SPARSEMEM is a pretty nice framework that unifies quite a bit of code over all the arches. It would be great if it could be the default so that we can get rid of various forms of DISCONTIG and other variations on memory maps. So far what has hindered this are the additional lookups that SPARSEMEM introduces for virt_to_page and page_address. This goes so far that the code to do this has to be kept in a separate function and cannot be used inline. This patch introduces a virtual memmap mode for SPARSEMEM, in which the memmap is mapped into a virtually contigious area, only the active sections are physically backed. This allows virt_to_page page_address and cohorts become simple shift/add operations. No page flag fields, no table lookups, nothing involving memory is required. The two key operations pfn_to_page and page_to_page become: #define __pfn_to_page(pfn) (vmemmap + (pfn)) #define __page_to_pfn(page) ((page) - vmemmap) By having a virtual mapping for the memmap we allow simple access without wasting physical memory. As kernel memory is typically already mapped 1:1 this introduces no additional overhead. The virtual mapping must be big enough to allow a struct page to be allocated and mapped for all valid physical pages. This vill make a virtual memmap difficult to use on 32 bit platforms that support 36 address bits. However, if there is enough virtual space available and the arch already maps its 1-1 kernel space using TLBs (f.e. true of IA64 and x86_64) then this technique makes SPARSEMEM lookups even more efficient than CONFIG_FLATMEM. FLATMEM needs to read the contents of the mem_map variable to get the start of the memmap and then add the offset to the required entry. vmemmap is a constant to which we can simply add the offset. This patch has the potential to allow us to make SPARSMEM the default (and even the only) option for most systems. It should be optimal on UP, SMP and NUMA on most platforms. Then we may even be able to remove the other memory models: FLATMEM, DISCONTIG etc. The current aim is to bring a common virtually mapped mem_map to all architectures. This should facilitate the removal of the bespoke implementations from the architectures. This also brings performance improvements for most architecture making sparsmem vmemmap the more desirable memory model. The ultimate aim of this work is to expand sparsemem support to encompass all the features of the other memory models. This could allow us to drop support for and remove the other models in the longer term. Below are some comparitive kernbench numbers for various architectures, comparing default memory model against SPARSEMEM VMEMMAP. All but ia64 show marginal improvement; we expect the ia64 figures to be sorted out when the larger mapping support returns. x86-64 non-NUMA Base VMEMAP % change (-ve good) User 85.07 84.84 -0.26 System 34.32 33.84 -1.39 Total 119.38 118.68 -0.59 ia64 Base VMEMAP % change (-ve good) User 1016.41 1016.93 0.05 System 50.83 51.02 0.36 Total 1067.25 1067.95 0.07 x86-64 NUMA Base VMEMAP % change (-ve good) User 30.77 431.73 0.22 System 45.39 43.98 -3.11 Total 476.17 475.71 -0.10 ppc64 Base VMEMAP % change (-ve good) User 488.77 488.35 -0.09 System 56.92 56.37 -0.97 Total 545.69 544.72 -0.18 Below are some AIM bencharks on IA64 and x86-64 (thank Bob). The seems pretty much flat as you would expect. ia64 results 2 cpu non-numa 4Gb SCSI disk Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" extreme Jun 1 07:17:24 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 98.9 100 58.9 1.3 1.6482 101 5547.1 95 106.0 79.4 0.9154 201 6377.7 95 183.4 158.3 0.5288 301 6932.2 95 252.7 237.3 0.3838 401 7075.8 93 329.8 316.7 0.2941 501 7235.6 94 403.0 396.2 0.2407 600 7387.5 94 472.7 475.0 0.2052 Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" vmemmap Jun 1 09:59:04 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 99.1 100 58.8 1.2 1.6509 101 5480.9 95 107.2 79.2 0.9044 201 6490.3 95 180.2 157.8 0.5382 301 6886.6 94 254.4 236.8 0.3813 401 7078.2 94 329.7 316.0 0.2942 501 7250.3 95 402.2 395.4 0.2412 600 7399.1 94 471.9 473.9 0.2055 open power 710 2 cpu, 4 Gb, SCSI and configured physically Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" extreme May 29 15:42:53 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 25.7 100 226.3 4.3 0.4286 101 1096.0 97 536.4 199.8 0.1809 201 1236.4 96 946.1 389.1 0.1025 301 1280.5 96 1368.0 582.3 0.0709 401 1270.2 95 1837.4 771.0 0.0528 501 1251.4 96 2330.1 955.9 0.0416 601 1252.6 96 2792.4 1139.2 0.0347 701 1245.2 96 3276.5 1334.6 0.0296 918 