[patch 3/3] add x86-64 support for memory hot-add

[patch 3/3] add x86-64 support for memory hot-add

[Matt Tolentino]

This patch provide the capability to hot-add memory on single node x86-64
systems.  Stub functions for adding and onlining pages as well as 
updating the kernel page tables when memory is added are provided.

Signed-off-by: Matt Tolentino (matthew.e.tolentino@intel.com>
---

diff -urNp linux-2.6.15-rc5/arch/x86_64/Kconfig linux-2.6.15-rc5-matt/arch/x86_64/Kconfig
--- linux-2.6.15-rc5/arch/x86_64/Kconfig	2005-12-04 00:10:42.000000000 -0500
+++ linux-2.6.15-rc5-matt/arch/x86_64/Kconfig	2005-12-08 15:11:49.000000000 -0500
@@ -283,7 +283,11 @@ config ARCH_DISCONTIGMEM_DEFAULT
 
 config ARCH_SPARSEMEM_ENABLE
 	def_bool y
-	depends on NUMA
+	depends on (NUMA || EXPERIMENTAL)
+
+config ARCH_MEMORY_PROBE
+	def_bool y
+	depends on MEMORY_HOTPLUG
 
 config ARCH_FLATMEM_ENABLE
 	def_bool y
@@ -293,6 +297,7 @@ source "mm/Kconfig"
 
 config HAVE_ARCH_EARLY_PFN_TO_NID
 	def_bool y
+	depends on NUMA
 
 config NR_CPUS
 	int "Maximum number of CPUs (2-256)"
diff -urNp linux-2.6.15-rc5/arch/x86_64/mm/init.c linux-2.6.15-rc5-matt/arch/x86_64/mm/init.c
--- linux-2.6.15-rc5/arch/x86_64/mm/init.c	2005-12-04 00:10:42.000000000 -0500
+++ linux-2.6.15-rc5-matt/arch/x86_64/mm/init.c	2005-12-08 15:02:30.000000000 -0500
@@ -23,6 +23,8 @@
 #include <linux/bootmem.h>
 #include <linux/proc_fs.h>
 #include <linux/pci.h>
+#include <linux/module.h>
+#include <linux/memory_hotplug.h>
 
 #include <asm/processor.h>
 #include <asm/system.h>
@@ -174,13 +176,19 @@ static  struct temp_map { 
 	{}
 }; 
 
-static __init void *alloc_low_page(int *index, unsigned long *phys) 
+static __devinit void *alloc_low_page(int *index, unsigned long *phys) 
 { 
 	struct temp_map *ti;
 	int i; 
 	unsigned long pfn = table_end++, paddr; 
 	void *adr;
 
+	if (after_bootmem) {
+		adr = (void *)get_zeroed_page(GFP_ATOMIC);
+		*phys = __pa(adr);
+		return adr;
+	}
+
 	if (pfn >= end_pfn) 
 		panic("alloc_low_page: ran out of memory"); 
 	for (i = 0; temp_mappings[i].allocated; i++) {
@@ -193,55 +201,86 @@ static __init void *alloc_low_page(int *
 	ti->allocated = 1; 
 	__flush_tlb(); 	       
 	adr = ti->address + ((pfn << PAGE_SHIFT) & ~PMD_MASK); 
+	memset(adr, 0, PAGE_SIZE);
 	*index = i; 
 	*phys  = pfn * PAGE_SIZE;  
 	return adr; 
 } 
 
-static __init void unmap_low_page(int i)
+static __devinit void unmap_low_page(int i)
 { 
-	struct temp_map *ti = &temp_mappings[i];
+	struct temp_map *ti;
+
+	if (after_bootmem)
+		return;
+
+	ti = &temp_mappings[i];
 	set_pmd(ti->pmd, __pmd(0));
 	ti->allocated = 0; 
 } 
 
