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63863}
This introduces {GetWasmFunctionOffset} to replace the above method,
since calculating offsets into the wire bytes is independent of the
concrete module object and hence only needs the shared decoded module.
R=clemensh@chromium.org
BUG=v8:6847
Change-Id: I6818de4589e26dd8f69dfb71d15bbca127c7ee3e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809368
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63862}
This introduces {DisassembleWasmFunction} to replace the above method,
since disassembling a function is independent of the concrete module
object and hence can be done for shared decoded modules.
R=clemensh@chromium.org
BUG=v8:6847
Change-Id: I5abea2a1381a9b8d3717a55d0b2b937dfbbafefd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809359
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63860}
This is a first step towards being able to share the same script for
multiple {WasmModuleObject} objects. In general it should be possible
for the inspector to debug (i.e. disassemble, set breakpoints) multiple
modules if they all have the same URL (and the same wire bytes). These
are the same conditions under which we can canonicalize the modules to
be based on the same underlying {NativeModule} as well. Hence it makes
sense to establish a link from {Script} to {NativeModule} in this CL.
Subsequent CLs will eventually deprecate the {wasm_module_object} field.
R=clemensh@chromium.org
BUG=v8:6847
Change-Id: I5cfb617e18d9b06682e6437b2a2146ea5665c1c6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807371
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63859}
This hopefully fix the regression test from c693e005a7
Bug: v8:1004766
Change-Id: If3c554070af6b795e4b3f99cd592a62453028874
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809363
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Victor Gomes <victorgomes@google.com>
Cr-Commit-Position: refs/heads/master@{#63858}
This is an unmodified reland of
9febc505bd. Nosnap bots do not block
LKGR any more: https://crbug.com/v8/9737#c10.
Original change's description:
> Reland "Remove all custom CopyCharsUnsigned implementations"
>
> This is a reland of 5d8c489000
>
> Original change's description:
> > Remove all custom CopyCharsUnsigned implementations
> >
> > It's unclear whether the custom implementation have any advantage over
> > the standard library one's.
> > Since we update our toolchain and standard library regularly, it might
> > well be the case that the custom implementations are slower by now.
> >
> > Thus this CL removes all {CopyCharsUnsigned} implementations and
> > implements {CopyChars} generically using {std::copy_n}.
> >
> > Note that this does not touch the {MemMove} and {MemCopy} functions
> > yet, as we have seen regressions when trying to remove them before
> > (https://crbug.com/v8/8675#c5).
> >
> > R=leszeks@chromium.org
> >
> > Bug: v8:9396
> > Change-Id: I97a183afebcccd2fbb567bdba02e827331475608
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1800577
> > Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
> > Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#63808}
>
> Bug: v8:9396
> Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
> Change-Id: I9cd754ebe6b802bb4aabd6d2a448de41da040874
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807357
> Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63823}
TBR=leszeks@chromium.org
Bug: v8:9396
Change-Id: I793524d76b8b9c93d2a98c73e8d72967880fe1cf
Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809362
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63857}
ArrayBuffer tracking has landed, turning on GrowMemory for Shared
WebAssembly.memory on by default. Enable all variants of tests based
on the new implementation.
Bug: v8:8564, v8:9221, v8:8832
Change-Id: I0ff8688636303896450b788b2ff5a7268d386050
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1808106
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63855}
This reverts commits 9febc505bd
(along with followup commit 60624b5692).
Reason for revert: Breaks win32 nosnap shared, blocking lkgr & roll:
https://ci.chromium.org/p/v8/builders/ci/V8%20Win32%20-%20nosnap%20-%20shared/35145
nosnap bots may be deprecated, but as long as they're in LKGR
we need to mind them.
Original change's description:
> Reland "Remove all custom CopyCharsUnsigned implementations"
>
> This is a reland of 5d8c489000
>
> Original change's description:
> > Remove all custom CopyCharsUnsigned implementations
> >
> > It's unclear whether the custom implementation have any advantage over
> > the standard library one's.
> > Since we update our toolchain and standard library regularly, it might
> > well be the case that the custom implementations are slower by now.
> >
> > Thus this CL removes all {CopyCharsUnsigned} implementations and
> > implements {CopyChars} generically using {std::copy_n}.
> >
> > Note that this does not touch the {MemMove} and {MemCopy} functions
> > yet, as we have seen regressions when trying to remove them before
> > (https://crbug.com/v8/8675#c5).
> >
> > R=leszeks@chromium.org
> >
> > Bug: v8:9396
> > Change-Id: I97a183afebcccd2fbb567bdba02e827331475608
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1800577
> > Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
> > Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#63808}
>
> Bug: v8:9396
> Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
> Change-Id: I9cd754ebe6b802bb4aabd6d2a448de41da040874
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807357
> Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63823}
TBR=leszeks@chromium.org,clemensh@chromium.org
Change-Id: Ic53ab2293d5dc7722a1121d1aa1159328a6ed8f5
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9396
Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1808035
Reviewed-by: Adam Klein <adamk@chromium.org>
Commit-Queue: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63854}
Runtime_CreateRegExpLiteral pretends to handle the case where we
already have a boilerplate but the code just passes an empty handle
to JSRegExp::Copy. This CL replaces the code with a CHECK that the
boilerplate doesn't exist.
