2.5.47-mjb3 (scalability / NUMA patchset)

2.5.47-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

I've been maintaining a patchset for internal use on and off
for a while, and have decided to publish it externally. It's 
based on 2.5.47 partly because I'm slow and boring, but also
because I don't particularly want anything from 49, and there's
some bleeding edge stuff in it I'd rather not cut myself on.

The patchset contains mainly scalability and NUMA stuff, and 
anything else that stops things from irritating me. It's meant 
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running 
large SMP or NUMA boxes.

Next additions will probably be timer fixes, moving NUMA-Q to
subarch, Summit support, and shared pagetables.

http://www.aracnet.com/~fletch/linux/2.5.47/patch-2.5.47-mjb3

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

rcu_stats					Dipankar Sarma
	Gives rcu statistics

dcache_rcu					Dipankar Sarma
	Use read copy update locking for the directory entry cache

noearlyirq					Martin Bligh
	Don't allow irqs on secondary cpus during IO-APIC init until __cpu_up
	Else your machine may go belly up on boot.

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

use_generic_topo				Matt Dobson
	Something to do with tolopology that I forget.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

frameptr					Martin Bligh
	Disable -fomit_frame_pointer

confighz					Dave Hansen
	Make HZ a config option

mjb3						Martin Bligh
	Add a tag to the makefile

2.5.50-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and
anything else that stops things from irritating me. It's meant
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running
large SMP or NUMA boxes.

http://www.aracnet.com/~fletch/linux/2.5.50/patch-2.5.50-mjb1.bz2

Since 2.5.47-mjb3:

merged by Linus:
- noearlyirq					Martin Bligh et al.

all screwed up at the moment:
- dcache_rcu					Dipankar Sarma

added:
merged up to 2.5.50 base
+ devclass_panic				Bill Irwin
+ shpte						Dave McCracken
+ subarch reorg					John Stultz
+ numaq_makefile				Martin Bligh
+ numaq_apic					James Cleverdon / Martin Bligh
+ numaq_mpparse1				James Cleverdon / Martin Bligh
+ config_page_offset				Dave Hansen / Andrea
+ vmalloc_stats					Dave Hansen
+ notsc						Martin Bligh

Pending:
Speed up page init on boot (Bill Irwin)
Move more of NUMA-Q to subarch (James C / Martin / John)
Full Summit support (James C / John)
RCU routecache (?)
Per-cpu interrupt stacks (bcrl / Dave)
4K Kernel stacks config option (Dave)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
The revenge of dcache_rcu (Dipankar)
/proc/meminfo.numa (Me and others).

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

rcu_stats					Dipankar Sarma
	Gives rcu statistics

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

use_generic_topo				Matt Dobson
	Something to do with tolopology that I forget.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

frameptr					Martin Bligh
	Disable -fomit_frame_pointer

confighz					Dave Hansen
	Make HZ a config option

devclass_panic					Bill Irwin
	Reorder sysfs init for topo to avoid panic

shpte						Dave McCracken
	Shared pagetables (as a config option)

subarch reorg					John Stultz
	Move the header files for subarch under include/asm-i386

numaq_makefile				Martin Bligh
	Create shell of NUMA-Q subarch support.

numaq_apic					James Cleverdon / Martin Bligh
	Break out some apic stuff for NUMA-Q into subarch

numaq_mpparse1					James Cleverdon / Martin Bligh
	Break out some mpparse stuff for NUMA-Q into subarch

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mjb1						Martin Bligh
	Add a tag to the makefile

Re: 2.5.50-mjb1 (scalability / NUMA patchset)

[William Lee Irwin III]

On Sun, Dec 08, 2002 at 10:11:19PM -0800, Martin J. Bligh wrote:
> Speed up page init on boot (Bill Irwin)

This needs a one-liner to go from fls() to ffs() since alignment is
required and things falling on alignment boundaries have fls(n) == ffs(n)

Very straightforward indeed; one-liners off from the 3rd or 4th boot
must be very obvious observations.


Bill

2.5.50-mjb2 (scalability / NUMA patchset)

[Martin J. Bligh]

Sorry, previous patch seemed to have some problems on non-NUMA boxen.
This one seems to work fine on a 4x machine at least ... next release
will have more substance ;-)

--------------------------

The patchset contains mainly scalability and NUMA stuff, and
anything else that stops things from irritating me. It's meant
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running
large SMP or NUMA boxes.

http://www.aracnet.com/~fletch/linux/2.5.50/patch-2.5.50-mjb2.bz2

Since 2.5.50-mjb1

Returned from the dead:
+ dcache_rcu					Dipankar Sarma

Gone to renegotiate with it's maker (seems to cause problems on standard 
SMP):
- numaq_makefile				Martin Bligh
- numaq_apic					James Cleverdon / Martin Bligh
- numaq_mpparse1				James Cleverdon / Martin Bligh

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
Move more of NUMA-Q to subarch (James C / Martin / John)
Full Summit support (James C / John)
RCU routecache (?)
Per-cpu interrupt stacks (bcrl / Dave)
4K Kernel stacks config option (Dave)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
/proc/meminfo.numa (Me and others).

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

rcu_stats					Dipankar Sarma
	Gives rcu statistics

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

use_generic_topo				Matt Dobson
	Something to do with tolopology that I forget.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

frameptr					Martin Bligh
	Disable -fomit_frame_pointer

confighz					Dave Hansen
	Make HZ a config option

devclass_panic					Bill Irwin
	Reorder sysfs init for topo to avoid panic

shpte						Dave McCracken
	Shared pagetables (as a config option)

subarch reorg					John Stultz
	Move the header files for subarch under include/asm-i386

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

dcache_rcu
	Use RCU type locking for the dentry cache.

mjb2						Martin Bligh
	Add a tag to the makefile

Re: 2.5.50-mjb2 (scalability / NUMA patchset)

[William Lee Irwin III]

On Mon, Dec 09, 2002 at 10:42:03PM -0800, Martin J. Bligh wrote:
> Speed up page init on boot (Bill Irwin)

Hang on until I push the fixes for this guy out (probably tonight).


On Mon, Dec 09, 2002 at 10:42:03PM -0800, Martin J. Bligh wrote:
> devclass_panic					Bill Irwin
> 	Reorder sysfs init for topo to avoid panic

That really needs to go straight to Linus and fast, I'll push.



Thanks,
Bill

2.5.52-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and
anything else that stops things from irritating me. It's meant
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running
large SMP or NUMA boxes.

http://www.aracnet.com/~fletch/linux/2.5.52/patch-2.5.52-mjb1.bz2

Since 2.5.50-mjb2
+ thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
+ interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
+ stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
+ 4k_stack            (4K stacks pt 4)		Dave Hansen
+ numameminfo					Martin Bligh / Keith Mannthey

Returned from the dead:
+ numaq_makefile				Martin Bligh
+ numaq_apic					James Cleverdon / Martin Bligh
+ numaq_mpparse1				James Cleverdon / Martin Bligh
+ numaq_mpparse2				James Cleverdon / Martin Bligh

Merged:
- devclass_panic				Bill Irwin

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
Move more of NUMA-Q to subarch (James C / Martin / John)
Full Summit support (James C / John)
RCU routecache (?)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

use_generic_topo				Matt Dobson
	Something to do with tolopology that I forget.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Dave Hansen
	Make HZ a config option

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

shpte						Dave McCracken
	Shared pagetables (as a config option)

subarch reorg					John Stultz
	Move the header files for subarch under include/asm-i386

numaq_makefile					James Cleverdon / Martin Bligh
	Numa-Q subarch splitup

numaq_apic					James Cleverdon / Martin Bligh
	Numa-Q subarch splitup

numaq_mpparse1					James Cleverdon / Martin Bligh
	Numa-Q subarch splitup

numaq_mpparse2					James Cleverdon / Martin Bligh
	Numa-Q subarch splitup

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

mjb1						Martin Bligh
	Add a tag to the makefile

Re: 2.5.52-mjb1 (scalability / NUMA patchset)

[William Lee Irwin III]

On Tue, Dec 17, 2002 at 12:10:29AM -0800, Martin J. Bligh wrote:
> The patchset contains mainly scalability and NUMA stuff, and
> anything else that stops things from irritating me. It's meant
> to be pretty stable, not so much a testing ground for new stuff.
> I'd be very interested in feedback from other people running
> large SMP or NUMA boxes.
> http://www.aracnet.com/~fletch/linux/2.5.52/patch-2.5.52-mjb1.bz2

BTW, the mem_map initialization stuff has been pretty well shaken down,
so it's safe to slurp that up now.


Bill

Re: 2.5.52-mjb1 (scalability / NUMA patchset)

[William Lee Irwin III]

On Tue, Dec 17, 2002 at 12:10:29AM -0800, Martin J. Bligh wrote:
> The patchset contains mainly scalability and NUMA stuff, and
> anything else that stops things from irritating me. It's meant
> to be pretty stable, not so much a testing ground for new stuff.
> I'd be very interested in feedback from other people running
> large SMP or NUMA boxes.
> http://www.aracnet.com/~fletch/linux/2.5.52/patch-2.5.52-mjb1.bz2


pfn_to_nid() got lots of icache misses. Try using a macro.

 arch/i386/kernel/i386_ksyms.c |    1 -
 arch/i386/kernel/numaq.c      |   15 ++-------------
 include/asm-i386/numaq.h      |    3 ++-
 3 files changed, 4 insertions(+), 15 deletions(-)


diff -urpN linux-2.5.52-mm1/arch/i386/kernel/i386_ksyms.c mm1-2.5.52-1/arch/i386/kernel/i386_ksyms.c
--- linux-2.5.52-mm1/arch/i386/kernel/i386_ksyms.c	2002-12-16 19:29:45.000000000 -0800
+++ mm1-2.5.52-1/arch/i386/kernel/i386_ksyms.c	2002-12-17 08:47:25.000000000 -0800
@@ -67,7 +67,6 @@ EXPORT_SYMBOL(EISA_bus);
 EXPORT_SYMBOL(MCA_bus);
 #ifdef CONFIG_DISCONTIGMEM
 EXPORT_SYMBOL(node_data);
-EXPORT_SYMBOL(pfn_to_nid);
 #endif
 #ifdef CONFIG_X86_NUMAQ
 EXPORT_SYMBOL(xquad_portio);
diff -urpN linux-2.5.52-mm1/arch/i386/kernel/numaq.c mm1-2.5.52-1/arch/i386/kernel/numaq.c
--- linux-2.5.52-mm1/arch/i386/kernel/numaq.c	2002-12-15 18:08:13.000000000 -0800
+++ mm1-2.5.52-1/arch/i386/kernel/numaq.c	2002-12-17 08:51:44.000000000 -0800
@@ -27,6 +27,7 @@
 #include <linux/mm.h>
 #include <linux/bootmem.h>
 #include <linux/mmzone.h>
+#include <linux/module.h>
 #include <asm/numaq.h>
 
 /* These are needed before the pgdat's are created */
@@ -82,19 +83,7 @@ static void __init smp_dump_qct(void)
  * physnode_map[8- ] = -1;
  */
 int physnode_map[MAX_ELEMENTS] = { [0 ... (MAX_ELEMENTS - 1)] = -1};
-
-#define PFN_TO_ELEMENT(pfn) (pfn / PAGES_PER_ELEMENT)
-#define PA_TO_ELEMENT(pa) (PFN_TO_ELEMENT(pa >> PAGE_SHIFT))
-
-int pfn_to_nid(unsigned long pfn)
-{
-	int nid = physnode_map[PFN_TO_ELEMENT(pfn)];
-
-	if (nid == -1)
-		BUG(); /* address is not present */
-
-	return nid;
-}
+EXPORT_SYMBOL(physnode_map);
 
 /*
  * for each node mark the regions
diff -urpN linux-2.5.52-mm1/include/asm-i386/numaq.h mm1-2.5.52-1/include/asm-i386/numaq.h
--- linux-2.5.52-mm1/include/asm-i386/numaq.h	2002-12-15 18:08:09.000000000 -0800
+++ mm1-2.5.52-1/include/asm-i386/numaq.h	2002-12-17 08:45:19.000000000 -0800
@@ -38,10 +38,11 @@
 #define MAX_ELEMENTS 256
 #define PAGES_PER_ELEMENT (16777216/256)
 
+extern int physnode_map[];
+#define pfn_to_nid(pfn)	({ physnode_map[(pfn) / PAGES_PER_ELEMENT]; })
 #define pfn_to_pgdat(pfn) NODE_DATA(pfn_to_nid(pfn))
 #define PHYSADDR_TO_NID(pa) pfn_to_nid(pa >> PAGE_SHIFT)
 #define MAX_NUMNODES		8
-extern int pfn_to_nid(unsigned long);
 extern void get_memcfg_numaq(void);
 #define get_memcfg_numa() get_memcfg_numaq()
 

Re: 2.5.52-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

>> The patchset contains mainly scalability and NUMA stuff, and
>> anything else that stops things from irritating me. It's meant
>> to be pretty stable, not so much a testing ground for new stuff.
>> I'd be very interested in feedback from other people running
>> large SMP or NUMA boxes.
>> http://www.aracnet.com/~fletch/linux/2.5.52/patch-2.5.52-mjb1.bz2
> 
> pfn_to_nid() got lots of icache misses. Try using a macro.

