// SPDX-License-Identifier: GPL-2.0-only /* * Simple NUMA memory policy for the Linux kernel. * * Copyright 2003,2004 Andi Kleen, SuSE Labs. * (C) Copyright 2005 Christoph Lameter, Silicon Graphics, Inc. * * NUMA policy allows the user to give hints in which node(s) memory should * be allocated. * * Support six 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. * * weighted interleave * Allocate memory interleaved over a set of nodes based on * a set of weights (per-node), with normal fallback if it * fails. Otherwise operates the same as interleave. * Example: nodeset(0,1) & weights (2,1) - 2 pages allocated * on node 0 for every 1 page allocated on node 1. * * bind Only allocate memory on a specific set of nodes, * no fallback. * 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 * * preferred Try a specific node first before normal fallback. * As a special case NUMA_NO_NODE 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. * * preferred many Try a set of nodes first before normal fallback. This is * similar to preferred without the special case. * * 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 shmem/tmpfs 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.
*/
/* * weightiness balances the tradeoff between small weights (cycles through nodes * faster, more fair/even distribution) and large weights (smaller errors * between actual bandwidth ratios and weight ratios). 32 is a number that has * been found to perform at a reasonable compromise between the two goals.
*/ staticconstint weightiness = 32;
/* * A null weighted_interleave_state is interpreted as having .mode="auto", * and .iw_table is interpreted as an array of 1s with length nr_node_ids.
*/ struct weighted_interleave_state { bool mode_auto;
u8 iw_table[];
}; staticstruct weighted_interleave_state __rcu *wi_state; staticunsignedint *node_bw_table;
/* * wi_state_lock protects both wi_state and node_bw_table. * node_bw_table is only used by writers to update wi_state.
*/ static DEFINE_MUTEX(wi_state_lock);
new_wi_state = kmalloc(struct_size(new_wi_state, iw_table, nr_node_ids),
GFP_KERNEL); if (!new_wi_state) {
kfree(new_bw); return -ENOMEM;
}
new_wi_state->mode_auto = true; for (i = 0; i < nr_node_ids; i++)
new_wi_state->iw_table[i] = 1;
/* * Update bandwidth info, even in manual mode. That way, when switching * to auto mode in the future, iw_table can be overwritten using * accurate bw data.
*/
mutex_lock(&wi_state_lock);
/** * numa_nearest_node - Find nearest node by state * @node: Node id to start the search * @state: State to filter the search * * Lookup the closest node by distance if @nid is not in state. * * Return: this @node if it is in state, otherwise the closest node by distance
*/ int numa_nearest_node(int node, unsignedint state)
{ int min_dist = INT_MAX, dist, n, min_node;
if (state >= NR_NODE_STATES) return -EINVAL;
if (node == NUMA_NO_NODE || node_state(node, state)) return node;
/** * nearest_node_nodemask - Find the node in @mask at the nearest distance * from @node. * * @node: a valid node ID to start the search from. * @mask: a pointer to a nodemask representing the allowed nodes. * * This function iterates over all nodes in @mask and calculates the * distance from the starting @node, then it returns the node ID that is * the closest to @node, or MAX_NUMNODES if no node is found. * * Note that @node must be a valid node ID usable with node_distance(), * providing an invalid node ID (e.g., NUMA_NO_NODE) may result in crashes * or unexpected behavior.
*/ int nearest_node_nodemask(int node, nodemask_t *mask)
{ int dist, n, min_dist = INT_MAX, min_node = MAX_NUMNODES;
node = numa_node_id(); if (node != NUMA_NO_NODE) {
pol = &preferred_node_policy[node]; /* preferred_node_policy is not initialised early in boot */ if (pol->mode) return pol;
}
/* * mpol_set_nodemask is called after mpol_new() to set up the nodemask, if * any, for the new policy. mpol_new() has already validated the nodes * parameter with respect to the policy mode and flags. * * Must be called holding task's alloc_lock to protect task's mems_allowed * and mempolicy. May also be called holding the mmap_lock for write.
*/ staticint mpol_set_nodemask(struct mempolicy *pol, const nodemask_t *nodes, struct nodemask_scratch *nsc)
{ int ret;
/* * Default (pol==NULL) resp. local memory policies are not a * subject of any remapping. They also do not need any special * constructor.
*/ if (!pol || pol->mode == MPOL_LOCAL) return 0;
if (mpol_store_user_nodemask(pol))
pol->w.user_nodemask = *nodes; else
pol->w.cpuset_mems_allowed = cpuset_current_mems_allowed;
ret = mpol_ops[pol->mode].create(pol, &nsc->mask2); return ret;
}
/* * This function just creates a new policy, does some check and simple * initialization. You must invoke mpol_set_nodemask() to set nodes.
*/ staticstruct mempolicy *mpol_new(unsignedshort mode, unsignedshort flags,
nodemask_t *nodes)
{ struct mempolicy *policy;
if (mode == MPOL_DEFAULT) { if (nodes && !nodes_empty(*nodes)) return ERR_PTR(-EINVAL); return NULL;
}
VM_BUG_ON(!nodes);
/* * MPOL_PREFERRED cannot be used with MPOL_F_STATIC_NODES or * MPOL_F_RELATIVE_NODES if the nodemask is empty (local allocation). * All other modes require a valid pointer to a non-empty nodemask.
*/ if (mode == MPOL_PREFERRED) { if (nodes_empty(*nodes)) { if (((flags & MPOL_F_STATIC_NODES) ||
(flags & MPOL_F_RELATIVE_NODES))) return ERR_PTR(-EINVAL);
/* * mpol_rebind_policy - Migrate a policy to a different set of nodes * * Per-vma policies are protected by mmap_lock. Allocations using per-task * policies are protected by task->mems_allowed_seq to prevent a premature * OOM/allocation failure due to parallel nodemask modification.
*/ staticvoid mpol_rebind_policy(struct mempolicy *pol, const nodemask_t *newmask)
{ if (!pol || pol->mode == MPOL_LOCAL) return; if (!mpol_store_user_nodemask(pol) &&
nodes_equal(pol->w.cpuset_mems_allowed, *newmask)) return;
mpol_ops[pol->mode].rebind(pol, newmask);
}
/* * Wrapper for mpol_rebind_policy() that just requires task * pointer, and updates task mempolicy. * * Called with task's alloc_lock held.
*/ void mpol_rebind_task(struct task_struct *tsk, const nodemask_t *new)
{
mpol_rebind_policy(tsk->mempolicy, new);
}
/* * Rebind each vma in mm to new nodemask. * * Call holding a reference to mm. Takes mm->mmap_lock during call.
*/ void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new)
{ struct vm_area_struct *vma;
VMA_ITERATOR(vmi, mm, 0);
struct queue_pages { struct list_head *pagelist; unsignedlong flags;
nodemask_t *nmask; unsignedlong start; unsignedlong end; struct vm_area_struct *first; struct folio *large; /* note last large folio encountered */ long nr_failed; /* could not be isolated at this time */
};
/* * Check if the folio's nid is in qp->nmask. * * If MPOL_MF_INVERT is set in qp->flags, check if the nid is * in the invert of qp->nmask.
