Re: [PATCH v4] ioctl_tty.2: Add example how to get or set baudrate on the serial port

["Pali Rohár" <pali@kernel.org>]

On Friday 10 September 2021 15:37:54 Alejandro Colomar (man-pages) wrote:
> Hi, Pali!
> 
> On 8/31/21 10:34 PM, Pali Rohár wrote:
> > On Tuesday 10 August 2021 21:49:28 Pali Rohár wrote:
> > > Setting custom baudrate for which is not defined Bnnn constant is possible
> > > via BOTHER flag and then filling speed in c_ospeed and c_ispeed fields.
> > > 
> > > These two fields are either in struct termios or struct termios2. Former
> > > belongs to TCGETS/TCSETS ioctls, latter to TCGETS2/TCSETS2 ioctls.
> > > 
> > > BOTHER flag with these two fields and new struct termios2 is not supported
> > > by older versions of include header files.
> > > 
> > > Some architectures (e.g. amd64) provide both struct termios and struct
> > > termios2, but c_ospeed and c_ispeed are only in struct termios2.
> > > 
> > > Some other architectures (e.g. alpha) provide both struct termios and struct
> > > termios2 and both have c_ospeed and c_ispeed fields.
> > > 
> > > And some other architectures (e.g. powerpc) provide only struct termios
> > > (no struct termios2) and it has c_ospeed and c_ispeed fields.
> > > 
> > > So basically to support all architectures it is needed to use
> > > struct termios2 when TCGETS2/TCSETS2 is supported. Otherwise it is needed
> > > to use struct termios with TCGETS/TCSETS (case for e.g. powerpc).
> > > 
> > > Setting input baudrate is done via IBSHIFT macro.
> > > 
> > > Signed-off-by: Pali Rohár <pali@kernel.org>
> > 
> > Hello! Do you have any comments on this patch?
> 
> Sorry, I started to fix the licensing issues Greg pointed out, and forgot
> about this.
> 
> I assume it's good since you have been doing a lot of related work in recent
> patches.  I'll only point out a cosmetic issue with the preprocessor stuff:
> I'd indent with a single space after '#' to clarify preprocessor #if #else
> #endif relationships and improve readability.  I also prefer '#if
> [!]defined()' rather than '#if[n]def' (except for include guards); and there
> seems to be a mix in the existing pages, so I'll standardize that way from
> now on.
> 
> For example:
> 
> #if !defined a
> #else
> # if defined b
> # endif
> #endif
> 
> However, I applied those changes myself in a following patch, so you don't
> worry about them.

Ok, perfect, thanks!

> Patch applied!
> 
> Thanks,
> 
> Alex