The patch titled mempolicy: mPOL_PREFERRED cleanups for "local allocation" has been added to the -mm tree. Its filename is mempolicy-mpol_preferred-cleanups-for-local-allocation.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: mempolicy: mPOL_PREFERRED cleanups for "local allocation" From: Lee Schermerhorn <lee.schermerhorn@xxxxxx> Here are a couple of "cleanups" for MPOL_PREFERRED behavior when v.preferred_node < 0 -- i.e., "local allocation": 1) [do_]get_mempolicy() calls the now renamed get_policy_nodemask() to fetch the nodemask associated with a policy. Currently, get_policy_nodemask() returns the set of nodes with memory, when the policy 'mode' is 'PREFERRED, and the preferred_node is < 0. Change to return an empty nodemask, as this is what was specified to achieve "local allocation". 2) When a task is moved into a [new] cpuset, mpol_rebind_policy() is called to adjust any task and vma policy nodes to be valid in the new cpuset. However, when the policy is MPOL_PREFERRED, and the preferred_node is <0, no rebind is necessary. The "local allocation" indication is valid in any cpuset. Existing code will "do the right thing" because node_remap() will just return the argument node when it is outside of the valid range of node ids. However, I think it is clearer and cleaner to skip the remap explicitly in this case. 3) mpol_to_str() produces a printable, "human readable" string from a struct mempolicy. For MPOL_PREFERRED with preferred_node <0, show "local", as this indicates local allocation, as the task migrates among nodes. Note that this matches the usage of "local allocation" in libnuma() and numactl. Without this change, I believe that node_set() [via set_bit()] will set bit 31, resulting in a misleading display. Signed-off-by: Lee Schermerhorn <lee.schermerhorn@xxxxxx> Cc: Christoph Lameter <clameter@xxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Cc: Mel Gorman <mel@xxxxxxxxx> Cc: Andi Kleen <ak@xxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/mempolicy.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff -puN mm/mempolicy.c~mempolicy-mpol_preferred-cleanups-for-local-allocation mm/mempolicy.c --- a/mm/mempolicy.c~mempolicy-mpol_preferred-cleanups-for-local-allocation +++ a/mm/mempolicy.c @@ -645,11 +645,9 @@ static void get_policy_nodemask(struct m *nodes = p->v.nodes; break; case MPOL_PREFERRED: - /* or use current node instead of memory_map? */ - if (p->v.preferred_node < 0) - *nodes = node_states[N_HIGH_MEMORY]; - else + if (p->v.preferred_node >= 0) node_set(p->v.preferred_node, *nodes); + /* else return empty node mask for local allocation */ break; default: BUG(); @@ -804,7 +802,7 @@ int do_migrate_pages(struct mm_struct *m int err = 0; nodemask_t tmp; - down_read(&mm->mmap_sem); + down_read(&mm->mmap_sem); err = migrate_vmas(mm, from_nodes, to_nodes, flags); if (err) @@ -1949,10 +1947,12 @@ void numa_default_policy(void) } /* - * Display pages allocated per node and memory policy via /proc. + * "local" is pseudo-policy: MPOL_PREFERRED with preferred_node == -1 + * Used only for mpol_to_str() */ +#define MPOL_LOCAL (MPOL_INTERLEAVE + 1) static const char * const policy_types[] = - { "default", "prefer", "bind", "interleave" }; + { "default", "prefer", "bind", "interleave", "local" }; /* * Convert a mempolicy into a string. @@ -1963,6 +1963,7 @@ static inline int mpol_to_str(char *buff { char *p = buffer; int l; + int nid; nodemask_t nodes; unsigned short mode; unsigned short flags = pol ? pol->flags : 0; @@ -1979,7 +1980,11 @@ static inline int mpol_to_str(char *buff case MPOL_PREFERRED: nodes_clear(nodes); - node_set(pol->v.preferred_node, nodes); + nid = pol->v.preferred_node; + if (nid < 0) + mode = MPOL_LOCAL; /* pseudo-policy */ + else + node_set(nid, nodes); break; case MPOL_BIND: @@ -1994,8 +1999,8 @@ static inline int mpol_to_str(char *buff } l = strlen(policy_types[mode]); - if (buffer + maxlen < p + l + 1) - return -ENOSPC; + if (buffer + maxlen < p + l + 1) + return -ENOSPC; strcpy(p, policy_types[mode]); p += l; @@ -2094,6 +2099,9 @@ static inline void check_huge_range(stru } #endif +/* + * Display pages allocated per node and memory policy via /proc. + */ int show_numa_map(struct seq_file *m, void *v) { struct proc_maps_private *priv = m->private; _ Patches currently in -mm which might be from lee.schermerhorn@xxxxxx are mm-use-zonelists-instead-of-zones-when-direct-reclaiming-pages.patch mm-introduce-node_zonelist-for-accessing-the-zonelist-for-a-gfp-mask.patch mm-remember-what-the-preferred-zone-is-for-zone_statistics.patch mm-use-two-zonelist-that-are-filtered-by-gfp-mask.patch mm-have-zonelist-contains-structs-with-both-a-zone-pointer-and-zone_idx.patch mm-have-zonelist-contains-structs-with-both-a-zone-pointer-and-zone_idx-fix-memcg-ooms.patch mm-have-zonelist-contains-structs-with-both-a-zone-pointer-and-zone_idx-just-return-do_try_to_free_pages.patch mm-have-zonelist-contains-structs-with-both-a-zone-pointer-and-zone_idx-just-return-do_try_to_free_pages-do_try_to_free_pages-gfp_mask-redundant.patch mm-filter-based-on-a-nodemask-as-well-as-a-gfp_mask.patch mm-filter-based-on-a-nodemask-as-well-as-a-gfp_mask-doc-fixes.patch mm-filter-based-on-a-nodemask-as-well-as-a-gfp_mask-make-dequeue_huge_page_vma-obey-mpol_bind-nodemask.patch mm-filter-based-on-a-nodemask-as-well-as-a-gfp_mask-make-dequeue_huge_page_vma-obey-mpol_bind-nodemask-rework.patch mm-filter-based-on-a-nodemask-as-well-as-a-gfp_mask-deporkify.patch mempolicy-add-bitmap_onto-and-bitmap_fold-operations.patch mempolicy-create-mempolicy_operations-structure-fix.patch mempolicy-disallow-static-or-relative-flags-for-local-preferred-mode.patch mempolicy-fix-parsing-of-tmpfs-mpol-mount-option.patch mempolicy-rename-mpol_free-to-mpol_put.patch mempolicy-rename-mpol_copy-to-mpol_dup.patch mempolicy-write-lock-mmap_sem-while-changing-task-mempolicy.patch mempolicy-fixup-fallback-for-default-shmem-policy.patch mempolicy-rename-struct-mempolicy-policy-member-to-mode.patch mempolicy-mark-shared-policies-for-unref.patch mempolicy-document-setget_policy-vm_ops-apis.patch mempolicy-rework-mempolicy-reference-counting.patch mempolicy-use-mpol_preferred-for-system-wide-default-policy.patch mempolicy-mpol_preferred-cleanups-for-local-allocation.patch mempolicy-use-mpol_f_local-to-indicate-preferred-local-policy.patch mempolicy-clean-up-mpol-to-str-mempolicy-formatting.patch mempolicy-rework-shmem-mpol-parsing-and-display.patch mempolicy-support-mpol=local-tmpfs-mount-option.patch mempolicy-use-struct-mempolicy-pointer-in-shmem_sb_info.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html