Am 17.06.2013 22:34, schrieb Richard Henderson: > On 06/16/2013 08:57 AM, Andreas Färber wrote: >> + cpu = CPU(alpha_env_get_cpu(cpu_single_env)); >> + cs = CPU(alpha_env_get_cpu(cpu_single_env)); >> + CPUState *cpu_single_cpu = CPU(alpha_env_get_cpu(cpu_single_env)); >> + cs = CPU(alpha_env_get_cpu(cpu_single_env)); > > ENV_GET_CPU.
Intentional, and all of these get replaced later on anyway. >> + cpu_class_set_do_unassigned_access(cc, alpha_cpu_unassigned_access); > > Wouldn't it be better to arrange for the function to exist always, with > differing implementation for user and system, much like do_interrupt? Sure, that would be possible. But determining what the user implementation should do would be up to the target, i.e. you for alpha. > What you've got is good enough for now, I suppose. Thanks, v3 coming up. Andreas -- SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
