SpiderMonkey Newsletter #9

46 views
Skip to first unread message

Jan de Mooij

unread,
Feb 26, 2021, 12:17:30 PMFeb 26
to dev-platform
(This newsletter is also available on our blog
<https://mozilla-spidermonkey.github.io/blog/2021/02/26/newsletter-9.html>.)

SpiderMonkey is the JavaScript engine used in Mozilla Firefox. This
newsletter gives an overview of the JavaScript and WebAssembly work we’ve
done as part of the Firefox 86 and 87 Nightly release cycles.

If you like these newsletters, you may also enjoy Yulia’s Compiler Compiler
live stream
<https://developer.mozilla.com/events/compiler-compiler-yulia-startsev/>.
🏆 New contributors

- Jonatan fixed a subtle bug
<https://bugzilla.mozilla.org/show_bug.cgi?id=1683784> in how our error
messages got printed around private fields.

👷🏽‍♀️ JS features

- Yulia attended the TC39 meeting in January and took notes
<https://github.com/codehag/TC39-news/blob/master/meetings/2021/tc39-01-2021.md>
.
- André implemented
<https://bugzilla.mozilla.org/show_bug.cgi?id=1670044> support for the
‘Arbitrary module namespace identifier names’ spec change
<https://github.com/tc39/ecma262/pull/2154>.
- Jan added support for large ArrayBuffers to structured cloning
<https://bugzilla.mozilla.org/show_bug.cgi?id=1686445>, JITs
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687441>, typed arrays
<https://bugzilla.mozilla.org/show_bug.cgi?id=1686936> and the JSAPI
<https://bugzilla.mozilla.org/show_bug.cgi?id=1674777>. This is now
available behind a pref in Nightly and will let us support
<https://bugzilla.mozilla.org/show_bug.cgi?id=1673619> WebAssembly
memories larger than 2 GB on 64-bit platforms.
- André changed <https://bugzilla.mozilla.org/show_bug.cgi?id=1672862>
typed array elements to appear configurable, following a recent spec
change <https://github.com/tc39/ecma262/pull/2164>.
- Jason optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1683746>
the bytecode generated for private fields.
- André updated <https://bugzilla.mozilla.org/show_bug.cgi?id=1685481>
timezone information twice
<https://bugzilla.mozilla.org/show_bug.cgi?id=1689294>.
- Yulia updated <https://bugzilla.mozilla.org/show_bug.cgi?id=1689499>
the Top Level Await implementation to fix a spec bug.
- Yulia reviewed <https://bugzilla.mozilla.org/show_bug.cgi?id=1693261>
an incoming potential change to the TLA specification.
- André enabled <https://bugzilla.mozilla.org/show_bug.cgi?id=1654116>
the Intl.DisplayNames constructor by default.

⚡ WebAssembly

- Lars and Yury made
<https://bugzilla.mozilla.org/show_bug.cgi?id=1625130> many changes for
SIMD support.
- Ioanna and Asumu from Igalia implemented
<https://bugzilla.mozilla.org/show_bug.cgi?id=1335652> support for the
exception handling proposal
<https://github.com/WebAssembly/exception-handling> in the Wasm Baseline
compiler.
- Lars fixed <https://bugzilla.mozilla.org/show_bug.cgi?id=1678542>
various issues to support the Ion backend for Wasm on ARM64.
- Luke Wagner added
<https://bugzilla.mozilla.org/show_bug.cgi?id=1647791> telemetry for
duplicate imports, to see if this could be disallowed in the future.
- Tom improved <https://bugzilla.mozilla.org/show_bug.cgi?id=1687063>
error messages when trying to consume a Response for Wasm.
- Ryan added <https://bugzilla.mozilla.org/show_bug.cgi?id=1686755>
support for optional parameters in the WebAssembly.Table API.
- Ryan continues to work on implementing the draft wasm GC specification.

