2005-04-17 02:20:36 +04:00
/*
* Simple NUMA memory policy for the Linux kernel .
*
* Copyright 2003 , 2004 Andi Kleen , SuSE Labs .
2005-10-30 04:16:59 +03:00
* ( C ) Copyright 2005 Christoph Lameter , Silicon Graphics , Inc .
2005-04-17 02:20:36 +04:00
* Subject to the GNU Public License , version 2.
*
* NUMA policy allows the user to give hints in which node ( s ) memory should
* be allocated .
*
* Support four policies per VMA and per process :
*
* The VMA policy has priority over the process policy for a page fault .
*
* interleave Allocate memory interleaved over a set of nodes ,
* with normal fallback if it fails .
* For VMA based allocations this interleaves based on the
* offset into the backing object or offset into the mapping
* for anonymous memory . For process policy an process counter
* is used .
2005-10-30 04:16:59 +03:00
*
2005-04-17 02:20:36 +04:00
* bind Only allocate memory on a specific set of nodes ,
* no fallback .
2005-10-30 04:16:59 +03:00
* FIXME : memory is allocated starting with the first node
* to the last . It would be better if bind would truly restrict
* the allocation to memory nodes instead
*
2005-04-17 02:20:36 +04:00
* preferred Try a specific node first before normal fallback .
* As a special case node - 1 here means do the allocation
* on the local CPU . This is normally identical to default ,
* but useful to set in a VMA when you have a non default
* process policy .
2005-10-30 04:16:59 +03:00
*
2005-04-17 02:20:36 +04:00
* default Allocate on the local node first , or when on a VMA
* use the process policy . This is what Linux always did
* in a NUMA aware kernel and still does by , ahem , default .
*
* The process policy is applied for most non interrupt memory allocations
* in that process ' context . Interrupts ignore the policies and always
* try to allocate on the local CPU . The VMA policy is only applied for memory
* allocations for a VMA in the VM .
*
* Currently there are a few corner cases in swapping where the policy
* is not applied , but the majority should be handled . When process policy
* is used it is not remembered over swap outs / swap ins .
*
* Only the highest zone in the zone hierarchy gets policied . Allocations
* requesting a lower zone just use default policy . This implies that
* on systems with highmem kernel lowmem allocation don ' t get policied .
* Same with GFP_DMA allocations .
*
* For shmfs / tmpfs / hugetlbfs shared memory the policy is shared between
* all users and remembered even when nobody has memory mapped .
*/
/* Notebook:
fix mmap readahead to honour policy and enable policy for any page cache
object
statistics for bigpages
global policy for page cache ? currently it uses process policy . Requires
first item above .
handle mremap for shared memory ( currently ignored for the policy )
grows down ?
make bind policy root only ? It can trigger oom much faster and the
kernel is not always grateful with that .
could replace all the switch ( ) es with a mempolicy_ops structure .
*/
# include <linux/mempolicy.h>
# include <linux/mm.h>
# include <linux/highmem.h>
# include <linux/hugetlb.h>
# include <linux/kernel.h>
# include <linux/sched.h>
# include <linux/mm.h>
# include <linux/nodemask.h>
# include <linux/cpuset.h>
# include <linux/gfp.h>
# include <linux/slab.h>
# include <linux/string.h>
# include <linux/module.h>
# include <linux/interrupt.h>
# include <linux/init.h>
# include <linux/compat.h>
# include <linux/mempolicy.h>
2006-01-08 12:00:50 +03:00
# include <linux/swap.h>
2006-01-08 12:01:02 +03:00
# include <linux/seq_file.h>
# include <linux/proc_fs.h>
2006-01-08 12:00:50 +03:00
2005-04-17 02:20:36 +04:00
# include <asm/tlbflush.h>
# include <asm/uaccess.h>
2006-01-08 12:01:01 +03:00
/* Internal flags */
2006-01-08 12:00:50 +03:00
# define MPOL_MF_DISCONTIG_OK (MPOL_MF_INTERNAL << 0) /* Skip checks for continuous vmas */
2006-01-08 12:01:01 +03:00
# define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1) /* Invert check for nodemask */
2006-01-08 12:01:02 +03:00
# define MPOL_MF_STATS (MPOL_MF_INTERNAL << 2) /* Gather statistics */
2006-01-08 12:00:50 +03:00
2006-02-01 14:05:40 +03:00
/* The number of pages to migrate per call to migrate_pages() */
# define MIGRATE_CHUNK_SIZE 256
2005-04-17 02:20:36 +04:00
static kmem_cache_t * policy_cache ;
static kmem_cache_t * sn_cache ;
# define PDprintk(fmt...)
/* Highest zone. An specific allocation for a zone below that is not
policied . */
2006-01-06 11:11:17 +03:00
int policy_zone = ZONE_DMA ;
2005-04-17 02:20:36 +04:00
2005-07-06 21:56:03 +04:00
struct mempolicy default_policy = {
2005-04-17 02:20:36 +04:00
. refcnt = ATOMIC_INIT ( 1 ) , /* never free it */
. policy = MPOL_DEFAULT ,
} ;
/* Do sanity checking on a policy */
2005-10-30 04:15:48 +03:00
static int mpol_check_policy ( int mode , nodemask_t * nodes )
2005-04-17 02:20:36 +04:00
{
2005-10-30 04:15:48 +03:00
int empty = nodes_empty ( * nodes ) ;
2005-04-17 02:20:36 +04:00
switch ( mode ) {
case MPOL_DEFAULT :
if ( ! empty )
return - EINVAL ;
break ;
case MPOL_BIND :
case MPOL_INTERLEAVE :
/* Preferred will only use the first bit, but allow
more for now . */
if ( empty )
return - EINVAL ;
break ;
}
2005-10-30 04:15:48 +03:00
return nodes_subset ( * nodes , node_online_map ) ? 0 : - EINVAL ;
2005-04-17 02:20:36 +04:00
}
2006-02-17 03:39:16 +03:00
2005-04-17 02:20:36 +04:00
/* Generate a custom zonelist for the BIND policy. */
2005-10-30 04:15:48 +03:00
static struct zonelist * bind_zonelist ( nodemask_t * nodes )
2005-04-17 02:20:36 +04:00
{
struct zonelist * zl ;
2006-02-17 03:39:16 +03:00
int num , max , nd , k ;
2005-04-17 02:20:36 +04:00
2005-10-30 04:15:48 +03:00
max = 1 + MAX_NR_ZONES * nodes_weight ( * nodes ) ;
2006-02-17 03:39:16 +03:00
zl = kmalloc ( sizeof ( struct zone * ) * max , GFP_KERNEL ) ;
2005-04-17 02:20:36 +04:00
if ( ! zl )
return NULL ;
num = 0 ;
2006-02-17 03:39:16 +03:00
/* First put in the highest zones from all nodes, then all the next
lower zones etc . Avoid empty zones because the memory allocator
doesn ' t like them . If you implement node hot removal you
have to fix that . */
for ( k = policy_zone ; k > = 0 ; k - - ) {
for_each_node_mask ( nd , * nodes ) {
struct zone * z = & NODE_DATA ( nd ) - > node_zones [ k ] ;
if ( z - > present_pages > 0 )
zl - > zones [ num + + ] = z ;
}
}
2005-04-17 02:20:36 +04:00
zl - > zones [ num ] = NULL ;
return zl ;
}
/* Create a new policy */
2005-10-30 04:15:48 +03:00
static struct mempolicy * mpol_new ( int mode , nodemask_t * nodes )
2005-04-17 02:20:36 +04:00
{
struct mempolicy * policy ;
2005-10-30 04:15:48 +03:00
PDprintk ( " setting mode %d nodes[0] %lx \n " , mode , nodes_addr ( * nodes ) [ 0 ] ) ;
2005-04-17 02:20:36 +04:00
if ( mode = = MPOL_DEFAULT )
return NULL ;
policy = kmem_cache_alloc ( policy_cache , GFP_KERNEL ) ;
if ( ! policy )
return ERR_PTR ( - ENOMEM ) ;
atomic_set ( & policy - > refcnt , 1 ) ;
switch ( mode ) {
case MPOL_INTERLEAVE :
2005-10-30 04:15:48 +03:00
policy - > v . nodes = * nodes ;
2006-01-03 02:07:28 +03:00
if ( nodes_weight ( * nodes ) = = 0 ) {
kmem_cache_free ( policy_cache , policy ) ;
return ERR_PTR ( - EINVAL ) ;
}
2005-04-17 02:20:36 +04:00
break ;
case MPOL_PREFERRED :
2005-10-30 04:15:48 +03:00
policy - > v . preferred_node = first_node ( * nodes ) ;
2005-04-17 02:20:36 +04:00
if ( policy - > v . preferred_node > = MAX_NUMNODES )
policy - > v . preferred_node = - 1 ;
break ;
case MPOL_BIND :
policy - > v . zonelist = bind_zonelist ( nodes ) ;
if ( policy - > v . zonelist = = NULL ) {
kmem_cache_free ( policy_cache , policy ) ;
return ERR_PTR ( - ENOMEM ) ;
}
break ;
}
policy - > policy = mode ;
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
policy - > cpuset_mems_allowed = cpuset_mems_allowed ( current ) ;
2005-04-17 02:20:36 +04:00
return policy ;
}
2006-01-08 12:01:02 +03:00
static void gather_stats ( struct page * , void * ) ;
2006-01-19 04:42:29 +03:00
static void migrate_page_add ( struct page * page , struct list_head * pagelist ,
unsigned long flags ) ;
2006-01-08 12:01:02 +03:00
2006-01-08 12:01:01 +03:00
/* Scan through pages checking if pages follow certain conditions. */
2005-10-30 04:16:12 +03:00
static int check_pte_range ( struct vm_area_struct * vma , pmd_t * pmd ,
2006-01-08 12:00:50 +03:00
unsigned long addr , unsigned long end ,
const nodemask_t * nodes , unsigned long flags ,
2006-01-08 12:01:01 +03:00
void * private )
2005-04-17 02:20:36 +04:00
{
2005-06-22 04:15:07 +04:00
pte_t * orig_pte ;
pte_t * pte ;
2005-10-30 04:16:27 +03:00
spinlock_t * ptl ;
2005-06-22 04:15:06 +04:00
2005-10-30 04:16:27 +03:00
orig_pte = pte = pte_offset_map_lock ( vma - > vm_mm , pmd , addr , & ptl ) ;
2005-06-22 04:15:07 +04:00
do {
2005-11-29 01:34:23 +03:00
struct page * page ;
2005-06-22 04:15:07 +04:00
unsigned int nid ;
if ( ! pte_present ( * pte ) )
2005-04-17 02:20:36 +04:00
continue ;
2005-11-29 01:34:23 +03:00
page = vm_normal_page ( vma , addr , * pte ) ;
if ( ! page )
2005-04-17 02:20:36 +04:00
continue ;
2006-01-19 04:42:27 +03:00
/*
* The check for PageReserved here is important to avoid
* handling zero pages and other pages that may have been
* marked special by the system .