How much difference does this make?

M.

2.5.53-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and
anything else that stops things from irritating me. It's meant
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running
large SMP or NUMA boxes.

http://www.aracnet.com/~fletch/linux/2.5.53/patch-2.5.53-mjb1.bz2

Since 2.5.52-mjb1
- config_hz					Dave Hansen
+ config_hz (new version) 			Andrew Morton / Dave Hansen
- i386_topo					Matt Dobson
+ i386_topo (new version)			Matt Dobson

Merged:
- numaq_makefile                                Martin Bligh
- numaq_apic                                    James Cleverdon / Martin 
Bligh
- numaq_mpparse1                                James Cleverdon / Martin 
Bligh
- numaq_mpparse2                                James Cleverdon / Martin 
Bligh

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
Final bits of NUMA-Q / clustered_apic_mode to subarch (Martin)
Full Summit support (James C / John)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)
Lockless xtime structures (Andi)

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

use_generic_topo				Matt Dobson
	Make non-NUMA PPC64 use generic topology functions

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

shpte						Dave McCracken
	Shared pagetables (as a config option)

more_numaq1					James Cleverdon / Martin Bligh
	yet more Numa-Q subarch splitup

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

mjb1						Martin Bligh
	Add a tag to the makefile

2.5.54-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and
anything else that stops things from irritating me. It's meant
to be pretty stable, not so much a testing ground for new stuff.
I'd be very interested in feedback from other people running
large SMP or NUMA boxes.

http://www.aracnet.com/~fletch/linux/2.5.54/patch-2.5.54-mjb1.bz2

Since 2.5.53-mjb1 (nothing very interesting going on, apart from
a critical bugfix to the config_hz stuff).

- config_hz		 			Andrew Morton / Dave Hansen
+ config_hz (new version) 			Andrew Morton / Dave Hansen
- use_generic_topo				Matt Dobson
+ apicid_to_node				Martin Bligh
- i386_topo					Matt Dobson
+ i386_topo (new version)			Matt Dobson / Martin Bligh
+ fix_starfire_warning				Martin Bligh
+ kallsyms fix					Andi Kleen / Daniel Ritz

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
Final bits of NUMA-Q / clustered_apic_mode to subarch (Martin)
Full Summit support (James C / John)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)
Lockless xtime structures (Andi)

kgdb						Various People
	The older version of kgdb, not the shiny new stuff in Andrew's tree.
	Yes, I'm boring and slow.

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

apicid_to_node					Martin Bligh
	Create an machine specific apicid_to_node for everyone

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

fix_starfire_warning				Martin Bligh
	Fix trivial starfire compile warning that keeps annoying me.

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

shpte						Dave McCracken
	Shared pagetables (as a config option)

more_numaq1					James Cleverdon / Martin Bligh
	yet more Numa-Q subarch splitup

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kallsyms					Andi Kleen / Daniel Ritz
	Fix some bug.

mjb1						Martin Bligh
	Add a tag to the makefile

2.5.54-mjb2 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything
else that stops things from irritating me. It's meant to be pretty stable,
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.54/patch-2.5.54-mjb2.bz2

Since 2.5.54-mjb1 (mainly finished moving NUMA-Q into subarch).

- kgdb
+ kgdb (new version)				Andrew Morton / Various People
~ i386_caching_topo (rejigged)
+ cleanup_cpu_apicid				Martin Bligh
+ smpboot_cam					Martin Bligh
+ nuke_clustered_apic				Martin Bligh
~ interrupt_stacks (fixed a warning)
~ stack_usage_check (fixed a warning)
+ do_boot_error (this was in -mjb1, but I forgot it in the notes)

Also reordered a bunch of stuff.

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
Full Summit support (James C / John)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)
Lockless xtime structures (Andi)

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

apicid_to_node					Martin Bligh
	Create an machine specific apicid_to_node for everyone

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

do_boot_error					James Cleverdon
	Change do_boot_cpu to return an error code instead of fishing globally

fix_starfire_warning				Martin Bligh
	Fix trivial starfire compile warning that keeps annoying me.

shpte						Dave McCracken
	Shared pagetables (as a config option)

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.

early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

more_numaq1					James Cleverdon / Martin Bligh
	yet more Numa-Q subarch splitup

cleanup_cpu_apicid				Martin J. Bligh
	Cleanup & simplify the apicid <-> cpu mapping stuff I put in ages ago.

smpboot_cam					Martin J. Bligh
	Remove clustered_apic_mode stuff from smpboot.c

nuke_clustered_apic				Martin J. Bligh
	Kill clustered_apic_mode and CONFIG_CLUSTERED_APIC forever.

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kallsyms					Andi Kleen / Daniel Ritz
	Fix some bug.

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.54-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

Yeah, I know I'm "fashionably late", but I wanted to clear down some
stuff before porting to 2.5.55. Will do that tommorow.

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.54/patch-2.5.54-mjb3.bz2

Since 2.5.54-mjb2 (tweaked NUMA-Q patches, added Summit support)

~ cleanup_cpu_apicid				Martin J. Bligh
~ smpboot_cam					Martin J. Bligh
+ summit1					James Cleverdon / John Stultz
+ summit2					James Cleverdon / John Stultz
+ summit3					James Cleverdon / John Stultz
+ summit4					James Cleverdon / John Stultz

Also reordered a bunch of stuff.

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)
Lockless xtime structures (Andi)

kallsyms					Andi Kleen / Daniel Ritz
	Fix stem compression bug.

apicid_to_node					Martin Bligh
	Create an machine specific apicid_to_node for everyone

i386_topo					Matt Dobson
	Some i386 topology cleanups to make it cache the data.

do_boot_error					James Cleverdon
	Change do_boot_cpu to return an error code instead of fishing globally

more_numaq1					James Cleverdon / Martin Bligh
	yet more Numa-Q subarch splitup

cleanup_cpu_apicid				Martin J. Bligh
	Cleanup & simplify the apicid <-> cpu mapping stuff I put in ages ago.

smpboot_cam					Martin J. Bligh
	Remove clustered_apic_mode stuff from smpboot.c

nuke_clustered_apic				Martin J. Bligh
	Kill clustered_apic_mode and CONFIG_CLUSTERED_APIC forever.

fix_starfire_warning				Martin Bligh
	Fix trivial starfire compile warning that keeps annoying me.

shpte						Dave McCracken
	Shared pagetables (as a config option)

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.55-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.55/patch-2.5.55-mjb1.bz2

Since 2.5.54-mjb3

merged with Linus:
- kallsyms					Andi Kleen / Daniel Ritz
- apicid_to_node				Martin Bligh
- i386_topo					Matt Dobson
- do_boot_error					James Cleverdon
- more_numaq1					James Cleverdon / Martin Bligh
- cleanup_cpu_apicid				Martin J. Bligh
- smpboot_cam					Martin J. Bligh
- nuke_clustered_apic				Martin J. Bligh
- fix_starfire_warning				Martin J. Bligh

Other:
~ summit2					James Cleverdon / John Stultz
- shpte						Dave McCracken
		(dropped temporarily until Dave merges up with 2.5.55)
~ interrupt_stacks				Dave Hansen / Ben LaHaise
~ stack_usage_check				Dave Hansen / Ben LaHaise
+ ksymsoff					Hugh Dickins

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Scheduler tunables (rml)
Lockless xtime structures (Andi)
P4 oprofile support (movement)

summit1						James Cleverdon / John Stultz
	Summit support part 1

summit2						James Cleverdon / John Stultz
	Summit support part 2

summit3						James Cleverdon / John Stultz
	Summit support part 3

summit4						James Cleverdon / John Stultz
	Summit support part 4

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

numasched1					Erich Focht
	Numa scheduler general foundation work + pooling

numasched2					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ksymoff						Hugh Dickins
	Fix off by one error in kksymoops

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.58-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.58/patch-2.5.58-mjb1.bz2

Since 2.5.55-mjb1

Notes:
The interrupt distribution on Summit should be fixed now.
I put the new minimal NUMA scheduler in, no node-balance code yet (real soon).
The summit code will break the new bigsmp subarch for now.

- ksymsoff					Hugh Dickens
			(kind of part merged, and I don't have an updated ver)
- numasched1					Erich Focht
+ summit1					James Cleverdon / John Stultz
~ summit2					John Stultz
+ min_numasched					Martin J. Bligh
+ sched_tunables				Robert Love

Pending:
NUMA node rebalancer (Erich Focht)
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Lockless xtime structures (Andi)
P4 oprofile support (movement)

summit1						James Cleverdon / John Stultz
	Summit support part 1

summit2						James Cleverdon / John Stultz
	Summit support part 2

summit3						James Cleverdon / John Stultz
	Summit support part 3

summit4						James Cleverdon / John Stultz
	Summit support part 4

summit5						James Cleverdon / John Stultz
	Summit support part 5

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

min_numasched					Martin J. Bligh
	Minimal NUMA scheduler to make balancing node-local

numasched_ilb					Michael Hohnbaum
	Numa scheduler lightweight initial load balancing.

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.58-mjb2 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.58/patch-2.5.58-mjb2.bz2

Since 2.5.58-mjb1 (~ = changed, + = added, - = dropped)

~ summit1					James Cleverdon / John Stultz
~ summit2					James Cleverdon / John Stultz
~ summit3					James Cleverdon / John Stultz
~ summit4					James Cleverdon / John Stultz
~ summit5					James Cleverdon / John Stultz
~ min_numasched					Martin J. Bligh
~ numasched_ilb					Michael Hohnbaum
+ numa_rebalancer				Erich Focht
+ mpc_apic_id					Martin J. Bligh
+ doaction					Martin J. Bligh
+ vm_enough_memory				Martin J. Bligh
+ discontig_x440				Pat Gaughen / Chandra
+ topo_hack					Pat Gaughen

Notes:
NUMA scheduler is fully there now. Please let me know of tuning data.
Summit code should be fully working, with no more interrupt distribution 
problems. Discontigmem code for x440 may not work without a small ACPI fix. 

Pending:
Speed up page init on boot (Bill Irwin)
Notsc automatic enablement (someone, please ... anyone?)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Lockless xtime structures (Andi)
P4 oprofile support (movement)
Child runs first (akpm)
New starfire driver (Ion)
New qlogic driver (Badari ??)


summit1						James Cleverdon / John Stultz
	Summit support part 1

summit2						James Cleverdon / John Stultz
	Summit support part 2

summit3						James Cleverdon / John Stultz
	Summit support part 3

summit4						James Cleverdon / John Stultz
	Summit support part 4

summit5						James Cleverdon / John Stultz
	Summit support part 5

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

min_numasched					Martin J. Bligh
	Minimal NUMA scheduler to make balancing node-local

numasched_ilb					Michael Hohnbaum
	NUMA scheduler lightweight initial load balancing.

numa_rebalancer					Erich Focht
	NUMA inter-node rebalancer tunable by architecture.

