Message ID | 7ec75007-2067-8c71-625c-b22ba5f0f0d9@citrix.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Tue, May 30, Andrew Cooper wrote: > Does > + domctl.u.address_size.size = 0; > help? Likely yes. I just scanned the buildlogs and found this failure. I do not have a ARM toolchain around to double check. Not sure why the marco, or its usage, is not like 'struct xen_domctl domctl = {}' Olaf
On 30/05/17 16:40, Olaf Hering wrote: > On Tue, May 30, Andrew Cooper wrote: > >> Does >> + domctl.u.address_size.size = 0; >> help? > Likely yes. I just scanned the buildlogs and found this failure. > I do not have a ARM toolchain around to double check. > Not sure why the marco, or its usage, is not like 'struct xen_domctl domctl = {}' I've never been convinced by the utility of those macros. All they do is make it cognitively harder to read the code. I think I even tried to kill them once, but got nacked. Either way, my position still stands that they should be relegated to /dev/null and we pretend that they never existed. ~Andrew
diff --git a/tools/libxc/xc_dom_arm.c b/tools/libxc/xc_dom_arm.c index e7d4bd0..e669fb0 100644 --- a/tools/libxc/xc_dom_arm.c +++ b/tools/libxc/xc_dom_arm.c @@ -223,6 +223,8 @@ static int set_mode(xc_interface *xch, domid_t domid, char *guest_type) domctl.domain = domid; domctl.cmd = XEN_DOMCTL_set_address_size; + domctl.u.address_size.size = 0; + for ( i = 0; i < ARRAY_SIZE(types); i++ ) if ( !strcmp(types[i].guest, guest_type) )