1229.5 96 4345.4 1728.7 0.0223 Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" vmemmap May 30 07:28:26 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 25.6 100 226.9 4.3 0.4275 101 1049.3 97 560.2 198.1 0.1731 201 1199.1 97 975.6 390.7 0.0994 301 1261.7 96 1388.5 591.5 0.0699 401 1256.1 96 1858.1 771.9 0.0522 501 1220.1 96 2389.7 955.3 0.0406 601 1224.6 96 2856.3 1133.4 0.0340 701 1252.0 96 3258.7 1314.1 0.0298 915 1232.8 96 4319.7 1704.0 0.0225 amd64 2 2-core, 4Gb and SATA Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" extreme Jun 2 03:59:48 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 13.0 100 446.4 2.1 0.2173 101 533.4 97 1102.0 110.2 0.0880 201 578.3 97 2022.8 220.8 0.0480 301 583.8 97 3000.6 332.3 0.0323 401 580.5 97 4020.1 442.2 0.0241 501 574.8 98 5072.8 558.8 0.0191 600 566.5 98 6163.8 671.0 0.0157 Benchmark Version Machine Run Date AIM Multiuser Benchmark - Suite VII "1.1" vmemmap Jun 3 04:19:31 2007 Tasks Jobs/Min JTI Real CPU Jobs/sec/task 1 13.0 100 447.8 2.0 0.2166 101 536.5 97 1095.6 109.7 0.0885 201 567.7 97 2060.5 219.3 0.0471 301 582.1 96 3009.4 330.2 0.0322 401 578.2 96 4036.4 442.4 0.0240 501 585.1 98 4983.2 555.1 0.0195 600 565.5 98 6175.2 660.6 0.0157 This patch: Fix some spelling errors. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andy Whitcroft <apw@shadowen.org> Acked-by: Mel Gorman <mel@csn.ul.ie> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: "David S. Miller" <davem@davemloft.net> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
349 lines
8.3 KiB
C
349 lines
8.3 KiB
C
/*
|
|
* sparse memory mappings.
|
|
*/
|
|
#include <linux/mm.h>
|
|
#include <linux/mmzone.h>
|
|
#include <linux/bootmem.h>
|
|
#include <linux/highmem.h>
|
|
#include <linux/module.h>
|
|
#include <linux/spinlock.h>
|
|
#include <linux/vmalloc.h>
|
|
#include <asm/dma.h>
|
|
|
|
/*
|
|
* Permanent SPARSEMEM data:
|
|
*
|
|
* 1) mem_section - memory sections, mem_map's for valid memory
|
|
*/
|
|
#ifdef CONFIG_SPARSEMEM_EXTREME
|
|
struct mem_section *mem_section[NR_SECTION_ROOTS]
|
|
____cacheline_internodealigned_in_smp;
|
|
#else
|
|
struct mem_section mem_section[NR_SECTION_ROOTS][SECTIONS_PER_ROOT]
|
|
____cacheline_internodealigned_in_smp;
|
|
#endif
|
|
EXPORT_SYMBOL(mem_section);
|
|
|
|
#ifdef NODE_NOT_IN_PAGE_FLAGS
|
|
/*
|
|
* If we did not store the node number in the page then we have to
|
|
* do a lookup in the section_to_node_table in order to find which
|
|
* node the page belongs to.
|
|
*/
|
|
#if MAX_NUMNODES <= 256
|
|
static u8 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned;
|
|
#else
|
|
static u16 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned;
|
|
#endif
|
|
|
|
int page_to_nid(struct page *page)
|
|
{
|
|
return section_to_node_table[page_to_section(page)];
|
|
}
|
|
EXPORT_SYMBOL(page_to_nid);
|
|
|
|
static void set_section_nid(unsigned long section_nr, int nid)
|
|
{
|
|
section_to_node_table[section_nr] = nid;
|
|
}
|
|
#else /* !NODE_NOT_IN_PAGE_FLAGS */
|
|
static inline void set_section_nid(unsigned long section_nr, int nid)
|
|
{
|
|
}
|
|
#endif
|
|
|
|
#ifdef CONFIG_SPARSEMEM_EXTREME
|
|
static struct mem_section noinline __init_refok *sparse_index_alloc(int nid)
|
|
{
|
|
struct mem_section *section = NULL;
|
|
unsigned long array_size = SECTIONS_PER_ROOT *
|
|
sizeof(struct mem_section);
|
|
|
|
if (slab_is_available())
|
|
section = kmalloc_node(array_size, GFP_KERNEL, nid);
|
|
else
|
|
section = alloc_bootmem_node(NODE_DATA(nid), array_size);
|
|
|
|
if (section)
|
|
memset(section, 0, array_size);
|
|
|
|
return section;
|
|
}
|
|
|
|
static int __meminit sparse_index_init(unsigned long section_nr, int nid)
|
|
{
|
|
static DEFINE_SPINLOCK(index_init_lock);
|
|
unsigned long root = SECTION_NR_TO_ROOT(section_nr);
|
|
struct mem_section *section;
|
|
int ret = 0;
|
|
|
|
if (mem_section[root])
|
|
return -EEXIST;
|
|
|
|
section = sparse_index_alloc(nid);
|
|
/*
|
|
* This lock keeps two different sections from
|
|
* reallocating for the same index
|
|
*/
|
|
spin_lock(&index_init_lock);
|
|
|
|
if (mem_section[root]) {
|
|
ret = -EEXIST;
|
|
goto out;
|
|
}
|
|
|
|
mem_section[root] = section;
|
|
out:
|
|
spin_unlock(&index_init_lock);
|
|
return ret;
|
|
}
|
|
#else /* !SPARSEMEM_EXTREME */
|
|
static inline int sparse_index_init(unsigned long section_nr, int nid)
|
|
{
|
|
return 0;
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Although written for the SPARSEMEM_EXTREME case, this happens
|
|
* to also work for the flat array case because
|
|
* NR_SECTION_ROOTS==NR_MEM_SECTIONS.
