Port-xen archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: [PATCH v3 7/7] libxl: Add interface for querying hypervisor about PCI topology



On Mon, 2015-02-16 at 10:54 -0500, Boris Ostrovsky wrote:

> I am not sure what the standard practice is for LIBXL_HAVE_ macros. I 
> see a couple of examples where '#ifdef LIBXL_HAVE_*" is used in libxl 
> code, which is why I have it here as well.
> 
Yes, that is the case for two of them, AFAICS:

 * LIBXL_HAVE_NO_SUSPEND_RESUME
 * LIBXL_HAVE_PSR_CMT

If you look at how these two are defined, you'll see that the definition
is actually conditional:

#if defined(__arm__) || defined(__aarch64__)
#define LIBXL_HAVE_NO_SUSPEND_RESUME 1
#endif

#if defined(__i386__) || defined(__x86_64__)
#define LIBXL_HAVE_PSR_CMT 1
#endif

Which is, IMO, why we need to take special care of them in libxl and xl
code. All other macros, which are always and unconditionally #defined to
1, we just forget about them, and leave it to the caller.

> This is probably question for maintainers.
> 
Sure! :-)

> > Also, ISTR that the first change that actually changes the API should
> > bump the MAJOR in libxl's Makefile. I'm not sure this change qualifies
> > as such, as you're adding stuff, not altering existing data structs
> > (e.g., by adding fields, etc). Personally, I think it does, but I'm
> > leaving this to tools maintainers.
> 
> libxl.h seems to suggest that API version is changed only when we make 
> an incompatible change to the library. In my mind new interface is does 
> not break compatibility so I didn't think a bump would be necessary.
> 
Yeah, that was why I was not sure. Re-thinking about it, you probably
are right, no need for it for now.

Regards,
Dario

Attachment: signature.asc
Description: This is a digitally signed message part



Home | Main Index | Thread Index | Old Index