>>> On 16.10.17 at 12:18, <julien.gr...@linaro.org> wrote: > On 16/10/17 10:02, Bhupinder Thakur wrote: >> --- a/tools/console/daemon/io.c >> +++ b/tools/console/daemon/io.c >> @@ -80,6 +80,7 @@ static unsigned int current_array_size; >> static unsigned int nr_fds; >> >> #define ROUNDUP(_x,_w) (((unsigned long)(_x)+(1UL<<(_w))-1) & >> ~((1UL<<(_w))-1)) >> +#define INVALID_RING_REF (~(xen_pfn_t)0) > > I am a bit surprised we don't have an INVALID_XEN_PFN definition in the > public header. Jan would you have an objection to introduce that?
If this was for a use with a hypercall, I probably wouldn't mind. But here this is a completely local constant, unrelated to anything Xen produces or consumes. Of course seeing mention of INVALID_GFN in a public header comment, there likely are hypercalls that could benefit from such a constant (or maybe more than one, for the various frame number flavors). Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