*
* If the PageReserved would not be checked here then f . e .
* the location of the zero page could have an influence
* on MPOL_MF_STRICT , zero pages would be counted for
* the per node stats , and there would be useless attempts
* to put zero pages on the migration list .
*/
2006-01-12 12:05:20 +03:00
if ( PageReserved ( page ) )
continue ;
2005-11-29 01:34:23 +03:00
nid = page_to_nid ( page ) ;
2006-01-08 12:01:01 +03:00
if ( node_isset ( nid , * nodes ) = = ! ! ( flags & MPOL_MF_INVERT ) )
continue ;
2006-01-08 12:01:02 +03:00
if ( flags & MPOL_MF_STATS )
gather_stats ( page , private ) ;
2006-01-19 04:42:27 +03:00
else if ( flags & ( MPOL_MF_MOVE | MPOL_MF_MOVE_ALL ) )
2006-01-19 04:42:29 +03:00
migrate_page_add ( page , private , flags ) ;
2006-01-08 12:01:01 +03:00
else
break ;
2005-06-22 04:15:07 +04:00
} while ( pte + + , addr + = PAGE_SIZE , addr ! = end ) ;
2005-10-30 04:16:27 +03:00
pte_unmap_unlock ( orig_pte , ptl ) ;
2005-06-22 04:15:07 +04:00
return addr ! = end ;
}
2005-10-30 04:16:12 +03:00
static inline int check_pmd_range ( struct vm_area_struct * vma , pud_t * pud ,
2006-01-08 12:00:50 +03:00
unsigned long addr , unsigned long end ,
const nodemask_t * nodes , unsigned long flags ,
2006-01-08 12:01:01 +03:00
void * private )
2005-06-22 04:15:07 +04:00
{
pmd_t * pmd ;
unsigned long next ;
pmd = pmd_offset ( pud , addr ) ;
do {
next = pmd_addr_end ( addr , end ) ;
if ( pmd_none_or_clear_bad ( pmd ) )
continue ;
2006-01-08 12:00:50 +03:00
if ( check_pte_range ( vma , pmd , addr , next , nodes ,
2006-01-08 12:01:01 +03:00
flags , private ) )
2005-06-22 04:15:07 +04:00
return - EIO ;
} while ( pmd + + , addr = next , addr ! = end ) ;
return 0 ;
}
2005-10-30 04:16:12 +03:00
static inline int check_pud_range ( struct vm_area_struct * vma , pgd_t * pgd ,
2006-01-08 12:00:50 +03:00
unsigned long addr , unsigned long end ,
const nodemask_t * nodes , unsigned long flags ,
2006-01-08 12:01:01 +03:00
void * private )
2005-06-22 04:15:07 +04:00
{
pud_t * pud ;
unsigned long next ;
pud = pud_offset ( pgd , addr ) ;
do {
next = pud_addr_end ( addr , end ) ;
if ( pud_none_or_clear_bad ( pud ) )
continue ;
2006-01-08 12:00:50 +03:00
if ( check_pmd_range ( vma , pud , addr , next , nodes ,
2006-01-08 12:01:01 +03:00
flags , private ) )
2005-06-22 04:15:07 +04:00
return - EIO ;
} while ( pud + + , addr = next , addr ! = end ) ;
return 0 ;
}
2005-10-30 04:16:12 +03:00
static inline int check_pgd_range ( struct vm_area_struct * vma ,
2006-01-08 12:00:50 +03:00
unsigned long addr , unsigned long end ,
const nodemask_t * nodes , unsigned long flags ,
2006-01-08 12:01:01 +03:00
void * private )
2005-06-22 04:15:07 +04:00
{
pgd_t * pgd ;
unsigned long next ;
2005-10-30 04:16:12 +03:00
pgd = pgd_offset ( vma - > vm_mm , addr ) ;
2005-06-22 04:15:07 +04:00
do {
next = pgd_addr_end ( addr , end ) ;
if ( pgd_none_or_clear_bad ( pgd ) )
continue ;
2006-01-08 12:00:50 +03:00
if ( check_pud_range ( vma , pgd , addr , next , nodes ,
2006-01-08 12:01:01 +03:00
flags , private ) )
2005-06-22 04:15:07 +04:00
return - EIO ;
} while ( pgd + + , addr = next , addr ! = end ) ;
return 0 ;
2005-04-17 02:20:36 +04:00
}
2006-01-08 12:00:50 +03:00
/* Check if a vma is migratable */
static inline int vma_migratable ( struct vm_area_struct * vma )
{
if ( vma - > vm_flags & (
2006-01-12 12:05:20 +03:00
VM_LOCKED | VM_IO | VM_HUGETLB | VM_PFNMAP | VM_RESERVED ) )
2006-01-08 12:00:50 +03:00
return 0 ;
return 1 ;
}
/*
* Check if all pages in a range are on a set of nodes .
* If pagelist ! = NULL then isolate pages from the LRU and
* put them on the pagelist .
*/
2005-04-17 02:20:36 +04:00
static struct vm_area_struct *
check_range ( struct mm_struct * mm , unsigned long start , unsigned long end ,
2006-01-08 12:01:01 +03:00
const nodemask_t * nodes , unsigned long flags , void * private )
2005-04-17 02:20:36 +04:00
{
int err ;
struct vm_area_struct * first , * vma , * prev ;
2006-01-19 04:42:27 +03:00
/* Clear the LRU lists so pages can be isolated */
if ( flags & ( MPOL_MF_MOVE | MPOL_MF_MOVE_ALL ) )
lru_add_drain_all ( ) ;
2005-04-17 02:20:36 +04:00
first = find_vma ( mm , start ) ;
if ( ! first )
return ERR_PTR ( - EFAULT ) ;
prev = NULL ;
for ( vma = first ; vma & & vma - > vm_start < end ; vma = vma - > vm_next ) {
2006-01-08 12:00:50 +03:00
if ( ! ( flags & MPOL_MF_DISCONTIG_OK ) ) {
if ( ! vma - > vm_next & & vma - > vm_end < end )
return ERR_PTR ( - EFAULT ) ;
if ( prev & & prev - > vm_end < vma - > vm_start )
return ERR_PTR ( - EFAULT ) ;
}
if ( ! is_vm_hugetlb_page ( vma ) & &
( ( flags & MPOL_MF_STRICT ) | |
( ( flags & ( MPOL_MF_MOVE | MPOL_MF_MOVE_ALL ) ) & &
vma_migratable ( vma ) ) ) ) {
2005-09-13 12:25:08 +04:00
unsigned long endvma = vma - > vm_end ;
2006-01-08 12:00:50 +03:00
2005-09-13 12:25:08 +04:00
if ( endvma > end )
endvma = end ;
if ( vma - > vm_start > start )
start = vma - > vm_start ;
2006-01-08 12:00:50 +03:00
err = check_pgd_range ( vma , start , endvma , nodes ,
2006-01-08 12:01:01 +03:00
flags , private ) ;
2005-04-17 02:20:36 +04:00
if ( err ) {
first = ERR_PTR ( err ) ;
break ;
}
}
prev = vma ;
}
return first ;
}
/* Apply policy to a single VMA */
static int policy_vma ( struct vm_area_struct * vma , struct mempolicy * new )
{
int err = 0 ;
struct mempolicy * old = vma - > vm_policy ;
PDprintk ( " vma %lx-%lx/%lx vm_ops %p vm_file %p set_policy %p \n " ,
vma - > vm_start , vma - > vm_end , vma - > vm_pgoff ,
vma - > vm_ops , vma - > vm_file ,
vma - > vm_ops ? vma - > vm_ops - > set_policy : NULL ) ;
if ( vma - > vm_ops & & vma - > vm_ops - > set_policy )
err = vma - > vm_ops - > set_policy ( vma , new ) ;
if ( ! err ) {
mpol_get ( new ) ;
vma - > vm_policy = new ;
mpol_free ( old ) ;
}
return err ;
}
/* Step 2: apply policy to a range and do splits. */
static int mbind_range ( struct vm_area_struct * vma , unsigned long start ,
unsigned long end , struct mempolicy * new )
{
struct vm_area_struct * next ;
int err ;
err = 0 ;
for ( ; vma & & vma - > vm_start < end ; vma = next ) {
next = vma - > vm_next ;
if ( vma - > vm_start < start )
err = split_vma ( vma - > vm_mm , vma , start , 1 ) ;
if ( ! err & & vma - > vm_end > end )
err = split_vma ( vma - > vm_mm , vma , end , 0 ) ;
if ( ! err )
err = policy_vma ( vma , new ) ;
if ( err )
break ;
}
return err ;
}
2005-10-30 04:16:59 +03:00
static int contextualize_policy ( int mode , nodemask_t * nodes )
{
if ( ! nodes )
return 0 ;
2006-01-08 12:01:54 +03:00
cpuset_update_task_memory_state ( ) ;
2006-01-08 12:01:47 +03:00
if ( ! cpuset_nodes_subset_current_mems_allowed ( * nodes ) )
return - EINVAL ;
2005-10-30 04:16:59 +03:00
return mpol_check_policy ( mode , nodes ) ;
}
2005-04-17 02:20:36 +04:00
/* Set the process memory policy */
2005-10-30 04:16:59 +03:00
long do_set_mempolicy ( int mode , nodemask_t * nodes )
2005-04-17 02:20:36 +04:00
{
struct mempolicy * new ;
2005-10-30 04:16:59 +03:00
if ( contextualize_policy ( mode , nodes ) )
2005-04-17 02:20:36 +04:00
return - EINVAL ;
2005-10-30 04:16:59 +03:00
new = mpol_new ( mode , nodes ) ;
2005-04-17 02:20:36 +04:00
if ( IS_ERR ( new ) )
return PTR_ERR ( new ) ;
mpol_free ( current - > mempolicy ) ;
current - > mempolicy = new ;
if ( new & & new - > policy = = MPOL_INTERLEAVE )
2005-10-30 04:15:48 +03:00
current - > il_next = first_node ( new - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
return 0 ;
}
/* Fill a zone bitmap for a policy */
2005-10-30 04:15:48 +03:00
static void get_zonemask ( struct mempolicy * p , nodemask_t * nodes )
2005-04-17 02:20:36 +04:00
{
int i ;
2005-10-30 04:15:48 +03:00
nodes_clear ( * nodes ) ;
2005-04-17 02:20:36 +04:00
switch ( p - > policy ) {
case MPOL_BIND :
for ( i = 0 ; p - > v . zonelist - > zones [ i ] ; i + + )
2005-10-30 04:16:59 +03:00
node_set ( p - > v . zonelist - > zones [ i ] - > zone_pgdat - > node_id ,
* nodes ) ;
2005-04-17 02:20:36 +04:00
