Patchwork [09/20,v3] target-i386: add x86cpu_vendor_words2str()

login
register
mail settings
Submitter Igor Mammedov
Date Jan. 4, 2013, 7:37 p.m.
Message ID <1357328244-4410-1-git-send-email-imammedo@redhat.com>
Download mbox | patch
Permalink /patch/209528/
State New
Headers show

Comments

Igor Mammedov - Jan. 4, 2013, 7:37 p.m.
Make for() cycle reusable for the next patch

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
  v3:
    fix/swap vendor2 and vendor3 order
        Spotted-By: Eduardo Habkost <ehabkost@redhat.com>
  v2:
    place x86cpu_vendor_words2str() a bit earlier, before feature
    arrays to avoid compile error when vendor property is converted
    static property.
---
 target-i386/cpu.c |   21 ++++++++++++++-------
 1 files changed, 14 insertions(+), 7 deletions(-)
Eduardo Habkost - Jan. 4, 2013, 8:02 p.m.
On Fri, Jan 04, 2013 at 08:37:24PM +0100, Igor Mammedov wrote:
> Make for() cycle reusable for the next patch
> 
> Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> ---
>   v3:
>     fix/swap vendor2 and vendor3 order
>         Spotted-By: Eduardo Habkost <ehabkost@redhat.com>
>   v2:
>     place x86cpu_vendor_words2str() a bit earlier, before feature
>     arrays to avoid compile error when vendor property is converted
>     static property.
> ---
>  target-i386/cpu.c |   21 ++++++++++++++-------
>  1 files changed, 14 insertions(+), 7 deletions(-)
> 
> diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> index d6e4e71..e26e631 100644
> --- a/target-i386/cpu.c
> +++ b/target-i386/cpu.c
> @@ -45,6 +45,18 @@
>  #include "hw/apic_internal.h"
>  #endif
>  
> +static void x86cpu_vendor_words2str(char *dst, uint32_t ebx, uint32_t ecx,
> +                                    uint32_t edx)
> +{
> +    int i;
> +    for (i = 0; i < 4; i++) {
> +        dst[i] = ebx >> (8 * i);
> +        dst[i + 4] = ecx >> (8 * i);
> +        dst[i + 8] = edx >> (8 * i);
> +    }
> +    dst[CPUID_VENDOR_SZ] = '\0';
> +}

Now the code seems to work as expected, but the parameter names are
misleading. String bytes 4-7 (vendor2) come from EDX on the CPUID
instruction, and bytes 8-11 (vendor3) come from ECX. Look at the
x86cpu_vendor_words2str() calls you added on patch 10/20.

> +
>  /* feature flags taken from "Intel Processor Identification and the CPUID
>   * Instruction" and AMD's "CPUID Specification".  In cases of disagreement
>   * between feature naming conventions, aliases may be added.
> @@ -1131,15 +1143,10 @@ static char *x86_cpuid_get_vendor(Object *obj, Error **errp)
>      X86CPU *cpu = X86_CPU(obj);
>      CPUX86State *env = &cpu->env;
>      char *value;
> -    int i;
>  
>      value = (char *)g_malloc(CPUID_VENDOR_SZ + 1);
> -    for (i = 0; i < 4; i++) {
> -        value[i    ] = env->cpuid_vendor1 >> (8 * i);
> -        value[i + 4] = env->cpuid_vendor2 >> (8 * i);
> -        value[i + 8] = env->cpuid_vendor3 >> (8 * i);
> -    }
> -    value[CPUID_VENDOR_SZ] = '\0';
> +    x86cpu_vendor_words2str(value, env->cpuid_vendor1, env->cpuid_vendor2,
> +                            env->cpuid_vendor3);
>      return value;
>  }
>  
> -- 
> 1.7.1
>
Igor Mammedov - Jan. 4, 2013, 8:46 p.m.
On Fri, 4 Jan 2013 18:02:33 -0200
Eduardo Habkost <ehabkost@redhat.com> wrote:

> On Fri, Jan 04, 2013 at 08:37:24PM +0100, Igor Mammedov wrote:
> > Make for() cycle reusable for the next patch
> > 
> > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > ---
> >   v3:
> >     fix/swap vendor2 and vendor3 order
> >         Spotted-By: Eduardo Habkost <ehabkost@redhat.com>
> >   v2:
> >     place x86cpu_vendor_words2str() a bit earlier, before feature
> >     arrays to avoid compile error when vendor property is converted
> >     static property.
> > ---
> >  target-i386/cpu.c |   21 ++++++++++++++-------
> >  1 files changed, 14 insertions(+), 7 deletions(-)
> > 
> > diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> > index d6e4e71..e26e631 100644
> > --- a/target-i386/cpu.c
> > +++ b/target-i386/cpu.c
> > @@ -45,6 +45,18 @@
> >  #include "hw/apic_internal.h"
> >  #endif
> >  
> > +static void x86cpu_vendor_words2str(char *dst, uint32_t ebx, uint32_t ecx,
> > +                                    uint32_t edx)
> > +{
> > +    int i;
> > +    for (i = 0; i < 4; i++) {
> > +        dst[i] = ebx >> (8 * i);
> > +        dst[i + 4] = ecx >> (8 * i);
> > +        dst[i + 8] = edx >> (8 * i);
> > +    }
> > +    dst[CPUID_VENDOR_SZ] = '\0';
> > +}
> 
> Now the code seems to work as expected, but the parameter names are
> misleading. String bytes 4-7 (vendor2) come from EDX on the CPUID
> instruction, and bytes 8-11 (vendor3) come from ECX. Look at the
> x86cpu_vendor_words2str() calls you added on patch 10/20.