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler

local_pgdat					Bill Irwin
	Move the pgdat structure into the remapped space with lmem_map

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

vm_enough_memory				Martin J. Bligh
	Make vm_enough_memory more efficient (for overcommit = 2)

discontig_x440					Pat Gaughen / Chandra
	SLIT/SRAT parsing for x440 discontigmem

topo_hack					Pat Gaughen
	Disable some topo stuff for Summit because we're cowards.

-mjb						Martin Bligh
	Add a tag to the makefile

Re: 2.5.58-mjb2 (scalability / NUMA patchset)

[William Lee Irwin III]

On Thu, Jan 16, 2003 at 11:11:54PM -0800, Martin J. Bligh wrote:
> Speed up page init on boot (Bill Irwin)
> local_pgdat					Bill Irwin
> 	Move the pgdat structure into the remapped space with lmem_map

Any chance you could push these Linus-ward? akpm appears to have
lost the intestinal fortitude to carry NUMA-Q/Summit -specific stuff
himself, which is fine, I'd just rather not see these lost in the
shuffle, esp. as a day or two was burned on each.


Thanks,
Bill

Re: 2.5.58-mjb2 (scalability / NUMA patchset)

[Martin J. Bligh]

>> Speed up page init on boot (Bill Irwin)
>> local_pgdat					Bill Irwin
>> 	Move the pgdat structure into the remapped space with lmem_map
> 
> Any chance you could push these Linus-ward? akpm appears to have
> lost the intestinal fortitude to carry NUMA-Q/Summit -specific stuff
> himself, which is fine, I'd just rather not see these lost in the
> shuffle, esp. as a day or two was burned on each.

The first one isn't in the tree as yet ... I just haven't been excited
enough about speeding up boot speed, to be perfectly honest, seeing as 
it still takes 5 minutes anyway. It might make more sense once kexec
works, and the percentage improvement would become significant.

The local_pgdat stuff definitely makes sense ... I'd kind of prefer it 
to go after the mem_map so I don't have to think about alignment
issues so much, but it's been stable for ages, so I guess I'll push it
as soon as Linus returns from vacation ...

M.

2.5.59-mjb1 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb1.bz2

Since 2.5.58-mjb2 (~ = changed, + = added, - = dropped)

Notes:
Lots of good stuff merged up with Linus. x440 distcontigmem seems to have
problems in some circumstances, but APCI should work in this release.

merged with Linus:

- summit1					James Cleverdon / John Stultz
- summit2					James Cleverdon / John Stultz
- summit3					James Cleverdon / John Stultz
- summit4					James Cleverdon / John Stultz
- summit5					James Cleverdon / John Stultz
- min_numasched					Martin J. Bligh
- numasched_ilb					Michael Hohnbaum
- numa_rebalancer				Erich Focht
- vm_enough_memory				Martin J. Bligh

Other:

+ ingosched					Ingo Molar
~ sched_tunables				Robert Love
+ acpi_x440_hack				Anonymous Coward
+ numaq_ioapicids				William Lee Irwin
+ oprofile_p4					John Levon
+ starfire					Ion Badulescu

Pending:
Notsc automatic enablement (someone, please ... anyone?)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Lockless xtime structures (Andi)
Child runs first (akpm)
New qlogic driver (Badari ??)


dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

discontig_x440					Pat Gaughen / Chandra
	SLIT/SRAT parsing for x440 discontigmem

topo_hack					Pat Gaughen
	Disable some topo stuff for Summit because we're cowards.

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.59-mjb2 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb2.bz2

Since 2.5.59-mjb1 (~ = changed, + = added, - = dropped)

Notes:
Added frlock xtime patches, cyclone timer fixes, sched stats. 
I have new code for 4K stacks, but haven't applied it yet (next release).

+ schedstat					Rick Lindsley
~ sched_tuneables				Robert Love
- topo_hack					Pat Gaughen
+ sysfs_fix					Pat Gaughen
+ cyclone_fixes					John Stultz
+ enable_cyclone				John Stultz
+ lost_tick					John Stultz
+ frlock_xtime					Stephen Hemminger et al.
+ frlock_xtime-i386				Stephen Hemminger et al.
+ frlock_xtime-ia64				Stephen Hemminger et al.
+ frlock_xtime-other				Stephen Hemminger et al.
+ tcp_fix					Alexey
+ numa_pci_fix					Dave Hansen

Pending:
Revised 4K stacks code
Notsc automatic enablement (someone, please ... anyone?)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Child runs first (akpm)
New qlogic driver (Badari ??)
Kexec
Linux Kernel Crash Dump

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

schedstat					Rick Lindsley
	Provide stats about the scheduler under /proc/stat

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

discontig_x440					Pat Gaughen / Chandra
	SLIT/SRAT parsing for x440 discontigmem

sysfs_fix					Pat Gaughen
	Fix sysfs for x440 machines instead of some topo hack ;-)

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

cyclone_fixes					John Stultz
	Fix up some stuff for the x440's cyclone timer

enable_cyclone					John Stultz
	Enable the x440's cyclone timer

lost_tick					John Stultz
	Detect lost timer ticks

frlock_xtime					Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-i386				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-ia64				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-other				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

tcp_fix						Alexey
	Stop some tcp problem with hardware checksumming (e1000?)

numa_pci_fix					Dave Hansen
	Fix a potential error in the numa pci code from Stanford Checker

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.59-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb3.bz2

Since 2.5.59-mjb2 (~ = changed, + = added, - = dropped)

Notes:
Added frlock xtime patches, cyclone timer fixes, sched stats. 
I have new code for 4K stacks, but haven't applied it yet (next release).

~ early_printk					Dave Hansen et al.
~ interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
- sysfs_fix					Pat Gaughen
+ srat_fixes					Pat Gaughen
+ summit_smp					John Stultz
+ pgd_ctor					William Lee Irwin
+ pfn_to_nid					William Lee Irwin
+ pidmaps_nodepages				Dave Hansen

Pending:
auto_disable_tsc (John Stultz)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Child runs first (akpm)
New qlogic driver (Badari ??)
Kexec
Linux Kernel Crash Dump
Kprobes
e1000 fixes

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

schedstat					Rick Lindsley
	Provide stats about the scheduler under /proc/stat

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

discontig_x440					Pat Gaughen / Chandra
	SLIT/SRAT parsing for x440 discontigmem

srat_fixes					Pat Gaughen
	Fix sysfs for x440 machines, and make srat conditional on NUMA

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

summit_smp					John Stultz
	Make Summit config options work on standard SMP

cyclone_fixes					John Stultz
	Fix up some stuff for the x440's cyclone timer

enable_cyclone					John Stultz
	Enable the x440's cyclone timer

lost_tick					John Stultz
	Detect lost timer ticks

frlock_xtime					Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-i386				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-ia64				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-other				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

tcp_fix						Alexey
	Stop some tcp problem with hardware checksumming (e1000?)

numa_pci_fix					Dave Hansen
	Fix a potential error in the numa pci code from Stanford Checker

pgd_ctor					William Lee Irwin
	Use slabs for pgd

pfn_to_nid					William Lee Irwin
	Turn pfn_to_nid into a macro

pidmaps_nodepages				Dave Hansen
	Display NUMA information in /proc/pid/maps

-mjb						Martin Bligh
	Add a tag to the makefile

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Mark Haverkamp]

[-- Attachment #1: Type: text/plain, Size: 650 bytes --]

On Sun, 2003-02-02 at 18:08, Martin J. Bligh wrote:
> The patchset contains mainly scalability and NUMA stuff, and anything 
> else that stops things from irritating me. It's meant to be pretty stable, 
> not so much a testing ground for new stuff.
> 
> I'd be very interested in feedback from anyone willing to test on any 
> platform, however large or small.
> 
> http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb3.bz2
> 

I tried 2.5.59-mjb3 today on our 16 processor numaq and got a boot
hang.  I had been running 2.5.59-mjb2 previously.  I have included the
console output and my config file.   


-- 
Mark Haverkamp <markh@osdl.org>

[-- Attachment #2: dev16-mjb3.log --]
[-- Type: text/plain, Size: 14765 bytes --]