-static void __init phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
+static void __devinit
+phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
+{
+	int i;
+
+	for (i = 0; i < PTRS_PER_PMD; pmd++, i++, address += PMD_SIZE) {
+		unsigned long entry;
+
+		if (address > end) {
+			for (; i < PTRS_PER_PMD; i++, pmd++)
+				set_pmd(pmd, __pmd(0));
+			break;
+		}
+		entry = _PAGE_NX|_PAGE_PSE|_KERNPG_TABLE|_PAGE_GLOBAL|address;
+		entry &= __supported_pte_mask;
+		set_pmd(pmd, __pmd(entry));
+	}
+}
+
+static void __devinit
+phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
+{
+	pmd_t *pmd = pmd_offset(pud, (unsigned long)__va(address));
+	
+	if (pmd_none(*pmd)) {
+		spin_lock(&init_mm.page_table_lock);
+		phys_pmd_init(pmd, address, end);
+		spin_unlock(&init_mm.page_table_lock);
+		__flush_tlb_all();
+	}
+}
+
+static void __devinit phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
 { 
-	long i, j; 
+	long i = pud_index(address);
 
-	i = pud_index(address);
 	pud = pud + i;
+
+	if (after_bootmem && pud_val(*pud)) {
+		phys_pmd_update(pud, address, end);
+		return;
+	}
+
 	for (; i < PTRS_PER_PUD; pud++, i++) {
 		int map; 
 		unsigned long paddr, pmd_phys;
 		pmd_t *pmd;
 
-		paddr = address + i*PUD_SIZE;
-		if (paddr >= end) { 
-			for (; i < PTRS_PER_PUD; i++, pud++) 
-				set_pud(pud, __pud(0)); 
+		paddr = (address & PGDIR_MASK) + i*PUD_SIZE;
+		if (paddr >= end)
 			break;
-		} 
 
-		if (!e820_mapped(paddr, paddr+PUD_SIZE, 0)) { 
+		if (!after_bootmem && !e820_mapped(paddr, paddr+PUD_SIZE, 0)) { 
 			set_pud(pud, __pud(0)); 
 			continue;
 		} 
 
 		pmd = alloc_low_page(&map, &pmd_phys);
+		if (after_bootmem) spin_lock(&init_mm.page_table_lock);
 		set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
-		for (j = 0; j < PTRS_PER_PMD; pmd++, j++, paddr += PMD_SIZE) {
-			unsigned long pe;
-
-			if (paddr >= end) { 
-				for (; j < PTRS_PER_PMD; j++, pmd++)
-					set_pmd(pmd,  __pmd(0)); 
-				break;
-		}
-			pe = _PAGE_NX|_PAGE_PSE | _KERNPG_TABLE | _PAGE_GLOBAL | paddr;
-			pe &= __supported_pte_mask;
-			set_pmd(pmd, __pmd(pe));
-		}
+		phys_pmd_init(pmd, paddr, end);
+		if (after_bootmem) spin_unlock(&init_mm.page_table_lock);
 		unmap_low_page(map);
 	}
 	__flush_tlb();
@@ -262,12 +301,15 @@ static void __init find_early_table_spac
 
 	table_start >>= PAGE_SHIFT;
 	table_end = table_start;
+
+	early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
+		end, table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
 }
 
 /* Setup the direct mapping of the physical memory at PAGE_OFFSET.
    This runs before bootmem is initialized and gets pages directly from the 
    physical memory. To access them they are temporarily mapped. */
-void __init init_memory_mapping(unsigned long start, unsigned long end)
+void __devinit init_memory_mapping(unsigned long start, unsigned long end)
 { 
 	unsigned long next; 
 
@@ -279,7 +321,8 @@ void __init init_memory_mapping(unsigned
 	 * mapped.  Unfortunately this is done currently before the nodes are 
 	 * discovered.
 	 */
-	find_early_table_space(end);
+	if (!after_bootmem)
+		find_early_table_space(end);
 
 	start = (unsigned long)__va(start);
 	end = (unsigned long)__va(end);
@@ -287,20 +330,26 @@ void __init init_memory_mapping(unsigned
 	for (; start < end; start = next) {
 		int map;
 		unsigned long pud_phys; 
-		pud_t *pud = alloc_low_page(&map, &pud_phys);
+		pgd_t *pgd = pgd_offset_k(start);
+		pud_t *pud;
+
+		if (after_bootmem)
+			pud = pud_offset_k(pgd, __PAGE_OFFSET);
+		else
+			pud = alloc_low_page(&map, &pud_phys);
+
 		next = start + PGDIR_SIZE;
 		if (next > end) 
 			next = end; 
 		phys_pud_init(pud, __pa(start), __pa(next));
-		set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
+		if (!after_bootmem)
+			set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
 		unmap_low_page(map);   
 	} 
 