break ;
case MPOL_DEFAULT :
break ;
case MPOL_INTERLEAVE :
2005-10-30 04:15:48 +03:00
* nodes = p - > v . nodes ;
2005-04-17 02:20:36 +04:00
break ;
case MPOL_PREFERRED :
/* or use current node instead of online map? */
if ( p - > v . preferred_node < 0 )
2005-10-30 04:15:48 +03:00
* nodes = node_online_map ;
2005-04-17 02:20:36 +04:00
else
2005-10-30 04:15:48 +03:00
node_set ( p - > v . preferred_node , * nodes ) ;
2005-04-17 02:20:36 +04:00
break ;
default :
BUG ( ) ;
}
}
static int lookup_node ( struct mm_struct * mm , unsigned long addr )
{
struct page * p ;
int err ;
err = get_user_pages ( current , mm , addr & PAGE_MASK , 1 , 0 , 0 , & p , NULL ) ;
if ( err > = 0 ) {
err = page_to_nid ( p ) ;
put_page ( p ) ;
}
return err ;
}
/* Retrieve NUMA policy */
2005-10-30 04:16:59 +03:00
long do_get_mempolicy ( int * policy , nodemask_t * nmask ,
unsigned long addr , unsigned long flags )
2005-04-17 02:20:36 +04:00
{
2005-10-30 04:16:59 +03:00
int err ;
2005-04-17 02:20:36 +04:00
struct mm_struct * mm = current - > mm ;
struct vm_area_struct * vma = NULL ;
struct mempolicy * pol = current - > mempolicy ;
2006-01-08 12:01:54 +03:00
cpuset_update_task_memory_state ( ) ;
2005-04-17 02:20:36 +04:00
if ( flags & ~ ( unsigned long ) ( MPOL_F_NODE | MPOL_F_ADDR ) )
return - EINVAL ;
if ( flags & MPOL_F_ADDR ) {
down_read ( & mm - > mmap_sem ) ;
vma = find_vma_intersection ( mm , addr , addr + 1 ) ;
if ( ! vma ) {
up_read ( & mm - > mmap_sem ) ;
return - EFAULT ;
}
if ( vma - > vm_ops & & vma - > vm_ops - > get_policy )
pol = vma - > vm_ops - > get_policy ( vma , addr ) ;
else
pol = vma - > vm_policy ;
} else if ( addr )
return - EINVAL ;
if ( ! pol )
pol = & default_policy ;
if ( flags & MPOL_F_NODE ) {
if ( flags & MPOL_F_ADDR ) {
err = lookup_node ( mm , addr ) ;
if ( err < 0 )
goto out ;
2005-10-30 04:16:59 +03:00
* policy = err ;
2005-04-17 02:20:36 +04:00
} else if ( pol = = current - > mempolicy & &
pol - > policy = = MPOL_INTERLEAVE ) {
2005-10-30 04:16:59 +03:00
* policy = current - > il_next ;
2005-04-17 02:20:36 +04:00
} else {
err = - EINVAL ;
goto out ;
}
} else
2005-10-30 04:16:59 +03:00
* policy = pol - > policy ;
2005-04-17 02:20:36 +04:00
if ( vma ) {
up_read ( & current - > mm - > mmap_sem ) ;
vma = NULL ;
}
err = 0 ;
2005-10-30 04:16:59 +03:00
if ( nmask )
get_zonemask ( pol , nmask ) ;
2005-04-17 02:20:36 +04:00
out :
if ( vma )
up_read ( & current - > mm - > mmap_sem ) ;
return err ;
}
2006-01-08 12:01:04 +03:00
/*
* page migration
*/
2006-01-19 04:42:29 +03:00
static void migrate_page_add ( struct page * page , struct list_head * pagelist ,
unsigned long flags )
2006-01-08 12:01:04 +03:00
{
/*
2006-01-19 04:42:29 +03:00
* Avoid migrating a page that is shared with others .
2006-01-08 12:01:04 +03:00
*/
2006-01-19 04:42:29 +03:00
if ( ( flags & MPOL_MF_MOVE_ALL ) | | page_mapcount ( page ) = = 1 ) {
2006-01-19 04:42:27 +03:00
if ( isolate_lru_page ( page ) )
2006-02-25 00:04:12 +03:00
list_add_tail ( & page - > lru , pagelist ) ;
2006-01-08 12:01:04 +03:00
}
}
2006-02-01 14:05:40 +03:00
/*
* Migrate the list ' pagelist ' of pages to a certain destination .
*
* Specify destination with either non - NULL vma or dest_node > = 0
* Return the number of pages not migrated or error code
*/
static int migrate_pages_to ( struct list_head * pagelist ,
struct vm_area_struct * vma , int dest )
2006-01-08 12:01:04 +03:00
{
2006-02-01 14:05:40 +03:00
LIST_HEAD ( newlist ) ;
2006-01-08 12:01:04 +03:00
LIST_HEAD ( moved ) ;
LIST_HEAD ( failed ) ;
2006-02-01 14:05:40 +03:00
int err = 0 ;
2006-02-25 00:04:12 +03:00
unsigned long offset = 0 ;
2006-02-01 14:05:40 +03:00
int nr_pages ;
struct page * page ;
struct list_head * p ;
redo :
nr_pages = 0 ;
list_for_each ( p , pagelist ) {
2006-02-25 00:04:12 +03:00
if ( vma ) {
/*
* The address passed to alloc_page_vma is used to
* generate the proper interleave behavior . We fake
* the address here by an increasing offset in order
* to get the proper distribution of pages .
*
* No decision has been made as to which page
* a certain old page is moved to so we cannot
* specify the correct address .
*/
page = alloc_page_vma ( GFP_HIGHUSER , vma ,
offset + vma - > vm_start ) ;
offset + = PAGE_SIZE ;
}
2006-02-01 14:05:40 +03:00
else
page = alloc_pages_node ( dest , GFP_HIGHUSER , 0 ) ;
if ( ! page ) {
err = - ENOMEM ;
goto out ;
}
2006-02-25 00:04:12 +03:00
list_add_tail ( & page - > lru , & newlist ) ;
2006-02-01 14:05:40 +03:00
nr_pages + + ;
2006-02-21 05:28:10 +03:00
if ( nr_pages > MIGRATE_CHUNK_SIZE )
2006-02-01 14:05:40 +03:00
break ;
}
err = migrate_pages ( pagelist , & newlist , & moved , & failed ) ;
putback_lru_pages ( & moved ) ; /* Call release pages instead ?? */
if ( err > = 0 & & list_empty ( & newlist ) & & ! list_empty ( pagelist ) )
goto redo ;
out :
/* Return leftover allocated pages */
while ( ! list_empty ( & newlist ) ) {
page = list_entry ( newlist . next , struct page , lru ) ;
list_del ( & page - > lru ) ;
__free_page ( page ) ;
}
list_splice ( & failed , pagelist ) ;
if ( err < 0 )
return err ;
/* Calculate number of leftover pages */
nr_pages = 0 ;
list_for_each ( p , pagelist )
nr_pages + + ;
return nr_pages ;
}
2006-01-08 12:01:04 +03:00
2006-02-01 14:05:40 +03:00
/*
* Migrate pages from one node to a target node .
* Returns error or the number of pages not migrated .
*/
int migrate_to_node ( struct mm_struct * mm , int source , int dest , int flags )
{
nodemask_t nmask ;
LIST_HEAD ( pagelist ) ;
int err = 0 ;
nodes_clear ( nmask ) ;
node_set ( source , nmask ) ;
2006-01-08 12:01:04 +03:00
2006-02-01 14:05:40 +03:00
check_range ( mm , mm - > mmap - > vm_start , TASK_SIZE , & nmask ,
flags | MPOL_MF_DISCONTIG_OK , & pagelist ) ;
if ( ! list_empty ( & pagelist ) ) {
err = migrate_pages_to ( & pagelist , NULL , dest ) ;
if ( ! list_empty ( & pagelist ) )
putback_lru_pages ( & pagelist ) ;
}
return err ;
2006-01-08 12:01:04 +03:00
}
2006-01-08 12:00:51 +03:00
/*
2006-02-01 14:05:40 +03:00
* Move pages between the two nodesets so as to preserve the physical
* layout as much as possible .
2006-01-08 12:00:51 +03:00
*
* Returns the number of page that could not be moved .
*/
int do_migrate_pages ( struct mm_struct * mm ,
const nodemask_t * from_nodes , const nodemask_t * to_nodes , int flags )
{
LIST_HEAD ( pagelist ) ;
2006-02-01 14:05:40 +03:00
int busy = 0 ;
int err = 0 ;
nodemask_t tmp ;
2006-01-08 12:00:51 +03:00
2006-02-01 14:05:40 +03:00
down_read ( & mm - > mmap_sem ) ;
2006-01-08 12:00:51 +03:00
2006-02-01 14:05:40 +03:00
/*
* Find a ' source ' bit set in ' tmp ' whose corresponding ' dest '
* bit in ' to ' is not also set in ' tmp ' . Clear the found ' source '
* bit in ' tmp ' , and return that < source , dest > pair for migration .
* The pair of nodemasks ' to ' and ' from ' define the map .
*
* If no pair of bits is found that way , fallback to picking some
* pair of ' source ' and ' dest ' bits that are not the same . If the
* ' source ' and ' dest ' bits are the same , this represents a node
* that will be migrating to itself , so no pages need move .
*
* If no bits are left in ' tmp ' , or if all remaining bits left
* in ' tmp ' correspond to the same bit in ' to ' , return false
* ( nothing left to migrate ) .
*
* This lets us pick a pair of nodes to migrate between , such that
* if possible the dest node is not already occupied by some other
* source node , minimizing the risk of overloading the memory on a
* node that would happen if we migrated incoming memory to a node
* before migrating outgoing memory source that same node .
*
* A single scan of tmp is sufficient . As we go , we remember the
* most recent < s , d > pair that moved ( s ! = d ) . If we find a pair
* that not only moved , but what ' s better , moved to an empty slot
* ( d is not set in tmp ) , then we break out then , with that pair .
