1687634 - Cleanup FindReservedWord

Status: RESOLVED FIXED

(Core :: JavaScript Engine, task, P3)

Description

[Tooru Fujisawa [:arai]]

After bug 1687428, we use TaggedParserAtomIndex almost everywhere, and
all reserved words are well-known atom.
So we don't have to look into the atom content, but just check the index.

Comment 1

[Tooru Fujisawa [:arai]]

WIP patch results in 2.4% improvement in parsemark

Comment 2

[Tooru Fujisawa [:arai]]

Attachment: Bug 1687634 - Part 1: Remove unused JSLinearString variant of IsKeyword/FindReservedWord.

Depends on D103546

Comment 3

[Tooru Fujisawa [:arai]]

Attachment: Bug 1687634 - Part 2: Add TaggedParserAtomIndex-specific variant of FindReservedWord.

Instead of looking into ParserAtomEntry content, compare tagged index against
well-known atoms.

visibility parameters are removed because they're unused in practice.

Character-based FindReservedWord used by TokenStreamSpecific::identifierName is
kept, because replaing it with atom-index based one results in ~0.7% regression
across parsemark.

Depends on D103547

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/0883c747def3
Part 1: Remove unused JSLinearString variant of IsKeyword/FindReservedWord. r=nbp
https://hg.mozilla.org/integration/autoland/rev/1d18648713ad
Part 2: Add TaggedParserAtomIndex-specific variant of FindReservedWord. r=nbp

Comment 5

[Dorel Luca [:dluca]]

https://hg.mozilla.org/mozilla-central/rev/0883c747def3
https://hg.mozilla.org/mozilla-central/rev/1d18648713ad