Perhaps naming params like this would be better: 
x86cpu_vendor_words2str(char *dst, uint32_t vendor1, uint32_t vendor2, uint32_t vendor3)

> 
[...]
> >  
> > -- 
> > 1.7.1
> > 
> 
> -- 
> Eduardo
Eduardo Habkost - Jan. 4, 2013, 10:12 p.m.
On Fri, Jan 04, 2013 at 09:46:30PM +0100, Igor Mammedov wrote:
> On Fri, 4 Jan 2013 18:02:33 -0200
> Eduardo Habkost <ehabkost@redhat.com> wrote:
> 
> > On Fri, Jan 04, 2013 at 08:37:24PM +0100, Igor Mammedov wrote:
> > > Make for() cycle reusable for the next patch
> > > 
> > > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > > ---
> > >   v3:
> > >     fix/swap vendor2 and vendor3 order
> > >         Spotted-By: Eduardo Habkost <ehabkost@redhat.com>
> > >   v2:
> > >     place x86cpu_vendor_words2str() a bit earlier, before feature
> > >     arrays to avoid compile error when vendor property is converted
> > >     static property.
> > > ---
> > >  target-i386/cpu.c |   21 ++++++++++++++-------
> > >  1 files changed, 14 insertions(+), 7 deletions(-)
> > > 
> > > diff --git a/target-i386/cpu.c b/target-i386/cpu.c
> > > index d6e4e71..e26e631 100644
> > > --- a/target-i386/cpu.c
> > > +++ b/target-i386/cpu.c
> > > @@ -45,6 +45,18 @@
> > >  #include "hw/apic_internal.h"
> > >  #endif
> > >  
> > > +static void x86cpu_vendor_words2str(char *dst, uint32_t ebx, uint32_t ecx,
> > > +                                    uint32_t edx)
> > > +{
> > > +    int i;
> > > +    for (i = 0; i < 4; i++) {
> > > +        dst[i] = ebx >> (8 * i);
> > > +        dst[i + 4] = ecx >> (8 * i);
> > > +        dst[i + 8] = edx >> (8 * i);
> > > +    }
> > > +    dst[CPUID_VENDOR_SZ] = '\0';
> > > +}
> > 
> > Now the code seems to work as expected, but the parameter names are
> > misleading. String bytes 4-7 (vendor2) come from EDX on the CPUID
> > instruction, and bytes 8-11 (vendor3) come from ECX. Look at the
> > x86cpu_vendor_words2str() calls you added on patch 10/20.
> 
> Perhaps naming params like this would be better: 
> x86cpu_vendor_words2str(char *dst, uint32_t vendor1, uint32_t vendor2, uint32_t vendor3)

Sounds good to me, considering that the x86_cpuid_get_vendor() code
doesn't know anything about ebx/ecx/edx, just about
vendor1/vendor2/vendor3. Then just the code that deals directly with
registers from the CPUID output will need to take care about the
ebx/edx/ecx ordering when calling the function.

Patch

diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index d6e4e71..e26e631 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -45,6 +45,18 @@ 
 #include "hw/apic_internal.h"
 #endif
 
+static void x86cpu_vendor_words2str(char *dst, uint32_t ebx, uint32_t ecx,
+                                    uint32_t edx)
+{
+    int i;
+    for (i = 0; i < 4; i++) {
+        dst[i] = ebx >> (8 * i);
+        dst[i + 4] = ecx >> (8 * i);
+        dst[i + 8] = edx >> (8 * i);
+    }
+    dst[CPUID_VENDOR_SZ] = '\0';
+}
+
 /* feature flags taken from "Intel Processor Identification and the CPUID
  * Instruction" and AMD's "CPUID Specification".  In cases of disagreement
  * between feature naming conventions, aliases may be added.
@@ -1131,15 +1143,10 @@  static char *x86_cpuid_get_vendor(Object *obj, Error **errp)
     X86CPU *cpu = X86_CPU(obj);
     CPUX86State *env = &cpu->env;
     char *value;
-    int i;
 
     value = (char *)g_malloc(CPUID_VENDOR_SZ + 1);
-    for (i = 0; i < 4; i++) {
-        value[i    ] = env->cpuid_vendor1 >> (8 * i);
-        value[i + 4] = env->cpuid_vendor2 >> (8 * i);
-        value[i + 8] = env->cpuid_vendor3 >> (8 * i);
-    }
-    value[CPUID_VENDOR_SZ] = '\0';
+    x86cpu_vendor_words2str(value, env->cpuid_vendor1, env->cpuid_vendor2,
+                            env->cpuid_vendor3);
     return value;
 }