* Otherwise when we finish scannng from_tmp , we at least have the
* most recent < s , d > pair that moved . If we get all the way through
* the scan of tmp without finding any node that moved , much less
* moved to an empty node , then there is nothing left worth migrating .
*/
2006-01-08 12:00:55 +03:00
2006-02-01 14:05:40 +03:00
tmp = * from_nodes ;
while ( ! nodes_empty ( tmp ) ) {
int s , d ;
int source = - 1 ;
int dest = 0 ;
for_each_node_mask ( s , tmp ) {
d = node_remap ( s , * from_nodes , * to_nodes ) ;
if ( s = = d )
continue ;
source = s ; /* Node moved. Memorize */
dest = d ;
/* dest not in remaining from nodes? */
if ( ! node_isset ( dest , tmp ) )
break ;
}
if ( source = = - 1 )
break ;
node_clear ( source , tmp ) ;
err = migrate_to_node ( mm , source , dest , flags ) ;
if ( err > 0 )
busy + = err ;
if ( err < 0 )
break ;
2006-01-08 12:00:51 +03:00
}
2006-01-08 12:00:55 +03:00
2006-01-08 12:00:51 +03:00
up_read ( & mm - > mmap_sem ) ;
2006-02-01 14:05:40 +03:00
if ( err < 0 )
return err ;
return busy ;
2006-01-08 12:00:51 +03:00
}
2006-01-08 12:01:04 +03:00
long do_mbind ( unsigned long start , unsigned long len ,
unsigned long mode , nodemask_t * nmask , unsigned long flags )
{
struct vm_area_struct * vma ;
struct mm_struct * mm = current - > mm ;
struct mempolicy * new ;
unsigned long end ;
int err ;
LIST_HEAD ( pagelist ) ;
if ( ( flags & ~ ( unsigned long ) ( MPOL_MF_STRICT |
MPOL_MF_MOVE | MPOL_MF_MOVE_ALL ) )
| | mode > MPOL_MAX )
return - EINVAL ;
if ( ( flags & MPOL_MF_MOVE_ALL ) & & ! capable ( CAP_SYS_RESOURCE ) )
return - EPERM ;
if ( start & ~ PAGE_MASK )
return - EINVAL ;
if ( mode = = MPOL_DEFAULT )
flags & = ~ MPOL_MF_STRICT ;
len = ( len + PAGE_SIZE - 1 ) & PAGE_MASK ;
end = start + len ;
if ( end < start )
return - EINVAL ;
if ( end = = start )
return 0 ;
if ( mpol_check_policy ( mode , nmask ) )
return - EINVAL ;
new = mpol_new ( mode , nmask ) ;
if ( IS_ERR ( new ) )
return PTR_ERR ( new ) ;
/*
* If we are using the default policy then operation
* on discontinuous address spaces is okay after all
*/
if ( ! new )
flags | = MPOL_MF_DISCONTIG_OK ;
PDprintk ( " mbind %lx-%lx mode:%ld nodes:%lx \n " , start , start + len ,
mode , nodes_addr ( nodes ) [ 0 ] ) ;
down_write ( & mm - > mmap_sem ) ;
vma = check_range ( mm , start , end , nmask ,
flags | MPOL_MF_INVERT , & pagelist ) ;
err = PTR_ERR ( vma ) ;
if ( ! IS_ERR ( vma ) ) {
int nr_failed = 0 ;
err = mbind_range ( vma , start , end , new ) ;
2006-02-01 14:05:40 +03:00
2006-01-08 12:01:04 +03:00
if ( ! list_empty ( & pagelist ) )
2006-02-01 14:05:40 +03:00
nr_failed = migrate_pages_to ( & pagelist , vma , - 1 ) ;
2006-01-08 12:01:04 +03:00
if ( ! err & & nr_failed & & ( flags & MPOL_MF_STRICT ) )
err = - EIO ;
}
if ( ! list_empty ( & pagelist ) )
putback_lru_pages ( & pagelist ) ;
up_write ( & mm - > mmap_sem ) ;
mpol_free ( new ) ;
return err ;
}
2005-10-30 04:16:59 +03:00
/*
* User space interface with variable sized bitmaps for nodelists .
*/
/* Copy a node mask from user space. */
2006-01-08 12:00:51 +03:00
static int get_nodes ( nodemask_t * nodes , const unsigned long __user * nmask ,
2005-10-30 04:16:59 +03:00
unsigned long maxnode )
{
unsigned long k ;
unsigned long nlongs ;
unsigned long endmask ;
- - maxnode ;
nodes_clear ( * nodes ) ;
if ( maxnode = = 0 | | ! nmask )
return 0 ;
2006-02-21 05:27:59 +03:00
if ( maxnode > PAGE_SIZE * BITS_PER_BYTE )
2006-02-18 00:59:36 +03:00
return - EINVAL ;
2005-10-30 04:16:59 +03:00
nlongs = BITS_TO_LONGS ( maxnode ) ;
if ( ( maxnode % BITS_PER_LONG ) = = 0 )
endmask = ~ 0UL ;
else
endmask = ( 1UL < < ( maxnode % BITS_PER_LONG ) ) - 1 ;
/* When the user specified more nodes than supported just check
if the non supported part is all zero . */
if ( nlongs > BITS_TO_LONGS ( MAX_NUMNODES ) ) {
if ( nlongs > PAGE_SIZE / sizeof ( long ) )
return - EINVAL ;
for ( k = BITS_TO_LONGS ( MAX_NUMNODES ) ; k < nlongs ; k + + ) {
unsigned long t ;
if ( get_user ( t , nmask + k ) )
return - EFAULT ;
if ( k = = nlongs - 1 ) {
if ( t & endmask )
return - EINVAL ;
} else if ( t )
return - EINVAL ;
}
nlongs = BITS_TO_LONGS ( MAX_NUMNODES ) ;
endmask = ~ 0UL ;
}
if ( copy_from_user ( nodes_addr ( * nodes ) , nmask , nlongs * sizeof ( unsigned long ) ) )
return - EFAULT ;
nodes_addr ( * nodes ) [ nlongs - 1 ] & = endmask ;
return 0 ;
}
/* Copy a kernel node mask to user space */
static int copy_nodes_to_user ( unsigned long __user * mask , unsigned long maxnode ,
nodemask_t * nodes )
{
unsigned long copy = ALIGN ( maxnode - 1 , 64 ) / 8 ;
const int nbytes = BITS_TO_LONGS ( MAX_NUMNODES ) * sizeof ( long ) ;
if ( copy > nbytes ) {
if ( copy > PAGE_SIZE )
return - EINVAL ;
if ( clear_user ( ( char __user * ) mask + nbytes , copy - nbytes ) )
return - EFAULT ;
copy = nbytes ;
}
return copy_to_user ( mask , nodes_addr ( * nodes ) , copy ) ? - EFAULT : 0 ;
}
asmlinkage long sys_mbind ( unsigned long start , unsigned long len ,
unsigned long mode ,
unsigned long __user * nmask , unsigned long maxnode ,
unsigned flags )
{
nodemask_t nodes ;
int err ;
err = get_nodes ( & nodes , nmask , maxnode ) ;
if ( err )
return err ;
return do_mbind ( start , len , mode , & nodes , flags ) ;
}
/* Set the process memory policy */
asmlinkage long sys_set_mempolicy ( int mode , unsigned long __user * nmask ,
unsigned long maxnode )
{
int err ;
nodemask_t nodes ;
if ( mode < 0 | | mode > MPOL_MAX )
return - EINVAL ;
err = get_nodes ( & nodes , nmask , maxnode ) ;
if ( err )
return err ;
return do_set_mempolicy ( mode , & nodes ) ;
}
2006-01-08 12:00:51 +03:00
asmlinkage long sys_migrate_pages ( pid_t pid , unsigned long maxnode ,
const unsigned long __user * old_nodes ,
const unsigned long __user * new_nodes )
{
struct mm_struct * mm ;
struct task_struct * task ;
nodemask_t old ;
nodemask_t new ;
nodemask_t task_nodes ;
int err ;
err = get_nodes ( & old , old_nodes , maxnode ) ;
if ( err )
return err ;
err = get_nodes ( & new , new_nodes , maxnode ) ;
if ( err )
return err ;
/* Find the mm_struct */
read_lock ( & tasklist_lock ) ;
task = pid ? find_task_by_pid ( pid ) : current ;
if ( ! task ) {
read_unlock ( & tasklist_lock ) ;
return - ESRCH ;
}
mm = get_task_mm ( task ) ;
read_unlock ( & tasklist_lock ) ;
if ( ! mm )
return - EINVAL ;
/*
* Check if this process has the right to modify the specified
* process . The right exists if the process has administrative
* capabilities , superuser priviledges or the same
* userid as the target process .
