1 bug cause the kdb-3.0-2.4.20 patch fail to compile

[Keith Owens]

On Sat, 8 Feb 2003 15:11:54 +0800 , 
"Zhang, Sonic" <sonic.zhang at intel.com> wrote:
>	Yes, your addition to VT100 character sequence works well. 
>	But I wonder why you don't add these translation in architecture
>dependent function "static int get_serial_char(void);". I think it is better
>to return the meaningful ASCII characters in these architecture dependent
>functions.

(1) vt100 is a global definition, not architecture specific.

(2) The timing algorithms for vt100 mean that you need a global view of
    all input sources to determine if the current character is part of
    the current vt100 sequence or something else from another data
    source.