🧹 Garbage Collection

- Jon improved <https://bugzilla.mozilla.org/show_bug.cgi?id=1677765>
arena unmarking to happen concurrently at the start of GC.
- Jon changed <https://bugzilla.mozilla.org/show_bug.cgi?id=1681533>
nursery collections during major GCs to happen only when necessary.
- Steve added <https://bugzilla.mozilla.org/show_bug.cgi?id=1658866>
counters for various string operations.
- Steve used <https://bugzilla.mozilla.org/show_bug.cgi?id=1682947>
MADV_FREE_REUSE on macOS when using decommitted memory to improve the
accuracy of memory usage data.
- Jon changed <https://bugzilla.mozilla.org/show_bug.cgi?id=1685128>
idle collections to collect and shrink the nursery when it’s underused.
- Jon moved <https://bugzilla.mozilla.org/show_bug.cgi?id=1686219> chunk
metadata from the end of the chunk to the start, so that it can be accessed
more efficiently, and cleaned up
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687956> the chunk code.
- Jon optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1689394>
the tracing/marking code more to fix a telemetry regression.
- Steve optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1691361>
nursery collections by not adding strings and BigInts without children to
the fixup list. He also split
<https://bugzilla.mozilla.org/show_bug.cgi?id=1692359>
collectToFixedPoint in separate object and string parts to improve GC
statistics.

❇️ Stencil

Stencil <https://bugzilla.mozilla.org/show_bug.cgi?id=1601332> is our
project to create an explicit interface between the frontend (parser,
bytecode emitter) and the rest of the VM, decoupling those components. This
lets us improve web-browsing performance, simplify a lot of code and
improve bytecode caching.

- Nicolas made <https://bugzilla.mozilla.org/show_bug.cgi?id=1668361#c4>
shell tests a lot faster by using cached bytecode for self-hosted code.
- Arai landed <https://bugzilla.mozilla.org/show_bug.cgi?id=1674306> a
large number of changes to improve performance of the Stencil XDR (bytecode
serialization) format.
- Ted enabled <https://bugzilla.mozilla.org/show_bug.cgi?id=1667804>
stencil-mvp by default. This switched bytecode caching to use the new
Stencil-based format and fixes performance-cliffs related to GC blocking
off-thread parsing.
- Matthew later removed
<https://bugzilla.mozilla.org/show_bug.cgi?id=1690115> the stencil-mvp
pref to unblock more optimization work.
- Ted optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1687338>
ScriptSource to XDR encoding and main-thread GC function allocations
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687602>.
- Arai optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1688534>
main-thread GC allocations for object literals.
- Arai changed <https://bugzilla.mozilla.org/show_bug.cgi?id=1660275>
the frontend to use a snapshot of scope information.
- Ted cleaned up <https://bugzilla.mozilla.org/show_bug.cgi?id=1690943>
code for missing class constructor by synthesizing these functions in the
parser instead of cloning self-hosted functions.
- Arai landed <https://bugzilla.mozilla.org/show_bug.cgi?id=1689892> many
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687428> performance and
memory usage improvements for parser atoms.
- Ted added <https://bugzilla.mozilla.org/show_bug.cgi?id=1691612> a
public API for Stencil data structures. This will be used in the browser to
improve bytecode caching more.