*/ staticinlinebool queue_folio_required(struct folio *folio, struct queue_pages *qp)
{ int nid = folio_nid(folio); unsignedlong flags = qp->flags;
if (unlikely(is_pmd_migration_entry(*pmd))) {
qp->nr_failed++; return;
}
folio = pmd_folio(*pmd); if (is_huge_zero_folio(folio)) {
walk->action = ACTION_CONTINUE; return;
} if (!queue_folio_required(folio, qp)) return; if (!(qp->flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) ||
!vma_migratable(walk->vma) ||
!migrate_folio_add(folio, qp->pagelist, qp->flags))
qp->nr_failed++;
}
/* * Scan through folios, checking if they satisfy the required conditions, * moving them from LRU to local pagelist for migration if they do (or not). * * queue_folios_pte_range() has two possible return values: * 0 - continue walking to scan for more, even if an existing folio on the * wrong node could not be isolated and queued for migration. * -EIO - only MPOL_MF_STRICT was specified, without MPOL_MF_MOVE or ..._ALL, * and an existing folio was on a node that does not follow the policy.
*/ staticint queue_folios_pte_range(pmd_t *pmd, unsignedlong addr, unsignedlong end, struct mm_walk *walk)
{ struct vm_area_struct *vma = walk->vma; struct folio *folio; struct queue_pages *qp = walk->private; unsignedlong flags = qp->flags;
pte_t *pte, *mapped_pte;
pte_t ptent;
spinlock_t *ptl; int max_nr, nr;
mapped_pte = pte = pte_offset_map_lock(walk->mm, pmd, addr, &ptl); if (!pte) {
walk->action = ACTION_AGAIN; return 0;
} for (; addr != end; pte += nr, addr += nr * PAGE_SIZE) {
max_nr = (end - addr) >> PAGE_SHIFT;
nr = 1;
ptent = ptep_get(pte); if (pte_none(ptent)) continue; if (!pte_present(ptent)) { if (is_migration_entry(pte_to_swp_entry(ptent)))
qp->nr_failed++; continue;
}
folio = vm_normal_folio(vma, addr, ptent); if (!folio || folio_is_zone_device(folio)) continue; if (folio_test_large(folio) && max_nr != 1)
nr = folio_pte_batch(folio, pte, ptent, max_nr); /* * vm_normal_folio() filters out zero pages, but there might * still be reserved folios to skip, perhaps in a VDSO.
*/ if (folio_test_reserved(folio)) continue; if (!queue_folio_required(folio, qp)) continue; if (folio_test_large(folio)) { /* * A large folio can only be isolated from LRU once, * but may be mapped by many PTEs (and Copy-On-Write may * intersperse PTEs of other, order 0, folios). This is * a common case, so don't mistake it for failure (but * there can be other cases of multi-mapped pages which * this quick check does not help to filter out - and a * search of the pagelist might grow to be prohibitive). * * migrate_pages(&pagelist) returns nr_failed folios, so * check "large" now so that queue_pages_range() returns * a comparable nr_failed folios. This does imply that * if folio could not be isolated for some racy reason * at its first PTE, later PTEs will not give it another * chance of isolation; but keeps the accounting simple.
*/ if (folio == qp->large) continue;
qp->large = folio;
} if (!(flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) ||
!vma_migratable(vma) ||
!migrate_folio_add(folio, qp->pagelist, flags)) {
qp->nr_failed += nr; if (strictly_unmovable(flags)) break;
}
}
pte_unmap_unlock(mapped_pte, ptl);
cond_resched();
out: if (qp->nr_failed && strictly_unmovable(flags)) return -EIO; return 0;
}
ptl = huge_pte_lock(hstate_vma(walk->vma), walk->mm, pte);
entry = huge_ptep_get(walk->mm, addr, pte); if (!pte_present(entry)) { if (unlikely(is_hugetlb_entry_migration(entry)))
qp->nr_failed++; goto unlock;
}
folio = pfn_folio(pte_pfn(entry)); if (!queue_folio_required(folio, qp)) goto unlock; if (!(flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) ||
!vma_migratable(walk->vma)) {
qp->nr_failed++; goto unlock;
} /* * Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio. * Choosing not to migrate a shared folio is not counted as a failure. * * See folio_maybe_mapped_shared() on possible imprecision when we * cannot easily detect if a folio is shared.
*/ if ((flags & MPOL_MF_MOVE_ALL) ||
(!folio_maybe_mapped_shared(folio) && !hugetlb_pmd_shared(pte))) if (!folio_isolate_hugetlb(folio, qp->pagelist))
qp->nr_failed++;
unlock:
spin_unlock(ptl); if (qp->nr_failed && strictly_unmovable(flags)) return -EIO; #endif return 0;
}
#ifdef CONFIG_NUMA_BALANCING /* * This is used to mark a range of virtual addresses to be inaccessible. * These are later cleared by a NUMA hinting fault. Depending on these * faults, pages may be migrated for better NUMA placement. * * This is assuming that NUMA faults are handled using PROT_NONE. If * an architecture makes a different choice, it will need further * changes to the core.
*/ unsignedlong change_prot_numa(struct vm_area_struct *vma, unsignedlong addr, unsignedlong end)
{ struct mmu_gather tlb; long nr_updated;
/* range check first */
VM_BUG_ON_VMA(!range_in_vma(vma, start, end), vma);
if (!qp->first) {
qp->first = vma; if (!(flags & MPOL_MF_DISCONTIG_OK) &&
(qp->start < vma->vm_start)) /* hole at head side of range */ return -EFAULT;
}
next = find_vma(vma->vm_mm, vma->vm_end); if (!(flags & MPOL_MF_DISCONTIG_OK) &&
((vma->vm_end < qp->end) &&
(!next || vma->vm_end < next->vm_start))) /* hole at middle or tail of range */ return -EFAULT;
/* * Need check MPOL_MF_STRICT to return -EIO if possible * regardless of vma_migratable
*/ if (!vma_migratable(vma) &&
!(flags & MPOL_MF_STRICT)) return 1;
/* * Check page nodes, and queue pages to move, in the current vma. * But if no moving, and no strict checking, the scan can be skipped.
*/ if (flags & (MPOL_MF_STRICT | MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) return 0; return 1;
}
/* * Walk through page tables and collect pages to be migrated. * * If pages found in a given range are not on the required set of @nodes, * and migration is allowed, they are isolated and queued to @pagelist. * * queue_pages_range() may return: * 0 - all pages already on the right node, or successfully queued for moving * (or neither strict checking nor moving requested: only range checking). * >0 - this number of misplaced folios could not be queued for moving * (a hugetlbfs page or a transparent huge page being counted as 1). * -EIO - a misplaced page found, when MPOL_MF_STRICT specified without MOVEs. * -EFAULT - a hole in the memory range, when MPOL_MF_DISCONTIG_OK unspecified.
*/ staticlong
queue_pages_range(struct mm_struct *mm, unsignedlong start, unsignedlong end,
nodemask_t *nodes, unsignedlong flags, struct list_head *pagelist)
{ int err; struct queue_pages qp = {
.pagelist = pagelist,
.flags = flags,
.nmask = nodes,
.start = start,
.end = end,
.first = NULL,
}; conststruct mm_walk_ops *ops = (flags & MPOL_MF_WRLOCK) ?
&queue_pages_lock_vma_walk_ops : &queue_pages_walk_ops;
err = walk_page_range(mm, start, end, ops, &qp);
if (!qp.first) /* whole range in hole */
err = -EFAULT;
return err ? : qp.nr_failed;
}
/* * Apply policy to a single VMA * This must be called with the mmap_lock held for writing.
