cfsetispeed and zero baudrate
Hello Alejandro!
Currently in cfsetispeed() documentation is written:
If the input baud rate is set to zero, the input baud rate will be equal to the output baud rate.
There is B0 constant which represents "zero baud rate". And also information that argument for cfsetispeed() (and also cfsetospeed()) must be Bnnn constant.
But above quote documents special handling, "zero" here is really integer 0 and not pre-defined constant B0.
I think that phrase "If the input baud rate is set to zero" could be confusing, as based on requirement to use Bnnn constants somebody could interpret that "zero" as B0.
What do you think, should be above documentation part extended to express that "zero" is really integer 0 and not constant B0?
Some references:
* glibc really checks for integer 0:
https://sourceware.org/git/?p=glibc.git;a=blob;f=sysdeps/unix/sysv/linux/speed.c;h=daac5ecde1f3695b8cb178002e8756895406087b;hb=HEAD#l86
* POSIX describes that it is integer 0:
https://pubs.opengroup.org/onlinepubs/9699919799/functions/tcsetattr.html
Hello Pali!
Quoted 18 lines · Pali Rohár
Hello Alejandro!
Currently in
cfsetispeed()documentation is written:If the input baud rate is set to zero, the input baud rate will be equal to the output baud rate.
There is B0 constant which represents "zero baud rate". And also information that argument for
cfsetispeed()(and alsocfsetospeed()) must be Bnnn constant.But above quote documents special handling, "zero" here is really integer 0 and not pre-defined constant B0.
I think that phrase "If the input baud rate is set to zero" could be confusing, as based on requirement to use Bnnn constants somebody could interpret that "zero" as B0.
Quoted 8 lines
What do you think, should be above documentation part extended to express that "zero" is really integer 0 and not constant B0?
Some references:
* glibc really checks for integer 0:
https://sourceware.org/git/?p=glibc.git;a=blob;f=sysdeps/unix/sysv/linux/speed.c;h=daac5ecde1f3695b8cb178002e8756895406087b;hb=HEAD#l86
* POSIX describes that it is integer 0:
https://pubs.opengroup.org/onlinepubs/9699919799/functions/tcsetattr.html
I think we should, as POSIX does, use literal 0 (and not mention 'zero' at all) to clearly differentiate it from the 'zero baud rate' B0.
Would you send a patch?
Thanks!
Alex
Signature · 3 lines
Alejandro Colomar Linux man-pages comaintainer; https://www.kernel.org/doc/man-pages/ http://www.alejandro-colomar.es/
PATCH 1/1 termios.3: Clarify zero argument for cfsetispeed()
Zero in this case refers to literal constant 0 and not symbolic constant B0.
--- man3/termios.3 | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
man3/termios.3 +5 -1
@@ -1017,7 +1017,11 @@ which must be specified as one of the .BI B nnn constants listed above for .BR cfsetospeed (). -If the input baud rate is set to zero, the input baud rate will be +If the input baud rate is set to literal constant +.B 0 +(not the symbolic constant +.BR B0 ), +the input baud rate will be equal to the output baud rate. .PP .BR cfsetspeed ()
Signature · 1 line
2.20.1
Re: [PATCH] termios.3: Clarify zero argument for cfsetispeed()
Hello Pali,
Quoted 3 lines · Pali Rohár
Zero in this case refers to literal constant 0 and not symbolic constant B0.
Signed-off-by Pali Rohár <[email protected]>
Thanks. Patch applied.
Cheers,
Michael
Quoted 22 lines
--- man3/termios.3 | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)man3/termios.3 +5
-1@@ -1017,7 +1017,11 @@ which must be specified as one of the .BI B nnn constants listed above for .BR cfsetospeed (). -If the input baud rate is set to zero, the input baud rate will be +If the input baud rate is set to literal constant +.B 0 +(not the symbolic constant +.BR B0 ), +the input baud rate will be equal to the output baud rate. .PP .BR cfsetspeed ()
Signature · 3 lines
Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Linux/UNIX System Programming Training: http://man7.org/training/