Closed Bug 2083707 Opened 28 days ago Closed 27 days ago

Update irregexp to new version 98f55fcc719a757173c470768f7bcd69af1b7ba6 from 2026-10-19 12:20:51

Categories

(Core :: JavaScript Engine, enhancement)

enhancement

Tracking

()

RESOLVED FIXED
160 Branch
Tracking Status
firefox160 --- fixed

People

(Reporter: update-bot, Assigned: iain)

References

Details

(Whiteboard: [3pl-filed][task_id: WGr4lVWxSey5nw35-hwKDQ])

Attachments

(8 external files)

This update covers 1216 commits, including 351 new upstream commits I've never filed a bug on before. (They're the top 351.). Here are the overall diff statistics.


js/src/irregexp/imported/gen-regexp-special-case.cc | 57 +++++-
js/src/irregexp/imported/regexp-ast.h | 8 +-
js/src/irregexp/imported/regexp-bytecode-generator-inl.h | 4 +-
js/src/irregexp/imported/regexp-bytecode-generator.cc | 2 +-
js/src/irregexp/imported/regexp-bytecodes-inl.h | 7 +-
js/src/irregexp/imported/regexp-code-generator.cc | 1 -
js/src/irregexp/imported/regexp-compiler-tonode.cc | 107 +++++++------
js/src/irregexp/imported/regexp-compiler.cc | 114 +++++++------
js/src/irregexp/imported/regexp-compiler.h | 6 +-
js/src/irregexp/imported/regexp-graph-printer.cc | 2 -
js/src/irregexp/imported/regexp-interpreter.cc | 7 +-
js/src/irregexp/imported/regexp-macro-assembler-tracer.cc | 2 +-
js/src/irregexp/imported/regexp-macro-assembler-tracer.h | 4 +-
js/src/irregexp/imported/regexp-macro-assembler.cc | 38 +++-
js/src/irregexp/imported/regexp-nodes.h | 7 +-
js/src/irregexp/imported/regexp-parser.cc | 8 +-
js/src/irregexp/imported/regexp-stack.cc | 4 -
js/src/irregexp/imported/regexp-stack.h | 47 ++++-
js/src/irregexp/imported/special-case.h | 110 +++++-------
js/src/irregexp/moz.yaml | 4 +-
20 files changed, 306 insertions(+), 233 deletions(-)

Duplicate of this bug: 2080700

I think updatebot is having some sort of issue where it is pushing patches based on a stale version of V8.

On Friday I was working on an update to irregexp. I ran ./mach vendor locally and wrote patches to make it build. This morning I used moz-phab patch to pull updatebot's latest version and rebased on top of it, but my patches don't build. Specifically, the failure is error: out-of-line definition of 'SupplementaryEquivalentsSet' does not match any declaration in 'v8::internal::regexp::CaseFolding'. If you look at special-case.h in updatebot's patch, you can see that it is indeed true that SupplementaryEquivalentsSet is not defined. But it was added last week! If you take the revision that updatebot is writing into moz.yaml and inspect special-case.h at that revision upstream,, you can see that it does define SupplementaryEquivalentsSet.

Now that I notice this, I seem to recall a few other times where a fix landed upstream, I updated irregexp, and I was surprised to later see that the bug hadn't been fixed. At the time I assumed that the problem was entirely user error on my end (and to be sure, I might have noticed this earlier if I'd landed the testcases immediately), but now I think it's possible that updatebot shares some of the blame.

I'll rebase these patches on a manual ./mach vendor, but we should probably figure out what's up with updatebot.

Flags: needinfo?(tom)
Assignee: nobody → iireland
Status: NEW → ASSIGNED

Andrew Gaul landed a patch in V8 to access all RegExp::Stack state through a single pinned register. This patch just rebases our current approach on top of his changes; I'll land a follow-up patch in a separate bug to make the equivalent change in our code.

The spec rules for case-insensitive comparison are finicky. Prior to this update, both engines had known issues. V8 fixed their issues, so we might as well use their implementation directly.

This fixes bug 2072420. I'll land the testcase from that bug separately so that it gets automatically closed when the stack lands.

The initial plan for special-case.cc was that on the rare occasion where we needed to update it, we could simply copy the output of V8's build process. It turns out that this isn't entirely true. In particular, the next Unicode version (Unicode 18) will add characters in the supplementary set that casefold to characters in the basic multilingual plane, meaning that for the first time the case-folding equivalence set for characters in the BMP may contain surrogate pairs. At the moment, Chrome has updated to use Unicode 18, but Firefox is still using Unicode 17. This means that our SupplementaryEquivalents set is empty, but we still need to define it. Since we're expecting to rebuild this again in the future, I got Claude to write a quick script to do so. It's not part of the regular build; it should be run manually when necessary.