*/
if ( ( current - > euid ! = task - > suid ) & & ( current - > euid ! = task - > uid ) & &
( current - > uid ! = task - > suid ) & & ( current - > uid ! = task - > uid ) & &
! capable ( CAP_SYS_ADMIN ) ) {
err = - EPERM ;
goto out ;
}
task_nodes = cpuset_mems_allowed ( task ) ;
/* Is the user allowed to access the target nodes? */
if ( ! nodes_subset ( new , task_nodes ) & & ! capable ( CAP_SYS_ADMIN ) ) {
err = - EPERM ;
goto out ;
}
2006-03-01 03:58:57 +03:00
err = do_migrate_pages ( mm , & old , & new ,
capable ( CAP_SYS_ADMIN ) ? MPOL_MF_MOVE_ALL : MPOL_MF_MOVE ) ;
2006-01-08 12:00:51 +03:00
out :
mmput ( mm ) ;
return err ;
}
2005-10-30 04:16:59 +03:00
/* Retrieve NUMA policy */
asmlinkage long sys_get_mempolicy ( int __user * policy ,
unsigned long __user * nmask ,
unsigned long maxnode ,
unsigned long addr , unsigned long flags )
{
int err , pval ;
nodemask_t nodes ;
if ( nmask ! = NULL & & maxnode < MAX_NUMNODES )
return - EINVAL ;
err = do_get_mempolicy ( & pval , & nodes , addr , flags ) ;
if ( err )
return err ;
if ( policy & & put_user ( pval , policy ) )
return - EFAULT ;
if ( nmask )
err = copy_nodes_to_user ( nmask , maxnode , & nodes ) ;
return err ;
}
2005-04-17 02:20:36 +04:00
# ifdef CONFIG_COMPAT
asmlinkage long compat_sys_get_mempolicy ( int __user * policy ,
compat_ulong_t __user * nmask ,
compat_ulong_t maxnode ,
compat_ulong_t addr , compat_ulong_t flags )
{
long err ;
unsigned long __user * nm = NULL ;
unsigned long nr_bits , alloc_size ;
DECLARE_BITMAP ( bm , MAX_NUMNODES ) ;
nr_bits = min_t ( unsigned long , maxnode - 1 , MAX_NUMNODES ) ;
alloc_size = ALIGN ( nr_bits , BITS_PER_LONG ) / 8 ;
if ( nmask )
nm = compat_alloc_user_space ( alloc_size ) ;
err = sys_get_mempolicy ( policy , nm , nr_bits + 1 , addr , flags ) ;
if ( ! err & & nmask ) {
err = copy_from_user ( bm , nm , alloc_size ) ;
/* ensure entire bitmap is zeroed */
err | = clear_user ( nmask , ALIGN ( maxnode - 1 , 8 ) / 8 ) ;
err | = compat_put_bitmap ( nmask , bm , nr_bits ) ;
}
return err ;
}
asmlinkage long compat_sys_set_mempolicy ( int mode , compat_ulong_t __user * nmask ,
compat_ulong_t maxnode )
{
long err = 0 ;
unsigned long __user * nm = NULL ;
unsigned long nr_bits , alloc_size ;
DECLARE_BITMAP ( bm , MAX_NUMNODES ) ;
nr_bits = min_t ( unsigned long , maxnode - 1 , MAX_NUMNODES ) ;
alloc_size = ALIGN ( nr_bits , BITS_PER_LONG ) / 8 ;
if ( nmask ) {
err = compat_get_bitmap ( bm , nmask , nr_bits ) ;
nm = compat_alloc_user_space ( alloc_size ) ;
err | = copy_to_user ( nm , bm , alloc_size ) ;
}
if ( err )
return - EFAULT ;
return sys_set_mempolicy ( mode , nm , nr_bits + 1 ) ;
}
asmlinkage long compat_sys_mbind ( compat_ulong_t start , compat_ulong_t len ,
compat_ulong_t mode , compat_ulong_t __user * nmask ,
compat_ulong_t maxnode , compat_ulong_t flags )
{
long err = 0 ;
unsigned long __user * nm = NULL ;
unsigned long nr_bits , alloc_size ;
2005-10-30 04:15:48 +03:00
nodemask_t bm ;
2005-04-17 02:20:36 +04:00
nr_bits = min_t ( unsigned long , maxnode - 1 , MAX_NUMNODES ) ;
alloc_size = ALIGN ( nr_bits , BITS_PER_LONG ) / 8 ;
if ( nmask ) {
2005-10-30 04:15:48 +03:00
err = compat_get_bitmap ( nodes_addr ( bm ) , nmask , nr_bits ) ;
2005-04-17 02:20:36 +04:00
nm = compat_alloc_user_space ( alloc_size ) ;
2005-10-30 04:15:48 +03:00
err | = copy_to_user ( nm , nodes_addr ( bm ) , alloc_size ) ;
2005-04-17 02:20:36 +04:00
}
if ( err )
return - EFAULT ;
return sys_mbind ( start , len , mode , nm , nr_bits + 1 , flags ) ;
}
# endif
/* Return effective policy for a VMA */
2006-01-08 12:01:03 +03:00
static struct mempolicy * get_vma_policy ( struct task_struct * task ,
struct vm_area_struct * vma , unsigned long addr )
2005-04-17 02:20:36 +04:00
{
2005-09-04 02:54:45 +04:00
struct mempolicy * pol = task - > mempolicy ;
2005-04-17 02:20:36 +04:00
if ( vma ) {
if ( vma - > vm_ops & & vma - > vm_ops - > get_policy )
2005-10-30 04:16:59 +03:00
pol = vma - > vm_ops - > get_policy ( vma , addr ) ;
2005-04-17 02:20:36 +04:00
else if ( vma - > vm_policy & &
vma - > vm_policy - > policy ! = MPOL_DEFAULT )
pol = vma - > vm_policy ;
}
if ( ! pol )
pol = & default_policy ;
return pol ;
}
/* Return a zonelist representing a mempolicy */
2005-10-07 10:46:04 +04:00
static struct zonelist * zonelist_policy ( gfp_t gfp , struct mempolicy * policy )
2005-04-17 02:20:36 +04:00
{
int nd ;
switch ( policy - > policy ) {
case MPOL_PREFERRED :
nd = policy - > v . preferred_node ;
if ( nd < 0 )
nd = numa_node_id ( ) ;
break ;
case MPOL_BIND :
/* Lower zones don't get a policy applied */
/* Careful: current->mems_allowed might have moved */
2005-10-21 10:55:38 +04:00
if ( gfp_zone ( gfp ) > = policy_zone )
2005-04-17 02:20:36 +04:00
if ( cpuset_zonelist_valid_mems_allowed ( policy - > v . zonelist ) )
return policy - > v . zonelist ;
/*FALL THROUGH*/
case MPOL_INTERLEAVE : /* should not happen */
case MPOL_DEFAULT :
nd = numa_node_id ( ) ;
break ;
default :
nd = 0 ;
BUG ( ) ;
}
2005-10-21 10:55:38 +04:00
return NODE_DATA ( nd ) - > node_zonelists + gfp_zone ( gfp ) ;
2005-04-17 02:20:36 +04:00
}
/* Do dynamic interleaving for a process */
static unsigned interleave_nodes ( struct mempolicy * policy )
{
unsigned nid , next ;
struct task_struct * me = current ;
nid = me - > il_next ;
2005-10-30 04:15:48 +03:00
next = next_node ( nid , policy - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
if ( next > = MAX_NUMNODES )
2005-10-30 04:15:48 +03:00
next = first_node ( policy - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
me - > il_next = next ;
return nid ;
}
2006-01-19 04:42:36 +03:00
/*
* Depending on the memory policy provide a node from which to allocate the
* next slab entry .
*/
unsigned slab_node ( struct mempolicy * policy )
{
switch ( policy - > policy ) {
case MPOL_INTERLEAVE :
return interleave_nodes ( policy ) ;
case MPOL_BIND :
/*
* Follow bind policy behavior and start allocation at the
* first node .
*/
return policy - > v . zonelist - > zones [ 0 ] - > zone_pgdat - > node_id ;
case MPOL_PREFERRED :
if ( policy - > v . preferred_node > = 0 )
return policy - > v . preferred_node ;
/* Fall through */
default :
return numa_node_id ( ) ;
}
}
2005-04-17 02:20:36 +04:00
/* Do static interleaving for a VMA with known offset. */
static unsigned offset_il_node ( struct mempolicy * pol ,
struct vm_area_struct * vma , unsigned long off )
{
2005-10-30 04:15:48 +03:00
unsigned nnodes = nodes_weight ( pol - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
unsigned target = ( unsigned ) off % nnodes ;
int c ;
int nid = - 1 ;
c = 0 ;
do {
2005-10-30 04:15:48 +03:00
nid = next_node ( nid , pol - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
c + + ;
} while ( c < = target ) ;
return nid ;
}
2006-01-06 11:10:46 +03:00
/* Determine a node number for interleave */
static inline unsigned interleave_nid ( struct mempolicy * pol ,
struct vm_area_struct * vma , unsigned long addr , int shift )
{
if ( vma ) {
unsigned long off ;
off = vma - > vm_pgoff ;
off + = ( addr - vma - > vm_start ) > > shift ;
return offset_il_node ( pol , vma , off ) ;
} else
return interleave_nodes ( pol ) ;
}
2006-02-03 23:51:14 +03:00
# ifdef CONFIG_HUGETLBFS
2006-01-06 11:10:46 +03:00
/* Return a zonelist suitable for a huge page allocation. */
struct zonelist * huge_zonelist ( struct vm_area_struct * vma , unsigned long addr )
{
struct mempolicy * pol = get_vma_policy ( current , vma , addr ) ;
if ( pol - > policy = = MPOL_INTERLEAVE ) {
unsigned nid ;
nid = interleave_nid ( pol , vma , addr , HPAGE_SHIFT ) ;
return NODE_DATA ( nid ) - > node_zonelists + gfp_zone ( GFP_HIGHUSER ) ;
}
return zonelist_policy ( GFP_HIGHUSER , pol ) ;
}
2006-02-03 23:51:14 +03:00
# endif
2006-01-06 11:10:46 +03:00
2005-04-17 02:20:36 +04:00
/* Allocate a page in interleaved policy.
Own path because it needs to do special accounting . */
2005-10-30 04:15:49 +03:00
static struct page * alloc_page_interleave ( gfp_t gfp , unsigned order ,
unsigned nid )
2005-04-17 02:20:36 +04:00
{
struct zonelist * zl ;
struct page * page ;
2005-10-21 10:55:38 +04:00
zl = NODE_DATA ( nid ) - > node_zonelists + gfp_zone ( gfp ) ;
2005-04-17 02:20:36 +04:00
page = __alloc_pages ( gfp , order , zl ) ;
if ( page & & page_zone ( page ) = = zl - > zones [ 0 ] ) {
2005-06-22 04:14:47 +04:00
zone_pcp ( zl - > zones [ 0 ] , get_cpu ( ) ) - > interleave_hit + + ;
2005-04-17 02:20:36 +04:00
put_cpu ( ) ;
}
return page ;
}
/**
* alloc_page_vma - Allocate a page for a VMA .
*
* @ gfp :
* % GFP_USER user allocation .
* % GFP_KERNEL kernel allocations ,
* % GFP_HIGHMEM highmem / user allocations ,
* % GFP_FS allocation should not call back into a file system .
* % GFP_ATOMIC don ' t sleep .
*
* @ vma : Pointer to VMA or NULL if not available .
* @ addr : Virtual Address of the allocation . Must be inside the VMA .
*
* This function allocates a page from the kernel page pool and applies
* a NUMA policy associated with the VMA or the current process .
* When VMA is not NULL caller must hold down_read on the mmap_sem of the
* mm_struct of the VMA to prevent it from going away . Should be used for
* all allocations for pages that will be mapped into
* user space . Returns NULL when no page can be allocated .
*
* Should be called with the mm_sem of the vma hold .
*/
struct page *
2005-10-07 10:46:04 +04:00
alloc_page_vma ( gfp_t gfp , struct vm_area_struct * vma , unsigned long addr )
2005-04-17 02:20:36 +04:00
{
2005-09-04 02:54:45 +04:00
struct mempolicy * pol = get_vma_policy ( current , vma , addr ) ;
2005-04-17 02:20:36 +04:00
2006-01-08 12:01:54 +03:00
cpuset_update_task_memory_state ( ) ;
2005-04-17 02:20:36 +04:00
if ( unlikely ( pol - > policy = = MPOL_INTERLEAVE ) ) {
unsigned nid ;
2006-01-06 11:10:46 +03:00
nid = interleave_nid ( pol , vma , addr , PAGE_SHIFT ) ;
2005-04-17 02:20:36 +04:00
return alloc_page_interleave ( gfp , 0 , nid ) ;
}
return __alloc_pages ( gfp , 0 , zonelist_policy ( gfp , pol ) ) ;
}
/**
* alloc_pages_current - Allocate pages .