*/ staticint vma_replace_policy(struct vm_area_struct *vma, struct mempolicy *pol)
{ int err; struct mempolicy *old; struct mempolicy *new;
vma_assert_write_locked(vma);
new = mpol_dup(pol); if (IS_ERR(new)) return PTR_ERR(new);
if (vma->vm_ops && vma->vm_ops->set_policy) {
err = vma->vm_ops->set_policy(vma, new); if (err) goto err_out;
}
old = vma->vm_policy;
vma->vm_policy = new; /* protected by mmap_lock */
mpol_put(old);
return 0;
err_out:
mpol_put(new); return err;
}
/* Split or merge the VMA (if required) and apply the new policy */ staticint mbind_range(struct vma_iterator *vmi, struct vm_area_struct *vma, struct vm_area_struct **prev, unsignedlong start, unsignedlong end, struct mempolicy *new_pol)
{ unsignedlong vmstart, vmend;
/* Set the process memory policy */ staticlong do_set_mempolicy(unsignedshort mode, unsignedshort flags,
nodemask_t *nodes)
{ struct mempolicy *new, *old;
NODEMASK_SCRATCH(scratch); int ret;
if (!scratch) return -ENOMEM;
new = mpol_new(mode, flags, nodes); if (IS_ERR(new)) {
ret = PTR_ERR(new); goto out;
}
task_lock(current);
ret = mpol_set_nodemask(new, nodes, scratch); if (ret) {
task_unlock(current);
mpol_put(new); goto out;
}
old = current->mempolicy;
current->mempolicy = new; if (new && (new->mode == MPOL_INTERLEAVE ||
new->mode == MPOL_WEIGHTED_INTERLEAVE)) {
current->il_prev = MAX_NUMNODES-1;
current->il_weight = 0;
}
task_unlock(current);
mpol_put(old);
ret = 0;
out:
NODEMASK_SCRATCH_FREE(scratch); return ret;
}
/* * Return nodemask for policy for get_mempolicy() query * * Called with task's alloc_lock held
*/ staticvoid get_policy_nodemask(struct mempolicy *pol, nodemask_t *nodes)
{
nodes_clear(*nodes); if (pol == &default_policy) return;
switch (pol->mode) { case MPOL_BIND: case MPOL_INTERLEAVE: case MPOL_PREFERRED: case MPOL_PREFERRED_MANY: case MPOL_WEIGHTED_INTERLEAVE:
*nodes = pol->nodes; break; case MPOL_LOCAL: /* return empty node mask for local allocation */ break; default:
BUG();
}
}
if (flags &
~(unsignedlong)(MPOL_F_NODE|MPOL_F_ADDR|MPOL_F_MEMS_ALLOWED)) return -EINVAL;
if (flags & MPOL_F_MEMS_ALLOWED) { if (flags & (MPOL_F_NODE|MPOL_F_ADDR)) return -EINVAL;
*policy = 0; /* just so it's initialized */
task_lock(current);
*nmask = cpuset_current_mems_allowed;
task_unlock(current); return 0;
}
if (flags & MPOL_F_ADDR) {
pgoff_t ilx; /* ignored here */ /* * Do NOT fall back to task policy if the * vma/shared policy at addr is NULL. We * want to return MPOL_DEFAULT in this case.
*/
mmap_read_lock(mm);
vma = vma_lookup(mm, addr); if (!vma) {
mmap_read_unlock(mm); return -EFAULT;
}
pol = __get_vma_policy(vma, addr, &ilx);
} elseif (addr) return -EINVAL;
if (!pol)
pol = &default_policy; /* indicates default behavior */
if (flags & MPOL_F_NODE) { if (flags & MPOL_F_ADDR) { /* * Take a refcount on the mpol, because we are about to * drop the mmap_lock, after which only "pol" remains * valid, "vma" is stale.
*/
pol_refcount = pol;
vma = NULL;
mpol_get(pol);
mmap_read_unlock(mm);
err = lookup_node(mm, addr); if (err < 0) goto out;
*policy = err;
} elseif (pol == current->mempolicy &&
pol->mode == MPOL_INTERLEAVE) {
*policy = next_node_in(current->il_prev, pol->nodes);
} elseif (pol == current->mempolicy &&
pol->mode == MPOL_WEIGHTED_INTERLEAVE) { if (current->il_weight)
*policy = current->il_prev; else
*policy = next_node_in(current->il_prev,
pol->nodes);
} else {
err = -EINVAL; goto out;
}
} else {
*policy = pol == &default_policy ? MPOL_DEFAULT :
pol->mode; /* * Internal mempolicy flags must be masked off before exposing * the policy to userspace.
*/
*policy |= (pol->flags & MPOL_MODE_FLAGS);
}
out:
mpol_cond_put(pol); if (vma)
mmap_read_unlock(mm); if (pol_refcount)
mpol_put(pol_refcount); return err;
}
#ifdef CONFIG_MIGRATION staticbool migrate_folio_add(struct folio *folio, struct list_head *foliolist, unsignedlong flags)
{ /* * Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio. * Choosing not to migrate a shared folio is not counted as a failure. * * See folio_maybe_mapped_shared() on possible imprecision when we * cannot easily detect if a folio is shared.
*/ if ((flags & MPOL_MF_MOVE_ALL) || !folio_maybe_mapped_shared(folio)) { if (folio_isolate_lru(folio)) {
list_add_tail(&folio->lru, foliolist);
node_stat_mod_folio(folio,
NR_ISOLATED_ANON + folio_is_file_lru(folio),
folio_nr_pages(folio));
} else { /* * Non-movable folio may reach here. And, there may be * temporary off LRU folios or non-LRU movable folios. * Treat them as unmovable folios since they can't be * isolated, so they can't be moved at the moment.
*/ returnfalse;
}
} returntrue;
}
/* * Migrate pages from one node to a target node. * Returns error or the number of pages not migrated.
*/ staticlong migrate_to_node(struct mm_struct *mm, int source, int dest, int flags)
{
nodemask_t nmask; struct vm_area_struct *vma;
LIST_HEAD(pagelist); long nr_failed; long err = 0; struct migration_target_control mtc = {
.nid = dest,
.gfp_mask = GFP_HIGHUSER_MOVABLE | __GFP_THISNODE,
.reason = MR_SYSCALL,
};
/* * This does not migrate the range, but isolates all pages that * need migration. Between passing in the full user address * space range and MPOL_MF_DISCONTIG_OK, this call cannot fail, * but passes back the count of pages which could not be isolated.
*/
nr_failed = queue_pages_range(mm, vma->vm_start, mm->task_size, &nmask,
flags | MPOL_MF_DISCONTIG_OK, &pagelist);
mmap_read_unlock(mm);
if (!list_empty(&pagelist)) {
err = migrate_pages(&pagelist, alloc_migration_target, NULL,
(unsignedlong)&mtc, MIGRATE_SYNC, MR_SYSCALL, NULL); if (err)
putback_movable_pages(&pagelist);
}
if (err >= 0)
err += nr_failed; return err;
}
/* * Move pages between the two nodesets so as to preserve the physical * layout as much as possible. * * Returns the number of page that could not be moved.
*/ int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from, const nodemask_t *to, int flags)
{ long nr_failed = 0; long err = 0;
nodemask_t tmp;
lru_cache_disable();
/* * 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 scanning 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.