Loading l-2.5.59-mjb3.....................
Linux version 2.5.59-mjb3 (markh@markh1.pdx.osdl.net) (gcc version 3.2 20020903 (Red Hat Linux 8.0 3.2-7)) #1 SMP Mon Feb 3 09:32:21 PST 2003
Video mode to be used for restore is ffff
BIOS-provided physical RAM map:
 BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
 BIOS-e820: 0000000000100000 - 00000000f8000000 (usable)
 BIOS-e820: 00000000fec00000 - 00000000fec09000 (reserved)
 BIOS-e820: 00000000ffe80000 - 0000000100000000 (reserved)
 BIOS-e820: 0000000100000000 - 0000000400000000 (usable)
Reserving 10752 pages of KVA for lmem_map of node 1
Shrinking node 1 from 2097152 pages to 2086400 pages
Reserving 10752 pages of KVA for lmem_map of node 2
Shrinking node 2 from 3145728 pages to 3134976 pages
Reserving 10752 pages of KVA for lmem_map of node 3
Shrinking node 3 from 4194304 pages to 4183552 pages
Reserving total of 32256 pages for numa KVA remap
15488MB HIGHMEM available.
770MB LOWMEM available.
min_low_pfn = 968, max_low_pfn = 197120, highstart_pfn = 229376
Low memory ends at vaddr f0200000
node 0 will remap to vaddr f8000000 - f8000000
node 1 will remap to vaddr f5600000 - f8000000
node 2 will remap to vaddr f2c00000 - f5600000
node 3 will remap to vaddr f0200000 - f2c00000
High memory starts at vaddr f8000000
found SMP MP-table at 000f6040
hm, page 000f6000 reserved twice.
hm, page 000f7000 reserved twice.
On node 0 totalpages: 1048576
  DMA zone: 4096 pages, LIFO batch:1
  Normal zone: 193024 pages, LIFO batch:16
  HighMem zone: 851456 pages, LIFO batch:16
BUG: wrong zone alignment, it will crash
On node 1 totalpages: 1037824
  DMA zone: 0 pages, LIFO batch:1
  Normal zone: 0 pages, LIFO batch:1
  HighMem zone: 1037824 pages, LIFO batch:16
On node 2 totalpages: 1037824
  DMA zone: 0 pages, LIFO batch:1
  Normal zone: 0 pages, LIFO batch:1
  HighMem zone: 1037824 pages, LIFO batch:16
On node 3 totalpages: 1037824
  DMA zone: 0 pages, LIFO batch:1
  Normal zone: 0 pages, LIFO batch:1
  HighMem zone: 1037824 pages, LIFO batch:16
Intel MultiProcessor Specification v1.4
    Virtual Wire compatibility mode.
OEM ID: IBM NUMA Product ID: SBB          Found an OEM MPC table at   6009b0 - parsing it ... 
Translation: record 0, type 1, quad 0, global 3, local 3
Translation: record 1, type 1, quad 0, global 1, local 1
Translation: record 2, type 1, quad 0, global 1, local 1
Translation: record 3, type 1, quad 0, global 1, local 1
Translation: record 4, type 1, quad 1, global 1, local 3
Translation: record 5, type 1, quad 1, global 1, local 1
Translation: record 6, type 1, quad 1, global 1, local 1
Translation: record 7, type 1, quad 1, global 1, local 1
Translation: record 8, type 1, quad 2, global 1, local 3
Translation: record 9, type 1, quad 2, global 1, local 1
Translation: record 10, type 1, quad 2, global 1, local 1
Translation: record 11, type 1, quad 2, global 1, local 1
Translation: record 12, type 1, quad 3, global 1, local 3
Translation: record 13, type 1, quad 3, global 1, local 1
Translation: record 14, type 1, quad 3, global 1, local 1
Translation: record 15, type 1, quad 3, global 1, local 1
Translation: record 16, type 3, quad 0, global 0, local 0
Translation: record 17, type 3, quad 0, global 1, local 1
Translation: record 18, type 4, quad 0, global 8, local 18
Translation: record 19, type 3, quad 1, global 2, local 0
Translation: record 20, type 3, quad 1, global 3, local 1
Translation: record 21, type 4, quad 1, global 9, local 18
Translation: record 22, type 3, quad 2, global 4, local 0
Translation: record 23, type 3, quad 2, global 5, local 1
Translation: record 24, type 4, quad 2, global 10, local 18
Translation: record 25, type 3, quad 3, global 6, local 0
Translation: record 26, type 3, quad 3, global 7, local 1
Translation: record 27, type 4, quad 3, global 11, local 18
Translation: record 28, type 2, quad 0, global 13, local 14
Translation: record 29, type 2, quad 0, global 14, local 13
Translation: record 30, type 2, quad 1, global 15, local 14
Translation: record 31, type 2, quad 1, global 16, local 13
Translation: record 32, type 2, quad 2, global 17, local 14
Translation: record 33, type 2, quad 2, global 18, local 13
Translation: record 34, type 2, quad 3, global 19, local 14
Translation: record 35, type 2, quad 3, global 20, local 13
APIC at: 0xFEC08000
Processor #0 6:7 APIC version 17 (quad 0, apic 1)
Processor #4 6:7 APIC version 17 (quad 0, apic 8)
Processor #1 6:7 APIC version 17 (quad 0, apic 2)
Processor #2 6:7 APIC version 17 (quad 0, apic 4)
Processor #0 6:7 APIC version 17 (quad 1, apic 17)
Processor #4 6:7 APIC version 17 (quad 1, apic 24)
Processor #1 6:7 APIC version 17 (quad 1, apic 18)
Processor #2 6:7 APIC version 17 (quad 1, apic 20)
Processor #0 6:7 APIC version 17 (quad 2, apic 33)
Processor #4 6:7 APIC version 17 (quad 2, apic 40)
Processor #1 6:7 APIC version 17 (quad 2, apic 34)
Processor #2 6:7 APIC version 17 (quad 2, apic 36)
Processor #0 6:7 APIC version 17 (quad 3, apic 49)
Processor #4 6:7 APIC version 17 (quad 3, apic 56)
Processor #1 6:7 APIC version 17 (quad 3, apic 50)
Processor #2 6:7 APIC version 17 (quad 3, apic 52)
Bus #0 is PCI    (node 0)
Bus #1 is PCI    (node 0)
Bus #8 is EISA   (node 0)
Bus #2 is PCI    (node 1)
Bus #3 is PCI    (node 1)
Bus #9 is EISA   (node 1)
Bus #4 is PCI    (node 2)
Bus #5 is PCI    (node 2)
Bus #10 is EISA   (node 2)
Bus #6 is PCI    (node 3)
Bus #7 is PCI    (node 3)
Bus #11 is EISA   (node 3)
I/O APIC #13 Version 17 at 0xFE800000.
I/O APIC #14 Version 17 at 0xFE801000.
I/O APIC #15 Version 17 at 0xFE840000.
I/O APIC #16 Version 17 at 0xFE841000.
I/O APIC #17 Version 17 at 0xFE880000.
I/O APIC #18 Version 17 at 0xFE881000.
I/O APIC #19 Version 17 at 0xFE8C0000.
I/O APIC #20 Version 17 at 0xFE8C1000.
Enabling APIC mode:  NUMA-Q.  Using 8 I/O APICs
Processors: 16
Building zonelist for node : 0
Building zonelist for node : 1
Building zonelist for node : 2
Building zonelist for node : 3
Kernel command line: BOOT_IMAGE=l-2.5.59-mjb3 ro root=802 BOOT_FILE=/boot/vmlinuz-2.5.59-mjb3 console=ttyS0,38400 console=tty1
Initializing CPU#0
PID hash table entries: 4096 (order 12: 32768 bytes)
Detected 495.202 MHz processor.
Console: colour VGA+ 80x25

Warning! Detected 1660695 micro-second gap between interrupts.
  Compensating for 1659 lost ticks.
Call Trace:
 [<c010bb38>] handle_IRQ_event+0x38/0x60
 [<c010bd3e>] do_IRQ+0x9e/0x110
 [<c0105000>] rest_init+0x0/0x60
 [<c010a30e>] common_interrupt+0x42/0x58

Calibrating delay loop... 974.84 BogoMIPS
Initializing highpages for node 0
Initializing highpages for node 1
Initializing highpages for node 2
Initializing highpages for node 3
Memory: 16469460k/16777216k available (1596k kernel code, 45388k reserved, 606k data, 288k init, 15728640k highmem)
Dentry cache hash table entries: 524288 (order: 10, 4194304 bytes)
Inode-cache hash table entries: 524288 (order: 10, 4194304 bytes)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
-> /dev
-> /dev/console
-> /root
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
CPU0: Intel Pentium III (Katmai) stepping 03
per-CPU timeslice cutoff: 1461.65 usecs.
task migration cache decay timeout: 2 msecs.
enabled ExtINT on CPU#0
Leaving ESR disabled.
Mapping cpu 0 to node 0
Remapping cross-quad port I/O for 4 quads
xquad_portio vaddr 0xf8800000, len 00100000
Booting processor 1/2 eip 2000
Storing NMI vector
Initializing CPU#1
masked ExtINT on CPU#1
Leaving ESR disabled.
Mapping cpu 1 to node 0
Calibrating delay loop... 987.13 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU1: Intel Pentium III (Katmai) stepping 03
Booting processor 2/4 eip 2000
Storing NMI vector
Initializing CPU#2
masked ExtINT on CPU#2
Leaving ESR disabled.
Mapping cpu 2 to node 0
Calibrating delay loop... 987.13 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU2: Intel Pentium III (Katmai) stepping 03
Booting processor 3/8 eip 2000
Storing NMI vector
Initializing CPU#3
masked ExtINT on CPU#3
Leaving ESR disabled.
Mapping cpu 3 to node 0
Calibrating delay loop... 989.18 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU3: Intel Pentium III (Katmai) stepping 03
Booting processor 4/17 eip 2000
Storing NMI vector
Initializing CPU#4
masked ExtINT on CPU#4
Leaving ESR disabled.
Mapping cpu 4 to node 1
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU4: Intel Pentium III (Katmai) stepping 03
Booting processor 5/18 eip 2000
Storing NMI vector
Initializing CPU#5
masked ExtINT on CPU#5
Leaving ESR disabled.
Mapping cpu 5 to node 1
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU5: Intel Pentium III (Katmai) stepping 03
Booting processor 6/20 eip 2000
Storing NMI vector
Initializing CPU#6
masked ExtINT on CPU#6
Leaving ESR disabled.
Mapping cpu 6 to node 1
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU6: Intel Pentium III (Katmai) stepping 03
Booting processor 7/24 eip 2000
Storing NMI vector
Initializing CPU#7
masked ExtINT on CPU#7
Leaving ESR disabled.
Mapping cpu 7 to node 1
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU7: Intel Pentium III (Katmai) stepping 03
Booting processor 8/33 eip 2000
Storing NMI vector
Initializing CPU#8
masked ExtINT on CPU#8
Leaving ESR disabled.
Mapping cpu 8 to node 2
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU8: Intel Pentium III (Katmai) stepping 03
Booting processor 9/34 eip 2000
Storing NMI vector
Initializing CPU#9
masked ExtINT on CPU#9
Leaving ESR disabled.
Mapping cpu 9 to node 2
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU9: Intel Pentium III (Katmai) stepping 03
Booting processor 10/36 eip 2000
Storing NMI vector
Initializing CPU#10
masked ExtINT on CPU#10
Leaving ESR disabled.
Mapping cpu 10 to node 2
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU10: Intel Pentium III (Katmai) stepping 03
Booting processor 11/40 eip 2000
Storing NMI vector
Initializing CPU#11
masked ExtINT on CPU#11
Leaving ESR disabled.
Mapping cpu 11 to node 2
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU11: Intel Pentium III (Katmai) stepping 03
Booting processor 12/49 eip 2000
Storing NMI vector
Initializing CPU#12
masked ExtINT on CPU#12
Leaving ESR disabled.
Mapping cpu 12 to node 3
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU12: Intel Pentium III (Katmai) stepping 03
Booting processor 13/50 eip 2000
Storing NMI vector
Initializing CPU#13
masked ExtINT on CPU#13
Leaving ESR disabled.
Mapping cpu 13 to node 3
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU13: Intel Pentium III (Katmai) stepping 03
Booting processor 14/52 eip 2000
Storing NMI vector
Initializing CPU#14
masked ExtINT on CPU#14
Leaving ESR disabled.
Mapping cpu 14 to node 3
Calibrating delay loop... 985.08 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU14: Intel Pentium III (Katmai) stepping 03
Booting processor 15/56 eip 2000
Storing NMI vector
Initializing CPU#15
masked ExtINT on CPU#15
Leaving ESR disabled.
Mapping cpu 15 to node 3
Calibrating delay loop... 987.13 BogoMIPS
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU serial number disabled.
CPU15: Intel Pentium III (Katmai) stepping 03
Total of 16 processors activated (15761.40 BogoMIPS).
ENABLING IO-APIC IRQs
..TIMER: vector=0x31 pin1=2 pin2=0
..MP-BIOS bug: 8254 timer not connected to IO-APIC
...trying to set up timer (IRQ0) through the 8259A ... 
..... (found pin 0) ... failed.
...trying to set up timer as Virtual Wire IRQ...<4>
Warning! Detected 99069 micro-second gap between interrupts.
  Compensating for 98 lost ticks.
Call Trace:
 [<c010bb38>] handle_IRQ_event+0x38/0x60
 [<c010bd3e>] do_IRQ+0x9e/0x110
 [<c010a30e>] common_interrupt+0x42/0x58

 works.
testing the IO APIC.......................








.................................... done.
Using local APIC timer interrupts.
calibrating APIC timer ...
..... CPU clock speed is 494.0860 MHz.
..... host bus clock speed is 89.0974 MHz.