|
|
*/
|
|
int __section_nr(struct mem_section* ms)
|
|
{
|
|
unsigned long root_nr;
|
|
struct mem_section* root;
|
|
|
|
for (root_nr = 0; root_nr < NR_SECTION_ROOTS; root_nr++) {
|
|
root = __nr_to_section(root_nr * SECTIONS_PER_ROOT);
|
|
if (!root)
|
|
continue;
|
|
|
|
if ((ms >= root) && (ms < (root + SECTIONS_PER_ROOT)))
|
|
break;
|
|
}
|
|
|
|
return (root_nr * SECTIONS_PER_ROOT) + (ms - root);
|
|
}
|
|
|
|
/*
|
|
* During early boot, before section_mem_map is used for an actual
|
|
* mem_map, we use section_mem_map to store the section's NUMA
|
|
* node. This keeps us from having to use another data structure. The
|
|
* node information is cleared just before we store the real mem_map.
|
|
*/
|
|
static inline unsigned long sparse_encode_early_nid(int nid)
|
|
{
|
|
return (nid << SECTION_NID_SHIFT);
|
|
}
|
|
|
|
static inline int sparse_early_nid(struct mem_section *section)
|
|
{
|
|
return (section->section_mem_map >> SECTION_NID_SHIFT);
|
|
}
|
|
|
|
/* Record a memory area against a node. */
|
|
void __init memory_present(int nid, unsigned long start, unsigned long end)
|
|
{
|
|
unsigned long pfn;
|
|
|
|
start &= PAGE_SECTION_MASK;
|
|
for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) {
|
|
unsigned long section = pfn_to_section_nr(pfn);
|
|
struct mem_section *ms;
|
|
|
|
sparse_index_init(section, nid);
|
|
set_section_nid(section, nid);
|
|
|
|
ms = __nr_to_section(section);
|
|
if (!ms->section_mem_map)
|
|
ms->section_mem_map = sparse_encode_early_nid(nid) |
|
|
SECTION_MARKED_PRESENT;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Only used by the i386 NUMA architecures, but relatively
|
|
* generic code.
|
|
*/
|
|
unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn,
|
|
unsigned long end_pfn)
|
|
{
|
|
unsigned long pfn;
|
|
unsigned long nr_pages = 0;
|
|
|
|
for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
|
|
if (nid != early_pfn_to_nid(pfn))
|
|
continue;
|
|
|
|
if (pfn_valid(pfn))
|
|
nr_pages += PAGES_PER_SECTION;
|
|
}
|
|
|
|
return nr_pages * sizeof(struct page);
|
|
}
|
|
|
|
/*
|
|
* Subtle, we encode the real pfn into the mem_map such that
|
|
* the identity pfn - section_mem_map will return the actual
|
|
* physical page frame number.
|
|
*/
|
|
static unsigned long sparse_encode_mem_map(struct page *mem_map, unsigned long pnum)
|
|
{
|
|
return (unsigned long)(mem_map - (section_nr_to_pfn(pnum)));
|
|
}
|
|
|
|
/*
|
|
* We need this if we ever free the mem_maps. While not implemented yet,
|
|
* this function is included for parity with its sibling.
