1879225 - Update irregexp to new version

Status: RESOLVED FIXED

(Core :: JavaScript Engine, enhancement, P3)

Description

[Update Bot]

This update covers 3526 commits, including 112 new upstream commits I've never filed a bug on before. (They're the top 112.). Here are the overall diff statistics.

---

js/src/irregexp/imported/gen-regexp-special-case.cc | 48 ---
js/src/irregexp/imported/regexp-ast.cc | 17 +-
js/src/irregexp/imported/regexp-ast.h | 23 +-
js/src/irregexp/imported/regexp-bytecode-generator.cc | 2 +-
js/src/irregexp/imported/regexp-bytecode-peephole.cc | 4 +-
js/src/irregexp/imported/regexp-compiler-tonode.cc | 86 +++--
js/src/irregexp/imported/regexp-compiler.cc | 43 ++-
js/src/irregexp/imported/regexp-compiler.h | 8 +-
js/src/irregexp/imported/regexp-dotprinter.cc | 4 +
js/src/irregexp/imported/regexp-interpreter.cc | 84 ++--
js/src/irregexp/imported/regexp-interpreter.h | 13 +-
js/src/irregexp/imported/regexp-macro-assembler.cc | 70 ++--
js/src/irregexp/imported/regexp-macro-assembler.h | 23 +-
js/src/irregexp/imported/regexp-nodes.h | 18 +-
js/src/irregexp/imported/regexp-parser.cc | 287 +++++++++++------
js/src/irregexp/imported/regexp.h | 12 +-
js/src/irregexp/imported/special-case.h | 10 -
js/src/irregexp/moz.yaml | 4 +-
18 files changed, 417 insertions(+), 339 deletions(-)

Comment 2

[Update Bot]

I've submitted a try run for this commit: https://treeherder.mozilla.org/jobs?repo=try&revision=6521f00fc41d9d02c140c2bda198653522c5eaa4

Comment 3

[Update Bot]

Attachment: Bug 1879225 - Update irregexp to ec89cca93594ef6ae683d6608904cd5e30378115

Comment 5

[Update Bot]

All jobs completed, we found the following issues.

Known Issues (From Taskcluster):

• spidermonkey-sm-asan-linux64/opt (PsFtIrWrTJul21qalkfUag) - new failure not classified
• spidermonkey-sm-tsan-linux64/opt (Pi9gYabgR5iM3_kw6wtHTw) - new failure not classified
• spidermonkey-sm-temporal-linux64/debug (SlgOWmnmR-mHcCMV3NDTPQ) - new failure not classified
• spidermonkey-sm-fuzzilli-linux64/debug (T_7PHwBNRCCLbFH9HilYOA) - new failure not classified
• spidermonkey-sm-nojit-linux64/opt (W8KjzTGaQ4OKWT9KiOFuXQ) - new failure not classified
• spidermonkey-sm-nonunified-linux64/debug (VTQXn59lRrqonXVql3IGPg) - new failure not classified
• spidermonkey-sm-linux64-wasi-pbl/opt (YXIePf9ESOClTLc6Tnt4Kw) - new failure not classified
• spidermonkey-sm-plain-linux64/debug (ahkNh0MfScuIuvLCQ3jo6A) - new failure not classified
• hazard-linux64-shell-haz/debug (Wrio3qsGR5uMGp2VU7qZKw) - new failure not classified
• spidermonkey-sm-plain-linux32/debug (Ze2HDv-MQ7yShpKZDUSaHQ) - new failure not classified
• spidermonkey-sm-linux64-wasi/opt (bl_Tm-FjSa2yhjOOmLPSlQ) - new failure not classified
• spidermonkey-sm-wasm-no-experimental-linux64/debug (d7ugXHtOSriBhT-lYckHlA) - new failure not classified
• spidermonkey-sm-compacting-linux64/debug (dfVBSEGOTOCP0xE_Dhpnag) - new failure not classified
• spidermonkey-sm-rt-linux64/debug (eb7pAE_YQZ6rxGyOQPhhPg) - new failure not classified
• spidermonkey-sm-rootanalysis-linux64/debug (ECIHJqqyQdKGUScx7-cGtg) - new failure not classified
• spidermonkey-sm-arm64-sim-linux64/debug (eOHrJpn0RFOmMIm5etpbqA) - new failure not classified
• spidermonkey-sm-fuzzing-linux64/opt (GCN5q7VeTX6iLbfLDwIAtw) - new failure not classified
• spidermonkey-sm-gdb-linux64/debug (GgQKqOh8QXKyZUlKakVAyA) - new failure not classified
• spidermonkey-sm-package-linux64/opt (IcjySo8MRGON1o9ApPW06A) - new failure not classified
• spidermonkey-sm-pbl-linux64/opt (Id5xl_V4SqKwFf-vxi8E-Q) - new failure not classified
• spidermonkey-sm-arm-sim-linux32/debug (JcFr-CruTPiO_8yMDnspJQ) - new failure not classified
• spidermonkey-sm-linux64-wasi-intl/opt (NhUL54VMRy6th0OREeyjmA) - new failure not classified
• spidermonkey-sm-plain-linux64/opt (KVYRnlawQmWMpr_7BHiC7g) - new failure not classified