Warning! Detected 130932 micro-second gap between interrupts.
  Compensating for 129 lost ticks.
Call Trace:
 [<c010bb38>] handle_IRQ_event+0x38/0x60
 [<c010bd3e>] do_IRQ+0x9e/0x110
 [<c010a30e>] common_interrupt+0x42/0x58

checking TSC synchronization across 16 CPUs: 

Warning! Detected 7720312 micro-second gap between interrupts.
  Compensating for 7719 lost ticks.
Call Trace:
 [<c010bb38>] handle_IRQ_event+0x38/0x60
 [<c010bd3e>] do_IRQ+0x9e/0x110
 [<c010a30e>] common_interrupt+0x42/0x58

BIOS BUG: CPU#0 improperly initialized, has -15606 usecs TSC skew! FIXED.
BIOS BUG: CPU#1 improperly initialized, has -15606 usecs TSC skew! FIXED.
BIOS BUG: CPU#2 improperly initialized, has -15606 usecs TSC skew! FIXED.
BIOS BUG: CPU#3 improperly initialized, has -15606 usecs TSC skew! FIXED.
BIOS BUG: CPU#4 improperly initialized, has -812 usecs TSC skew! FIXED.
BIOS BUG: CPU#5 improperly initialized, has -812 usecs TSC skew! FIXED.
BIOS BUG: CPU#6 improperly initialized, has -812 usecs TSC skew! FIXED.
BIOS BUG: CPU#7 improperly initialized, has -812 usecs TSC skew! FIXED.
BIOS BUG: CPU#8 improperly initialized, has 14807 usecs TSC skew! FIXED.
BIOS BUG: CPU#9 improperly initialized, has 14807 usecs TSC skew! FIXED.
BIOS BUG: CPU#10 improperly initialized, has 14807 usecs TSC skew! FIXED.
BIOS BUG: CPU#11 improperly initialized, has 14807 usecs TSC skew! FIXED.
BIOS BUG: CPU#12 improperly initialized, has 1611 usecs TSC skew! FIXED.
BIOS BUG: CPU#13 improperly initialized, has 1611 usecs TSC skew! FIXED.
BIOS BUG: CPU#14 improperly initialized, has 1611 usecs TSC skew! FIXED.
BIOS BUG: CPU#15 improperly initialized, has 1611 usecs TSC skew! FIXED.
Starting migration thread for cpu 0
Bringing up 1
CPU 1 IS NOW UP!
Starting migration thread for cpu 1




[-- Attachment #3: .config --]
[-- Type: text/plain, Size: 13424 bytes --]

#
# Automatically generated make config: don't edit
#
CONFIG_X86=y
CONFIG_MMU=y
CONFIG_SWAP=y
CONFIG_UID16=y
CONFIG_GENERIC_ISA_DMA=y

#
# Code maturity level options
#
CONFIG_EXPERIMENTAL=y

#
# General setup
#
CONFIG_SYSVIPC=y
# CONFIG_BSD_PROCESS_ACCT is not set
CONFIG_SYSCTL=y
CONFIG_LOG_BUF_SHIFT=16

#
# Loadable module support
#
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
CONFIG_OBSOLETE_MODPARM=y
CONFIG_KMOD=y

#
# Processor type and features
#
# CONFIG_X86_PC is not set
# CONFIG_X86_VOYAGER is not set
CONFIG_X86_NUMAQ=y
# CONFIG_X86_SUMMIT is not set
# CONFIG_X86_BIGSMP is not set
# CONFIG_M386 is not set
# CONFIG_M486 is not set
# CONFIG_M586 is not set
# CONFIG_M586TSC is not set
# CONFIG_M586MMX is not set
# CONFIG_M686 is not set
# CONFIG_MPENTIUMII is not set
CONFIG_MPENTIUMIII=y
# CONFIG_MPENTIUM4 is not set
# CONFIG_MK6 is not set
# CONFIG_MK7 is not set
# CONFIG_MK8 is not set
# CONFIG_MELAN is not set
# CONFIG_MCRUSOE is not set
# CONFIG_MWINCHIPC6 is not set
# CONFIG_MWINCHIP2 is not set
# CONFIG_MWINCHIP3D is not set
# CONFIG_MCYRIXIII is not set
CONFIG_X86_CMPXCHG=y
CONFIG_X86_XADD=y
CONFIG_X86_L1_CACHE_SHIFT=5
CONFIG_RWSEM_XCHGADD_ALGORITHM=y
CONFIG_X86_WP_WORKS_OK=y
CONFIG_X86_INVLPG=y
CONFIG_X86_BSWAP=y
CONFIG_X86_POPAD_OK=y
CONFIG_X86_GOOD_APIC=y
CONFIG_X86_INTEL_USERCOPY=y
CONFIG_X86_USE_PPRO_CHECKSUM=y
CONFIG_X86_PREFETCH=y
CONFIG_X86_CMOV=y
# CONFIG_HUGETLB_PAGE is not set
CONFIG_SMP=y
# CONFIG_PREEMPT is not set
CONFIG_X86_LOCAL_APIC=y
CONFIG_X86_IO_APIC=y
CONFIG_NR_CPUS=32
CONFIG_NUMA=y
CONFIG_DISCONTIGMEM=y
CONFIG_HAVE_ARCH_BOOTMEM_NODE=y
# CONFIG_X86_MCE is not set
# CONFIG_TOSHIBA is not set
# CONFIG_I8K is not set
# CONFIG_MICROCODE is not set
# CONFIG_X86_MSR is not set
CONFIG_X86_CPUID=y
# CONFIG_EDD is not set
# CONFIG_NOHIGHMEM is not set
# CONFIG_HIGHMEM4G is not set
CONFIG_HIGHMEM64G=y
# CONFIG_05GB is not set
CONFIG_1GB=y
# CONFIG_2GB is not set
# CONFIG_3GB is not set
CONFIG_HIGHMEM=y
CONFIG_X86_PAE=y
# CONFIG_HIGHPTE is not set
# CONFIG_4K_STACK is not set
# CONFIG_MATH_EMULATION is not set
CONFIG_MTRR=y
# CONFIG_100HZ is not set
CONFIG_1000HZ=y
CONFIG_HAVE_DEC_LOCK=y

#
# Power management options (ACPI, APM)
#
# CONFIG_PM is not set

#
# ACPI Support
#
# CONFIG_ACPI is not set
# CONFIG_CPU_FREQ is not set

#
# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
#
CONFIG_PCI=y
# CONFIG_PCI_GOBIOS is not set
# CONFIG_PCI_GODIRECT is not set
CONFIG_PCI_GOANY=y
CONFIG_PCI_BIOS=y
CONFIG_PCI_DIRECT=y
# CONFIG_SCx200 is not set
# CONFIG_PCI_LEGACY_PROC is not set
CONFIG_PCI_NAMES=y
CONFIG_ISA=y
# CONFIG_EISA is not set
# CONFIG_MCA is not set
# CONFIG_HOTPLUG is not set

#
# Executable file formats
#
CONFIG_KCORE_ELF=y
# CONFIG_KCORE_AOUT is not set
CONFIG_BINFMT_AOUT=y
CONFIG_BINFMT_ELF=y
CONFIG_BINFMT_MISC=y

#
# Memory Technology Devices (MTD)
#
# CONFIG_MTD is not set

#
# Parallel port support
#
# CONFIG_PARPORT is not set

#
# Plug and Play support
#
# CONFIG_PNP is not set

#
# Block devices
#
# CONFIG_BLK_DEV_FD is not set
# CONFIG_BLK_DEV_XD is not set
# CONFIG_BLK_CPQ_DA is not set
# CONFIG_BLK_CPQ_CISS_DA is not set
# CONFIG_BLK_DEV_DAC960 is not set
# CONFIG_BLK_DEV_UMEM is not set
# CONFIG_BLK_DEV_LOOP is not set
# CONFIG_BLK_DEV_NBD is not set
# CONFIG_BLK_DEV_RAM is not set
CONFIG_LBD=y

#
# ATA/ATAPI/MFM/RLL device support
#
# CONFIG_IDE is not set

#
# SCSI device support
#
CONFIG_SCSI=y

#
# SCSI support type (disk, tape, CD-ROM)
#
CONFIG_BLK_DEV_SD=y
CONFIG_CHR_DEV_ST=y
# CONFIG_CHR_DEV_OSST is not set
CONFIG_BLK_DEV_SR=y
# CONFIG_BLK_DEV_SR_VENDOR is not set
# CONFIG_CHR_DEV_SG is not set

#
# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
#
CONFIG_SCSI_MULTI_LUN=y
CONFIG_SCSI_REPORT_LUNS=y
# CONFIG_SCSI_CONSTANTS is not set
# CONFIG_SCSI_LOGGING is not set

#
# SCSI low-level drivers
#
# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
# CONFIG_SCSI_7000FASST is not set
# CONFIG_SCSI_ACARD is not set
# CONFIG_SCSI_AHA152X is not set
# CONFIG_SCSI_AHA1542 is not set
# CONFIG_SCSI_AACRAID is not set
# CONFIG_SCSI_AIC7XXX is not set
# CONFIG_SCSI_AIC7XXX_OLD is not set
# CONFIG_SCSI_AIC79XX is not set
# CONFIG_SCSI_DPT_I2O is not set
# CONFIG_SCSI_ADVANSYS is not set
# CONFIG_SCSI_IN2000 is not set
# CONFIG_SCSI_AM53C974 is not set
# CONFIG_SCSI_MEGARAID is not set
# CONFIG_SCSI_BUSLOGIC is not set
# CONFIG_SCSI_CPQFCTS is not set
# CONFIG_SCSI_DMX3191D is not set
# CONFIG_SCSI_DTC3280 is not set
# CONFIG_SCSI_EATA is not set
# CONFIG_SCSI_EATA_DMA is not set
# CONFIG_SCSI_EATA_PIO is not set
# CONFIG_SCSI_FUTURE_DOMAIN is not set
# CONFIG_SCSI_GDTH is not set
# CONFIG_SCSI_GENERIC_NCR5380 is not set
# CONFIG_SCSI_GENERIC_NCR5380_MMIO is not set
# CONFIG_SCSI_IPS is not set
# CONFIG_SCSI_INITIO is not set
# CONFIG_SCSI_INIA100 is not set
# CONFIG_SCSI_NCR53C406A is not set
# CONFIG_SCSI_NCR53C7xx is not set
# CONFIG_SCSI_SYM53C8XX_2 is not set
# CONFIG_SCSI_NCR53C8XX is not set
# CONFIG_SCSI_SYM53C8XX is not set
# CONFIG_SCSI_PAS16 is not set
# CONFIG_SCSI_PCI2000 is not set
# CONFIG_SCSI_PCI2220I is not set
# CONFIG_SCSI_PSI240I is not set
# CONFIG_SCSI_QLOGIC_FAS is not set
CONFIG_SCSI_QLOGIC_ISP=y
# CONFIG_SCSI_QLOGIC_FC is not set
# CONFIG_SCSI_QLOGIC_1280 is not set
# CONFIG_SCSI_SEAGATE is not set
# CONFIG_SCSI_SIM710 is not set
# CONFIG_SCSI_SYM53C416 is not set
# CONFIG_SCSI_DC390T is not set
# CONFIG_SCSI_T128 is not set
# CONFIG_SCSI_U14_34F is not set
# CONFIG_SCSI_ULTRASTOR is not set
# CONFIG_SCSI_NSP32 is not set
# CONFIG_SCSI_DEBUG is not set

#
# Old CD-ROM drivers (not SCSI, not IDE)
#
# CONFIG_CD_NO_IDESCSI is not set

#
# Multi-device support (RAID and LVM)
#
# CONFIG_MD is not set

#
# Fusion MPT device support
#
# CONFIG_FUSION is not set

#
# IEEE 1394 (FireWire) support (EXPERIMENTAL)
#
# CONFIG_IEEE1394 is not set

#
# I2O device support
#
# CONFIG_I2O is not set

#
# Networking support
#
CONFIG_NET=y

#
# Networking options
#
CONFIG_PACKET=y
# CONFIG_PACKET_MMAP is not set
# CONFIG_NETLINK_DEV is not set
# CONFIG_NETFILTER is not set
# CONFIG_FILTER is not set
CONFIG_UNIX=y
# CONFIG_NET_KEY is not set
CONFIG_INET=y
# CONFIG_IP_MULTICAST is not set
# CONFIG_IP_ADVANCED_ROUTER is not set
# CONFIG_IP_PNP is not set
# CONFIG_NET_IPIP is not set
# CONFIG_NET_IPGRE is not set
# CONFIG_ARPD is not set
# CONFIG_INET_ECN is not set
# CONFIG_SYN_COOKIES is not set
# CONFIG_INET_AH is not set
# CONFIG_INET_ESP is not set
# CONFIG_XFRM_USER is not set
# CONFIG_IPV6 is not set

#
# SCTP Configuration (EXPERIMENTAL)
#
CONFIG_IPV6_SCTP__=y
# CONFIG_IP_SCTP is not set
# CONFIG_ATM is not set
# CONFIG_VLAN_8021Q is not set
# CONFIG_LLC is not set
# CONFIG_DECNET is not set
# CONFIG_BRIDGE is not set
# CONFIG_X25 is not set
# CONFIG_LAPB is not set
# CONFIG_NET_DIVERT is not set
# CONFIG_ECONET is not set
# CONFIG_WAN_ROUTER is not set
# CONFIG_NET_FASTROUTE is not set
# CONFIG_NET_HW_FLOWCONTROL is not set