*/
tmp = *from; while (!nodes_empty(tmp)) { int s, d; int source = NUMA_NO_NODE; int dest = 0;
for_each_node_mask(s, tmp) {
/* * do_migrate_pages() tries to maintain the relative * node relationship of the pages established between * threads and memory areas. * * However if the number of source nodes is not equal to * the number of destination nodes we can not preserve * this node relative relationship. In that case, skip * copying memory from a node that is in the destination * mask. * * Example: [2,3,4] -> [3,4,5] moves everything. * [0-7] - > [3,4,5] moves only 0,1,2,6,7.
*/
if ((nodes_weight(*from) != nodes_weight(*to)) &&
(node_isset(s, *to))) continue;
d = node_remap(s, *from, *to); 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 == NUMA_NO_NODE) break;
if (flags & ~(unsignedlong)MPOL_MF_VALID) return -EINVAL; if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_NICE)) return -EPERM;
if (start & ~PAGE_MASK) return -EINVAL;
if (mode == MPOL_DEFAULT)
flags &= ~MPOL_MF_STRICT;
len = PAGE_ALIGN(len);
end = start + len;
if (end < start) return -EINVAL; if (end == start) return 0;
new = mpol_new(mode, mode_flags, 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;
if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
lru_cache_disable();
{
NODEMASK_SCRATCH(scratch); if (scratch) {
mmap_write_lock(mm);
err = mpol_set_nodemask(new, nmask, scratch); if (err)
mmap_write_unlock(mm);
} else
err = -ENOMEM;
NODEMASK_SCRATCH_FREE(scratch);
} if (err) goto mpol_out;
/* * Lock the VMAs before scanning for pages to migrate, * to ensure we don't miss a concurrently inserted page.
*/
nr_failed = queue_pages_range(mm, start, end, nmask,
flags | MPOL_MF_INVERT | MPOL_MF_WRLOCK, &pagelist);
if (!err && !list_empty(&pagelist)) { /* Convert MPOL_DEFAULT's NULL to task or default policy */ if (!new) { new = get_task_policy(current);
mpol_get(new);
}
mmpol.pol = new;
mmpol.ilx = 0;
/* * In the interleaved case, attempt to allocate on exactly the * targeted nodes, for the first VMA to be migrated; for later * VMAs, the nodes will still be interleaved from the targeted * nodemask, but one by one may be selected differently.
*/ if (new->mode == MPOL_INTERLEAVE ||
new->mode == MPOL_WEIGHTED_INTERLEAVE) { struct folio *folio; unsignedint order; unsignedlong addr = -EFAULT;
list_for_each_entry(folio, &pagelist, lru) { if (!folio_test_ksm(folio)) break;
} if (!list_entry_is_head(folio, &pagelist, lru)) {
vma_iter_init(&vmi, mm, start);
for_each_vma_range(vmi, vma, end) {
addr = page_address_in_vma(folio,
folio_page(folio, 0), vma); if (addr != -EFAULT) break;
}
} if (addr != -EFAULT) {
order = folio_order(folio); /* We already know the pol, but not the ilx */
mpol_cond_put(get_vma_policy(vma, addr, order,
&mmpol.ilx)); /* Set base from which to increment by index */
mmpol.ilx -= folio->index >> order;
}
}
}
/* Copy a node mask from user space. */ staticint get_nodes(nodemask_t *nodes, constunsignedlong __user *nmask, unsignedlong maxnode)
{
--maxnode;
nodes_clear(*nodes); if (maxnode == 0 || !nmask) return 0; if (maxnode > PAGE_SIZE*BITS_PER_BYTE) return -EINVAL;
/* * When the user specified more nodes than supported just check * if the non supported part is all zero, one word at a time, * starting at the end.
*/ while (maxnode > MAX_NUMNODES) { unsignedlong bits = min_t(unsignedlong, maxnode, BITS_PER_LONG); unsignedlong t;
if (get_bitmap(&t, &nmask[(maxnode - 1) / BITS_PER_LONG], bits)) return -EFAULT;
start = untagged_addr(start); if (start & ~PAGE_MASK) return -EINVAL; /* * flags is used for future extension if any.
*/ if (flags != 0) return -EINVAL;
/* * Check home_node is online to avoid accessing uninitialized * NODE_DATA.
*/ if (home_node >= MAX_NUMNODES || !node_online(home_node)) return -EINVAL;
len = PAGE_ALIGN(len);
end = start + len;
if (end < start) return -EINVAL; if (end == start) return 0;
mmap_write_lock(mm);
prev = vma_prev(&vmi);
for_each_vma_range(vmi, vma, end) { /* * If any vma in the range got policy other than MPOL_BIND * or MPOL_PREFERRED_MANY we return error. We don't reset * the home node for vmas we already updated before.
*/
old = vma_policy(vma); if (!old) {
prev = vma; continue;
} if (old->mode != MPOL_BIND && old->mode != MPOL_PREFERRED_MANY) {
err = -EOPNOTSUPP; break;
} new = mpol_dup(old); if (IS_ERR(new)) {
err = PTR_ERR(new); break;
}
/* * Check if this process has the right to modify the specified process. * Use the regular "ptrace_may_access()" checks.
*/ if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS)) {
rcu_read_unlock();
err = -EPERM; goto out_put;
}
rcu_read_unlock();
task_nodes = cpuset_mems_allowed(task); /* Is the user allowed to access the target nodes? */ if (!nodes_subset(*new, task_nodes) && !capable(CAP_SYS_NICE)) {
err = -EPERM; goto out_put;
}
task_nodes = cpuset_mems_allowed(current);
nodes_and(*new, *new, task_nodes); if (nodes_empty(*new)) goto out_put;
err = security_task_movememory(task); if (err) goto out_put;
/* * DAX device mappings require predictable access latency, so avoid * incurring periodic faults.
*/ if (vma_is_dax(vma)) returnfalse;
if (is_vm_hugetlb_page(vma) &&
!hugepage_migration_supported(hstate_vma(vma))) returnfalse;
/* * Migration allocates pages in the highest zone. If we cannot * do so then migration (at least from node to node) is not * possible.
*/ if (vma->vm_file &&
gfp_zone(mapping_gfp_mask(vma->vm_file->f_mapping))
< policy_zone) returnfalse; returntrue;
}
/* * get_vma_policy(@vma, @addr, @order, @ilx) * @vma: virtual memory area whose policy is sought * @addr: address in @vma for shared policy lookup * @order: 0, or appropriate huge_page_order for interleaving * @ilx: interleave index (output), for use only when MPOL_INTERLEAVE or * MPOL_WEIGHTED_INTERLEAVE * * Returns effective policy for a VMA at specified address. * Falls back to current->mempolicy or system default policy, as necessary. * Shared policies [those marked as MPOL_F_SHARED] require an extra reference * count--added by the get_policy() vm_op, as appropriate--to protect against * freeing by another task. It is the caller's responsibility to free the * extra reference for shared policies.
*/ struct mempolicy *get_vma_policy(struct vm_area_struct *vma, unsignedlong addr, int order, pgoff_t *ilx)
{ struct mempolicy *pol;
pol = __get_vma_policy(vma, addr, ilx); if (!pol)
pol = get_task_policy(current); if (pol->mode == MPOL_INTERLEAVE ||
pol->mode == MPOL_WEIGHTED_INTERLEAVE) {
*ilx += vma->vm_pgoff >> order;
*ilx += (addr - vma->vm_start) >> (PAGE_SHIFT + order);
} return pol;
}
/* * if policy->nodes has movable memory only, * we apply policy when gfp_zone(gfp) = ZONE_MOVABLE only. * * policy->nodes is intersect with node_states[N_MEMORY]. * so if the following test fails, it implies * policy->nodes has movable memory only.
