[PATCH] remove CONFIG_X86_NUMA

[PATCH] remove CONFIG_X86_NUMA

[Christoph Hellwig]

It's only used to hide two entries in arch/i386/Kconfig.

ACKed my mbligh.


--- 1.14/arch/i386/Kconfig	Tue Dec 17 19:50:36 2002
+++ edited/arch/i386/Kconfig	Sat Dec 21 18:10:12 2002
@@ -385,14 +385,8 @@
 	  This is purely to save memory - each supported CPU adds
 	  approximately eight kilobytes to the kernel image.
 
-config X86_NUMA
-	bool "Multi-node NUMA system support"
-	depends on SMP
-
-#Platform Choices
 config X86_NUMAQ
 	bool "Multiquad (IBM/Sequent) NUMAQ support"
-	depends on X86_NUMA
 	help
 	  This option is used for getting Linux to run on a (IBM/Sequent) NUMA 
 	  multiquad box. This changes the way that processors are bootstrapped,
@@ -402,7 +396,6 @@
 
 config X86_SUMMIT
 	bool "IBM x440 (Summit/EXA) support"
-	depends on X86_NUMA
 	help
 	  This option is needed for IBM systems that use the Summit/EXA chipset.
 	  In particular, it is needed for the x440.
@@ -411,7 +404,7 @@
 
 config CLUSTERED_APIC
 	bool
-	depends on X86_NUMA && (X86_NUMAQ || X86_SUMMIT)
+	depends on X86_NUMAQ || X86_SUMMIT
 	default y
 
 # Common NUMA Features

Re: [PATCH] remove CONFIG_X86_NUMA

[James Bottomley]

> It's only used to hide two entries in arch/i386/Kconfig.

The patch looks good.  If it's OK to get rid of X86_NUMA, could you also move 
X86_NUMAQ under the subarch menu?

Thanks,

James

Re: [PATCH] remove CONFIG_X86_NUMA

[Christoph Hellwig]

On Sun, Dec 29, 2002 at 04:39:25PM -0600, James Bottomley wrote:
> > It's only used to hide two entries in arch/i386/Kconfig.
> 
> The patch looks good.  If it's OK to get rid of X86_NUMA, could you also move 
> X86_NUMAQ under the subarch menu?

I already wondered about that, but AFAIK a kernel with X86_NUMAQ set
still boots on a PeeCee, so it's really an option, not a choice.

Re: [PATCH] remove CONFIG_X86_NUMA

[James Bottomley]

hch@lst.de said:
> I already wondered about that, but AFAIK a kernel with X86_NUMAQ set
> still boots on a PeeCee, so it's really an option, not a choice.

It alters the mflags-y and mcore-y variables in arch/i386/Makefile, so it's 
one of the subarch choices and thus should really be under the menu options.

James

Re: [PATCH] remove CONFIG_X86_NUMA

[Christoph Hellwig]

On Sun, Dec 29, 2002 at 04:51:44PM -0600, James Bottomley wrote:
> hch@lst.de said:
> > I already wondered about that, but AFAIK a kernel with X86_NUMAQ set
> > still boots on a PeeCee, so it's really an option, not a choice.
> 
> It alters the mflags-y and mcore-y variables in arch/i386/Makefile, so it's 
> one of the subarch choices and thus should really be under the menu options.

Okay, does this patch look better?


--- 1.19/arch/i386/Kconfig	Sat Dec 28 23:18:10 2002
+++ edited/arch/i386/Kconfig	Sun Dec 29 23:10:14 2002
@@ -49,7 +49,7 @@
 
 config VOYAGER
 	bool "NCR Voyager Architecture"
-	---help---
+	help
 	  Voyager is a MCA based 32 way capable SMP architecture proprietary
 	  to NCR Corp.  Machine classes 345x/35xx/4100/51xx are voyager based.
 	  
@@ -58,9 +58,19 @@
 	  If you do not specifically know you have a Voyager based machine,
 	  say N here otherwise the kernel you build will not be bootable.
 
+config X86_NUMAQ
+	bool "IBM/Sequent NUMAQ"
+	help
+	  This option is used for getting Linux to run on a (IBM/Sequent) NUMA 
+	  multiquad box. This changes the way that processors are bootstrapped,
+	  and uses Clustered Logical APIC addressing mode instead of Flat Logical.
+	  You will need a new lynxer.elf file to flash your firmware with - send
+	  email to Martin.Bligh@us.ibm.com
+
 # Visual Workstation support is utterly broken.
 # If you want to see it working mail an VW540 to hch@infradead.org 8)
-#bool 'SGI Visual Workstation support' CONFIG_VISWS
+#config VISWS
+#	bool "SGI Visual Workstation support"
 
 endchoice
 
@@ -430,24 +440,8 @@
 	  This is purely to save memory - each supported CPU adds
 	  approximately eight kilobytes to the kernel image.
 
-config X86_NUMA
-	bool "Multi-node NUMA system support"
-	depends on SMP
-
-#Platform Choices
-config X86_NUMAQ
-	bool "Multiquad (IBM/Sequent) NUMAQ support"
-	depends on X86_NUMA
-	help
-	  This option is used for getting Linux to run on a (IBM/Sequent) NUMA 
-	  multiquad box. This changes the way that processors are bootstrapped,
-	  and uses Clustered Logical APIC addressing mode instead of Flat Logical.
-	  You will need a new lynxer.elf file to flash your firmware with - send
-	  email to Martin.Bligh@us.ibm.com
-
 config X86_SUMMIT
 	bool "IBM x440 (Summit/EXA) support"
-	depends on X86_NUMA
 	help
 	  This option is needed for IBM systems that use the Summit/EXA chipset.
 	  In particular, it is needed for the x440.
@@ -456,7 +450,7 @@
 
 config CLUSTERED_APIC
 	bool
-	depends on X86_NUMA && (X86_NUMAQ || X86_SUMMIT)
+	depends on X86_NUMAQ || X86_SUMMIT
 	default y
 
 # Common NUMA Features

Re: [PATCH] remove CONFIG_X86_NUMA

[James Bottomley]

hch@lst.de said:
> Okay, does this patch look better? 

Perfect, Thanks!

James

Re: [PATCH] remove CONFIG_X86_NUMA

[Martin J. Bligh]

>> > It's only used to hide two entries in arch/i386/Kconfig.
>>
>> The patch looks good.  If it's OK to get rid of X86_NUMA, could you also
>> move  X86_NUMAQ under the subarch menu?
>
> I already wondered about that, but AFAIK a kernel with X86_NUMAQ set
> still boots on a PeeCee, so it's really an option, not a choice.

Nope, it won't boot on a PC - you're probably thinking of Summit,
which should. I think Bill had a patch to move NUMA-Q already ...
want to publish that one?

M.

Re: [PATCH] remove CONFIG_X86_NUMA

[William Lee Irwin III]

At some point in the past, someone had their attribution removed from:
>> I already wondered about that, but AFAIK a kernel with X86_NUMAQ set
>> still boots on a PeeCee, so it's really an option, not a choice.

On Sun, Dec 29, 2002 at 02:56:40PM -0800, Martin J. Bligh wrote:
> Nope, it won't boot on a PC - you're probably thinking of Summit,
> which should. I think Bill had a patch to move NUMA-Q already ...
> want to publish that one?

It's just moving the 2 config decls inside the choice/endchoice bracket,
not worth bothering with if it's getting fixed anyway.


Bill