🚀 JIT

- Matthew implemented
<https://bugzilla.mozilla.org/show_bug.cgi?id=1680602> support for
compiling async functions and generators with Warp. Enabling this by
default showed a 4% win on Ares6
<https://bugzilla.mozilla.org/show_bug.cgi?id=1682623#c5>.
- Iain improved the bailout code and added
<https://bugzilla.mozilla.org/show_bug.cgi?id=1673497> a mechanism to
catch bailout loops in debug builds. Bailout loops can result in
performance cliffs, and this work caught
<https://bugzilla.mozilla.org/show_bug.cgi?id=1686207> various
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687672> issues.
- Nicolas enabled <https://bugzilla.mozilla.org/show_bug.cgi?id=1679922>
use of the JSCVT instruction on Apple ARM64 hardware.
- Caroline added <https://bugzilla.mozilla.org/show_bug.cgi?id=1672787>
filtering for the JSON output used by the CacheIR health report tool, to
get it down to reasonable size for complex websites.
- André, Tom and Jan cleaned up
<https://bugzilla.mozilla.org/show_bug.cgi?id=1682767> and
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687255> deleted
<https://bugzilla.mozilla.org/show_bug.cgi?id=1686692> even
<https://bugzilla.mozilla.org/show_bug.cgi?id=1689990> more
<https://bugzilla.mozilla.org/show_bug.cgi?id=1657367> code
<https://bugzilla.mozilla.org/show_bug.cgi?id=1683093> following the
IonBuilder and TI removal in Firefox 85.
- Iain added <https://bugzilla.mozilla.org/show_bug.cgi?id=1674143>
CacheIR and Warp optimizations for Math.min/max spread calls with arrays.
- André added CacheIR and Warp optimizations for
Object.prototype.toString
<https://bugzilla.mozilla.org/show_bug.cgi?id=1687229>, Atomics functions
<https://bugzilla.mozilla.org/show_bug.cgi?id=1638295> with BigInt64Array
and BigUint64Array, and various
<https://bugzilla.mozilla.org/show_bug.cgi?id=1692517> TypedArray,
DataView and RegExp getters.
- Tom added <https://bugzilla.mozilla.org/show_bug.cgi?id=1666009> a
mechanism to inline certain self-hosted functions in Warp even if they’re
normally too big to be considered for inlining.
- André optimized <https://bugzilla.mozilla.org/show_bug.cgi?id=1679750>
BigInt operations in Warp by supporting them in the CacheIR-to-MIR
transpiler.
- Iain is making <https://bugzilla.mozilla.org/show_bug.cgi?id=1688033>
progress on replacing the heavyweight lazy-arguments optimization with a
simpler implementation based on scalar replacement in the JIT backend.

📚 Miscellaneous Documentation

- Ted added <https://bugzilla.mozilla.org/show_bug.cgi?id=1685958> a
SpiderMonkey overview page
<https://firefox-source-docs.mozilla.org/js/index.html> to Firefox
source docs.
- Jon added <https://bugzilla.mozilla.org/show_bug.cgi?id=1686772> GC
documentation <https://firefox-source-docs.mozilla.org/js/gc.html>.
- Iain added <https://bugzilla.mozilla.org/show_bug.cgi?id=1690092> a
high-level overview
<https://firefox-source-docs.mozilla.org/js/MIR-optimizations/index.html>
of MIR optimizations.
- Matthew improved <https://bugzilla.mozilla.org/show_bug.cgi?id=1682846>
the JIT frame layout comment.

Old bugs

- Jan changed <https://bugzilla.mozilla.org/show_bug.cgi?id=944846>
Number.prototype.{toFixed,
toExponential, toPrecision} to use the double-conversion code in MFBT
instead of dtoa. This fixes a toExponential rounding issue and is also
faster in most cases.
- Calixte Denizet fixed
<https://bugzilla.mozilla.org/show_bug.cgi?id=1676708> a date parsing
bug for GMT values starting with ‘00’.
- arai fixed <https://bugzilla.mozilla.org/show_bug.cgi?id=1579792> an
old bug where function stringification could include too much source code.

Modern C++

- Chris Peterson replaced
<https://bugzilla.mozilla.org/show_bug.cgi?id=1684092> MOZ_MUST_USE with
the C++17 [[nodiscard]] attribute.
- André replaced <https://bugzilla.mozilla.org/show_bug.cgi?id=1687463>
calls to various array-related functions in MFBT with STL equivalents.

Other

- Jason fixed <https://bugzilla.mozilla.org/show_bug.cgi?id=1671391>
debugger regressions from generator optimizations.
- Jan moved <https://bugzilla.mozilla.org/show_bug.cgi?id=1693483> the
TypeDescr for typed objects from the ObjectGroup to typed objects to
prepare for upcoming object layout changes.
- Julian improved <https://bugzilla.mozilla.org/show_bug.cgi?id=1686658>
the IONFLAGS logging output.
- Jan renamed <https://bugzilla.mozilla.org/show_bug.cgi?id=1592712> the
confusingly-named IfEq and IfNe bytecode instructions to JumpIfFalse and
JumpIfTrue.
- Jon added <https://bugzilla.mozilla.org/show_bug.cgi?id=1691004> a
flag to make it easier to run shell tests with a non-standard ‘GC zeal’.
Reply all
Reply to author
Forward
0 new messages