Re: [PATCH] A slightly smarter dmi_scan.c

Re: [PATCH] A slightly smarter dmi_scan.c

[Alan Cox]

> > APM can also be compiled as a module.
> 
> Right. Thanks. I blatantly ignored that APM can be built as modules ;(
> The fixed patch is here.

I think I prefer it without all the ifdefs being in the code like that. It is
cleaner to read before the change. I agree the messages being printed might
be confusing in some cases but I don't like the cure.

[PATCH] A slightly smarter dmi_scan.c

[Ishan Oshadi Jayawardena]

[-- Attachment #1: Type: text/plain, Size: 363 bytes --]

> APM can also be compiled as a module.

Right. Thanks. I blatantly ignored that APM can be built as modules ;(
The fixed patch is here.

Cheerio !

	- ioj

~~~~
    Ask not of race, but ask of conduct: 
    From the stick is born the sacred fire:
    The wise ascetic, though lowly born,
    Is noble in his modest self-control.
                - Gotama Buddha
.

[-- Attachment #2: dmi.patch --]
[-- Type: text/plain, Size: 1982 bytes --]

--- linux/arch/i386/kernel/dmi_scan.c	Mon Dec 24 22:50:41 2001
+++ linux/arch/i386/kernel/dmi_scan.c.c	Mon Dec 24 22:53:21 2001
@@ -255,26 +255,29 @@
 
 static __init int set_realmode_power_off(struct dmi_blacklist *d)
 {
+#if defined (CONFIG_APM) || defined (CONFIG_APM_MODULE)
        if (apm_info.realmode_power_off == 0)
        {
                apm_info.realmode_power_off = 1;
                printk(KERN_INFO "%s bios detected. Using realmode poweroff only.\n", d->ident);
        }
+#endif
        return 0;
 }
 
-
 /* 
  * Some laptops require interrupts to be enabled during APM calls 
  */
 
 static __init int set_apm_ints(struct dmi_blacklist *d)
 {
+#if defined (CONFIG_APM) || defined (CONFIG_APM_MODULE)
 	if (apm_info.allow_ints == 0)
 	{
 		apm_info.allow_ints = 1;
 		printk(KERN_INFO "%s machine detected. Enabling interrupts during APM calls.\n", d->ident);
 	}
+#endif
 	return 0;
 }
 
@@ -284,15 +287,16 @@
 
 static __init int apm_is_horked(struct dmi_blacklist *d)
 {
+#if defined (CONFIG_APM) || defined (CONFIG_APM_MODULE)
 	if (apm_info.disabled == 0)
 	{
 		apm_info.disabled = 1;
 		printk(KERN_INFO "%s machine detected. Disabling APM.\n", d->ident);
 	}
+#endif
 	return 0;
 }
 
-
 /*
  *  Check for clue free BIOS implementations who use
  *  the following QA technique
@@ -311,10 +315,12 @@
 
 static __init int broken_apm_power(struct dmi_blacklist *d)
 {
+#if defined (CONFIG_APM) || defined (CONFIG_APM_MODULE)
 	apm_info.get_power_status_broken = 1;
 	printk(KERN_WARNING "BIOS strings suggest APM bugs, disabling power status reporting.\n");
+#endif
 	return 0;
-}		
+}
 
 /*
  * Check for a Sony Vaio system
@@ -341,8 +347,10 @@
  
 static __init int swab_apm_power_in_minutes(struct dmi_blacklist *d)
 {
+#if defined (CONFIG_APM) || defined (CONFIG_APM_MODULE)
 	apm_info.get_power_status_swabinminutes = 1;
 	printk(KERN_WARNING "BIOS strings suggest APM reports battery life in minutes and wrong byte order.\n");
+#endif
 	return 0;
 }
 

Re: [PATCH] A slightly smarter dmi_scan.c

[Alvin of Diaspar]

> > > APM can also be compiled as a module.
> >
> > Right. Thanks. I blatantly ignored that APM can be built as modules ;(
> > The fixed patch is here.
  
> I think I prefer it without all the ifdefs being in the code like that. It is
> cleaner to read before the change. I agree the messages being printed might
> be confusing in some cases but I don't like the cure.

	I agree. The code in dmi_scan.c runs very elegantly now, and changing
that flow, although might be more efficient by something like 0.001%,
isn't worth the touble.
	Sorry about adding to the ambient background radiation :)

	- ioj

~~~~
    Ask not of race, but ask of conduct: 
    From the stick is born the sacred fire:
    The wise ascetic, though lowly born,
    Is noble in his modest self-control.
                - Gotama Buddha
.