---
- branch: MAIN
  date: Thu Apr  4 00:46:42 UTC 2024
  files:
  - new: '1.18'
    old: '1.17'
    path: src/lib/libc/arch/i386/gen/setjmp.S
    pathrev: src/lib/libc/arch/i386/gen/setjmp.S@1.18
    type: modified
  - new: '1.19'
    old: '1.18'
    path: src/lib/libc/arch/i386/gen/sigsetjmp.S
    pathrev: src/lib/libc/arch/i386/gen/sigsetjmp.S@1.19
    type: modified
  - new: '1.9'
    old: '1.8'
    path: src/tests/lib/libc/setjmp/t_sigstack.c
    pathrev: src/tests/lib/libc/setjmp/t_sigstack.c@1.9
    type: modified
  id: 20240404T004642Z.c7a9473dc42ed01b1b05958050daca52d17e1337
  log: |
    i386 longjmp: Restore stack first, then signal mask.

    Otherwise, a pending signal may be delivered on the wrong stack when
    we restore the signal mask.

    While here:

    - Tidy the code a little bit.
    - Sprinkle comments to explain what's going on.
    - Use forward branches for statically predicted not-taken.
      => val==0 is unlikely in longjmp

    PR lib/57946
  module: src
  subject: 'CVS commit: src'
  unixtime: '1712191602'
  user: riastradh