A couple of small shim updates that don't need their own patch.

The RoundUpToPowerOfTwo code is no longer needed because V8 replaced its uses with std::bit_ceil to reduce custom bit-twiddling code.

The try push is done, we found jobs with unclassified failures.

Needs Close Investigation:

  • spidermonkey-sm-rootanalysis-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: NYcT952EQImVbCrg0hh3hQ, Goad5IqZTYOOIxiAVotNoQ, OQ4eJgw3RiqB4tvDMfn-Zw, WYjS8yhxStK58wmvBajNqw)
  • spidermonkey-sm-arm64-sim-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: HoRs6B1RTh-uKYk80O9mag, YXmuJt6PR9u3-HAKIKYcIw, YuulvfGBRfKRbnFpSIfRNw, AACAso2KQ4yiGl_ASmWW_A)
  • spidermonkey-sm-fuzzing-linux32/opt - 4 of 4 failed on the same (retriggered) task (failed: facybl3AT7-akqOr2onLbw, XaN9G-vAQkWMolpnaGmHJA, RMnghrqWQQCcaavDmYyR4Q, RjkPT-lkQ0a9GNhk-2AoVw)
  • spidermonkey-sm-linux64-wasi-intl/opt - 4 of 4 failed on the same (retriggered) task (failed: dOH3SRdiSsGIpgvML-KMyQ, TLHItqr3QtOiRb7iFbXMsg, YucJLAtxTuOGEeG09aJusA, FmH91uc6RY6N6ZDvh5yncw)
  • spidermonkey-sm-wasm-no-experimental-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: K355nxZURBy29868r45yXA, IkoJ6nksSgGNE7e-DebGPQ, JZ0oeGlrToyOP3zsIbOA6w, Srj1JDAQQGGbJ2BFJiPn5g)
  • spidermonkey-sm-nojit-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: EJq5embiSASUZq69YAEE2A, QtK-a7keTPy6pTthH8XsmQ, ImpD1GHQRjiXwCRJJ6BiQw, cgyJjPiySE-97rihHYu50g)
  • spidermonkey-sm-fuzzing-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: d5Ou0DfGS92_4LpeSKZ7Kg, BrujP4iVTqqX_uDhmJYDzw, dH-njz1QQgSvfvqZwvnKJw, WuA5UoKyQka-1cqLKkHGLw)
  • spidermonkey-sm-gdb-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: JpztWbc3Rc-oEy3zaIH0mg, Z7DSj7o9T1SrRLiTAwpcQg, WIZJIxOLTUiMTiQzlqktfA, WgTcjt02R06AEmV8DLgh8A)
  • spidermonkey-sm-concurrent-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: LEafnW9jQZyKzu0Zrv08OA, BsqwiCbSQ7q21YpRuz3ARg, BTSDmc5QSSWXDIfrvugFuQ, LYIZgAi9RwGa7QUL3sf7EQ)
  • spidermonkey-sm-linux64-wasi/opt - 4 of 4 failed on the same (retriggered) task (failed: WNIcGDZdTWyfJeBOhalrHg, JygTOnroTiy6FF70C__Khg, bPyvI4pRQU-2x2EFXfFXSQ, VgLkUJirQyW_TzWo2aT4tA)
  • spidermonkey-sm-linux64-wasi-pbl/opt - 4 of 4 failed on the same (retriggered) task (failed: Yub137t8Rhar2zcXv0RYxQ, AzfqPQUnQYqEznEkEdj0sg, RvO_xPx3QimX4RunsF_p1g, e4-yUeHfTiKuop2brfeK6g)
  • spidermonkey-sm-arm-sim-linux32/debug - 4 of 4 failed on the same (retriggered) task (failed: dpIL1ZCwRwiBJXTjS5SRHA, dccXbxv7RwuJB-fAR2-sgw, H6qP5K3gSD6D2SIL-1dHxA, HNnDk4-RT6CLrKQEdOoqKQ)
  • spidermonkey-sm-fuzzilli-linux64-aarch64/debug - 4 of 4 failed on the same (retriggered) task (failed: cwnx3XU-SbGTfl7gA5whcw, PJrDuAiyQrmBoxrSHHttgA, D6s1tHPJTy-yjnMFcUP9TQ, CeHA1IulStWqjI8JfAGe7g)
  • spidermonkey-sm-tsan-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: GUBSjE9lSvSdSoqf3x3rqQ, BkfJaEPfTmqg0fFs0r3bDQ, bK3wQ-viQgaZksHb8E4LGA, UiTbwwAVQkaoodkZClIDPg)
  • spidermonkey-sm-plain-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: KJdKNwzoTniHndLX0lel5Q, U0ACFlCtTNiDW7JqUBTBIA, DCjUeZa2RZiWA5uXyScD8g, Ip1ulN4aR02-_Z53uJjFvA)
