On Wed, 2019-08-21 at 17:55 -0700, Sean Christopherson wrote: > On Thu, Aug 22, 2019 at 02:29:02AM +0300, Jarkko Sakkinen wrote: > > Using address resolution of any kind is obviously an overkill for > > anything that we know at compile time. Those sites should not hard > > bind how we store SECS. > > > > This commit adds @secs_child to sgx_encl_eldu() and sgx_encl_ewb() > > and replaces sgx_encl_get_index() with a macro SGX_ENCL_PAGE_INDEX() > > Why use a macro instead of an inline function? Just for consistency and nothing else. It falls into same category as the two other macros in encl.h: unconditionally give attributes of the page with some calculation appied (the earlier function had a real flow). We could consider separately converting all of the three as inline functions though (not in the context of this patch). > Rather than pass a bool, we can pass 'struct sgx_epc_page *parent_secs'. > That'd avoid having to name a bool, and would clean up this type of code. > In general, I dislike boolean parameters where the caller uses true/false > as I can never remember the context of true/false. Sure, but why not just @secs? It is always a parent. /Jarkko