These failures may mean that the library update succeeded; you'll need to review
them yourself and decide. If there are lint failures, you will need to fix them in
a follow-up patch. (Or ignore the patch I made, and recreate it yourself with
./mach vendor js/src/irregexp/moz.yaml.)

In either event, I have done all I can, so you will need to take it from here.
When reviewing, please note that this is external code, which needs a full and
careful inspection - not a rubberstamp.

Comment 6

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Add stub JitOption for regexp modifiers r=dminor

This flag gates the V8 implementation of this stage 3 proposal: https://github.com/tc39/proposal-regexp-modifiers

I've left it turned off for now. We can expose it in a separate bug.

Comment 7

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Add more irregexp support to RegExpFlags r=dminor

V8 started doing more things using RegExpFlags as part of supporting regexp modifiers.

Comment 8

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Add Tagged wrapper r=dminor

As part of this bug in which they move object definitions back from Torque to C++, V8 added a Tagged<T> wrapper, which stores a pointer that is low-bit tagged (V8's values are either a 0-tagged small integer or a 1-tagged pointer).

We don't have the same object representation, so we just need a transparent wrapper.

Comment 9

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Refactor IsFoo methods r=dminor

In this commit, V8 changed its IsFoo predicates from methods into free functions as part of the Tagged<T> work.

At first I was simply going to move the current implementation, but as I was writing a comment to explain why bool IsByteArray(Object obj) { return true; } was not as dumb as it looked, I decided that we could do better.

V8's ByteArrays are managed by the GC, and the typecheck can be done by looking at the map (shape). Our implementation of ByteArray is a length-prefixed array allocated by malloc, so we can't check the type the same way, but in debug builds we can stick a magic number in the header and validate that it hasn't been messed with.

We can use a dedicated MagicValue for the exception we throw when interrupting (which maps closely to V8's approach).

Comment 10

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Update special-case.cc r=dminor

This is an auto-generated file in V8 that changes very rarely (see https://bugzilla.mozilla.org/show_bug.cgi?id=1624015#c3). I copied this version over from a local build of V8.

Comment 11

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Rename ByteArray::GetDataStartAddress to ByteArray::begin r=dminor

V8 refactored ByteArray as part of this bug: https://bugs.chromium.org/p/v8/issues/detail?id=14345.

Comment 12

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Remove obsolete argument r=dminor

The RegExpCompiler stores the flags internally, so there isn't any need to pass them in here. Upstream V8 fixed the API.

Comment 13

[Iain Ireland [:iain]]

Attachment: Bug 1879225: Disable narrowing warning for irregexp r=dminor

This is necessary because of an awkward bit of code here in regexp-nodes.h that uses a struct initializer to create a RegExpFlags. C++ initializers do not do type conversions, so if we initialize RegExpFlags with an int, then the member field of RegExpFlags must also be an int. V8's RegExpFlags is backed by an int, but we use a uint8_t.

I based the clang/gcc flags on the #pragmas here.