*
* @ gfp :
* % GFP_USER user allocation ,
* % GFP_KERNEL kernel allocation ,
* % GFP_HIGHMEM highmem allocation ,
* % GFP_FS don ' t call back into a file system .
* % GFP_ATOMIC don ' t sleep .
* @ order : Power of two of allocation size in pages . 0 is a single page .
*
* Allocate a page from the kernel page pool . When not in
* interrupt context and apply the current process NUMA policy .
* Returns NULL when no page can be allocated .
*
2006-01-08 12:01:54 +03:00
* Don ' t call cpuset_update_task_memory_state ( ) unless
2005-04-17 02:20:36 +04:00
* 1 ) it ' s ok to take cpuset_sem ( can WAIT ) , and
* 2 ) allocating for current task ( not interrupt ) .
*/
2005-10-07 10:46:04 +04:00
struct page * alloc_pages_current ( gfp_t gfp , unsigned order )
2005-04-17 02:20:36 +04:00
{
struct mempolicy * pol = current - > mempolicy ;
if ( ( gfp & __GFP_WAIT ) & & ! in_interrupt ( ) )
2006-01-08 12:01:54 +03:00
cpuset_update_task_memory_state ( ) ;
2005-04-17 02:20:36 +04:00
if ( ! pol | | in_interrupt ( ) )
pol = & default_policy ;
if ( pol - > policy = = MPOL_INTERLEAVE )
return alloc_page_interleave ( gfp , order , interleave_nodes ( pol ) ) ;
return __alloc_pages ( gfp , order , zonelist_policy ( gfp , pol ) ) ;
}
EXPORT_SYMBOL ( alloc_pages_current ) ;
[PATCH] cpuset: rebind vma mempolicies fix
Fix more of longstanding bug in cpuset/mempolicy interaction.
NUMA mempolicies (mm/mempolicy.c) are constrained by the current tasks cpuset
to just the Memory Nodes allowed by that cpuset. The kernel maintains
internal state for each mempolicy, tracking what nodes are used for the
MPOL_INTERLEAVE, MPOL_BIND or MPOL_PREFERRED policies.
When a tasks cpuset memory placement changes, whether because the cpuset
changed, or because the task was attached to a different cpuset, then the
tasks mempolicies have to be rebound to the new cpuset placement, so as to
preserve the cpuset-relative numbering of the nodes in that policy.
An earlier fix handled such mempolicy rebinding for mempolicies attached to a
task.
This fix rebinds mempolicies attached to vma's (address ranges in a tasks
address space.) Due to the need to hold the task->mm->mmap_sem semaphore while
updating vma's, the rebinding of vma mempolicies has to be done when the
cpuset memory placement is changed, at which time mmap_sem can be safely
acquired. The tasks mempolicy is rebound later, when the task next attempts
to allocate memory and notices that its task->cpuset_mems_generation is
out-of-date with its cpusets mems_generation.
Because walking the tasklist to find all tasks attached to a changing cpuset
requires holding tasklist_lock, a spinlock, one cannot update the vma's of the
affected tasks while doing the tasklist scan. In general, one cannot acquire
a semaphore (which can sleep) while already holding a spinlock (such as
tasklist_lock). So a list of mm references has to be built up during the
tasklist scan, then the tasklist lock dropped, then for each mm, its mmap_sem
acquired, and the vma's in that mm rebound.
Once the tasklist lock is dropped, affected tasks may fork new tasks, before
their mm's are rebound. A kernel global 'cpuset_being_rebound' is set to
point to the cpuset being rebound (there can only be one; cpuset modifications
are done under a global 'manage_sem' semaphore), and the mpol_copy code that
is used to copy a tasks mempolicies during fork catches such forking tasks,
and ensures their children are also rebound.
When a task is moved to a different cpuset, it is easier, as there is only one
task involved. It's mm->vma's are scanned, using the same
mpol_rebind_policy() as used above.
It may happen that both the mpol_copy hook and the update done via the
tasklist scan update the same mm twice. This is ok, as the mempolicies of
each vma in an mm keep track of what mems_allowed they are relative to, and
safely no-op a second request to rebind to the same nodes.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:59 +03:00
/*
* If mpol_copy ( ) sees current - > cpuset = = cpuset_being_rebound , then it
* rebinds the mempolicy its copying by calling mpol_rebind_policy ( )
* with the mems_allowed returned by cpuset_mems_allowed ( ) . This
* keeps mempolicies cpuset relative after its cpuset moves . See
* further kernel / cpuset . c update_nodemask ( ) .
*/
void * cpuset_being_rebound ;
2005-04-17 02:20:36 +04:00
/* Slow path of a mempolicy copy */
struct mempolicy * __mpol_copy ( struct mempolicy * old )
{
struct mempolicy * new = kmem_cache_alloc ( policy_cache , GFP_KERNEL ) ;
if ( ! new )
return ERR_PTR ( - ENOMEM ) ;
[PATCH] cpuset: rebind vma mempolicies fix
Fix more of longstanding bug in cpuset/mempolicy interaction.
NUMA mempolicies (mm/mempolicy.c) are constrained by the current tasks cpuset
to just the Memory Nodes allowed by that cpuset. The kernel maintains
internal state for each mempolicy, tracking what nodes are used for the
MPOL_INTERLEAVE, MPOL_BIND or MPOL_PREFERRED policies.
When a tasks cpuset memory placement changes, whether because the cpuset
changed, or because the task was attached to a different cpuset, then the
tasks mempolicies have to be rebound to the new cpuset placement, so as to
preserve the cpuset-relative numbering of the nodes in that policy.
An earlier fix handled such mempolicy rebinding for mempolicies attached to a
task.
This fix rebinds mempolicies attached to vma's (address ranges in a tasks
address space.) Due to the need to hold the task->mm->mmap_sem semaphore while
updating vma's, the rebinding of vma mempolicies has to be done when the
cpuset memory placement is changed, at which time mmap_sem can be safely
acquired. The tasks mempolicy is rebound later, when the task next attempts
to allocate memory and notices that its task->cpuset_mems_generation is
out-of-date with its cpusets mems_generation.
Because walking the tasklist to find all tasks attached to a changing cpuset
requires holding tasklist_lock, a spinlock, one cannot update the vma's of the
affected tasks while doing the tasklist scan. In general, one cannot acquire
a semaphore (which can sleep) while already holding a spinlock (such as
tasklist_lock). So a list of mm references has to be built up during the
tasklist scan, then the tasklist lock dropped, then for each mm, its mmap_sem
acquired, and the vma's in that mm rebound.
Once the tasklist lock is dropped, affected tasks may fork new tasks, before
their mm's are rebound. A kernel global 'cpuset_being_rebound' is set to
point to the cpuset being rebound (there can only be one; cpuset modifications
are done under a global 'manage_sem' semaphore), and the mpol_copy code that
is used to copy a tasks mempolicies during fork catches such forking tasks,
and ensures their children are also rebound.
When a task is moved to a different cpuset, it is easier, as there is only one
task involved. It's mm->vma's are scanned, using the same
mpol_rebind_policy() as used above.
It may happen that both the mpol_copy hook and the update done via the
tasklist scan update the same mm twice. This is ok, as the mempolicies of
each vma in an mm keep track of what mems_allowed they are relative to, and
safely no-op a second request to rebind to the same nodes.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:59 +03:00
if ( current_cpuset_is_being_rebound ( ) ) {
nodemask_t mems = cpuset_mems_allowed ( current ) ;
mpol_rebind_policy ( old , & mems ) ;
}
2005-04-17 02:20:36 +04:00
* new = * old ;
atomic_set ( & new - > refcnt , 1 ) ;
if ( new - > policy = = MPOL_BIND ) {
int sz = ksize ( old - > v . zonelist ) ;
new - > v . zonelist = kmalloc ( sz , SLAB_KERNEL ) ;
if ( ! new - > v . zonelist ) {
kmem_cache_free ( policy_cache , new ) ;
return ERR_PTR ( - ENOMEM ) ;
}
memcpy ( new - > v . zonelist , old - > v . zonelist , sz ) ;
}
return new ;
}
/* Slow path of a mempolicy comparison */
int __mpol_equal ( struct mempolicy * a , struct mempolicy * b )
{
if ( ! a | | ! b )
return 0 ;
if ( a - > policy ! = b - > policy )
return 0 ;
switch ( a - > policy ) {
case MPOL_DEFAULT :
return 1 ;
case MPOL_INTERLEAVE :
2005-10-30 04:15:48 +03:00
return nodes_equal ( a - > v . nodes , b - > v . nodes ) ;
2005-04-17 02:20:36 +04:00
case MPOL_PREFERRED :
return a - > v . preferred_node = = b - > v . preferred_node ;
case MPOL_BIND : {
int i ;
for ( i = 0 ; a - > v . zonelist - > zones [ i ] ; i + + )
if ( a - > v . zonelist - > zones [ i ] ! = b - > v . zonelist - > zones [ i ] )
return 0 ;
return b - > v . zonelist - > zones [ i ] = = NULL ;
}
default :
BUG ( ) ;
return 0 ;
}
}
/* Slow path of a mpol destructor. */
void __mpol_free ( struct mempolicy * p )
{
if ( ! atomic_dec_and_test ( & p - > refcnt ) )
return ;
if ( p - > policy = = MPOL_BIND )
kfree ( p - > v . zonelist ) ;
p - > policy = MPOL_DEFAULT ;
kmem_cache_free ( policy_cache , p ) ;
}
/*
* Shared memory backing store policy support .
*
* Remember policies even when nobody has shared memory mapped .
* The policies are kept in Red - Black tree linked from the inode .
* They are protected by the sp - > lock spinlock , which should be held
* for any accesses to the tree .