#
# QoS and/or fair queueing
#
# CONFIG_NET_SCHED is not set

#
# Network testing
#
# CONFIG_NET_PKTGEN is not set
CONFIG_NETDEVICES=y

#
# ARCnet devices
#
# CONFIG_ARCNET is not set
# CONFIG_DUMMY is not set
# CONFIG_BONDING is not set
# CONFIG_EQUALIZER is not set
# CONFIG_TUN is not set
# CONFIG_ETHERTAP is not set

#
# Ethernet (10 or 100Mbit)
#
CONFIG_NET_ETHERNET=y
# CONFIG_HAPPYMEAL is not set
# CONFIG_SUNGEM is not set
# CONFIG_NET_VENDOR_3COM is not set
# CONFIG_LANCE is not set
# CONFIG_NET_VENDOR_SMC is not set
# CONFIG_NET_VENDOR_RACAL is not set

#
# Tulip family network device support
#
# CONFIG_NET_TULIP is not set
# CONFIG_AT1700 is not set
# CONFIG_DEPCA is not set
# CONFIG_HP100 is not set
# CONFIG_NET_ISA is not set
CONFIG_NET_PCI=y
# CONFIG_PCNET32 is not set
# CONFIG_AMD8111_ETH is not set
CONFIG_ADAPTEC_STARFIRE=y
# CONFIG_AC3200 is not set
# CONFIG_APRICOT is not set
# CONFIG_B44 is not set
# CONFIG_CS89x0 is not set
# CONFIG_DGRS is not set
CONFIG_EEPRO100=y
# CONFIG_E100 is not set
# CONFIG_FEALNX is not set
# CONFIG_NATSEMI is not set
# CONFIG_NE2K_PCI is not set
# CONFIG_8139CP is not set
# CONFIG_8139TOO is not set
# CONFIG_SIS900 is not set
# CONFIG_EPIC100 is not set
# CONFIG_SUNDANCE is not set
# CONFIG_TLAN is not set
# CONFIG_VIA_RHINE is not set
# CONFIG_NET_POCKET is not set

#
# Ethernet (1000 Mbit)
#
# CONFIG_ACENIC is not set
# CONFIG_DL2K is not set
# CONFIG_E1000 is not set
# CONFIG_NS83820 is not set
# CONFIG_HAMACHI is not set
# CONFIG_YELLOWFIN is not set
# CONFIG_R8169 is not set
# CONFIG_SK98LIN is not set
# CONFIG_TIGON3 is not set
# CONFIG_FDDI is not set
# CONFIG_HIPPI is not set
# CONFIG_PPP is not set
# CONFIG_SLIP is not set

#
# Wireless LAN (non-hamradio)
#
# CONFIG_NET_RADIO is not set

#
# Token Ring devices (depends on LLC=y)
#
# CONFIG_NET_FC is not set
# CONFIG_RCPCI is not set
# CONFIG_SHAPER is not set

#
# Wan interfaces
#
# CONFIG_WAN is not set

#
# Amateur Radio support
#
# CONFIG_HAMRADIO is not set

#
# IrDA (infrared) support
#
# CONFIG_IRDA is not set

#
# ISDN subsystem
#
# CONFIG_ISDN_BOOL is not set

#
# Telephony Support
#
# CONFIG_PHONE is not set

#
# Input device support
#
CONFIG_INPUT=y

#
# Userland interfaces
#
# CONFIG_INPUT_MOUSEDEV is not set
# CONFIG_INPUT_JOYDEV is not set
# CONFIG_INPUT_TSDEV is not set
# CONFIG_INPUT_EVDEV is not set
# CONFIG_INPUT_EVBUG is not set

#
# Input I/O drivers
#
# CONFIG_GAMEPORT is not set
CONFIG_SOUND_GAMEPORT=y
# CONFIG_SERIO is not set

#
# Input Device Drivers
#
CONFIG_INPUT_KEYBOARD=y
CONFIG_INPUT_MOUSE=y
# CONFIG_MOUSE_INPORT is not set
# CONFIG_MOUSE_LOGIBM is not set
# CONFIG_MOUSE_PC110PAD is not set
# CONFIG_INPUT_JOYSTICK is not set
# CONFIG_INPUT_TOUCHSCREEN is not set
# CONFIG_INPUT_MISC is not set

#
# Character devices
#
CONFIG_VT=y
CONFIG_VT_CONSOLE=y
CONFIG_HW_CONSOLE=y
# CONFIG_SERIAL_NONSTANDARD is not set

#
# Serial drivers
#
CONFIG_SERIAL_8250=y
CONFIG_SERIAL_8250_CONSOLE=y
# CONFIG_SERIAL_8250_EXTENDED is not set

#
# Non-8250 serial port support
#
CONFIG_SERIAL_CORE=y
CONFIG_SERIAL_CORE_CONSOLE=y
CONFIG_UNIX98_PTYS=y
CONFIG_UNIX98_PTY_COUNT=256

#
# I2C support
#
# CONFIG_I2C is not set

#
# I2C Hardware Sensors Mainboard support
#

#
# I2C Hardware Sensors Chip support
#

#
# Mice
#
# CONFIG_BUSMOUSE is not set
# CONFIG_QIC02_TAPE is not set

#
# IPMI
#
# CONFIG_IPMI_HANDLER is not set

#
# Watchdog Cards
#
# CONFIG_WATCHDOG is not set
CONFIG_INTEL_RNG=y
# CONFIG_AMD_RNG is not set
# CONFIG_NVRAM is not set
# CONFIG_RTC is not set
# CONFIG_GEN_RTC is not set
# CONFIG_DTLK is not set
# CONFIG_R3964 is not set
# CONFIG_APPLICOM is not set
# CONFIG_SONYPI is not set

#
# Ftape, the floppy tape device driver
#
# CONFIG_FTAPE is not set
# CONFIG_AGP is not set
# CONFIG_DRM is not set
# CONFIG_MWAVE is not set
# CONFIG_RAW_DRIVER is not set

#
# Multimedia devices
#
# CONFIG_VIDEO_DEV is not set

#
# File systems
#
# CONFIG_QUOTA is not set
# CONFIG_AUTOFS_FS is not set
# CONFIG_AUTOFS4_FS is not set
# CONFIG_REISERFS_FS is not set
# CONFIG_ADFS_FS is not set
# CONFIG_AFFS_FS is not set
# CONFIG_HFS_FS is not set
# CONFIG_BEFS_FS is not set
# CONFIG_BFS_FS is not set
CONFIG_EXT3_FS=y
# CONFIG_EXT3_FS_XATTR is not set
CONFIG_JBD=y
# CONFIG_JBD_DEBUG is not set
# CONFIG_FAT_FS is not set
# CONFIG_EFS_FS is not set
# CONFIG_CRAMFS is not set
# CONFIG_TMPFS is not set
CONFIG_RAMFS=y
CONFIG_ISO9660_FS=y
# CONFIG_JOLIET is not set
# CONFIG_ZISOFS is not set
# CONFIG_JFS_FS is not set
# CONFIG_MINIX_FS is not set
# CONFIG_VXFS_FS is not set
# CONFIG_NTFS_FS is not set
# CONFIG_HPFS_FS is not set
CONFIG_PROC_FS=y
# CONFIG_DEVFS_FS is not set
CONFIG_DEVPTS_FS=y
# CONFIG_QNX4FS_FS is not set
# CONFIG_ROMFS_FS is not set
CONFIG_EXT2_FS=y
# CONFIG_EXT2_FS_XATTR is not set
# CONFIG_SYSV_FS is not set
# CONFIG_UDF_FS is not set
# CONFIG_UFS_FS is not set
# CONFIG_XFS_FS is not set

#
# Network File Systems
#
# CONFIG_CODA_FS is not set
# CONFIG_INTERMEZZO_FS is not set
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
# CONFIG_NFS_V4 is not set
CONFIG_NFSD=y
CONFIG_NFSD_V3=y
# CONFIG_NFSD_V4 is not set
# CONFIG_NFSD_TCP is not set
CONFIG_SUNRPC=y
# CONFIG_SUNRPC_GSS is not set
CONFIG_LOCKD=y
CONFIG_LOCKD_V4=y
CONFIG_EXPORTFS=y
# CONFIG_CIFS is not set
# CONFIG_SMB_FS is not set
# CONFIG_NCP_FS is not set
# CONFIG_AFS_FS is not set

#
# Partition Types
#
# CONFIG_PARTITION_ADVANCED is not set
CONFIG_MSDOS_PARTITION=y

#
# Graphics support
#
# CONFIG_FB is not set
# CONFIG_VIDEO_SELECT is not set

#
# Console display driver support
#
CONFIG_VGA_CONSOLE=y
# CONFIG_MDA_CONSOLE is not set
CONFIG_DUMMY_CONSOLE=y

#
# Sound
#
# CONFIG_SOUND is not set

#
# USB support
#
# CONFIG_USB is not set

#
# Bluetooth support
#
# CONFIG_BT is not set

#
# Profiling support
#
CONFIG_PROFILING=y
CONFIG_OPROFILE=y

#
# Kernel hacking
#
CONFIG_DEBUG_KERNEL=y
# CONFIG_DEBUG_STACKOVERFLOW is not set
# CONFIG_DEBUG_SLAB is not set
# CONFIG_X86_REMOTE_DEBUG is not set
# CONFIG_DEBUG_IOVIRT is not set
CONFIG_MAGIC_SYSRQ=y
# CONFIG_EARLY_PRINTK is not set
# CONFIG_DEBUG_SPINLOCK is not set
# CONFIG_DEBUG_HIGHMEM is not set
CONFIG_KALLSYMS=y
# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
# CONFIG_FRAME_POINTER is not set
CONFIG_X86_EXTRA_IRQS=y
CONFIG_X86_FIND_SMP_CONFIG=y
CONFIG_X86_MPPARSE=y

#
# Security options
#
# CONFIG_SECURITY is not set

#
# Cryptographic options
#
# CONFIG_CRYPTO is not set

#
# Library routines
#
CONFIG_CRC32=y
CONFIG_X86_SMP=y
CONFIG_X86_HT=y
CONFIG_X86_BIOS_REBOOT=y
CONFIG_X86_TRAMPOLINE=y

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

>> The patchset contains mainly scalability and NUMA stuff, and anything 
>> else that stops things from irritating me. It's meant to be pretty stable, 
>> not so much a testing ground for new stuff.
>> 
>> I'd be very interested in feedback from anyone willing to test on any 
>> platform, however large or small.
>> 
>> http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb3.bz2
>> 
> 
> I tried 2.5.59-mjb3 today on our 16 processor numaq and got a boot
> hang.  I had been running 2.5.59-mjb2 previously.  I have included the
> console output and my config file.   

What gcc are you using? I'm betting 3.2 ... 2.95 seems to work fine.
(still might be an issue with the patch, just trying to track it down)

M.

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Mark Haverkamp]

On Mon, 2003-02-03 at 10:42, Martin J. Bligh wrote:
> >> The patchset contains mainly scalability and NUMA stuff, and anything 
> >> else that stops things from irritating me. It's meant to be pretty stable, 
> >> not so much a testing ground for new stuff.
> >> 
> >> I'd be very interested in feedback from anyone willing to test on any 
> >> platform, however large or small.
> >> 
> >> http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb3.bz2
> >> 
> > 
> > I tried 2.5.59-mjb3 today on our 16 processor numaq and got a boot
> > hang.  I had been running 2.5.59-mjb2 previously.  I have included the
> > console output and my config file.   
> 
> What gcc are you using? I'm betting 3.2 ... 2.95 seems to work fine.

