Skip to content

tty: disassociate_ctty() sends the extra SIGCONT

Oleg Nesterov lkml stable 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest1 patch 11 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/1tty: disassociate_ctty() sends the extra SIGCONTdrivers/tty/tty_io.c+2 -18 replies

Reviews trailers per patch

The patch has a Reviewed-by or Acked-by.

PatchPeter HurleyOleg Nesterov
SeriesT-b
1/1R-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH 0/1] tty: disassociate_ctty() sends the extra SIGCONT

Hello.

I know nothing about tty's, I have no idea why disassociate_ctty() checked on_exit before kill_pgrp(SIGCONT). And why this depends on tty/old_pgrp/TTY_DRIVER_TYPE_PTY.

But this was changed in v3.10, probably by f91e2590 "tty: Signal foreground group processes in hangup", and this breaks LSB tests which do not expect the "wrong" SIGCONT.

The patch cc's stable, but it needs the ack or nack from someone who these HUP/CONT rules.

IOW, I won't argue if the user-visible (and undocumented) change added by f91e2590 is actually fine, in this case LSB should be fixed.

Oleg.