*/
/* lookup first element intersecting start-end */
/* Caller holds sp->lock */
static struct sp_node *
sp_lookup ( struct shared_policy * sp , unsigned long start , unsigned long end )
{
struct rb_node * n = sp - > root . rb_node ;
while ( n ) {
struct sp_node * p = rb_entry ( n , struct sp_node , nd ) ;
if ( start > = p - > end )
n = n - > rb_right ;
else if ( end < = p - > start )
n = n - > rb_left ;
else
break ;
}
if ( ! n )
return NULL ;
for ( ; ; ) {
struct sp_node * w = NULL ;
struct rb_node * prev = rb_prev ( n ) ;
if ( ! prev )
break ;
w = rb_entry ( prev , struct sp_node , nd ) ;
if ( w - > end < = start )
break ;
n = prev ;
}
return rb_entry ( n , struct sp_node , nd ) ;
}
/* Insert a new shared policy into the list. */
/* Caller holds sp->lock */
static void sp_insert ( struct shared_policy * sp , struct sp_node * new )
{
struct rb_node * * p = & sp - > root . rb_node ;
struct rb_node * parent = NULL ;
struct sp_node * nd ;
while ( * p ) {
parent = * p ;
nd = rb_entry ( parent , struct sp_node , nd ) ;
if ( new - > start < nd - > start )
p = & ( * p ) - > rb_left ;
else if ( new - > end > nd - > end )
p = & ( * p ) - > rb_right ;
else
BUG ( ) ;
}
rb_link_node ( & new - > nd , parent , p ) ;
rb_insert_color ( & new - > nd , & sp - > root ) ;
PDprintk ( " inserting %lx-%lx: %d \n " , new - > start , new - > end ,
new - > policy ? new - > policy - > policy : 0 ) ;
}
/* Find shared policy intersecting idx */
struct mempolicy *
mpol_shared_policy_lookup ( struct shared_policy * sp , unsigned long idx )
{
struct mempolicy * pol = NULL ;
struct sp_node * sn ;
if ( ! sp - > root . rb_node )
return NULL ;
spin_lock ( & sp - > lock ) ;
sn = sp_lookup ( sp , idx , idx + 1 ) ;
if ( sn ) {
mpol_get ( sn - > policy ) ;
pol = sn - > policy ;
}
spin_unlock ( & sp - > lock ) ;
return pol ;
}
static void sp_delete ( struct shared_policy * sp , struct sp_node * n )
{
PDprintk ( " deleting %lx-l%x \n " , n - > start , n - > end ) ;
rb_erase ( & n - > nd , & sp - > root ) ;
mpol_free ( n - > policy ) ;
kmem_cache_free ( sn_cache , n ) ;
}
struct sp_node *
sp_alloc ( unsigned long start , unsigned long end , struct mempolicy * pol )
{
struct sp_node * n = kmem_cache_alloc ( sn_cache , GFP_KERNEL ) ;
if ( ! n )
return NULL ;
n - > start = start ;
n - > end = end ;
mpol_get ( pol ) ;
n - > policy = pol ;
return n ;
}
/* Replace a policy range. */
static int shared_policy_replace ( struct shared_policy * sp , unsigned long start ,
unsigned long end , struct sp_node * new )
{
struct sp_node * n , * new2 = NULL ;
restart :
spin_lock ( & sp - > lock ) ;
n = sp_lookup ( sp , start , end ) ;
/* Take care of old policies in the same range. */
while ( n & & n - > start < end ) {
struct rb_node * next = rb_next ( & n - > nd ) ;
if ( n - > start > = start ) {
if ( n - > end < = end )
sp_delete ( sp , n ) ;
else
n - > start = end ;
} else {
/* Old policy spanning whole new range. */
if ( n - > end > end ) {
if ( ! new2 ) {
spin_unlock ( & sp - > lock ) ;
new2 = sp_alloc ( end , n - > end , n - > policy ) ;
if ( ! new2 )
return - ENOMEM ;
goto restart ;
}
n - > end = start ;
sp_insert ( sp , new2 ) ;
new2 = NULL ;
break ;
} else
n - > end = start ;
}
if ( ! next )
break ;
n = rb_entry ( next , struct sp_node , nd ) ;
}
if ( new )
sp_insert ( sp , new ) ;
spin_unlock ( & sp - > lock ) ;
if ( new2 ) {
mpol_free ( new2 - > policy ) ;
kmem_cache_free ( sn_cache , new2 ) ;
}
return 0 ;
}
2006-01-15 00:20:48 +03:00
void mpol_shared_policy_init ( struct shared_policy * info , int policy ,
nodemask_t * policy_nodes )
{
info - > root = RB_ROOT ;
spin_lock_init ( & info - > lock ) ;
if ( policy ! = MPOL_DEFAULT ) {
struct mempolicy * newpol ;
/* Falls back to MPOL_DEFAULT on any error */
newpol = mpol_new ( policy , policy_nodes ) ;
if ( ! IS_ERR ( newpol ) ) {
/* Create pseudo-vma that contains just the policy */
struct vm_area_struct pvma ;
memset ( & pvma , 0 , sizeof ( struct vm_area_struct ) ) ;
/* Policy covers entire file */
pvma . vm_end = TASK_SIZE ;
mpol_set_shared_policy ( info , & pvma , newpol ) ;
mpol_free ( newpol ) ;
}
}
}
2005-04-17 02:20:36 +04:00
int mpol_set_shared_policy ( struct shared_policy * info ,
struct vm_area_struct * vma , struct mempolicy * npol )
{
int err ;
struct sp_node * new = NULL ;
unsigned long sz = vma_pages ( vma ) ;
PDprintk ( " set_shared_policy %lx sz %lu %d %lx \n " ,
vma - > vm_pgoff ,
sz , npol ? npol - > policy : - 1 ,
2005-10-30 04:15:48 +03:00
npol ? nodes_addr ( npol - > v . nodes ) [ 0 ] : - 1 ) ;
2005-04-17 02:20:36 +04:00
if ( npol ) {
new = sp_alloc ( vma - > vm_pgoff , vma - > vm_pgoff + sz , npol ) ;
if ( ! new )
return - ENOMEM ;
}
err = shared_policy_replace ( info , vma - > vm_pgoff , vma - > vm_pgoff + sz , new ) ;
if ( err & & new )
kmem_cache_free ( sn_cache , new ) ;
return err ;
}
/* Free a backing policy store on inode delete. */
void mpol_free_shared_policy ( struct shared_policy * p )
{
struct sp_node * n ;
struct rb_node * next ;
if ( ! p - > root . rb_node )
return ;
spin_lock ( & p - > lock ) ;
next = rb_first ( & p - > root ) ;
while ( next ) {
n = rb_entry ( next , struct sp_node , nd ) ;
next = rb_next ( & n - > nd ) ;
2005-07-27 22:43:50 +04:00
rb_erase ( & n - > nd , & p - > root ) ;
2005-04-17 02:20:36 +04:00
mpol_free ( n - > policy ) ;
kmem_cache_free ( sn_cache , n ) ;
}
spin_unlock ( & p - > lock ) ;
}
/* assumes fs == KERNEL_DS */
void __init numa_policy_init ( void )
{
policy_cache = kmem_cache_create ( " numa_policy " ,
sizeof ( struct mempolicy ) ,
0 , SLAB_PANIC , NULL , NULL ) ;
sn_cache = kmem_cache_create ( " shared_policy_node " ,
sizeof ( struct sp_node ) ,
0 , SLAB_PANIC , NULL , NULL ) ;
/* Set interleaving policy for system init. This way not all
the data structures allocated at system boot end up in node zero . */
2005-10-30 04:16:59 +03:00
if ( do_set_mempolicy ( MPOL_INTERLEAVE , & node_online_map ) )
2005-04-17 02:20:36 +04:00
printk ( " numa_policy_init: interleaving failed \n " ) ;
}
2005-10-30 04:16:59 +03:00
/* Reset policy of current process to default */
2005-04-17 02:20:36 +04:00
void numa_default_policy ( void )
{
2005-10-30 04:16:59 +03:00
do_set_mempolicy ( MPOL_DEFAULT , NULL ) ;
2005-04-17 02:20:36 +04:00
}
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
/* Migrate a policy to a different set of nodes */
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
void mpol_rebind_policy ( struct mempolicy * pol , const nodemask_t * newmask )
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
{
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
nodemask_t * mpolmask ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
nodemask_t tmp ;
if ( ! pol )
return ;
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
mpolmask = & pol - > cpuset_mems_allowed ;
if ( nodes_equal ( * mpolmask , * newmask ) )
return ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
switch ( pol - > policy ) {
case MPOL_DEFAULT :
break ;
case MPOL_INTERLEAVE :
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
nodes_remap ( tmp , pol - > v . nodes , * mpolmask , * newmask ) ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
pol - > v . nodes = tmp ;
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
* mpolmask = * newmask ;
current - > il_next = node_remap ( current - > il_next ,
* mpolmask , * newmask ) ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
break ;
case MPOL_PREFERRED :
pol - > v . preferred_node = node_remap ( pol - > v . preferred_node ,
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
* mpolmask , * newmask ) ;
* mpolmask = * newmask ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
break ;
case MPOL_BIND : {
nodemask_t nodes ;
struct zone * * z ;
struct zonelist * zonelist ;
nodes_clear ( nodes ) ;
for ( z = pol - > v . zonelist - > zones ; * z ; z + + )
node_set ( ( * z ) - > zone_pgdat - > node_id , nodes ) ;
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
nodes_remap ( tmp , nodes , * mpolmask , * newmask ) ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
nodes = tmp ;
zonelist = bind_zonelist ( & nodes ) ;
/* If no mem, then zonelist is NULL and we keep old zonelist.
* If that old zonelist has no remaining mems_allowed nodes ,
* then zonelist_policy ( ) will " FALL THROUGH " to MPOL_DEFAULT .
*/
if ( zonelist ) {
/* Good - got mem - substitute new zonelist */
kfree ( pol - > v . zonelist ) ;
pol - > v . zonelist = zonelist ;
}
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
* mpolmask = * newmask ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
break ;
}
default :
BUG ( ) ;
break ;
}
}
/*
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
* Wrapper for mpol_rebind_policy ( ) that just requires task
* pointer , and updates task mempolicy .
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
*/
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
void mpol_rebind_task ( struct task_struct * tsk , const nodemask_t * new )
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
{
[PATCH] cpuset: numa_policy_rebind cleanup
Cleanup, reorganize and make more robust the mempolicy.c code to rebind
mempolicies relative to the containing cpuset after a tasks memory placement
changes.
The real motivator for this cleanup patch is to lay more groundwork for the
upcoming patch to correctly rebind NUMA mempolicies that are attached to vma's
after the containing cpuset memory placement changes.
NUMA mempolicies are constrained by the cpuset their task is a member of.
When either (1) a task is moved to a different cpuset, or (2) the 'mems'
mems_allowed of a cpuset is changed, then the NUMA mempolicies have embedded
node numbers (for MPOL_BIND, MPOL_INTERLEAVE and MPOL_PREFERRED) that need to
be recalculated, relative to their new cpuset placement.
