The patch titled cpusets: randomize node rotor used in cpuset_mem_spread_node() has been removed from the -mm tree. Its filename was cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: cpusets: randomize node rotor used in cpuset_mem_spread_node() From: Michal Hocko <mhocko@xxxxxxx> [This patch has already been accepted as 0ac0c0d but later reverted (35926ff) because it itroduced arch specific __node_random which was defined only for x86 code so it broke other archs. This is a followup without any arch specific code. Other than that there are no functional changes.] Some workloads that create a large number of small files tend to assign too many pages to node 0 (multi-node systems). Part of the reason is that the rotor (in cpuset_mem_spread_node()) used to assign nodes starts at node 0 for newly created tasks. This patch changes the rotor to be initialized to a random node number of the cpuset. [akpm@xxxxxxxxxxxxxxxxxxxx: fix layout] [Lee.Schermerhorn@xxxxxx: Define stub numa_random() for !NUMA configuration] [mhocko@xxxxxxx: Make it arch independent] [akpm@xxxxxxxxxxxxxxxxxxxx: fix CONFIG_NUMA=y, MAX_NUMNODES>1 build] Signed-off-by: Jack Steiner <steiner@xxxxxxx> Signed-off-by: Lee Schermerhorn <lee.schermerhorn@xxxxxx> Signed-off-by: Michal Hocko <mhocko@xxxxxxx> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx> Cc: Christoph Lameter <cl@xxxxxxxxxxxxxxxxxxxx> Cc: Pekka Enberg <penberg@xxxxxxxxxxxxxx> Cc: Paul Menage <menage@xxxxxxxxxx> Cc: Jack Steiner <steiner@xxxxxxx> Cc: Robin Holt <holt@xxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Cc: Christoph Lameter <cl@xxxxxxxxxxxxxxxxxxxx> Cc: David Rientjes <rientjes@xxxxxxxxxx> Cc: Jack Steiner <steiner@xxxxxxx> Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx> Cc: Lee Schermerhorn <lee.schermerhorn@xxxxxx> Cc: Michal Hocko <mhocko@xxxxxxx> Cc: Paul Menage <menage@xxxxxxxxxx> Cc: Pekka Enberg <penberg@xxxxxxxxxxxxxx> Cc: Robin Holt <holt@xxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- include/linux/bitmap.h | 1 + include/linux/nodemask.h | 13 +++++++++++++ kernel/cpuset.c | 8 ++++++++ kernel/fork.c | 4 ++++ lib/bitmap.c | 2 +- mm/mempolicy.c | 16 ++++++++++++++++ 6 files changed, 43 insertions(+), 1 deletion(-) diff -puN include/linux/bitmap.h~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node include/linux/bitmap.h --- a/include/linux/bitmap.h~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/include/linux/bitmap.h @@ -144,6 +144,7 @@ extern int bitmap_find_free_region(unsig extern void bitmap_release_region(unsigned long *bitmap, int pos, int order); extern int bitmap_allocate_region(unsigned long *bitmap, int pos, int order); extern void bitmap_copy_le(void *dst, const unsigned long *src, int nbits); +extern int bitmap_ord_to_pos(const unsigned long *bitmap, int n, int bits); #define BITMAP_LAST_WORD_MASK(nbits) \ ( \ diff -puN include/linux/nodemask.h~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node include/linux/nodemask.h --- a/include/linux/nodemask.h~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/include/linux/nodemask.h @@ -66,6 +66,8 @@ * int num_online_nodes() Number of online Nodes * int num_possible_nodes() Number of all possible Nodes * + * int node_random(mask) Random node with set bit in mask + * * int node_online(node) Is some node online? * int node_possible(node) Is some node possible? * @@ -430,6 +432,7 @@ static inline void node_set_offline(int node_clear_state(nid, N_ONLINE); nr_online_nodes = num_node_state(N_ONLINE); } + #else static inline int node_state(int node, enum node_states state) @@ -460,6 +463,16 @@ static inline int num_node_state(enum no #define node_set_online(node) node_set_state((node), N_ONLINE) #define node_set_offline(node) node_clear_state((node), N_ONLINE) + +#endif + +#if defined(CONFIG_NUMA) && (MAX_NUMNODES > 1) +extern int node_random(const nodemask_t *maskp); +#else +static inline int node_random(const nodemask_t *mask) +{ + return 0; +} #endif #define node_online_map node_states[N_ONLINE] diff -puN kernel/cpuset.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node kernel/cpuset.c --- a/kernel/cpuset.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/kernel/cpuset.c @@ -2460,11 +2460,19 @@ static int cpuset_spread_node(int *rotor int cpuset_mem_spread_node(void) { + if (current->cpuset_mem_spread_rotor == NUMA_NO_NODE) + current->cpuset_mem_spread_rotor = + node_random(¤t->mems_allowed); + return cpuset_spread_node(¤t->cpuset_mem_spread_rotor); } int cpuset_slab_spread_node(void) { + if (current->cpuset_slab_spread_rotor == NUMA_NO_NODE) + current->cpuset_slab_spread_rotor = + node_random(¤t->mems_allowed); + return cpuset_spread_node(¤t->cpuset_slab_spread_rotor); } diff -puN kernel/fork.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node kernel/fork.c --- a/kernel/fork.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/kernel/fork.c @@ -1173,6 +1173,10 @@ static struct task_struct *copy_process( } mpol_fix_fork_child_flag(p); #endif +#ifdef CONFIG_CPUSETS + p->cpuset_mem_spread_rotor = NUMA_NO_NODE; + p->cpuset_slab_spread_rotor = NUMA_NO_NODE; +#endif #ifdef CONFIG_TRACE_IRQFLAGS p->irq_events = 0; #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW diff -puN lib/bitmap.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node lib/bitmap.c --- a/lib/bitmap.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/lib/bitmap.c @@ -756,7 +756,7 @@ static int bitmap_pos_to_ord(const unsig * * The bit positions 0 through @bits are valid positions in @buf. */ -static int bitmap_ord_to_pos(const unsigned long *buf, int ord, int bits) +int bitmap_ord_to_pos(const unsigned long *buf, int ord, int bits) { int pos = 0; diff -puN mm/mempolicy.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node mm/mempolicy.c --- a/mm/mempolicy.c~cpusets-randomize-node-rotor-used-in-cpuset_mem_spread_node +++ a/mm/mempolicy.c @@ -93,6 +93,7 @@ #include <asm/tlbflush.h> #include <asm/uaccess.h> +#include <linux/random.h> #include "internal.h" @@ -1645,6 +1646,21 @@ static inline unsigned interleave_nid(st return interleave_nodes(pol); } +/* + * Return the bit number of a random bit set in the nodemask. + * (returns -1 if nodemask is empty) + */ +int node_random(const nodemask_t *maskp) +{ + int w, bit = -1; + + w = nodes_weight(*maskp); + if (w) + bit = bitmap_ord_to_pos(maskp->bits, + get_random_int() % w, MAX_NUMNODES); + return bit; +} + #ifdef CONFIG_HUGETLBFS /* * huge_zonelist(@vma, @addr, @gfp_flags, @mpol) _ Patches currently in -mm which might be from mhocko@xxxxxxx are origin.patch memcg-do-not-expose-uninitialized-mem_cgroup_per_node-to-world.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