Change in flashrom[master]: improve termios settings on some OSes

[Nico Huber (Code Review)]

Nico Huber has posted comments on this change. ( https://review.coreboot.org/23693 )

Change subject: improve termios settings on some OSes
......................................................................

Patch Set 1:

(1 comment)

I have generally trouble to understand why everybody modifies an
existing terminal config instead of writing his own. Are there
too many implementation specific pitfalls?

https://review.coreboot.org/#/c/23693/1/serial.c
File serial.c:

https://review.coreboot.org/#/c/23693/1/serial.c@216
PS1, Line 216: 	wanted.c_cflag &= ~(PARENB | CSTOPB | CSIZE | CRTSCTS);
             : 	wanted.c_cflag |= (CS8 | CLOCAL | CREAD);
             : 	wanted.c_lflag &= ~(ICANON | ECHO | ECHOE | ECHOK | ECHONL | ISIG | IEXTEN);
             : 	wanted.c_lflag &= ~(ECHOCTL | ECHOKE);
             : 	wanted.c_iflag &= ~(IXON | IXOFF | IXANY | ICRNL | IGNCR | INLCR);
             : 	wanted.c_iflag &= ~(PARMRK);
             : 	wanted.c_oflag &= ~OPOST;
> It might be wise to start with cfmakeraw(3).  Although it does not
> appear to be standardized by Open Group, I see cfmakeraw(3)
> available on my GNU/Linux installations.
>
> https://nxr.netbsd.org/xref/src/lib/libc/termios/cfmakeraw.c#51

Given that flashrom is supposed to work on much more than Linux/BSD,
I'd rather not try to use non-standard functions.
> But, if you've covered all the bits that NetBSD's cfmakeraw(3)
> does, I have no objection.

It's actually your patch ;) I merely ported it from patchwork [1].
No idea if anybody or why nobody considered it earlier.

I have trouble understanding every detail in the man pages and
have no way for regression testing (lack of any serial based
programmer). That's also why a related commit, 22334, didn't make
it into the 1.0 release.

If you can test this on at least two different OSs (e.g. Linux,
BSD) that would be very nice.

[1] https://patchwork.coreboot.org/patch/2757/