Re: [PATCH 4/5] Handle Windows CE and rewrite NT version handling.

[Paolo Bonzini]

> > At the same time, Windows 95/98/ME never supported anything but 32-bit
> > x86, so its section can be simplified.
>
> There is no need to "simplify" here. The goal of this uname() function is
> to support all kinds of machines, even those to which we don't have
> access to.

And even operating systems that were never ported?  Windows 95/98/ME 
_cannot_ be ported to other processors than x86 because of its 
architecture (for the same reason why Wine cannot be ported to other 
processors than x86).

> > +       default: /* 3.5, 4.0, versions not yet known */
> > +         sprintf (buf->release, "Windows NT %s %u.%u",
> > +                  is_server ? "Server" : "Workstation",
> > +                  (unsigned int) version.dwMajorVersion,
> > +                  (unsigned int) version.dwMinorVersion);
>
> I object against outputting "Windows NT" for the versions not yet known
> (Windows 7, 8, who knows?). Returning "Windows" is suboptimal, but
> acceptable. Returning "Windows NT" will be confusing for end users.

I'll make it only Windows Server/Windows.

> > +      switch ((version.dwMajorVersion<<  8)
> > +             | version.dwMinorVersion)
>
> An idiom like this may save a couple of instructions at the expense of
> readability of the code. When I look at the MS documentation of
> OSVERSIONINFOEX, I see values 5, 6 mentioned, but I don't see 0x502,
> 0x601 etc. Consequence: This idiom makes the code harder to maintain.

But you do see the 5/0, 5/1, 5/2, 6/0, 6/1 pairs.  I did it all but to 
save instructions.  It was more readable for me.

Paolo