You are right, I am using:

gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)


> (still might be an issue with the patch, just trying to track it down)
> 
> M.
-- 
Mark Haverkamp <markh@osdl.org>

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

>> What gcc are you using? I'm betting 3.2 ... 2.95 seems to work fine.
> 
> You are right, I am using:
> 
> gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
> 
> 
>> (still might be an issue with the patch, just trying to track it down)

Could you find the definition of cpu_online_map, and remove the "=1"
initialisation from it ... see if that fixes it? (I just added that)
Seems suspiciously closely related ... if that doesn't do it, I'll 
try to diagnose here.

Thanks,

M.

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Mark Haverkamp]

On Mon, 2003-02-03 at 11:03, Martin J. Bligh wrote:
> >> What gcc are you using? I'm betting 3.2 ... 2.95 seems to work fine.
> > 
> > You are right, I am using:
> > 
> > gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
> > 
> > 
> >> (still might be an issue with the patch, just trying to track it down)
> 
> Could you find the definition of cpu_online_map, and remove the "=1"
> initialisation from it ... see if that fixes it? (I just added that)
> Seems suspiciously closely related ... if that doesn't do it, I'll 
> try to diagnose here.
> 
> Thanks,
> 
> M.

I removed the = 1 from cpu_online_map in arch/i386/kernel/smpboot.c and
got the boot hang in the same place.

Mark.
-- 
Mark Haverkamp <markh@osdl.org>

Re: 2.5.59-mjb3 (scalability / NUMA patchset)

[Martin J. Bligh]

[-- Attachment #1: Type: text/plain, Size: 604 bytes --]

>> > I tried 2.5.59-mjb3 today on our 16 processor numaq and got a boot
>> > hang.  I had been running 2.5.59-mjb2 previously.  I have included the
>> > console output and my config file.   
>> 
>> What gcc are you using? I'm betting 3.2 ... 2.95 seems to work fine.
> 
> You are right, I am using:
> 
> gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
>
>> (still might be an issue with the patch, just trying to track it down)

OK, it's the attatched patch's fault. Can you patch -R it, and confirm that
fixes it? (does for me).

Dave, can you take a look at why this breaks with gcc 3.2?

Thanks,

M.

[-- Attachment #2: B1-irqstackfix-2.5.59-mjb1-0.txt --]
[-- Type: text/plain, Size: 8550 bytes --]

diff -ur linux-2.5.59-mjb1/arch/i386/Kconfig linux-2.5.59-numaq-mjb1/arch/i386/Kconfig
--- linux-2.5.59-mjb1/arch/i386/Kconfig	Sun Jan 26 23:16:19 2003
+++ linux-2.5.59-numaq-mjb1/arch/i386/Kconfig	Mon Jan 27 10:59:46 2003
@@ -363,6 +363,11 @@
 	depends on MK8 || MPENTIUM4
 	default y
 
+config X86_CMOV
+	bool
+	depends on M686 || MPENTIUMII || MPENTIUMIII || MPENTIUM4 || MK8 || MCRUSOE
+	default y
+
 config HUGETLB_PAGE
 	bool "Huge TLB Page Support"
 	help
diff -ur linux-2.5.59-mjb1/arch/i386/kernel/apic.c linux-2.5.59-numaq-mjb1/arch/i386/kernel/apic.c
--- linux-2.5.59-mjb1/arch/i386/kernel/apic.c	Sun Jan 26 23:14:51 2003
+++ linux-2.5.59-numaq-mjb1/arch/i386/kernel/apic.c	Sun Jan 26 14:23:22 2003
@@ -1038,7 +1038,8 @@
  *   interrupt as well. Thus we cannot inline the local irq ... ]
  */
 
-void smp_apic_timer_interrupt(struct pt_regs regs)
+struct pt_regs * smp_apic_timer_interrupt(struct pt_regs* regs) __attribute__((regparm(1)));
+struct pt_regs * smp_apic_timer_interrupt(struct pt_regs* regs)
 {
 	int cpu = smp_processor_id();
 
@@ -1058,14 +1059,16 @@
 	 * interrupt lock, which is the WrongThing (tm) to do.
 	 */
 	irq_enter();
-	smp_local_timer_interrupt(&regs);
+	smp_local_timer_interrupt(regs);
 	irq_exit();
+	return regs;
 }
 
 /*
  * This interrupt should _never_ happen with our APIC/SMP architecture
  */
-asmlinkage void smp_spurious_interrupt(void)
+struct pt_regs * smp_spurious_interrupt(struct pt_regs* regs) __attribute__((regparm(1)));
+struct pt_regs * smp_spurious_interrupt(struct pt_regs* regs)
 {
 	unsigned long v;
 
@@ -1083,13 +1086,15 @@
 	printk(KERN_INFO "spurious APIC interrupt on CPU#%d, should never happen.\n",
 			smp_processor_id());
 	irq_exit();
+	return regs;
 }
 
 /*
  * This interrupt should never happen with our APIC/SMP architecture
  */
 
-asmlinkage void smp_error_interrupt(void)
+struct pt_regs * smp_error_interrupt(struct pt_regs* regs) __attribute__((regparm(1)));
+struct pt_regs * smp_error_interrupt(struct pt_regs* regs)
 {
 	unsigned long v, v1;
 
@@ -1114,6 +1119,7 @@
 	printk (KERN_INFO "APIC error on CPU%d: %02lx(%02lx)\n",
 	        smp_processor_id(), v , v1);
 	irq_exit();
+	return regs;
 }
 
 /*
diff -ur linux-2.5.59-mjb1/arch/i386/kernel/cpu/mcheck/p4.c linux-2.5.59-numaq-mjb1/arch/i386/kernel/cpu/mcheck/p4.c
--- linux-2.5.59-mjb1/arch/i386/kernel/cpu/mcheck/p4.c	Sun Jan 26 23:14:51 2003
+++ linux-2.5.59-numaq-mjb1/arch/i386/kernel/cpu/mcheck/p4.c	Sun Jan 26 23:23:08 2003
@@ -61,7 +61,8 @@
 /* Thermal interrupt handler for this CPU setup */
 static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = unexpected_thermal_interrupt;
 
-asmlinkage void smp_thermal_interrupt(struct pt_regs regs)
+asmlinkage struct pt_regs * smp_thermal_interrupt(struct pt_regs *regs) __attribute__((regparm(1)));
+struct pt_regs* smp_thermal_interrupt(struct pt_regs* regs) 
 {
 	irq_enter();
 	vendor_thermal_interrupt(&regs);
diff -ur linux-2.5.59-mjb1/arch/i386/kernel/entry.S linux-2.5.59-numaq-mjb1/arch/i386/kernel/entry.S
--- linux-2.5.59-mjb1/arch/i386/kernel/entry.S	Sun Jan 26 23:16:19 2003
+++ linux-2.5.59-numaq-mjb1/arch/i386/kernel/entry.S	Mon Jan 27 11:14:08 2003
@@ -138,8 +138,6 @@
 	.long 1b,2b;	\
 .previous
 
-
-
 ENTRY(lcall7)
 	pushfl			# We get a different stack layout with call
 				# gates, which has to be cleaned up later..
@@ -388,55 +386,76 @@
 vector=vector+1
 .endr
 
-	ALIGN
-common_interrupt:
-	SAVE_ALL
-
 
-	GET_THREAD_INFO(%ebx)
-	movl TI_IRQ_STACK(%ebx),%ecx
-	movl TI_TASK(%ebx),%edx
-	movl %esp,%eax
-	leal (THREAD_SIZE-4)(%ecx),%esi # %ecx+THREAD_SIZE is next stack
-	                                # -4 keeps us in the right one
-	testl %ecx,%ecx                 # is there a valid irq_stack?
-
-	# switch to the irq stack
-#ifdef CONFIG_X86_HAVE_CMOV
-	cmovnz %esi,%esp
+# lets play optimizing compiler...
+#ifdef CONFIG_X86_CMOV
+#define COND_MOVE	cmovnz %esi,%esp;
 #else
-	jz 1f
-	mov %esi,%esp
+#define COND_MOVE	\
+	jz 1f;		\
+	mov %esi,%esp;	\
 1:
 #endif
-
-	# update the task pointer in the irq stack
-	GET_THREAD_INFO(%esi)
-	movl %edx,TI_TASK(%esi)
-
-	# update the preempt count in the irq stack
-	movl TI_PRE_COUNT(%ebx),%ecx
-	movl %ecx,TI_PRE_COUNT(%esi)
-
-	call do_IRQ
 
-	movl %eax,%esp                  # potentially restore non-irq stack
-
-	# copy flags from the irq stack back into the task's thread_info
-	# %esi is saved over the do_IRQ call and contains the irq stack
-	# thread_info pointer
-	# %ebx contains the original thread_info pointer
-	movl TI_FLAGS(%esi),%eax
-	movl $0,TI_FLAGS(%esi)
-	LOCK orl %eax,TI_FLAGS(%ebx)
+# These macros will switch you to, and from a per-cpu interrupt stack
+# They take the pt_regs arg and move it from the normal place on the 
+# stack to %eax.  Any handler function can retrieve it using regparm(1). 
+# The handlers are expected to return the stack to switch back to in 
+# the same register. 
+#
+# This means that the irq handlers need to return their arg
+#
+# SWITCH_TO_IRQSTACK clobbers %ebx, %ecx, %edx, %esi
+# old stack in %eax
+
+#define SWITCH_TO_IRQSTACK 							\
+	GET_THREAD_INFO(%ebx);							\
+	movl TI_IRQ_STACK(%ebx),%ecx;						\
+	movl TI_TASK(%ebx),%edx;						\
+	movl %esp,%eax;								\
+										\
+	/* %ecx+THREAD_SIZE is next stack -4 keeps us in the right one */	\
+	leal (THREAD_SIZE-4)(%ecx),%esi; 					\
+										\
+	/* is there a valid irq_stack? */					\
+	testl %ecx,%ecx;							\
+	COND_MOVE;								\
+										\
+	/* update the task pointer in the irq stack */ 				\
+	GET_THREAD_INFO(%esi);							\
+	movl %edx,TI_TASK(%esi);						\
+										\
+	/* update the preempt count in the irq stack */ 			\
+	movl TI_PRE_COUNT(%ebx),%ecx;						\
+	movl %ecx,TI_PRE_COUNT(%esi);
+
+# copy flags from the irq stack back into the task's thread_info
+# %esi is saved over the irq handler call and contains the irq stack
+#      thread_info pointer
+# %eax was returned from the handler, as described above
+# %ebx contains the original thread_info pointer
+
+#define RESTORE_FROM_IRQSTACK 		\
+	movl %eax,%esp;			\
+	movl TI_FLAGS(%esi),%eax;	\
+	movl $0,TI_FLAGS(%esi);		\
+	LOCK orl %eax,TI_FLAGS(%ebx);
 
+	ALIGN
+common_interrupt:
+	SAVE_ALL
+	SWITCH_TO_IRQSTACK
+	call do_IRQ
+	RESTORE_FROM_IRQSTACK
 	jmp ret_from_intr
 
 #define BUILD_INTERRUPT(name, nr)	\
 ENTRY(name)				\
-	pushl $nr-256;			\
+	pushl $nr-256;                  \
 	SAVE_ALL			\
-	call smp_/**/name;	\
+	SWITCH_TO_IRQSTACK		\
+	call smp_/**/name;		\
+	RESTORE_FROM_IRQSTACK		\
 	jmp ret_from_intr;
 
 /* The include is where all of the SMP etc. interrupts come from */
diff -ur linux-2.5.59-mjb1/arch/i386/kernel/smp.c linux-2.5.59-numaq-mjb1/arch/i386/kernel/smp.c
--- linux-2.5.59-mjb1/arch/i386/kernel/smp.c	Sun Jan 26 23:16:19 2003
+++ linux-2.5.59-numaq-mjb1/arch/i386/kernel/smp.c	Sun Jan 26 14:37:09 2003
@@ -305,7 +305,8 @@
  * 2) Leave the mm if we are in the lazy tlb mode.
  */
 