-	asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
+	if (!after_bootmem)
+		asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
 	__flush_tlb_all();
-	early_printk("kernel direct mapping tables upto %lx @ %lx-%lx\n", end, 
-	       table_start<<PAGE_SHIFT, 
-	       table_end<<PAGE_SHIFT);
 }
 
 void __cpuinit zap_low_mappings(int cpu)
@@ -375,6 +424,9 @@ size_zones(unsigned long *z, unsigned lo
 void __init paging_init(void)
 {
 	unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
+
+	memory_present(0, 0, end_pfn);
+	sparse_init();
 	size_zones(zones, holes, 0, end_pfn);
 	free_area_init_node(0, NODE_DATA(0), zones,
 			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
@@ -415,6 +467,50 @@ void __init clear_kernel_mapping(unsigne
 	__flush_tlb_all();
 } 
 
+/* 
+ * Memory hotplug specific functions
+ * These are only for non-NUMA machines right now.
+ */
+#ifdef CONFIG_MEMORY_HOTPLUG
+
+void online_page(struct page *page)
+{
+	ClearPageReserved(page);
+	set_page_count(page, 1);
+	__free_page(page);
+	totalram_pages++;
+	num_physpages++;
+}
+
+int add_memory(u64 start, u64 size)
+{
+	struct pglist_data *pgdat = NODE_DATA(0);
+	struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2;
+	unsigned long start_pfn = start >> PAGE_SHIFT;
+	unsigned long nr_pages = size >> PAGE_SHIFT;
+	int ret;
+
+	ret = __add_pages(zone, start_pfn, nr_pages);
+	if (ret)
+		goto error;
+	
+	init_memory_mapping(start, (start + size -1));
+
+	return ret;
+error:
+	printk("%s: Problem encountered in __add_pages!\n", __func__);
+	return ret;
+}
+EXPORT_SYMBOL(add_memory);
+
+int remove_memory(u64 start, u64 size)
+{
+	return -EINVAL;
+}
+EXPORT_SYMBOL(remove_memory);
+
+#endif
+
 static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
 			 kcore_vsyscall;
 

Re: [patch 3/3] add x86-64 support for memory hot-add

[Andi Kleen]

On Fri, Dec 09, 2005 at 10:23:49AM -0500, Matt Tolentino wrote:
> --- linux-2.6.15-rc5/arch/x86_64/mm/init.c	2005-12-04 00:10:42.000000000 -0500
> +++ linux-2.6.15-rc5-matt/arch/x86_64/mm/init.c	2005-12-08 15:02:30.000000000 -0500
> @@ -23,6 +23,8 @@
>  #include <linux/bootmem.h>
>  #include <linux/proc_fs.h>
>  #include <linux/pci.h>
> +#include <linux/module.h>
> +#include <linux/memory_hotplug.h>
>  
>  #include <asm/processor.h>
>  #include <asm/system.h>
> @@ -174,13 +176,19 @@ static  struct temp_map { 
>  	{}
>  }; 
>  
> -static __init void *alloc_low_page(int *index, unsigned long *phys) 
> +static __devinit void *alloc_low_page(int *index, unsigned long *phys) 

These should be all __cpuinit.

In general SRAT has a hotplug memory bit so it's possible
to predict how much memory there will be in advance. Since
the overhead of the kernel page tables should be very
low I would prefer if you just used instead.

(i.e. instead of extending the kernel mapping preallocate
the direct mapping and just clear the P bits) 

That should be much simpler.

