RE: [BK PATCH] ACPI Update (resend 1)

-----Original Message-----
From: Dominik Brodowski [mailto:linux-JhLEnvuH02M@public.gmane.org] 
Sent: Thursday, February 06, 2003 1:27 AM
To: Linus Torvalds; Grover, Andrew
Cc: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: Re: [BK PATCH] ACPI Update (resend 1)


Sorry ...

On Wed, Feb 05, 2003 at 06:42:20PM -0800, Grover, Andrew wrote:

quoted

Dominik, I can come up with a quick fix (put a 

CONFIG_CPU_FREQ around

quoted

the code block at line 1044 in processor.c) but is there a 

better way?
No, there's no better way ... even though this code block can 
only be called
when CONFIG_CPU_FREQ is enabled -- pr->flags.performance is 
zero otherwise.

quoted

quoted

-----Original Message-----
From: Linus Torvalds [mailto:torvalds-Lhe3bsMrZseB+jHODAdFcQ@public.gmane.org] 

On Wed, 5 Feb 2003, Grover, Andrew wrote:

quoted

	bk pull http://linux-acpi.bkbits.net/linux-acpi

Hmm..

  drivers/built-in.o(.text+0x32f93): In function 
`acpi_processor_apply_limit':
  : undefined reference to `cpufreq_get_policy'

This seems to happen if ACPI is on, but CPUFREQ isn't.. Which 
is my normal 
dual P4-HT desktop requirement.

		Linus

--- linux/drivers/acpi/processor.c.original	2003-02-06 

10:13:50.000000000 +0100

+++ linux/drivers/acpi/processor.c	2003-02-06 

10:16:55.000000000 +0100

@@ -1041,6 +1041,7 @@
 		return_VALUE(-ENODEV);
 
 	if (pr->flags.performance) {
+#ifdef CONFIG_CPU_FREQ
 		px = pr->performance_platform_limit;
 		if (pr->limit.user.px > px)
 			px = pr->limit.user.px;

@@ -1055,6 +1056,7 @@
 		}
 		if (result)
 			goto end;
+#endif
 	} else if (pr->performance_platform_limit) {
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Platform limit 

event detected. Consider using ACPI P-States CPUfreq driver\n"));
 	}