speedstep-smi detection broken.

speedstep-smi detection broken.

[Dave Jones]

If someone inserts speedstep-smi on a mobile P4,
it prevents other cpufreq modules from loading until it
is unloaded.

Opinions on this patch ?
(I'm tempted to just remove the printk too).

	Dave

--- local-git/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c~	2006-10-17 01:53:29.000000000 -0400
+++ local-git/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c	2006-10-17 01:54:02.000000000 -0400
@@ -362,7 +362,7 @@ static int __init speedstep_init(void)
 		break;
 	case SPEEDSTEP_PROCESSOR_P4M:
 		printk(KERN_INFO "speedstep-smi: you're trying to use this cpufreq driver on a Pentium 4-based CPU. Most likely it will not work.\n");
-		break;
+		return -ENODEV;
 	default:
 		speedstep_processor = 0;
 	}


-- 
http://www.codemonkey.org.uk

Re: speedstep-smi detection broken.

[Hiroshi Miura]

I prefer removing all of P4M case  block  if you remove the printk too.
This means simply not to support P4M with this driver and
return -ENODEV.
If user turn on debug flag, they may see
	dprintk ("No supported Intel CPU detected.\n");
as same as other CPUs.

Hiroshi

--- linux-2.6.18/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c~	2006-09-20
12:42:06.000000000 +0900
+++ linux-2.6.18/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c	2006-10-17
22:59:50.553975632 +0900
@@ -360,9 +360,6 @@
 	case SPEEDSTEP_PROCESSOR_PIII_C:
 	case SPEEDSTEP_PROCESSOR_PIII_C_EARLY:
 		break;
-	case SPEEDSTEP_PROCESSOR_P4M:
-		printk(KERN_INFO "speedstep-smi: you're trying to use this cpufreq
driver on a Pentium 4-based CPU. Most likely it will not work.\n");
-		break;
 	default:
 		speedstep_processor = 0;
 	}

2006/10/17, Dave Jones <davej@redhat.com>:
> If someone inserts speedstep-smi on a mobile P4,
> it prevents other cpufreq modules from loading until it
> is unloaded.
>
> Opinions on this patch ?
> (I'm tempted to just remove the printk too).
>
>         Dave
>
> --- local-git/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c~     2006-10-17 01:53:29.000000000 -0400
> +++ local-git/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c      2006-10-17 01:54:02.000000000 -0400
> @@ -362,7 +362,7 @@ static int __init speedstep_init(void)
>                 break;
>         case SPEEDSTEP_PROCESSOR_P4M:
>                 printk(KERN_INFO "speedstep-smi: you're trying to use this cpufreq driver on a Pentium 4-based CPU. Most likely it will not work.\n");
> -               break;
> +               return -ENODEV;
>         default:
>                 speedstep_processor = 0;
>         }
>
>
> --
> http://www.codemonkey.org.uk
>


-- 
HIroshi Miura

Re: speedstep-smi detection broken.

[Dave Jones]

On Tue, Oct 17, 2006 at 11:00:42PM +0900, Hiroshi Miura wrote:
 > I prefer removing all of P4M case  block  if you remove the printk too.
 > This means simply not to support P4M with this driver and
 > return -ENODEV.
 > If user turn on debug flag, they may see
 > 	dprintk ("No supported Intel CPU detected.\n");
 > as same as other CPUs.

Yes, even better. I'll apply this.
Want me to fake a Signed-off-by: from you, as you forgot to add one? :)

	Dave 

-- 
http://www.codemonkey.org.uk

Re: speedstep-smi detection broken.

[Hiroshi Miura]

2006/10/18, Dave Jones <davej@redhat.com>:
> On Tue, Oct 17, 2006 at 11:00:42PM +0900, Hiroshi Miura wrote:
>  > I prefer removing all of P4M case  block  if you remove the printk too.
>  > This means simply not to support P4M with this driver and
>  > return -ENODEV.
>  > If user turn on debug flag, they may see
>  >      dprintk ("No supported Intel CPU detected.\n");
>  > as same as other CPUs.
>
> Yes, even better. I'll apply this.
> Want me to fake a Signed-off-by: from you, as you forgot to add one? :)

Exactly I forgot, sorry.

Hiroshi

Signed-off-by: Hiroshi Miura <miura@da-cha.org>
--- linux-2.6.18/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c~  2006-09-20
12:42:06.000000000 +0900
+++ linux-2.6.18/arch/i386/kernel/cpu/cpufreq/speedstep-smi.c   2006-10-17
22:59:50.553975632 +0900
@@ -360,9 +360,6 @@
       case SPEEDSTEP_PROCESSOR_PIII_C:
       case SPEEDSTEP_PROCESSOR_PIII_C_EARLY:
               break;
-       case SPEEDSTEP_PROCESSOR_P4M:
-               printk(KERN_INFO "speedstep-smi: you're trying to use
this cpufreq driver on a Pentium 4-based CPU. Most likely it will not
work.\n");
-               break;
       default:
               speedstep_processor = 0;
       }

-- 
HIroshi Miura