Re: [Qemu-devel] [PATCH 3/6] target-i386: Register QOM properties for feature flags

[Eduardo Habkost]

On Wed, Apr 08, 2015 at 01:36:29PM +0200, Igor Mammedov wrote:
> On Tue,  7 Apr 2015 17:46:40 -0300
> Eduardo Habkost <address@hidden> wrote:
> 
> > This uses the feature name arrays to register "feat-*" QOM properties
> > for feature flags. This simply adds the properties so they can be
> > configured using -global, but doesn't change x86_cpu_parse_featurestr()
> > to use them yet.
> > 
> > Signed-off-by: Eduardo Habkost <address@hidden>
> > ---
> >  target-i386/cpu.c | 98 
> > +++++++++++++++++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 98 insertions(+)
> > 
> > diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> > index 099ed03..f29e55e 100644
> > --- a/target-i386/cpu.c
> > +++ b/target-i386/cpu.c
> > @@ -2883,12 +2883,103 @@ out:
> >      }
> >  }
> >  
> > +typedef struct FeatureProperty {
> > +    FeatureWord word;
> > +    uint32_t mask;
> > +} FeatureProperty;
> > +
> > +
> > +static void x86_cpu_get_feature_prop(Object *obj,
> > +                                     struct Visitor *v,
> > +                                     void *opaque,
> > +                                     const char *name,
> > +                                     Error **errp)
> > +{
> > +    X86CPU *cpu = X86_CPU(obj);
> > +    CPUX86State *env = &cpu->env;
> > +    FeatureProperty *fp = opaque;
> > +    bool value = (env->features[fp->word] & fp->mask) == fp->mask;
> > +    visit_type_bool(v, &value, name, errp);
> > +}
> > +
> > +static void x86_cpu_set_feature_prop(Object *obj,
> > +                                     struct Visitor *v,
> > +                                     void *opaque,
> > +                                     const char *name,
> > +                                     Error **errp)
> > +{
> > +    X86CPU *cpu = X86_CPU(obj);
> > +    CPUX86State *env = &cpu->env;
> > +    FeatureProperty *fp = opaque;
> > +    bool value;
> > +    visit_type_bool(v, &value, name, errp);
> > +    if (value) {
> > +        env->features[fp->word] |= fp->mask;
> > +    } else {
> > +        env->features[fp->word] &= ~fp->mask;
> > +    }
> > +}
> > +
> > +/* Register a boolean feature-bits property.
> > + * If mask has multiple bits, all must be set for the property to return 
> > true.
> > + * The same property name can be registered multiple times to make it 
> > affect
> > + * multiple bits in the same FeatureWord.
> > + */
> > +static void x86_cpu_register_feature_prop(X86CPU *cpu,
> > +                                          const char *prop_name,
> > +                                          FeatureWord w,
> > +                                          uint32_t mask)
> > +{
> > +    FeatureProperty *fp;
> > +    ObjectProperty *op;
> > +    op = object_property_find(OBJECT(cpu), prop_name, NULL);
> > +    if (op) {
> > +        fp = op->opaque;
> > +        assert(fp->word == w);
> > +        fp->mask |= mask;
> > +    } else {
> > +        fp = g_new0(FeatureProperty, 1);
> > +        fp->word = w;
> > +        fp->mask = mask;
> > +        object_property_add(OBJECT(cpu), prop_name, "bool",
> > +                            x86_cpu_get_feature_prop,
> > +                            x86_cpu_set_feature_prop,
> > +                            NULL, fp, &error_abort);
> > +    }
> > +}
> it would be better to create generic bit property and replace above code with 
> it
> something similar to object_property_add_uint32_ptr()

object_property_add_*_ptr() adds read-only properties, and I didn't want
to make object_property_add_bit_ptr() inconsistent with the other
functions. But maybe it is better to have an inconsistent but reusable
API than making the new code non-reusable by keeping it inside
target-i386/cpu.c. I will give it a try.

BTW, it is on my wishlist to remove the existing duplication in
DEFINE_PROP_*(), QAPI, and object_property_add_*(), that are supposed to
support the same data types without duplicating code, but this may take
a while.



> 
> 
> > +
> > +static void x86_cpu_register_feature_bit_props(X86CPU *cpu,
> > +                                               FeatureWord w,
> > +                                               int bit)
> > +{
> > +    int i;
> > +    char **names;
> > +    FeatureWordInfo *fi = &feature_word_info[w];
> > +
> > +    if (!fi->feat_names) {
> > +        return;
> > +    }
> > +    if (!fi->feat_names[bit]) {
> > +        return;
> > +    }
> > +
> > +    names = g_strsplit(fi->feat_names[bit], "|", 0);
> > +    for (i = 0; names[i]; i++) {
> > +        char *feat_name = names[i];
> > +        char *prop_name = g_strdup_printf("feat-%s", feat_name);
> > +        x86_cpu_register_feature_prop(cpu, prop_name, w, (1UL << bit));
> it might be better instead of creating duplicate property to make an alias

I wasn't aware of property aliases. I will take a look. Thanks!

-- 
Eduardo