Re: isprint/isspace segfaulting

[Raj <raju at mailandnews dot com>]

Ulrich Drepper wrote:

> Why would you expect the program to work?  You pass a partially
> initialized value to isspace.  Three of the four bytes are
> uninitialized, and which depends on the endianess.
>
>  
hmm, that's true. but since i do not know what is happening internally , 
i safely assumed that the macros would mask off bits which they do not 
need.

Still the question bugs me. Why cant isspace() and family mask out bits 
which they are not bothered about ? if this is an implementaion decision 
take for efficiency etc, then i guess i am totally wrong in asking such 
a question. I am quite new to glibc internals.

Thx.

/Raj