On Fri, Jan 27, 2017 at 9:29 AM, Julien Grall <julien.gr...@arm.com> wrote: > Hi Tamas, > > On 27/01/17 16:23, Tamas K Lengyel wrote: >> >> On Fri, Jan 27, 2017 at 9:15 AM, Julien Grall <julien.gr...@arm.com> >> wrote: >>> >>> Hello, >>> >>> On 27/01/17 15:52, Tamas K Lengyel wrote: >>>> >>>> >>>> Well, yes, only ARM could _should_ call this function. The comment I >>>> think is important to tell the user don't expect it to do anything on >>>> x86. Doesn't mean they can't call it though - if that was the case it >>>> would be wrapped in an ifdef like all the other architecture specific >>>> bits in the header. I would think that's pretty straight forward. No >>>> objection to clarifing the comment though if it helps. >>> >>> >>> >>> If you looked at the commit log, the #ifdef was added to avoid calling >>> the >>> hypervisor for nothing and therefore saving few hundred cycles bit of >>> time. >>> Technically speaking, this helper abstracts the architectural behavior of >>> the cache. So it makes sense to call it on x86 even if it is a nop. >> >> >> Except that on x86 the user should be aware that it returns an error, >> which is normal and can be ignored. > > > It looks like the current callers does not check the return. However, it > would more make sense to return 0 if we expect nothing to be done rather > than -ENOSYS. >
That would be fine by me. Wei, what do you think? Tamas _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