*/ if (!nodes_intersects(policy->nodes, node_states[N_HIGH_MEMORY]))
dynamic_policy_zone = ZONE_MOVABLE;
retry: /* to prevent miscount use tsk->mems_allowed_seq to detect rebind */
cpuset_mems_cookie = read_mems_allowed_begin();
node = current->il_prev; if (!current->il_weight || !node_isset(node, policy->nodes)) {
node = next_node_in(node, policy->nodes); if (read_mems_allowed_retry(cpuset_mems_cookie)) goto retry; if (node == MAX_NUMNODES) return node;
current->il_prev = node;
current->il_weight = get_il_weight(node);
}
current->il_weight--; return node;
}
/* Do dynamic interleaving for a process */ staticunsignedint interleave_nodes(struct mempolicy *policy)
{ unsignedint nid; unsignedint cpuset_mems_cookie;
/* to prevent miscount, use tsk->mems_allowed_seq to detect rebind */ do {
cpuset_mems_cookie = read_mems_allowed_begin();
nid = next_node_in(current->il_prev, policy->nodes);
} while (read_mems_allowed_retry(cpuset_mems_cookie));
if (nid < MAX_NUMNODES)
current->il_prev = nid; return nid;
}
/* * Depending on the memory policy provide a node from which to allocate the * next slab entry.
*/ unsignedint mempolicy_slab_node(void)
{ struct mempolicy *policy; int node = numa_mem_id();
if (!in_task()) return node;
policy = current->mempolicy; if (!policy) return node;
switch (policy->mode) { case MPOL_PREFERRED: return first_node(policy->nodes);
case MPOL_INTERLEAVE: return interleave_nodes(policy);
case MPOL_WEIGHTED_INTERLEAVE: return weighted_interleave_nodes(policy);
case MPOL_BIND: case MPOL_PREFERRED_MANY:
{ struct zoneref *z;
/* * Follow bind policy behavior and start allocation at the * first node.
*/ struct zonelist *zonelist; enum zone_type highest_zoneidx = gfp_zone(GFP_KERNEL);
zonelist = &NODE_DATA(node)->node_zonelists[ZONELIST_FALLBACK];
z = first_zones_zonelist(zonelist, highest_zoneidx,
&policy->nodes); return zonelist_zone(z) ? zonelist_node_idx(z) : node;
} case MPOL_LOCAL: return node;
default:
BUG();
}
}
staticunsignedint read_once_policy_nodemask(struct mempolicy *pol,
nodemask_t *mask)
{ /* * barrier stabilizes the nodemask locally so that it can be iterated * over safely without concern for changes. Allocators validate node * selection does not violate mems_allowed, so this is safe.
*/
barrier();
memcpy(mask, &pol->nodes, sizeof(nodemask_t));
barrier(); return nodes_weight(*mask);
}
nr_nodes = read_once_policy_nodemask(pol, &nodemask); if (!nr_nodes) return numa_node_id();
rcu_read_lock();
state = rcu_dereference(wi_state); /* Uninitialized wi_state means we should assume all weights are 1 */ if (state)
table = state->iw_table;
/* calculate the total weight */
for_each_node_mask(nid, nodemask)
weight_total += table ? table[nid] : 1;
/* Calculate the node offset based on totals */
target = ilx % weight_total;
nid = first_node(nodemask); while (target) { /* detect system default usage */
weight = table ? table[nid] : 1; if (target < weight) break;
target -= weight;
nid = next_node_in(nid, nodemask);
}
rcu_read_unlock(); return nid;
}
/* * Do static interleaving for interleave index @ilx. Returns the ilx'th * node in pol->nodes (starting from ilx=0), wrapping around if ilx * exceeds the number of present nodes.
*/ staticunsignedint interleave_nid(struct mempolicy *pol, pgoff_t ilx)
{
nodemask_t nodemask; unsignedint target, nnodes; int i; int nid;
nnodes = read_once_policy_nodemask(pol, &nodemask); if (!nnodes) return numa_node_id();
target = ilx % nnodes;
nid = first_node(nodemask); for (i = 0; i < target; i++)
nid = next_node(nid, nodemask); return nid;
}
/* * Return a nodemask representing a mempolicy for filtering nodes for * page allocation, together with preferred node id (or the input node id).
*/ static nodemask_t *policy_nodemask(gfp_t gfp, struct mempolicy *pol,
pgoff_t ilx, int *nid)
{
nodemask_t *nodemask = NULL;
switch (pol->mode) { case MPOL_PREFERRED: /* Override input node id */
*nid = first_node(pol->nodes); break; case MPOL_PREFERRED_MANY:
nodemask = &pol->nodes; if (pol->home_node != NUMA_NO_NODE)
*nid = pol->home_node; break; case MPOL_BIND: /* Restrict to nodemask (but not on lower zones) */ if (apply_policy_zone(pol, gfp_zone(gfp)) &&
cpuset_nodemask_valid_mems_allowed(&pol->nodes))
nodemask = &pol->nodes; if (pol->home_node != NUMA_NO_NODE)
*nid = pol->home_node; /* * __GFP_THISNODE shouldn't even be used with the bind policy * because we might easily break the expectation to stay on the * requested node and not break the policy.
*/
WARN_ON_ONCE(gfp & __GFP_THISNODE); break; case MPOL_INTERLEAVE: /* Override input node id */
*nid = (ilx == NO_INTERLEAVE_INDEX) ?
interleave_nodes(pol) : interleave_nid(pol, ilx); break; case MPOL_WEIGHTED_INTERLEAVE:
*nid = (ilx == NO_INTERLEAVE_INDEX) ?
weighted_interleave_nodes(pol) :
weighted_interleave_nid(pol, ilx); break;
}
return nodemask;
}
#ifdef CONFIG_HUGETLBFS /* * huge_node(@vma, @addr, @gfp_flags, @mpol) * @vma: virtual memory area whose policy is sought * @addr: address in @vma for shared policy lookup and interleave policy * @gfp_flags: for requested zone * @mpol: pointer to mempolicy pointer for reference counted mempolicy * @nodemask: pointer to nodemask pointer for 'bind' and 'prefer-many' policy * * Returns a nid suitable for a huge page allocation and a pointer * to the struct mempolicy for conditional unref after allocation. * If the effective policy is 'bind' or 'prefer-many', returns a pointer * to the mempolicy's @nodemask for filtering the zonelist.
*/ int huge_node(struct vm_area_struct *vma, unsignedlong addr, gfp_t gfp_flags, struct mempolicy **mpol, nodemask_t **nodemask)
{
pgoff_t ilx; int nid;
/* * init_nodemask_of_mempolicy * * If the current task's mempolicy is "default" [NULL], return 'false' * to indicate default policy. Otherwise, extract the policy nodemask * for 'bind' or 'interleave' policy into the argument nodemask, or * initialize the argument nodemask to contain the single node for * 'preferred' or 'local' policy and return 'true' to indicate presence * of non-default mempolicy. * * We don't bother with reference counting the mempolicy [mpol_get/put] * because the current task is examining it's own mempolicy and a task's * mempolicy is only ever changed by the task itself. * * N.B., it is the caller's responsibility to free a returned nodemask.