  • spidermonkey-sm-plain-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: VPgHyA0gSbedobj8IYNTwg, TItqe41aRluEnHpLabAbiw, H4WMGSJBShqqAJa7olrGRQ, NranFt2pRi2IfIhNfeQ4eA)
  • spidermonkey-sm-asan-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: LwukXyvSQ4GxzsydqdELMA, LZJOCuyXR-WG2TDZBb2aXA, CubuYJzjQ4apxlae2jgG2g, GyR5DIaGRRm-exTIujWtMg)
  • spidermonkey-sm-plain-linux32/debug - 4 of 4 failed on the same (retriggered) task (failed: ORjZzQT7QG6gNawvBCnuoQ, aBy5Ol8OTKmn6D5HSem_wQ, XHnRxhBrTt-jfQlAP-M6Gw, RS7HmA54SQqIBT8vrLt57w)
  • spidermonkey-sm-package-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: JmYhE7imSfay2IBVntf6bg, WtavRiGXQ_qUtfB6KJGuBQ, EJ_dPjYsTrCoTNpJ3ETJ2w, N3S-mzWlRdGdQrbX2OXnIQ)
  • spidermonkey-sm-nonunified-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: bZYBdRKXRv-VEHSbh9aNQg, DmNEiOVtQu-J6LJo9fDQvQ, AIhTitLJRviJGdvbfFv1Hw, dWcT9wB1RnW6bXMIWbiThQ)
  • spidermonkey-sm-fuzzing-linux32/debug - 4 of 4 failed on the same (retriggered) task (failed: ObLyZYYeTmentTquBktChA, etxlFoXcQRWXBSVyHG66MA, eYOmqMyARjab2xCx010UtA, OVrjtyyfTOyaAJkWBD-U_w)
  • spidermonkey-sm-fuzzilli-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: FiS-4-P9TZel75ySPp_RSA, eGbw-p5PQTa08bSmPDd-dQ, fds1aZJiQWuLf3ZX8G6LMA, eXMIKpzfRZKdrfR5tvPlmQ)
  • spidermonkey-sm-pbl-linux64/opt - 4 of 4 failed on the same (retriggered) task (failed: FrDfjAO4SDOAq4DnUUvrGQ, Z9RU6xlxSl2aWbH7TKABuQ, dB3EH-jSRD2cF0pVAfKJbg, Zp7yUT0jQJO8xE36pal6oA)
  • spidermonkey-sm-pbl-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: FVerzRMwTP6PsQRc2KmSpA, G-EyCyKqS2-zPcbmY6TxVA, X4BvOla7TsSFnBMnAj1Ocw, bH66gqBlTwW6IzH7hfTqvQ)
  • hazard-linux64-shell-haz/debug - 4 of 4 failed on the same (retriggered) task (failed: Fozn4YzXSsCrzwTosXX2pA, GQ-8AbILSGiD4wCIFmnhfg, FJGV1vqVQ-6LmD4j7TcjqQ, PZ6fyPy_T7S2K3dGT1Odtw)
  • spidermonkey-sm-compacting-linux64/debug - 4 of 4 failed on the same (retriggered) task (failed: WXfMPB1bRl2AWwH8vVhLbQ, CaPvzpKEQiyQFQYpPdcheA, TFtSGaZmRsmsdIbRazc5xQ, W_58AVlHTUSelfTJlZn2DQ)
  • spidermonkey-sm-fuzzing-asan-linux32/opt - 4 of 4 failed on the same (retriggered) task (failed: XZhvhSW6RjGFfTimOIkV1A, FFNc8j3YTviAs0RHu4m8cQ, NiwgexidTkmcznGyx1n9_A, TgYC4fuCR16z98PwJEOVBg)

These failures could mean that the library update changed something and caused
tests to fail. You'll need to review them yourself and decide where to go from here.

In either event, I have done all I can and you will need to take it from here. If you
don't want to land my patch, you can replicate it locally for editing with
./mach vendor js/src/irregexp/moz.yaml

When reviewing, please note that this is external code, which needs a full and
careful inspection - not a rubberstamp.

Flags: needinfo?(tom) → needinfo?(iireland)

(Restoring the ni on tjr for comment 5)

Flags: needinfo?(iireland) → needinfo?(tom)

Bug 2084753 should fix it

Flags: needinfo?(tom)
QA Whiteboard: [qa-triage-done-c161/b160]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: