Just Launched: You can now import projects and releases from Google Code onto SourceForge
We are excited to release new functionality to enable a 1-click import from Google Code onto the Allura platform on SourceForge. You can import tickets, wikis, source, releases, and more with a few simple steps. Read More
I'm forwarding a new copy of a programmer's question I sent yesterday to Freedos-devel
but for some reason apparently didn't reach to the list server - maybe stalled somewhere, or awaiting processing by the US NSA ?
Of course there is a risk that you guys will suffer receiving 2 or more copies,
in advance of which I send my most sincere apology.
--- On Thursday, 7.4.13, Bertho Grandpied <y31415926536@...> a wrote :
> Objet: (Q) recommended way to reliably check for FreeDOS vs. AnyDOS ?
> Programmers, Kernel people, hi !
> What is (are) the recommended way(s) to assert
> programmatically one is running under FreeDOS, versus other
> common DOS brands and versions ?
> Precisions :
> - I am going for the simplest, yet reliable identification
> method. The problem posed here is to /reliably/ single out
> FreeDOS from foreign DOSes, NOT necessarily to get at
> FreeDOS (sub)version numbers (real or faked!).
> - Method should work in any reasonably current versions of
> FreeDOS - say 0.9, 1.0 and 1.1, whatever compilation
> options, and also be future proof w/ respect to FreeDOS, if
> - Must be available at system initialisation time (callable
> from a device driver's 'init' routine, that is).
> Requirement 1 - simplicity - suggests that we should use a
> FreeDOS-kernel-specific API, if such exists. But reliability
> of the detection is /primordial/, so, if reliabity means we
> have to 'peek' into the kernel and 'walk' binary structures,
> so be it.
> Of course we can't trust blandly the general documented DOS API
> calls for "version" or even "true ver", can we ? I believe
> true ver /should/ not lie but I also think (may be wrongly)
> that FreeDOS truever /does/ lie in certain circumstances,
> Obviously I could hack-and-design something by myself,
> however I would much much preffer to do things in the
> sanctionned way whenever possible.
> Thanks in advance. If this is documented well somewhere, a
> pointer will do for an answer.