*/ bool init_nodemask_of_mempolicy(nodemask_t *mask)
{ struct mempolicy *mempolicy;
if (!(mask && current->mempolicy)) returnfalse;
task_lock(current);
mempolicy = current->mempolicy; switch (mempolicy->mode) { case MPOL_PREFERRED: case MPOL_PREFERRED_MANY: case MPOL_BIND: case MPOL_INTERLEAVE: case MPOL_WEIGHTED_INTERLEAVE:
*mask = mempolicy->nodes; break;
case MPOL_LOCAL:
init_nodemask_of_node(mask, numa_node_id()); break;
default:
BUG();
}
task_unlock(current);
returntrue;
} #endif
/* * mempolicy_in_oom_domain * * If tsk's mempolicy is "bind", check for intersection between mask and * the policy nodemask. Otherwise, return true for all other policies * including "interleave", as a tsk with "interleave" policy may have * memory allocated from all nodes in system. * * Takes task_lock(tsk) to prevent freeing of its mempolicy.
*/ bool mempolicy_in_oom_domain(struct task_struct *tsk, const nodemask_t *mask)
{ struct mempolicy *mempolicy; bool ret = true;
if (!mask) return ret;
task_lock(tsk);
mempolicy = tsk->mempolicy; if (mempolicy && mempolicy->mode == MPOL_BIND)
ret = nodes_intersects(mempolicy->nodes, *mask);
task_unlock(tsk);
/* * This is a two pass approach. The first pass will only try the * preferred nodes but skip the direct reclaim and allow the * allocation to fail, while the second pass will try all the * nodes in system.
*/
preferred_gfp = gfp | __GFP_NOWARN;
preferred_gfp &= ~(__GFP_DIRECT_RECLAIM | __GFP_NOFAIL);
page = __alloc_frozen_pages_noprof(preferred_gfp, order, nid, nodemask); if (!page)
page = __alloc_frozen_pages_noprof(gfp, order, nid, NULL);
return page;
}
/** * alloc_pages_mpol - Allocate pages according to NUMA mempolicy. * @gfp: GFP flags. * @order: Order of the page allocation. * @pol: Pointer to the NUMA mempolicy. * @ilx: Index for interleave mempolicy (also distinguishes alloc_pages()). * @nid: Preferred node (usually numa_node_id() but @mpol may override it). * * Return: The page on success or NULL if allocation fails.
*/ staticstruct page *alloc_pages_mpol(gfp_t gfp, unsignedint order, struct mempolicy *pol, pgoff_t ilx, int nid)
{
nodemask_t *nodemask; struct page *page;
nodemask = policy_nodemask(gfp, pol, ilx, &nid);
if (pol->mode == MPOL_PREFERRED_MANY) return alloc_pages_preferred_many(gfp, order, nid, nodemask);
if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && /* filter "hugepage" allocation, unless from alloc_pages() */
order == HPAGE_PMD_ORDER && ilx != NO_INTERLEAVE_INDEX) { /* * For hugepage allocation and non-interleave policy which * allows the current node (or other explicitly preferred * node) we only try to allocate from the current/preferred * node and don't fall back to other nodes, as the cost of * remote accesses would likely offset THP benefits. * * If the policy is interleave or does not allow the current * node in its nodemask, we allocate the standard way.
*/ if (pol->mode != MPOL_INTERLEAVE &&
pol->mode != MPOL_WEIGHTED_INTERLEAVE &&
(!nodemask || node_isset(nid, *nodemask))) { /* * First, try to allocate THP only on local node, but * don't reclaim unnecessarily, just compact.
*/
page = __alloc_frozen_pages_noprof(
gfp | __GFP_THISNODE | __GFP_NORETRY, order,
nid, NULL); if (page || !(gfp & __GFP_DIRECT_RECLAIM)) return page; /* * If hugepage allocations are configured to always * synchronous compact or the vma has been madvised * to prefer hugepage backing, retry allowing remote * memory with both reclaim and compact as well.
*/
}
}
/** * vma_alloc_folio - Allocate a folio for a VMA. * @gfp: GFP flags. * @order: Order of the folio. * @vma: Pointer to VMA. * @addr: Virtual address of the allocation. Must be inside @vma. * * Allocate a folio for a specific address in @vma, using the appropriate * NUMA policy. The caller must hold the mmap_lock of the mm_struct of the * VMA to prevent it from going away. Should be used for all allocations * for folios that will be mapped into user space, excepting hugetlbfs, and * excepting where direct use of folio_alloc_mpol() is more appropriate. * * Return: The folio on success or NULL if allocation fails.
*/ struct folio *vma_alloc_folio_noprof(gfp_t gfp, int order, struct vm_area_struct *vma, unsignedlong addr)
{ struct mempolicy *pol;
pgoff_t ilx; struct folio *folio;
if (vma->vm_flags & VM_DROPPABLE)
gfp |= __GFP_NOWARN;
/* * No reference counting needed for current->mempolicy * nor system default_policy
*/ if (!in_interrupt() && !(gfp & __GFP_THISNODE))
pol = get_task_policy(current);
/** * alloc_pages - Allocate pages. * @gfp: GFP flags. * @order: Power of two of number of pages to allocate. * * Allocate 1 << @order contiguous pages. The physical address of the * first page is naturally aligned (eg an order-3 allocation will be aligned * to a multiple of 8 * PAGE_SIZE bytes). The NUMA policy of the current * process is honoured when in process context. * * Context: Can be called from any context, providing the appropriate GFP * flags are used. * Return: The page on success or NULL if allocation fails.
*/ struct page *alloc_pages_noprof(gfp_t gfp, unsignedint order)
{ struct page *page = alloc_frozen_pages_noprof(gfp, order);
if (page)
set_page_refcounted(page); return page;
}
EXPORT_SYMBOL(alloc_pages_noprof);
/* read the nodes onto the stack, retry if done during rebind */ do {
cpuset_mems_cookie = read_mems_allowed_begin();
nnodes = read_once_policy_nodemask(pol, &nodes);
} while (read_mems_allowed_retry(cpuset_mems_cookie));
/* if the nodemask has become invalid, we cannot do anything */ if (!nnodes) return 0;
/* Continue allocating from most recent node and adjust the nr_pages */
node = me->il_prev;
weight = me->il_weight; if (weight && node_isset(node, nodes)) {
node_pages = min(rem_pages, weight);
nr_allocated = __alloc_pages_bulk(gfp, node, NULL, node_pages,
page_array);
page_array += nr_allocated;
total_allocated += nr_allocated; /* if that's all the pages, no need to interleave */ if (rem_pages <= weight) {
me->il_weight -= rem_pages; return total_allocated;
} /* Otherwise we adjust remaining pages, continue from there */
rem_pages -= weight;
} /* clear active weight in case of an allocation failure */
me->il_weight = 0;
prev_node = node;
/* create a local copy of node weights to operate on outside rcu */
weights = kzalloc(nr_node_ids, GFP_KERNEL); if (!weights) return total_allocated;
rcu_read_lock();
state = rcu_dereference(wi_state); if (state) {
memcpy(weights, state->iw_table, nr_node_ids * sizeof(u8));
rcu_read_unlock();
} else {
rcu_read_unlock(); for (i = 0; i < nr_node_ids; i++)
weights[i] = 1;
}
/* alloc pages bulk and mempolicy should be considered at the * same time in some situation such as vmalloc. * * It can accelerate memory allocation especially interleaving * allocate memory.
*/ unsignedlong alloc_pages_bulk_mempolicy_noprof(gfp_t gfp, unsignedlong nr_pages, struct page **page_array)
{ struct mempolicy *pol = &default_policy;
nodemask_t *nodemask; int nid;
if (!in_interrupt() && !(gfp & __GFP_THISNODE))
pol = get_task_policy(current);
if (pol->mode == MPOL_INTERLEAVE) return alloc_pages_bulk_interleave(gfp, pol,
nr_pages, page_array);
if (pol->mode == MPOL_WEIGHTED_INTERLEAVE) return alloc_pages_bulk_weighted_interleave(
gfp, pol, nr_pages, page_array);
if (pol->mode == MPOL_PREFERRED_MANY) return alloc_pages_bulk_preferred_many(gfp,
numa_node_id(), pol, nr_pages, page_array);
if (IS_ERR(pol)) return PTR_ERR(pol);
dst->vm_policy = pol; return 0;
}
/* * If mpol_dup() 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(). * * current's mempolicy may be rebinded by the other task(the task that changes * cpuset's mems), so we needn't do rebind work for current task.
*/
/* Slow path of a mempolicy duplicate */ struct mempolicy *__mpol_dup(struct mempolicy *old)
{ struct mempolicy *new = kmem_cache_alloc(policy_cache, GFP_KERNEL);
if (!new) return ERR_PTR(-ENOMEM);
/* task's mempolicy is protected by alloc_lock */ if (old == current->mempolicy) {
task_lock(current);
*new = *old;
task_unlock(current);
} else
*new = *old;
/* Slow path of a mempolicy comparison */ bool __mpol_equal(struct mempolicy *a, struct mempolicy *b)
{ if (!a || !b) returnfalse; if (a->mode != b->mode) returnfalse; if (a->flags != b->flags) returnfalse; if (a->home_node != b->home_node) returnfalse; if (mpol_store_user_nodemask(a)) if (!nodes_equal(a->w.user_nodemask, b->w.user_nodemask)) returnfalse;
switch (a->mode) { case MPOL_BIND: case MPOL_INTERLEAVE: case MPOL_PREFERRED: case MPOL_PREFERRED_MANY: case MPOL_WEIGHTED_INTERLEAVE: return !!nodes_equal(a->nodes, b->nodes); case MPOL_LOCAL: returntrue; default:
BUG(); returnfalse;
}
}
/* * 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 rwlock, which should be held * for any accesses to the tree.
*/
/* * lookup first element intersecting start-end. Caller holds sp->lock for * reading or for writing
*/ staticstruct sp_node *sp_lookup(struct shared_policy *sp,
pgoff_t start, pgoff_t end)
{ struct rb_node *n = sp->root.rb_node;
/** * mpol_misplaced - check whether current folio node is valid in policy * * @folio: folio to be checked * @vmf: structure describing the fault * @addr: virtual address in @vma for shared policy lookup and interleave policy * * Lookup current policy node id for vma,addr and "compare to" folio's * node id. Policy determination "mimics" alloc_page_vma(). * Called from fault path where we know the vma and faulting address. * * Return: NUMA_NO_NODE if the page is in a node that is valid for this * policy, or a suitable node ID to allocate a replacement folio from.
*/ int mpol_misplaced(struct folio *folio, struct vm_fault *vmf, unsignedlong addr)
{ struct mempolicy *pol;
pgoff_t ilx; struct zoneref *z; int curnid = folio_nid(folio); struct vm_area_struct *vma = vmf->vma; int thiscpu = raw_smp_processor_id(); int thisnid = numa_node_id(); int polnid = NUMA_NO_NODE; int ret = NUMA_NO_NODE;
/* * Make sure ptl is held so that we don't preempt and we * have a stable smp processor id
*/
lockdep_assert_held(vmf->ptl);
pol = get_vma_policy(vma, addr, folio_order(folio), &ilx); if (!(pol->flags & MPOL_F_MOF)) goto out;
switch (pol->mode) { case MPOL_INTERLEAVE:
polnid = interleave_nid(pol, ilx); break;
case MPOL_WEIGHTED_INTERLEAVE:
polnid = weighted_interleave_nid(pol, ilx); break;
case MPOL_PREFERRED: if (node_isset(curnid, pol->nodes)) goto out;
polnid = first_node(pol->nodes); break;
case MPOL_LOCAL:
polnid = numa_node_id(); break;
case MPOL_BIND: case MPOL_PREFERRED_MANY: /* * Even though MPOL_PREFERRED_MANY can allocate pages outside * policy nodemask we don't allow numa migration to nodes * outside policy nodemask for now. This is done so that if we * want demotion to slow memory to happen, before allocating * from some DRAM node say 'x', we will end up using a * MPOL_PREFERRED_MANY mask excluding node 'x'. In such scenario * we should not promote to node 'x' from slow memory node.
*/ if (pol->flags & MPOL_F_MORON) { /* * Optimize placement among multiple nodes * via NUMA balancing
*/ if (node_isset(thisnid, pol->nodes)) break; goto out;
}
/* * use current page if in policy nodemask, * else select nearest allowed node, if any. * If no allowed nodes, use current [!misplaced].
*/ if (node_isset(curnid, pol->nodes)) goto out;
z = first_zones_zonelist(
node_zonelist(thisnid, GFP_HIGHUSER),
gfp_zone(GFP_HIGHUSER),
&pol->nodes);
polnid = zonelist_node_idx(z); break;
default:
BUG();
}
/* Migrate the folio towards the node whose CPU is referencing it */ if (pol->flags & MPOL_F_MORON) {
polnid = thisnid;
if (!should_numa_migrate_memory(current, folio, curnid,
thiscpu)) goto out;
}
if (curnid != polnid)
ret = polnid;
out:
mpol_cond_put(pol);
return ret;
}
/* * Drop the (possibly final) reference to task->mempolicy. It needs to be * dropped after task->mempolicy is set to NULL so that any allocation done as * part of its kmem_cache_free(), such as by KASAN, doesn't reference a freed * policy.
*/ void mpol_put_task_policy(struct task_struct *task)
{ struct mempolicy *pol;
task_lock(task);
pol = task->mempolicy;
task->mempolicy = NULL;
task_unlock(task);
mpol_put(pol);
}
restart:
write_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 (!n_new) goto alloc_new;
err_out: if (mpol_new)
mpol_put(mpol_new); if (n_new)
kmem_cache_free(sn_cache, n_new);
return ret;
alloc_new:
write_unlock(&sp->lock);
ret = -ENOMEM;
n_new = kmem_cache_alloc(sn_cache, GFP_KERNEL); if (!n_new) goto err_out;
mpol_new = kmem_cache_alloc(policy_cache, GFP_KERNEL); if (!mpol_new) goto err_out;
atomic_set(&mpol_new->refcnt, 1); goto restart;
}
/** * mpol_shared_policy_init - initialize shared policy for inode * @sp: pointer to inode shared policy * @mpol: struct mempolicy to install * * Install non-NULL @mpol in inode's shared policy rb-tree. * On entry, the current task has a reference on a non-NULL @mpol. * This must be released on exit. * This is called at get_inode() calls and we can use GFP_KERNEL.
*/ void mpol_shared_policy_init(struct shared_policy *sp, struct mempolicy *mpol)
{ int ret;
if (mpol) { struct sp_node *sn; struct mempolicy *npol;
NODEMASK_SCRATCH(scratch);
if (!scratch) goto put_mpol;
/* contextualize the tmpfs mount point mempolicy to this file */
npol = mpol_new(mpol->mode, mpol->flags, &mpol->w.user_nodemask); if (IS_ERR(npol)) goto free_scratch; /* no valid nodemask intersection */
task_lock(current);
ret = mpol_set_nodemask(npol, &mpol->w.user_nodemask, scratch);
task_unlock(current); if (ret) goto put_npol;
/* alloc node covering entire file; adds ref to file's npol */
sn = sp_alloc(0, MAX_LFS_FILESIZE >> PAGE_SHIFT, npol); if (sn)
sp_insert(sp, sn);
put_npol:
mpol_put(npol); /* drop initial ref on file's npol */
free_scratch:
NODEMASK_SCRATCH_FREE(scratch);
put_mpol:
mpol_put(mpol); /* drop our incoming ref on sb mpol */
}
}
if (pol) { new = sp_alloc(vma->vm_pgoff, vma->vm_pgoff + sz, pol); if (!new) return -ENOMEM;
}
err = shared_policy_replace(sp, vma->vm_pgoff, vma->vm_pgoff + sz, new); if (err && new)
sp_free(new); return err;
}
/* Free a backing policy store on inode delete. */ void mpol_free_shared_policy(struct shared_policy *sp)
{ struct sp_node *n; struct rb_node *next;
if (!sp->root.rb_node) return;
write_lock(&sp->lock);
next = rb_first(&sp->root); while (next) {
n = rb_entry(next, struct sp_node, nd);
next = rb_next(&n->nd);
sp_delete(sp, n);
}
write_unlock(&sp->lock);
}
/* * Set interleaving policy for system init. Interleaving is only * enabled across suitably sized nodes (default is >= 16MB), or * fall back to the largest node if they're all smaller.
*/
nodes_clear(interleave_nodes);
for_each_node_state(nid, N_MEMORY) { unsignedlong total_pages = node_present_pages(nid);
/* Preserve the largest node */ if (largest < total_pages) {
largest = total_pages;
prefer = nid;
}
/* Interleave this node? */ if ((total_pages << PAGE_SHIFT) >= (16 << 20))
node_set(nid, interleave_nodes);
}
/* All too small, use the largest */ if (unlikely(nodes_empty(interleave_nodes)))
node_set(prefer, interleave_nodes);
if (do_set_mempolicy(MPOL_INTERLEAVE, 0, &interleave_nodes))
pr_err("%s: interleaving failed\n", __func__);
check_numabalancing_enable();
}
/* Reset policy of current process to default */ void numa_default_policy(void)
{
do_set_mempolicy(MPOL_DEFAULT, 0, NULL);
}
switch (mode) { case MPOL_PREFERRED: /* * Insist on a nodelist of one node only, although later * we use first_node(nodes) to grab a single node, so here * nodelist (or nodes) cannot be empty.
*/ if (nodelist) { char *rest = nodelist; while (isdigit(*rest))
rest++; if (*rest) goto out; if (nodes_empty(nodes)) goto out;
} break; case MPOL_INTERLEAVE: case MPOL_WEIGHTED_INTERLEAVE: /* * Default to online nodes with memory if no nodelist
*/ if (!nodelist)
nodes = node_states[N_MEMORY]; break; case MPOL_LOCAL: /* * Don't allow a nodelist; mpol_new() checks flags
*/ if (nodelist) goto out; break; case MPOL_DEFAULT: /* * Insist on a empty nodelist
*/ if (!nodelist)
err = 0; goto out; case MPOL_PREFERRED_MANY: case MPOL_BIND: /* * Insist on a nodelist
*/ if (!nodelist) goto out;
}
mode_flags = 0; if (flags) { /* * Currently, we only support two mutually exclusive * mode flags.
*/ if (!strcmp(flags, "static"))
mode_flags |= MPOL_F_STATIC_NODES; elseif (!strcmp(flags, "relative"))
mode_flags |= MPOL_F_RELATIVE_NODES; else goto out;
}
new = mpol_new(mode, mode_flags, &nodes); if (IS_ERR(new)) goto out;
/* * Save nodes for mpol_to_str() to show the tmpfs mount options * for /proc/mounts, /proc/pid/mounts and /proc/pid/mountinfo.
*/ if (mode != MPOL_PREFERRED) {
new->nodes = nodes;
} elseif (nodelist) {
nodes_clear(new->nodes);
node_set(first_node(nodes), new->nodes);
} else {
new->mode = MPOL_LOCAL;
}
/* * Save nodes for contextualization: this will be used to "clone" * the mempolicy in a specific context [cpuset] at a later time.
*/
new->w.user_nodemask = nodes;
err = 0;
out: /* Restore string for error message */ if (nodelist)
*--nodelist = ':'; if (flags)
*--flags = '='; if (!err)
*mpol = new; return err;
} #endif/* CONFIG_TMPFS */
/** * mpol_to_str - format a mempolicy structure for printing * @buffer: to contain formatted mempolicy string * @maxlen: length of @buffer * @pol: pointer to mempolicy to be formatted * * Convert @pol into a string. If @buffer is too short, truncate the string. * Recommend a @maxlen of at least 51 for the longest mode, "weighted * interleave", plus the longest flag flags, "relative|balancing", and to * display at least a few node ids.
*/ void mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol)
{ char *p = buffer;
nodemask_t nodes = NODE_MASK_NONE; unsignedshort mode = MPOL_DEFAULT; unsignedshort flags = 0;
if (pol &&
pol != &default_policy &&
!(pol >= &preferred_node_policy[0] &&
pol <= &preferred_node_policy[ARRAY_SIZE(preferred_node_policy) - 1])) {
mode = pol->mode;
flags = pol->flags;
}
switch (mode) { case MPOL_DEFAULT: case MPOL_LOCAL: break; case MPOL_PREFERRED: case MPOL_PREFERRED_MANY: case MPOL_BIND: case MPOL_INTERLEAVE: case MPOL_WEIGHTED_INTERLEAVE:
nodes = pol->nodes; break; default:
WARN_ON_ONCE(1);
snprintf(p, maxlen, "unknown"); return;
}
p += snprintf(p, maxlen, "%s", policy_modes[mode]);
if (flags & MPOL_MODE_FLAGS) {
p += snprintf(p, buffer + maxlen - p, "=");
/* * Static and relative are mutually exclusive.
*/ if (flags & MPOL_F_STATIC_NODES)
p += snprintf(p, buffer + maxlen - p, "static"); elseif (flags & MPOL_F_RELATIVE_NODES)
p += snprintf(p, buffer + maxlen - p, "relative");
if (flags & MPOL_F_NUMA_BALANCING) { if (!is_power_of_2(flags & MPOL_MODE_FLAGS))
p += snprintf(p, buffer + maxlen - p, "|");
p += snprintf(p, buffer + maxlen - p, "balancing");
}
}
if (!nodes_empty(nodes))
p += scnprintf(p, buffer + maxlen - p, ":%*pbl",
nodemask_pr_args(&nodes));
}
staticint wi_node_notifier(struct notifier_block *nb, unsignedlong action, void *data)
{ int err; struct node_notify *nn = data; int nid = nn->nid;
switch (action) { case NODE_ADDED_FIRST_MEMORY:
err = sysfs_wi_node_add(nid); if (err)
pr_err("failed to add sysfs for node%d during hotplug: %d\n",
nid, err); break; case NODE_REMOVED_LAST_MEMORY:
sysfs_wi_node_delete(nid); break;
}
return NOTIFY_OK;
}
staticint __init add_weighted_interleave_group(struct kobject *mempolicy_kobj)
{ int nid, err;
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.