-Andi

Re: [discuss] Re: [patch 3/3] add x86-64 support for memory hot-add II

[Andi Kleen]

> In general SRAT has a hotplug memory bit so it's possible
> to predict how much memory there will be in advance. Since
> the overhead of the kernel page tables should be very
> low I would prefer if you just used instead.
> 
> (i.e. instead of extending the kernel mapping preallocate
> the direct mapping and just clear the P bits) 
> 
> That should be much simpler.

Looking at it again - accessing SRAT currently relies on the 
direct mapping already. Untangling that would be possible,
but require an bt_ioremap which would also add lots of code.

Ok I retract that objection. I guess your way is better
for now.

In addition to the __cpuinit comment

+if (after_bootmem) spin_unlock(&init_mm.page_table_lock);

Conditional locking is evil. spinlocking in the boot
case should just work too I think.

The EXPORTs should be probably EXPORT_SYMBOL_GPL.

With these changes it would look ok for me.

-Andi

RE: [discuss] Re: [patch 3/3] add x86-64 support for memory hot-add II

[Tolentino, Matthew E]

Andi Kleen <mailto:ak@suse.de> wrote:
>> In general SRAT has a hotplug memory bit so it's possible
>> to predict how much memory there will be in advance. Since
>> the overhead of the kernel page tables should be very
>> low I would prefer if you just used instead.
>> 
>> (i.e. instead of extending the kernel mapping preallocate
>> the direct mapping and just clear the P bits)
>> 
>> That should be much simpler.
> 
> Looking at it again - accessing SRAT currently relies on the
> direct mapping already. Untangling that would be possible,
> but require an bt_ioremap which would also add lots of code.
> 
> Ok I retract that objection. I guess your way is better
> for now.

Thanks for considering this Andi.  
 
> In addition to the __cpuinit comment
> 
> +if (after_bootmem) spin_unlock(&init_mm.page_table_lock);
> 
> Conditional locking is evil. spinlocking in the boot
> case should just work too I think.
> 
> The EXPORTs should be probably EXPORT_SYMBOL_GPL.
> 
> With these changes it would look ok for me.

Excellent points.  Thanks for the review and suggestions.  I'm
testing a revised patch now and will repost in a bit.  

matt

Re: [patch 3/3] add x86-64 support for memory hot-add

[Keith Mannthey]

> These should be all __cpuinit.
>
> In general SRAT has a hotplug memory bit so it's possible
> to predict how much memory there will be in advance. Since
> the overhead of the kernel page tables should be very
> low I would prefer if you just used instead.

How much overhead would there be?

> (i.e. instead of extending the kernel mapping preallocate
> the direct mapping and just clear the P bits)

On my box the SRAT for hot-add areas exposed are from the end
installed memory to way out in outerspace.
SRAT: hot plug zone found 280000000 - 2300000000
I can't hot add that sort of range on my box but the bios didn't want
to limit or is planing for really really big dimms.

I wouldn't want to waste resource for areas that will never be added.

Thanks,
 Keith

Re: [discuss] Re: [patch 3/3] add x86-64 support for memory hot-add

[Andi Kleen]

On Fri, Dec 09, 2005 at 04:16:41PM -0800, Keith Mannthey wrote:
> > These should be all __cpuinit.
> >
> > In general SRAT has a hotplug memory bit so it's possible
> > to predict how much memory there will be in advance. Since
> > the overhead of the kernel page tables should be very
> > low I would prefer if you just used instead.
> 
> How much overhead would there be?

It's 2MB pages in 3levels, so roughly 3*8=24 bytes per 2MB or roughly
512 bytes per GB (rounded up always to the next page)

> 
> > (i.e. instead of extending the kernel mapping preallocate
> > the direct mapping and just clear the P bits)
> 
> On my box the SRAT for hot-add areas exposed are from the end
> installed memory to way out in outerspace.
> SRAT: hot plug zone found 280000000 - 2300000000
> I can't hot add that sort of range on my box but the bios didn't want
> to limit or is planing for really really big dimms.

You're just proving someone's (anyone want to volunteer their name? ;-]
I think Linus pointed it out originally, so let's call it Linus') 
law - as soon as we use some BIOS feature we soon find a BIOS 
that will get it totally wrong.

Anyways, I retracted anyways because of some other issues so Matt's 
original approach should be ok.

-Andi