Comment 16

[Matthew Gaudet (he/him) [:mgaudet]]

Question about update bot: Iain, do you feel it's worthwhile to have Update bot run on the irregexp changes, or should we look into getting that turned off to reduce the noise for you?

Comment 17

[Iain Ireland [:iain]]

I appreciate having updatebot around. I don't always (ever?) land non-trivial changes immediately, but updatebot cleans up after itself if I ignore it, so there's no backlog building up. I'm already watching the entire JS component, so it's not a significant increase in my bugmail.

Comment 18

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/1b309a9c237b
Update irregexp to ec89cca93594ef6ae683d6608904cd5e30378115 r=dminor
https://hg.mozilla.org/integration/autoland/rev/6a96960cbfb0
Add stub JitOption for regexp modifiers r=dminor
https://hg.mozilla.org/integration/autoland/rev/9db3b5ee9eb1
Add more irregexp support to RegExpFlags r=dminor
https://hg.mozilla.org/integration/autoland/rev/4e39e690f2b8
Add Tagged wrapper r=dminor
https://hg.mozilla.org/integration/autoland/rev/8efe813ab556
Refactor IsFoo methods r=dminor
https://hg.mozilla.org/integration/autoland/rev/a0e12e764dab
Update special-case.cc r=dminor
https://hg.mozilla.org/integration/autoland/rev/e06159bc2d35
Rename ByteArray::GetDataStartAddress to ByteArray::begin r=dminor
https://hg.mozilla.org/integration/autoland/rev/85a16b24ff70
Remove obsolete argument r=dminor
https://hg.mozilla.org/integration/autoland/rev/589758807f24
Disable narrowing warning for irregexp r=dminor

Comment 19

[Narcis Beleuzu [:NarcisB]]

Backed out for bustages on regexp-nodes.h

Backout link: https://hg.mozilla.org/integration/autoland/rev/2e6f9604cc356e96fe89ec8f7670830b07f30104
Log link: https://treeherder.mozilla.org/logviewer?job_id=448828603&repo=autoland&lineNumber=32478

Please also check this SM bustage

Comment 20

[Iain Ireland [:iain]]

I disabled the narrowing warning for gcc and clang, but not clang-cl, which we use on Windows. Fixing and relanding.

Comment 21

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/fb35e3dc4f11
Update irregexp to ec89cca93594ef6ae683d6608904cd5e30378115 r=dminor
https://hg.mozilla.org/integration/autoland/rev/cc4dfb967ba7
Add stub JitOption for regexp modifiers r=dminor
https://hg.mozilla.org/integration/autoland/rev/aee1020e5dab
Add more irregexp support to RegExpFlags r=dminor
https://hg.mozilla.org/integration/autoland/rev/2c09b27de401
Add Tagged wrapper r=dminor
https://hg.mozilla.org/integration/autoland/rev/02cbfa48ac5e
Refactor IsFoo methods r=dminor
https://hg.mozilla.org/integration/autoland/rev/3a403f55d5b9
Update special-case.cc r=dminor
https://hg.mozilla.org/integration/autoland/rev/bd893ce35a99
Rename ByteArray::GetDataStartAddress to ByteArray::begin r=dminor
https://hg.mozilla.org/integration/autoland/rev/d521718e5c3e
Remove obsolete argument r=dminor
https://hg.mozilla.org/integration/autoland/rev/a55d20473169
Disable narrowing warning for irregexp r=dminor

Comment 22

[Serban Stanca [:SerbanS]]

https://hg.mozilla.org/mozilla-central/rev/fb35e3dc4f11
https://hg.mozilla.org/mozilla-central/rev/cc4dfb967ba7
https://hg.mozilla.org/mozilla-central/rev/aee1020e5dab
https://hg.mozilla.org/mozilla-central/rev/2c09b27de401
https://hg.mozilla.org/mozilla-central/rev/02cbfa48ac5e
https://hg.mozilla.org/mozilla-central/rev/3a403f55d5b9
https://hg.mozilla.org/mozilla-central/rev/bd893ce35a99
https://hg.mozilla.org/mozilla-central/rev/d521718e5c3e
https://hg.mozilla.org/mozilla-central/rev/a55d20473169