>>> On 24.06.16 at 18:29, <ta...@tklengyel.com> wrote: > On Fri, Jun 24, 2016 at 5:24 AM, Jan Beulich <jbeul...@suse.com> wrote: >>>>> On 24.06.16 at 13:20, <kevin.t...@intel.com> wrote: >>>> From: Tamas K Lengyel [mailto:ta...@tklengyel.com] >>>> + * rc < error, fall-through to exit_and_crash >>>> + */ >>>> + if ( !rc ) >>>> + { >>>> + vmx_propagate_intr(intr_info); >>>> + break; >>>> + } >>>> + if ( rc > 0 ) >>>> + break; >>>> + } >>>> else >>>> + { >>>> domain_pause_for_debugger(); >>>> - break; >>>> + break; >>>> + } >>>> + >>>> + goto exit_and_crash; >>> >>> Putting goto as the last line within a 'case' looks a bit strange. What >>> about putting goto directly under a "if ( rc < 0 )" check earlier? >>> >>> if ( !rc ) >>> ... >>> if ( rc < 0 ) >>> goto exit_and_crash; >>> } >>> else >>> domain_pause_for_debugger(); >>> break; >> >> Thanks, Kevin - indeed that's exactly what I had asked for already >> on the previous iteration. >> > > I'm OK with adding the rc < 0 case, it's just that the fall-through > style is already used for handling the int3 events for example. Should > I fix that too while I'm at it so the code is consistent?
Especially if you did it in a separate patch, I for one would appreciate that. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
