qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH for-2.11 4/6] ppc: replace inter-function cyclic


From: David Gibson
Subject: Re: [Qemu-devel] [PATCH for-2.11 4/6] ppc: replace inter-function cyclic dependency/recurssion with 2 simple lookups
Date: Wed, 30 Aug 2017 13:05:01 +1000
User-agent: Mutt/1.8.3 (2017-05-23)

On Fri, Aug 25, 2017 at 09:34:32AM +0200, Igor Mammedov wrote:
> On Fri, 25 Aug 2017 14:12:08 +1000
> David Gibson <address@hidden> wrote:
> 
> > On Thu, Aug 24, 2017 at 10:21:49AM +0200, Igor Mammedov wrote:
> > > previous patches cleaned up cpu model/alias naming which
> > > allows to simplify cpu model/alias to cpu type lookup a bit
> > > byt removing recurssion and dependency of ppc_cpu_class_by_name() /
> > > ppc_cpu_class_by_alias() on each other.
> > > Besides of simplifying code it reduces it by ~15LOC.
> > > 
> > > Signed-off-by: Igor Mammedov <address@hidden>  
> > 
> > Urm.. I think this is probably good.  But I'm having a little trouble
> > convincing myself this really has the same effect as before.
> It's hard to wrap brain around current cyclic recursion and
> how to 2 simple linear lookups could replace it.

I noticed :)

> By itself this patch won't work, it depends on 2-3/6 for
> normalized cpu type names and recursion-less alias table.
> 
> The only change in behavior here is that it does alias
> translation first and only then cpu_model to type translation.

I've had a closer look and convinced myself of that now.

Reviewed-by: David Gibson <address@hidden>

I'm sorry, I've lost track of these patches a bit, since I wasn't
originally expecting to queue them.  I can't actually remember if
there were any comments needing a respin.

Regardless, can you resend the series (including my R-bs) and I'll
queue it for 2.11.

> 
> > 
> > > ---
> > >  target/ppc/translate_init.c | 43 
> > > +++++++++++++------------------------------
> > >  1 file changed, 13 insertions(+), 30 deletions(-)
> > > 
> > > diff --git a/target/ppc/translate_init.c b/target/ppc/translate_init.c
> > > index 0325226..f1a559d 100644
> > > --- a/target/ppc/translate_init.c
> > > +++ b/target/ppc/translate_init.c
> > > @@ -10176,22 +10176,6 @@ PowerPCCPUClass 
> > > *ppc_cpu_class_by_pvr_mask(uint32_t pvr)
> > >      return pcc;
> > >  }
> > >  
> > > -static gint ppc_cpu_compare_class_name(gconstpointer a, gconstpointer b)
> > > -{
> > > -    ObjectClass *oc = (ObjectClass *)a;
> > > -    const char *name = b;
> > > -    PowerPCCPUClass *pcc = POWERPC_CPU_CLASS(oc);
> > > -
> > > -    if (strncasecmp(name, object_class_get_name(oc), strlen(name)) == 0 
> > > &&
> > > -        ppc_cpu_is_valid(pcc) &&
> > > -        strcmp(object_class_get_name(oc) + strlen(name),
> > > -               POWERPC_CPU_TYPE_SUFFIX) == 0) {
> > > -        return 0;
> > > -    }
> > > -    return -1;
> > > -}
> > > -
> > > -
> > >  static ObjectClass *ppc_cpu_class_by_name(const char *name);
> > >  
> > >  static ObjectClass *ppc_cpu_class_by_alias(PowerPCCPUAlias *alias)
> > > @@ -10216,8 +10200,8 @@ static ObjectClass 
> > > *ppc_cpu_class_by_alias(PowerPCCPUAlias *alias)
> > >  
> > >  static ObjectClass *ppc_cpu_class_by_name(const char *name)
> > >  {
> > > -    GSList *list, *item;
> > > -    ObjectClass *ret = NULL;
> > > +    char *cpu_model, *typename;
> > > +    ObjectClass *oc;
> > >      const char *p;
> > >      int i, len;
> > >  
> > > @@ -10238,21 +10222,20 @@ static ObjectClass *ppc_cpu_class_by_name(const 
> > > char *name)
> > >          }
> > >      }
> > >  
> > > -    list = object_class_get_list(TYPE_POWERPC_CPU, false);
> > > -    item = g_slist_find_custom(list, name, ppc_cpu_compare_class_name);
> > > -    if (item != NULL) {
> > > -        ret = OBJECT_CLASS(item->data);
> > > +    cpu_model = g_ascii_strup(name, -1);
> > > +    p = ppc_cpu_lookup_alias(cpu_model);
> > > +    if (p) {
> > > +        g_free(cpu_model);
> > > +        cpu_model = g_strdup(p);
> > >      }
> > > -    g_slist_free(list);
> > >  
> > > -    if (ret) {
> > > -        return ret;
> > > -    }
> > > +    typename = g_strdup_printf("%s" POWERPC_CPU_TYPE_SUFFIX, cpu_model);
> > > +    oc = object_class_by_name(typename);
> > > +    g_free(typename);
> > > +    g_free(cpu_model);
> > >  
> > > -    for (i = 0; ppc_cpu_aliases[i].alias != NULL; i++) {
> > > -        if (strcasecmp(ppc_cpu_aliases[i].alias, name) == 0) {
> > > -            return ppc_cpu_class_by_alias(&ppc_cpu_aliases[i]);
> > > -        }
> > > +    if (oc && ppc_cpu_is_valid(POWERPC_CPU_CLASS(oc))) {
> > > +        return oc;
> > >      }
> > >  
> > >      return NULL;  
> > 
> 

-- 
David Gibson                    | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
                                | _way_ _around_!
http://www.ozlabs.org/~dgibson

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]