-asmlinkage void smp_invalidate_interrupt (void)
+struct pt_regs *smp_invalidate_interrupt(struct pt_regs *regs) __attribute__((regparm(1)));
+struct pt_regs *smp_invalidate_interrupt(struct pt_regs *regs)
 {
 	unsigned long cpu;
 
@@ -336,6 +337,7 @@
 
 out:
 	put_cpu_no_resched();
+	return regs;
 }
 
 static void flush_tlb_others (unsigned long cpumask, struct mm_struct *mm,
@@ -598,12 +600,17 @@
  * all the work is done automatically when
  * we return from the interrupt.
  */
-asmlinkage void smp_reschedule_interrupt(void)
+
+asmlinkage struct pt_regs * smp_reschedule_interrupt(struct pt_regs *regs) __attribute__((regparm(1)));
+struct pt_regs * smp_reschedule_interrupt(struct pt_regs *regs)
 {
 	ack_APIC_irq();
+	return regs;
 }
 
-asmlinkage void smp_call_function_interrupt(struct pt_regs regs)
+
+asmlinkage struct pt_regs * smp_call_function_interrupt(struct pt_regs *regs) __attribute__((regparm(1)));
+struct pt_regs * smp_call_function_interrupt(struct pt_regs *regs)
 {
 	void (*func) (void *info, struct pt_regs *) = (void (*)(void *, struct pt_regs*))call_data->func;
 	void *info = call_data->info;
@@ -620,12 +627,13 @@
 	 * At this point the info structure may be out of scope unless wait==1
 	 */
 	irq_enter();
-	(*func)(info, &regs);
+	(*func)(info, regs);
 	irq_exit();
 
 	if (wait) {
 		mb();
 		atomic_inc(&call_data->finished);
 	}
+	return regs;
 }
 

2.5.59-mjb4 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb4.bz2

additional:

http://www.aracnet.com/~fletch/linux/2.5.59/pidmaps_nodepages
http://www.aracnet.com/~fletch/linux/2.5.59/mjb4-stacks

Since 2.5.59-mjb3 (~ = changed, + = added, - = dropped)

Notes:
Very, very boring ... just getting one (hopefully) rock-solid stabilisation
point for lots of stuff that's pending I have to merge somehow ...  
I've broken out the pidmaps_nodepages and 4K stacks patches that were
causing some problems ... will return soon. Available seperately (as above)

- thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
- interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
- stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
- 4k_stack            (4K stacks pt 4)		Dave Hansen
- pidmaps_nodepages				Dave Hansen

Pending:
Kprobes
Linux Kernel Crash Dump
kirq interrupt balancing patch (Nitin Kamble)
auto_disable_tsc (John Stultz)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Child runs first (akpm)
Feral qlogic driver (Matthew Jacob and the SCSI crew)
Kexec
e1000 fixes
Non-PAE aligned kernel splits (Dave Hansen)


dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

schedstat					Rick Lindsley
	Provide stats about the scheduler under /proc/stat

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

# thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
# interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

# stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

# 4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

discontig_x440					Pat Gaughen / Chandra
	SLIT/SRAT parsing for x440 discontigmem

srat_fixes					Pat Gaughen
	Fix sysfs for x440 machines, and make srat conditional on NUMA

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

summit_smp					John Stultz
	Make Summit config options work on standard SMP

cyclone_fixes					John Stultz
	Fix up some stuff for the x440's cyclone timer

enable_cyclone					John Stultz
	Enable the x440's cyclone timer

lost_tick					John Stultz
	Detect lost timer ticks

frlock_xtime					Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-i386				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-ia64				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-other				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

tcp_fix						Alexey
	Stop some tcp problem with hardware checksumming (e1000?)

numa_pci_fix					Dave Hansen
	Fix a potential error in the numa pci code from Stanford Checker

pgd_ctor					William Lee Irwin
	Use slabs for pgd

pfn_to_nid					William Lee Irwin
	Turn pfn_to_nid into a macro

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.59-mjb5 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb5.bz2

additional:

http://www.aracnet.com/~fletch/linux/2.5.59/pidmaps_nodepages
http://www.aracnet.com/~fletch/linux/2.5.59/mjb4-stacks

Since 2.5.59-mjb4 (~ = changed, + = added, - = dropped)

Notes:
Shared pagetables pt. 2 ... the revenge of shared pagetables.

~ discontig_x440				Pat Gaughen / IBM NUMA team
- srat_fixes					Pat Gaughen
+ kprobes					Vamsi Krishna S
+ dmc_exit1					Dave McCracken
+ dmc_exit2					Dave McCracken
+ shpte						Dave McCracken

Pending:
Linux Kernel Crash Dump
kirq interrupt balancing patch (Nitin Kamble)
auto_disable_tsc (John Stultz)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Child runs first (akpm)
Feral qlogic driver (Matthew Jacob and the SCSI crew)
Kexec
e1000 fixes
Non-PAE aligned kernel splits (Dave Hansen)


dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.
 
early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

schedstat					Rick Lindsley
	Provide stats about the scheduler under /proc/stat

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

# thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
# interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

# stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

# 4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

discontig_x440					Pat Gaughen / IBM NUMA team
	SLIT/SRAT parsing for x440 discontigmem

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

summit_smp					John Stultz
	Make Summit config options work on standard SMP

cyclone_fixes					John Stultz
	Fix up some stuff for the x440's cyclone timer

enable_cyclone					John Stultz
	Enable the x440's cyclone timer

lost_tick					John Stultz
	Detect lost timer ticks

frlock_xtime					Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-i386				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-ia64				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-other				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

tcp_fix						Alexey
	Stop some tcp problem with hardware checksumming (e1000?)

numa_pci_fix					Dave Hansen
	Fix a potential error in the numa pci code from Stanford Checker

pgd_ctor					William Lee Irwin
	Use slabs for pgd

pfn_to_nid					William Lee Irwin
	Turn pfn_to_nid into a macro

-mjb						Martin Bligh
	Add a tag to the makefile

2.5.59-mjb6 (scalability / NUMA patchset)

[Martin J. Bligh]

The patchset contains mainly scalability and NUMA stuff, and anything 
else that stops things from irritating me. It's meant to be pretty stable, 
not so much a testing ground for new stuff.

I'd be very interested in feedback from anyone willing to test on any 
platform, however large or small.

http://www.aracnet.com/~fletch/linux/2.5.59/patch-2.5.59-mjb6.bz2

additional:

http://www.aracnet.com/~fletch/linux/2.5.59/pidmaps_nodepages

Since 2.5.59-mjb5 (~ = changed, + = added, - = dropped)

Notes:
Fixed bug between NFS and dcache, and some oprofile bug on UP. 
4K stacks are back and crash dump support has arrived (thanks to Suparna 
for help merging and bufixing)

+ dcache_sunrpc					Maneesh
~ cyclone_fixes					John Stultz
+ oprofile_fixes				John Levon
+ thread_info_cleanup				Dave Hansen / Ben LaHaise
+ interrupt_stacks    				Dave Hansen / Ben LaHaise
+ stack_usage_check   				Dave Hansen / Ben LaHaise
+ 4k_stack            				Dave Hansen
+ fix_kgdb					Dave Hansen
+ stacks_from_slab				Bill Irwin
+ thread_under_page				Bill Irwin
+ lkcd						LKCD team
+ alt_sysrq_t					Russell King

Pending:
kirq interrupt balancing patch (Nitin Kamble)
auto_disable_tsc (John Stultz)
scheduler callers profiling (Anton)
PPC64 NUMA patches (Anton)
Child runs first (akpm)
Kexec
e1000 fixes
Non-PAE aligned kernel splits (Dave Hansen)
Update the lost timer ticks code
Ingo scheduler updates

dcache_rcu					Dipankar / Maneesh
	Use RCU type locking for the dentry cache.

dcache_sunrpc					Maneesh
	Fix up NFS to work properly with dcache

early_printk					Dave Hansen et al.
	Allow printk before console_init

confighz					Andrew Morton / Dave Hansen
	Make HZ a config option of 100 Hz or 1000 Hz

config_page_offset				Dave Hansen / Andrea
	Make PAGE_OFFSET a config option

vmalloc_stats					Dave Hansen
	Expose useful vmalloc statistics

local_pgdat					William Lee Irwin
	Move the pgdat structure into the remapped space with lmem_map

numameminfo					Martin Bligh / Keith Mannthey
	Expose NUMA meminfo information under /proc/meminfo.numa

notsc						Martin Bligh
	Enable notsc option for NUMA-Q (new version for new config system)

mpc_apic_id					Martin J. Bligh
	Fix null ptr dereference (optimised away, but ...)

doaction					Martin J. Bligh
	Fix cruel torture of macros and small furry animals in io_apic.c

kgdb						Andrew Morton / Various People
	The older version of kgdb, synched with 2.5.54-mm1

noframeptr					Martin Bligh
	Disable -fomit_frame_pointer

ingosched					Ingo Molnar
	Modify NUMA scheduler to have independant tick basis.

schedstat					Rick Lindsley
	Provide stats about the scheduler under /proc/stat

sched_tunables					Robert Love
	Provide tunable parameters for the scheduler (+ NUMA scheduler)

discontig_x440					Pat Gaughen / IBM NUMA team
	SLIT/SRAT parsing for x440 discontigmem

acpi_x440_hack					Anonymous Coward
	Stops x440 crashing, but owner is ashamed of it ;-)

summit_smp					John Stultz
	Make Summit config options work on standard SMP

cyclone_fixes					John Stultz
	Fix up some stuff for the x440's cyclone timer

enable_cyclone					John Stultz
	Enable the x440's cyclone timer

lost_tick					John Stultz
	Detect lost timer ticks

frlock_xtime					Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-i386				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-ia64				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

frlock-xtime-other				Stephen Hemminger et al.
	Turn xtime_lock into an frlock to reduce contention 

numaq_ioapicids					William Lee Irwin
	Stop 8 quad NUMA-Qs from panicing due to phys apicid "exhaustion".

oprofile_p4					John Levon
	Updates for oprofile for P4s. Needs new userspace tools.

starfire					Ion Badulescu
	64 bit aware starfire driver	

tcp_fix						Alexey
	Stop some tcp problem with hardware checksumming (e1000?)

numa_pci_fix					Dave Hansen
	Fix a potential error in the numa pci code from Stanford Checker

pgd_ctor					William Lee Irwin
	Use slabs for pgd

pfn_to_nid					William Lee Irwin
	Turn pfn_to_nid into a macro

oprofile_fixes					John Levon
	fix a couple of bugs in oprofile

kprobes						Vamsi Krishna S
	Add kernel probes hooks to the kernel

dmc_exit1					Dave McCracken
	Speed up the exit path, pt 1.

dmc_exit2					Dave McCracken
	Speed up the exit path, pt 1.

shpte						Dave McCracken
	Shared pagetables (as a config option)

thread_info_cleanup (4K stacks pt 1)		Dave Hansen / Ben LaHaise
	Prep work to reduce kernel stacks to 4K
	
interrupt_stacks    (4K stacks pt 2)		Dave Hansen / Ben LaHaise
	Create a per-cpu interrupt stack.

stack_usage_check   (4K stacks pt 3)		Dave Hansen / Ben LaHaise
	Check for kernel stack overflows.

4k_stack            (4K stacks pt 4)		Dave Hansen
	Config option to reduce kernel stacks to 4K

fix_kgdb					Dave Hansen
	Fix interaction between kgdb and 4K stacks

stacks_from_slab				William Lee Irwin
	Take kernel stacks from the slab cache, not page allocation.

thread_under_page				William Lee Irwin
	Fix THREAD_SIZE < PAGE_SIZE case

lkcd						LKCD team
	Linux kernel crash dump support

alt_sysrq_t					Russell King
	Fix up ALT+sysrq+t

-mjb						Martin Bligh
	Add a tag to the makefile