|
|
*/
|
|
static __attribute((unused))
|
|
struct page *sparse_decode_mem_map(unsigned long coded_mem_map, unsigned long pnum)
|
|
{
|
|
return ((struct page *)coded_mem_map) + section_nr_to_pfn(pnum);
|
|
}
|
|
|
|
static int __meminit sparse_init_one_section(struct mem_section *ms,
|
|
unsigned long pnum, struct page *mem_map)
|
|
{
|
|
if (!valid_section(ms))
|
|
return -EINVAL;
|
|
|
|
ms->section_mem_map &= ~SECTION_MAP_MASK;
|
|
ms->section_mem_map |= sparse_encode_mem_map(mem_map, pnum);
|
|
|
|
return 1;
|
|
}
|
|
|
|
__attribute__((weak)) __init
|
|
void *alloc_bootmem_high_node(pg_data_t *pgdat, unsigned long size)
|
|
{
|
|
return NULL;
|
|
}
|
|
|
|
static struct page __init *sparse_early_mem_map_alloc(unsigned long pnum)
|
|
{
|
|
struct page *map;
|
|
struct mem_section *ms = __nr_to_section(pnum);
|
|
int nid = sparse_early_nid(ms);
|
|
|
|
map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION);
|
|
if (map)
|
|
return map;
|
|
|
|
map = alloc_bootmem_high_node(NODE_DATA(nid),
|
|
sizeof(struct page) * PAGES_PER_SECTION);
|
|
if (map)
|
|
return map;
|
|
|
|
map = alloc_bootmem_node(NODE_DATA(nid),
|
|
sizeof(struct page) * PAGES_PER_SECTION);
|
|
if (map)
|
|
return map;
|
|
|
|
printk(KERN_WARNING "%s: allocation failed\n", __FUNCTION__);
|
|
ms->section_mem_map = 0;
|
|
return NULL;
|
|
}
|
|
|
|
/*
|
|
* Allocate the accumulated non-linear sections, allocate a mem_map
|
|
* for each and record the physical to section mapping.
|
|
*/
|
|
void __init sparse_init(void)
|
|
{
|
|
unsigned long pnum;
|
|
struct page *map;
|
|
|
|
for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
|
|
if (!valid_section_nr(pnum))
|
|
continue;
|
|
|
|
map = sparse_early_mem_map_alloc(pnum);
|
|
if (!map)
|
|
continue;
|
|
sparse_init_one_section(__nr_to_section(pnum), pnum, map);
|
|
}
|
|
}
|
|
|
|
#ifdef CONFIG_MEMORY_HOTPLUG
|
|
static struct page *__kmalloc_section_memmap(unsigned long nr_pages)
|
|
{
|
|
struct page *page, *ret;
|
|
unsigned long memmap_size = sizeof(struct page) * nr_pages;
|
|
|
|
page = alloc_pages(GFP_KERNEL|__GFP_NOWARN, get_order(memmap_size));
|
|
if (page)
|
|
goto got_map_page;
|
|
|
|
ret = vmalloc(memmap_size);
|
|
if (ret)
|
|
goto got_map_ptr;
|
|
|
|
return NULL;
|
|
got_map_page:
|
|
ret = (struct page *)pfn_to_kaddr(page_to_pfn(page));
|
|
got_map_ptr:
|
|
memset(ret, 0, memmap_size);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static int vaddr_in_vmalloc_area(void *addr)
|
|
{
|
|
if (addr >= (void *)VMALLOC_START &&
|
|
addr < (void *)VMALLOC_END)
|
|
return 1;
|
|
return 0;
|
|
}
|
|
|
|
static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages)
|
|
{
|
|
if (vaddr_in_vmalloc_area(memmap))
|
|
vfree(memmap);
|
|
else
|
|
free_pages((unsigned long)memmap,
|
|
get_order(sizeof(struct page) * nr_pages));
|
|
}
|
|
|
|
/*
|
|
* returns the number of sections whose mem_maps were properly
|
|
* set. If this is <=0, then that means that the passed-in
|
|
* map was not consumed and must be freed.
|
|
*/
|
|
int sparse_add_one_section(struct zone *zone, unsigned long start_pfn,
|
|
int nr_pages)
|
|
{
|
|
unsigned long section_nr = pfn_to_section_nr(start_pfn);
|
|
struct pglist_data *pgdat = zone->zone_pgdat;
|
|
struct mem_section *ms;
|
|
struct page *memmap;
|
|
unsigned long flags;
|
|
int ret;
|
|
|
|
/*
|
|
* no locking for this, because it does its own
|
|
* plus, it does a kmalloc
|
|
*/
|
|
sparse_index_init(section_nr, pgdat->node_id);
|
|
memmap = __kmalloc_section_memmap(nr_pages);
|
|
|
|
pgdat_resize_lock(pgdat, &flags);
|
|
|
|
ms = __pfn_to_section(start_pfn);
|
|
if (ms->section_mem_map & SECTION_MARKED_PRESENT) {
|
|
ret = -EEXIST;
|
|
goto out;
|
|
}
|
|
ms->section_mem_map |= SECTION_MARKED_PRESENT;
|
|
|
|
ret = sparse_init_one_section(ms, section_nr, memmap);
|
|
|
|
out:
|
|
pgdat_resize_unlock(pgdat, &flags);
|
|
if (ret <= 0)
|
|
__kfree_section_memmap(memmap, nr_pages);
|
|
return ret;
|
|
}
|
|
#endif
|