[Bug libc/10339] Terminal interface: non-standard baudrates are not handled properly on Linux-systems

["joseph at codesourcery dot com" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=10339

--- Comment #19 from joseph at codesourcery dot com <joseph at codesourcery dot com> ---
On Mon, 8 Oct 2018, hpa at zytor dot com wrote:

> Neither mips nor sparc defined the c_[io]speed in the *user space* struct
> termios.  MIPS has a fair number of unused bytes in c_cc, so my current
> thinking is to drop NCCS to 24 on MIPS (the kernel uses 23) and add those
> fields.

Could that change for MIPS cause a program built with glibc 2.29, and 
using the new fields, but not using any new functions, to misbehave when 
run with glibc 2.28 because the fields would not be set?  If so, I think 
you need new symbol versions for the relevant functions, even if the 
functions with the old and new symbol versions are aliases, so that 
programs expecting the new fields don't try to run with older libc.  If 
the change of NCCS could also break valid expectations for programs built 
with the old libc, you need new symbol versions which aren't aliases.

-- 
You are receiving this mail because:
You are on the CC list for the bug.