Re: [uml-user] Problem with mkstemp

[Jeff D. <jd...@ad...>]

pd...@ex... said:
> It's not trying very hard, is it? (the first write is my debug). 

No it's not.  You'd think that it would at least try to open a file.

> write(2, "ccs=", 4ccs=)                     = 4 

Do you know what this is?

I checked the usage of it against the man page in case UML got tripped up
by stricter argument checking, but it looks OK.  The man page says:

       EINVAL The  last  six characters of template were not XXXXXX.  Now tem-
              plate is unchanged.

but it seems pretty clear that the template is right.

So I'd chalk this up to being libc wierdness rather than a UML bug.

				Jeff