View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
00019321003.1(2024)/Issue8Base Definitions and Headerspublic2025-06-24 23:462025-06-26 11:22
Reporterhpa Assigned To 
PrioritynormalSeverityEditorialTypeClarification Requested
Status NewResolutionOpen 
NameH. Peter Anvin
Organization
User Reference
SectionXBD 11.1.10, 11.2.4, termios.h
Page NumberN/A
Line NumberN/A
Interp Status
Final Accepted Text
Summary0001932: termios: should cfsetospeed(B0) trigger SIGHUP?
DescriptionIt is unclear to me if setting the output baud rate to 0 using cfsetospeed() followed by tcsetattr() should cause SIGHUP to be sent to the controlling process; in particular in the case of a pseudo-terminal.

Linux currently does not send SIGHUP in this case, but has a separate, privileged, ioctl (TIOCVHANGUP) which can do so.
Desired ActionClarification as to the intended meaning of "hang up" using the zero baud rate, in particular in the case of a pseudo-terminal.

TagsNo tags attached.

Activities

geoffclare

2025-06-26 11:22

manager   bugnote:0007210

The description of tcsetattr() includes:
If the output baud rate stored in the termios structure pointed to by termios_p is the zero baud rate, B0, the modem control lines shall no longer be asserted. Normally, this shall disconnect the line.


How this applies to a pseudo-terminal would depend on whether the implementation emulates modem control lines or not.

The way this is handled in the UNIX conformance tests is that there is a config parameter to say whether or not modem control is supported. Certifications can be done using either hardware terminals or pseudo-terminals for the General Terminal Interface tests, and as far as I'm aware, all certifications that used pseudo-terminals have been done with that parameter set to "N".

Issue History

Date Modified Username Field Change
2025-06-24 23:46 hpa New Issue
2025-06-26 11:22 geoffclare Note Added: 0007210