A few mostly cosmetic changes along the way.
Change-Id: I856c77c1cdf58f09e920b773b66b037c5178edc7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1809357
Reviewed-by: Peter Marshall <petermarshall@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63852}
Fix m(ax|in)imumFractionDigits of Intl.NumberFormat
resolvedOptions are set to 0. For example, currency
instance for CPY or KRW.
Bug: chromium:1003748
Change-Id: Ia1963d8d070b066bd5afa61f8c4716a21450af05
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807742
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Frank Tang <ftang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63850}
The promise file is too big so I am splitting it in several CLs.
This is the first one.
TNodified:
* AllocateAndInitJSPromise (three versions)
* PerformPromiseThen
* AllocateJSPromise
Bug: v8:6949
Change-Id: I57ae8de3f929c00a9127ea4be51ffe7703b44959
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807370
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63849}
This moves PointerToSeqStringData from CSA to IntlBuiltinsAssembler
since that the only place it's used and converts a few Node*s to TNodes.
Bug: v8:6949, v8:9396
Change-Id: I2511c9f807fa0bc3101dd531f6724a170fed6bd9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807372
Commit-Queue: Dan Elphick <delphick@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63847}
One Node* remains since it is crashing when TNodified since it
has the empty Context. When this happens, it means the Context
might not be needed. In this case it's hard to see if that's the
case since it is propagated all over the place. Created a TODO
so we can solve it later, when TNodification is more complete.
Bug: v8:6949
Change-Id: I7db8c507689e30779f102a8272e1b13ecdc3e5e9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807274
Reviewed-by: Mythri Alle <mythria@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63846}
Port 1dd791fca2
Original Commit Message:
Uses templates to dispath the allocation flag statically.
Change-Id: I1d6a0f2c6ca04ac0f03afe392584e9f1e8dcfb9a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1806680
Auto-Submit: Mu Tao <pamilty@gmail.com>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63842}
- Makes accessor and iteration methods on Arguments and derived
classes const.
Bug: v8:9429,v8:9396
Change-Id: I47b3d95ab72e689327a0d7b6a36a08b4e63f6d95
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1803336
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63841}
This allows to remove special casing for the {count == 0} case, which
was needed because {memmove} does not accept {nullptr} arguments even
if the {count} is zero.
R=leszeks@chromium.org
Bug: v8:9396
Change-Id: Iaef3cdbbffa74c2ba1c4e4501dafd943282cbcd9
Cq-Include-Trybots: luci.v8.try:v8_linux64_ubsan_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807366
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63838}
The new indicator behaves like verbose without
printing the names of tests that passed. Also
a new option (--ci-test-completion=/path) was
added to represent a file where we can collect
test completion messages.
Bug: v8:9146
Change-Id: I0f1bbef4036a3019b60b094687b734d3d33a5915
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1806916
Reviewed-by: Tamer Tas <tmrts@chromium.org>
Commit-Queue: Liviu Rau <liviurau@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63837}
The element segment encoding in the bulk memory proposal changed
recently. With this CL the V8 implementation gets up to date again.
R=thibaudm@chromium.org
Bug: v8:9658
Change-Id: I4f45d04369400356a6f3aaed9570c7870f5f97bd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1778022
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63836}
For unknown reasons, WasmInterpreterInternals was a zone object. However
WasmInterpreterInternals indirectly owns a global handle and a
unique_ptr, both for the interpreter stack. As a zone object,
WasmInterpreterInternals is never destructed, and therefore never frees
the unique_ptr. With this CL I make WasmInterpreterInternals a normal
object and allocate it into a unique_ptr, so that it gets destructed
properly.
R=mstarzinger@chromium.org
Bug: chromium:1000610
Change-Id: Ie08c5627393a434521f5c32702bf9945db2c7811
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807361
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63835}
This will give us much more test coverage and fuzzer coverage.
R=mstarzinger@chromium.org
Bug: v8:9477
Change-Id: Iad76c2b5b8c7a29b4168bbefa38bac7d92a30599
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807367
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63832}
The {CountTrailingZeros} function is at least on one hot code path,
and there it causes significant overhead. With this CL I just call the
base::bit:: version of {CountTrailingZeros} directly. This allows the
compiler to compile it to a single hardware instruction.
R=v8-arm-ports@googlegroups.com
Bug: v8:9396
Change-Id: I81eccc5fce9b9856d41c503bd1e4a07287eb6e1e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1803648
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Martyn Capewell <martyn.capewell@arm.com>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63831}
TNodifies most of ic-accessor-assember. Most of the remaining Node* are
because of the Parameters.
Bug: v8:6949, v8:9396
Change-Id: Ife9fd96c5e46dee02fdc60e5825562d7ae89f8f9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1803634
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63830}
Adding a %SimulateNewspaceFull runtime function speeds up this test
from 7m21s to 0.3s (on arm.optdebug with --jitless).
Bonus content:
- speed up mjsunit/md5 by 23x (5m25s -> 7.5s)
- speed up mjsunit/string-replace-gc by 8x (1m37s -> 12s)
Bug: v8:9700, v8:9396
Change-Id: Id00d0b83b51192edf1d5493b49b79b5d76e78087
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1807355
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63829}
