Adds another optional mode flag, MPOL_F_RELATIVE_NODES, that specifies nodemasks passed via set_mempolicy() or mbind() should be considered relative to the current task's mems_allowed.
When the mempolicy is created, the passed nodemask is folded and mapped onto the current task's mems_allowed. For example, consider a task using set_mempolicy() to pass MPOL_INTERLEAVE | MPOL_F_RELATIVE_NODES with a nodemask of 1-3. If current's mems_allowed is 4-7, the effected nodemask is 5-7 (the second, third, and fourth node of mems_allowed). If the same task is attached to a cpuset, the mempolicy nodemask is rebound each time the mems are changed. Some possible rebinds and results are: mems result 1-3 1-3 1-7 2-4 1,5-6 1,5-6 1,5-7 5-7 Likewise, the zonelist built for MPOL_BIND acts on the set of zones assigned to the resultant nodemask from the relative remap. In the MPOL_PREFERRED case, the preferred node is remapped from the currently effected nodemask to the relative nodemask. This mempolicy mode flag was conceived of by Paul Jackson <[EMAIL PROTECTED]>. Cc: Paul Jackson <[EMAIL PROTECTED]> Cc: Christoph Lameter <[EMAIL PROTECTED]> Cc: Lee Schermerhorn <[EMAIL PROTECTED]> Cc: Andi Kleen <[EMAIL PROTECTED]> Signed-off-by: David Rientjes <[EMAIL PROTECTED]> --- include/linux/mempolicy.h | 3 ++- mm/mempolicy.c | 40 ++++++++++++++++++++++++++++++++++++++-- mm/shmem.c | 6 ++++++ 3 files changed, 46 insertions(+), 3 deletions(-) diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -25,12 +25,13 @@ enum { /* Flags for set_mempolicy */ #define MPOL_F_STATIC_NODES (1 << 15) +#define MPOL_F_RELATIVE_NODES (1 << 14) /* * MPOL_MODE_FLAGS is the union of all possible optional mode flags passed to * either set_mempolicy() or mbind(). */ -#define MPOL_MODE_FLAGS (MPOL_F_STATIC_NODES) +#define MPOL_MODE_FLAGS (MPOL_F_STATIC_NODES | MPOL_F_RELATIVE_NODES) /* Flags for get_mempolicy */ #define MPOL_F_NODE (1<<0) /* return next IL mode instead of node mask */ diff --git a/mm/mempolicy.c b/mm/mempolicy.c --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -152,7 +152,18 @@ static struct zonelist *bind_zonelist(nodemask_t *nodes) static inline int mpol_store_user_nodemask(const struct mempolicy *pol) { - return !!(pol->flags & MPOL_F_STATIC_NODES); + return !!(pol->flags & (MPOL_F_STATIC_NODES | MPOL_F_RELATIVE_NODES)); +} + +static nodemask_t mpol_relative_nodemask(const nodemask_t *orig, + const nodemask_t *rel) +{ + nodemask_t ret; + nodemask_t tmp; + + nodes_fold(tmp, *orig, nodes_weight(*rel)); + nodes_onto(ret, tmp, *rel); + return ret; } /* Create a new policy */ @@ -173,7 +184,12 @@ static struct mempolicy *mpol_new(unsigned short mode, unsigned short flags, return ERR_PTR(-ENOMEM); atomic_set(&policy->refcnt, 1); cpuset_update_task_memory_state(); - nodes_and(cpuset_context_nmask, *nodes, cpuset_current_mems_allowed); + if (flags & MPOL_F_RELATIVE_NODES) + cpuset_context_nmask = mpol_relative_nodemask(nodes, + &cpuset_current_mems_allowed); + else + nodes_and(cpuset_context_nmask, *nodes, + cpuset_current_mems_allowed); switch (mode) { case MPOL_INTERLEAVE: if (nodes_empty(*nodes) || nodes_empty(cpuset_context_nmask)) @@ -898,6 +914,9 @@ asmlinkage long sys_mbind(unsigned long start, unsigned long len, mode &= ~MPOL_MODE_FLAGS; if (mode >= MPOL_MAX) return -EINVAL; + if ((mode_flags & MPOL_F_STATIC_NODES) && + (mode_flags & MPOL_F_RELATIVE_NODES)) + return -EINVAL; err = get_nodes(&nodes, nmask, maxnode); if (err) return err; @@ -916,6 +935,8 @@ asmlinkage long sys_set_mempolicy(int mode, unsigned long __user *nmask, mode &= ~MPOL_MODE_FLAGS; if ((unsigned int)mode >= MPOL_MAX) return -EINVAL; + if ((flags & MPOL_F_STATIC_NODES) && (flags & MPOL_F_RELATIVE_NODES)) + return -EINVAL; err = get_nodes(&nodes, nmask, maxnode); if (err) return err; @@ -1747,10 +1768,12 @@ static void mpol_rebind_policy(struct mempolicy *pol, { nodemask_t tmp; int static_nodes; + int relative_nodes; if (!pol) return; static_nodes = pol->flags & MPOL_F_STATIC_NODES; + relative_nodes = pol->flags & MPOL_F_RELATIVE_NODES; if (!mpol_store_user_nodemask(pol) && nodes_equal(pol->w.cpuset_mems_allowed, *newmask)) return; @@ -1761,6 +1784,9 @@ static void mpol_rebind_policy(struct mempolicy *pol, case MPOL_INTERLEAVE: if (static_nodes) nodes_and(tmp, pol->w.user_nodemask, *newmask); + else if (relative_nodes) + tmp = mpol_relative_nodemask(&pol->w.user_nodemask, + newmask); else { nodes_remap(tmp, pol->v.nodes, pol->w.cpuset_mems_allowed, *newmask); @@ -1783,6 +1809,11 @@ static void mpol_rebind_policy(struct mempolicy *pol, pol->v.preferred_node = node; else pol->v.preferred_node = -1; + } else if (relative_nodes) { + tmp = mpol_relative_nodemask(&pol->w.user_nodemask, + newmask); + pol->v.preferred_node = node_remap(pol->v.preferred_node, + pol->w.cpuset_mems_allowed, tmp); } else { pol->v.preferred_node = node_remap(pol->v.preferred_node, pol->w.cpuset_mems_allowed, *newmask); @@ -1794,6 +1825,9 @@ static void mpol_rebind_policy(struct mempolicy *pol, if (static_nodes) nodes_and(tmp, pol->w.user_nodemask, *newmask); + else if (relative_nodes) + tmp = mpol_relative_nodemask(&pol->w.user_nodemask, + newmask); else { nodemask_t nodes; struct zone **z; @@ -1911,6 +1945,8 @@ static inline int mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol) if (flags & MPOL_F_STATIC_NODES) p += sprintf(p, "%sstatic", need_bar++ ? "|" : ""); + if (flags & MPOL_F_RELATIVE_NODES) + p += sprintf(p, "%srelative", need_bar++ ? "|" : ""); } if (!nodes_empty(nodes)) { diff --git a/mm/shmem.c b/mm/shmem.c --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1131,6 +1131,12 @@ static int shmem_parse_mpol(char *value, unsigned short *policy, if (flags) { if (!strcmp(flags, "static")) *mode_flags |= MPOL_F_STATIC_NODES; + if (!strcmp(flags, "relative")) + *mode_flags |= MPOL_F_RELATIVE_NODES; + + if ((*mode_flags & MPOL_F_STATIC_NODES) && + (*mode_flags & MPOL_F_RELATIVE_NODES)) + err = 1; } out: /* Restore string for error message */ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/