On Tue, Aug 05, 2014 at 05:47:03AM -0600, Eric Blake wrote: > On 08/04/2014 11:56 PM, Hu Tao wrote: > > These two are almost the same as memory_region_init_ram() and > > memory_region_init_ram_ptr() except that they have an extra errp > > parameter to let callers handle error. The purpose is to fix the bug > > described below. > > > > We should have added errp directly to memory_region_ram(), but that > > mixes updates to calls to memory_region_ram() and this bug fix and make > > it hard to review. > > > > We will rename _may_fail variants later so that we will have two versions > > of API: one with errp parameter(memory_region_init_ram(), > > memory_region_init_ram_ptr()), one without errp parameter and with > > suffix _nofail. > > That feels like overkill. Every caller that calls the _nofail variant > can instead call memory_region_init_ram(..., &error_abort), and then you > don't need the _nofail version.
Right. v5 is coming soon. Regards, Hu
