Public issue detail

Runaway CPU investigation for clang: unknown userspace loop at clang::Lexer::LexTokenInternal(clang::Token&, bool)

clang is stuck in a likely unclassified userspace loop: 1.91% of sampled CPU passed through clang::Lexer::LexTokenInternal(clang::Token&, bool), with repeated write x2, close x1, exit_group x1.

highscore: 106reports: 1

Last seen: 2026-10-20 04:50 UTC. Public JSON: /v1/issues/019ff123-7d66-7253-bc62-d7ba8c5bd3e8

Technical snapshot

This is the clearest retained userspace thread cluster Fixer captured while the process was spinning.

Representative thread backtrace

  • Command: /usr/bin/clang-19 -Wno-ignored-attributes -fno-strict-aliasing -fwrapv -fexcess-precision=standard -Wno-unused-command-line-argument -Wno-compound-token-split-by-macro -Wno-format-truncation -O2 -I../../liblwgeom -I../liblwgeom -I../../libpgcommon -I../../deps/flatgeobuf -I/usr/local/include -I/usr/include/x86_64-linux-gnu -I/usr/include/webp -isystem /usr/include/mit-krb5 -I/usr/include/p11-kit-1 -I/usr/include/libxml2 -I/usr/include -I/usr/include/json-c -DNDEBUG -I. -I./ -I/usr/include/postgresql/18/server -I/usr/include/postgresql/internal -Wdate-time -D_FORTIFY_SOURCE=2 -D_GNU_SOURCE -I/usr/include/libxml2 -flto=thin -emit-llvm -c -o gserialized_estimate.bc ../../postgis/gserialized_estimate.c
  • Why Fixer classified it this way: The process is demonstrably CPU-hot, but the current syscall and symbol sample does not point to a single dominant loop family yet.
  • Wait site: 0
  • Hot path: clang::Lexer::LexTokenInternal(clang::Token&, bool) (1.91% sampled CPU)
  • Repeated loop: fstat -> write -> write
  • Top syscalls: write x2, close x1, exit_group x1, fstat x1
  • Kernel: 7.1.8+deb14.1-amd64
  • Distribution: debian

Possible duplicates

These are suggestions based on sanitized trigram similarity plus structured fields like package, subsystem, classification, and wait site. They are not auto-merged.

cc1 is stuck in a likely unclassified userspace loop: 2.04% of sampled CPU passed through bitmap_set_bit(bitmap_head*, int), with repeated write x88, mmap x1.

highscore: 106reports: 1similarity: 75%

Why this looks related: same classification, same subsystem, same wait site

Last seen: 2026-10-13 06:02 UTC. Public page: /issues/019fa11d-a82b-79e0-891c-85e9b734e4a0. Public JSON: /v1/issues/019fa11d-a82b-79e0-891c-85e9b734e4a0

po2xml is stuck in a likely unclassified userspace loop: 53.94% of sampled CPU passed through unresolved offset, with repeated write x2, exit_group x1, fstat x1.

highscore: 106reports: 1similarity: 71%

Why this looks related: same classification, same subsystem, same wait site

Last seen: 2026-10-19 03:46 UTC. Public page: /issues/01a0f656-df4b-7fc2-acb4-c03fc6235e66. Public JSON: /v1/issues/01a0f656-df4b-7fc2-acb4-c03fc6235e66

llama-server is stuck in a likely unclassified userspace loop: 23.95% of sampled CPU passed through uvm_spin_loop, with repeated no dominant syscall sample was captured.

highscore: 106reports: 1similarity: 70%

Why this looks related: same classification, same subsystem, same wait site

Last seen: 2026-10-22 21:36 UTC. Public page: /issues/01a0b753-7de3-74d2-bfee-879445fd371b. Public JSON: /v1/issues/01a0b753-7de3-74d2-bfee-879445fd371b

lto1 is stuck in a likely unclassified userspace loop: 2.51% of sampled CPU passed through bitmap_set_bit(bitmap_head*, int), with repeated thread backtraces show 1 thread(s) around 0x000000000162a9f7 in constrain_operands(int, unsigned long) ().

highscore: 106reports: 1similarity: 67%

Why this looks related: same classification, same subsystem, same wait site

Last seen: 2026-09-30 11:46 UTC. Public page: /issues/019fa772-cf3b-7890-82d2-60e7dba39362. Public JSON: /v1/issues/019fa772-cf3b-7890-82d2-60e7dba39362

Published attempts

No public attempts have been published for this issue yet.