The old code used an unreliable method of determining what was the old
mems_allowed constraining the mempolicy. It just looked at the tasks
mems_allowed value. This sort of worked with the present code, that just
rebinds the -task- mempolicy, and leaves any -vma- mempolicies broken,
referring to the old nodes. But in an upcoming patch, the vma mempolicies
will be rebound as well. Then the order in which the various task and vma
mempolicies are updated will no longer be deterministic, and one can no longer
count on the task->mems_allowed holding the old value for as long as needed.
It's not even clear if the current code was guaranteed to work reliably for
task mempolicies.
So I added a mems_allowed field to each mempolicy, stating exactly what
mems_allowed the policy is relative to, and updated synchronously and reliably
anytime that the mempolicy is rebound.
Also removed a useless wrapper routine, numa_policy_rebind(), and had its
caller, cpuset_update_task_memory_state(), call directly to the rewritten
policy_rebind() routine, and made that rebind routine extern instead of
static, and added a "mpol_" prefix to its name, making it
mpol_rebind_policy().
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:56 +03:00
mpol_rebind_policy ( tsk - > mempolicy , new ) ;
[PATCH] cpusets: automatic numa mempolicy rebinding
This patch automatically updates a tasks NUMA mempolicy when its cpuset
memory placement changes. It does so within the context of the task,
without any need to support low level external mempolicy manipulation.
If a system is not using cpusets, or if running on a system with just the
root (all-encompassing) cpuset, then this remap is a no-op. Only when a
task is moved between cpusets, or a cpusets memory placement is changed
does the following apply. Otherwise, the main routine below,
rebind_policy() is not even called.
When mixing cpusets, scheduler affinity, and NUMA mempolicies, the
essential role of cpusets is to place jobs (several related tasks) on a set
of CPUs and Memory Nodes, the essential role of sched_setaffinity is to
manage a jobs processor placement within its allowed cpuset, and the
essential role of NUMA mempolicy (mbind, set_mempolicy) is to manage a jobs
memory placement within its allowed cpuset.
However, CPU affinity and NUMA memory placement are managed within the
kernel using absolute system wide numbering, not cpuset relative numbering.
This is ok until a job is migrated to a different cpuset, or what's the
same, a jobs cpuset is moved to different CPUs and Memory Nodes.
Then the CPU affinity and NUMA memory placement of the tasks in the job
need to be updated, to preserve their cpuset-relative position. This can
be done for CPU affinity using sched_setaffinity() from user code, as one
task can modify anothers CPU affinity. This cannot be done from an
external task for NUMA memory placement, as that can only be modified in
the context of the task using it.
However, it easy enough to remap a tasks NUMA mempolicy automatically when
a task is migrated, using the existing cpuset mechanism to trigger a
refresh of a tasks memory placement after its cpuset has changed. All that
is needed is the old and new nodemask, and notice to the task that it needs
to rebind its mempolicy. The tasks mems_allowed has the old mask, the
tasks cpuset has the new mask, and the existing
cpuset_update_current_mems_allowed() mechanism provides the notice. The
bitmap/cpumask/nodemask remap operators provide the cpuset relative
calculations.
This patch leaves open a couple of issues:
1) Updating vma and shmfs/tmpfs/hugetlbfs memory policies:
These mempolicies may reference nodes outside of those allowed to
the current task by its cpuset. Tasks are migrated as part of jobs,
which reside on what might be several cpusets in a subtree. When such
a job is migrated, all NUMA memory policy references to nodes within
that cpuset subtree should be translated, and references to any nodes
outside that subtree should be left untouched. A future patch will
provide the cpuset mechanism needed to mark such subtrees. With that
patch, we will be able to correctly migrate these other memory policies
across a job migration.
2) Updating cpuset, affinity and memory policies in user space:
This is harder. Any placement state stored in user space using
system-wide numbering will be invalidated across a migration. More
work will be required to provide user code with a migration-safe means
to manage its cpuset relative placement, while preserving the current
API's that pass system wide numbers, not cpuset relative numbers across
the kernel-user boundary.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-31 02:02:36 +03:00
}
2006-01-08 12:01:02 +03:00
[PATCH] cpuset: rebind vma mempolicies fix
Fix more of longstanding bug in cpuset/mempolicy interaction.
NUMA mempolicies (mm/mempolicy.c) are constrained by the current tasks cpuset
to just the Memory Nodes allowed by that cpuset. The kernel maintains
internal state for each mempolicy, tracking what nodes are used for the
MPOL_INTERLEAVE, MPOL_BIND or MPOL_PREFERRED policies.
When a tasks cpuset memory placement changes, whether because the cpuset
changed, or because the task was attached to a different cpuset, then the
tasks mempolicies have to be rebound to the new cpuset placement, so as to
preserve the cpuset-relative numbering of the nodes in that policy.
An earlier fix handled such mempolicy rebinding for mempolicies attached to a
task.
This fix rebinds mempolicies attached to vma's (address ranges in a tasks
address space.) Due to the need to hold the task->mm->mmap_sem semaphore while
updating vma's, the rebinding of vma mempolicies has to be done when the
cpuset memory placement is changed, at which time mmap_sem can be safely
acquired. The tasks mempolicy is rebound later, when the task next attempts
to allocate memory and notices that its task->cpuset_mems_generation is
out-of-date with its cpusets mems_generation.
Because walking the tasklist to find all tasks attached to a changing cpuset
requires holding tasklist_lock, a spinlock, one cannot update the vma's of the
affected tasks while doing the tasklist scan. In general, one cannot acquire
a semaphore (which can sleep) while already holding a spinlock (such as
tasklist_lock). So a list of mm references has to be built up during the
tasklist scan, then the tasklist lock dropped, then for each mm, its mmap_sem
acquired, and the vma's in that mm rebound.
Once the tasklist lock is dropped, affected tasks may fork new tasks, before
their mm's are rebound. A kernel global 'cpuset_being_rebound' is set to
point to the cpuset being rebound (there can only be one; cpuset modifications
are done under a global 'manage_sem' semaphore), and the mpol_copy code that
is used to copy a tasks mempolicies during fork catches such forking tasks,
and ensures their children are also rebound.
When a task is moved to a different cpuset, it is easier, as there is only one
task involved. It's mm->vma's are scanned, using the same
mpol_rebind_policy() as used above.
It may happen that both the mpol_copy hook and the update done via the
tasklist scan update the same mm twice. This is ok, as the mempolicies of
each vma in an mm keep track of what mems_allowed they are relative to, and
safely no-op a second request to rebind to the same nodes.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 12:01:59 +03:00
/*
* Rebind each vma in mm to new nodemask .
*
* Call holding a reference to mm . Takes mm - > mmap_sem during call .
*/
void mpol_rebind_mm ( struct mm_struct * mm , nodemask_t * new )
{
struct vm_area_struct * vma ;
down_write ( & mm - > mmap_sem ) ;
for ( vma = mm - > mmap ; vma ; vma = vma - > vm_next )
mpol_rebind_policy ( vma - > vm_policy , new ) ;
up_write ( & mm - > mmap_sem ) ;
}
2006-01-08 12:01:02 +03:00
/*
* Display pages allocated per node and memory policy via / proc .
*/
static const char * policy_types [ ] = { " default " , " prefer " , " bind " ,
" interleave " } ;
/*
* Convert a mempolicy into a string .
* Returns the number of characters in buffer ( if positive )
* or an error ( negative )
*/
static inline int mpol_to_str ( char * buffer , int maxlen , struct mempolicy * pol )
{
char * p = buffer ;
int l ;
nodemask_t nodes ;
int mode = pol ? pol - > policy : MPOL_DEFAULT ;
switch ( mode ) {
case MPOL_DEFAULT :
nodes_clear ( nodes ) ;
break ;
case MPOL_PREFERRED :
nodes_clear ( nodes ) ;
node_set ( pol - > v . preferred_node , nodes ) ;
break ;
case MPOL_BIND :
get_zonemask ( pol , & nodes ) ;
break ;
case MPOL_INTERLEAVE :
nodes = pol - > v . nodes ;
break ;
default :
BUG ( ) ;
return - EFAULT ;
}
l = strlen ( policy_types [ mode ] ) ;
if ( buffer + maxlen < p + l + 1 )
return - ENOSPC ;
strcpy ( p , policy_types [ mode ] ) ;
p + = l ;
if ( ! nodes_empty ( nodes ) ) {
if ( buffer + maxlen < p + 2 )
return - ENOSPC ;
* p + + = ' = ' ;
p + = nodelist_scnprintf ( p , buffer + maxlen - p , nodes ) ;
}
return p - buffer ;
}
struct numa_maps {
unsigned long pages ;
unsigned long anon ;
unsigned long mapped ;
unsigned long mapcount_max ;
unsigned long node [ MAX_NUMNODES ] ;
} ;
static void gather_stats ( struct page * page , void * private )
{
struct numa_maps * md = private ;
int count = page_mapcount ( page ) ;
if ( count )
md - > mapped + + ;
if ( count > md - > mapcount_max )
md - > mapcount_max = count ;
md - > pages + + ;
if ( PageAnon ( page ) )
md - > anon + + ;
md - > node [ page_to_nid ( page ) ] + + ;
cond_resched ( ) ;
}
int show_numa_map ( struct seq_file * m , void * v )
{
struct task_struct * task = m - > private ;
struct vm_area_struct * vma = v ;
struct numa_maps * md ;
int n ;
char buffer [ 50 ] ;
if ( ! vma - > vm_mm )
return 0 ;
md = kzalloc ( sizeof ( struct numa_maps ) , GFP_KERNEL ) ;
if ( ! md )
return 0 ;
check_pgd_range ( vma , vma - > vm_start , vma - > vm_end ,
& node_online_map , MPOL_MF_STATS , md ) ;
if ( md - > pages ) {
mpol_to_str ( buffer , sizeof ( buffer ) ,
get_vma_policy ( task , vma , vma - > vm_start ) ) ;
seq_printf ( m , " %08lx %s pages=%lu mapped=%lu maxref=%lu " ,
vma - > vm_start , buffer , md - > pages ,
md - > mapped , md - > mapcount_max ) ;
if ( md - > anon )
seq_printf ( m , " anon=%lu " , md - > anon ) ;
for_each_online_node ( n )
if ( md - > node [ n ] )
seq_printf ( m , " N%d=%lu " , n , md - > node [ n ] ) ;
seq_putc ( m , ' \n ' ) ;
}
kfree ( md ) ;
if ( m - > count < m - > size )
m - > version = ( vma ! = get_gate_vma ( task ) ) ? vma - > vm_start : 0 ;
return 0 ;
}