task_numa_find_cpu helps to find the cpu to swap/move the task.
Its guarded by numa_has_capacity(). However node not having capacity
shouldn't deter a task swapping if it helps numa placement.

Further load_too_imbalanced, which evaluates possibilities of move/swap,
provides similar checks as numa_has_capacity.

Hence remove numa_has_capacity() to enhance possibilities of task
swapping even if load is imbalanced.

Testcase       Time:         Min         Max         Avg      StdDev
numa01.sh      Real:      674.61      997.71      785.01      115.95
numa01.sh       Sys:      180.87      318.88      270.13       51.32
numa01.sh      User:    54001.30    71936.50    60495.48     6237.55
numa02.sh      Real:       60.62       62.30       61.46        0.62
numa02.sh       Sys:       15.01       33.63       24.38        6.81
numa02.sh      User:     5234.20     5325.60     5276.23       38.85
numa03.sh      Real:      827.62      946.85      914.48       44.58
numa03.sh       Sys:      135.55      172.40      158.46       12.75
numa03.sh      User:    64839.42    73195.44    70805.96     3061.20
numa04.sh      Real:      481.01      608.76      521.14       47.28
numa04.sh       Sys:      329.59      373.15      353.20       14.20
numa04.sh      User:    37649.09    40722.94    38806.32     1072.32
numa05.sh      Real:      399.21      415.38      409.88        5.54
numa05.sh       Sys:      319.46      418.57      363.31       37.62
numa05.sh      User:    33727.77    34732.68    34127.41      447.11

Testcase       Time:         Min         Max         Avg      StdDev     %Change
numa01.sh      Real:      435.67      707.28      527.49       97.85     48.81%
numa01.sh       Sys:       76.41      231.19      162.49       56.13     66.24%
numa01.sh      User:    38247.36    59033.52    45129.31     7642.69     34.04%
numa02.sh      Real:       60.35       62.09       61.09        0.69     0.605%
numa02.sh       Sys:       15.01       30.20       20.64        5.56     18.12%
numa02.sh      User:     5195.93     5294.82     5240.99       40.55     0.672%
numa03.sh      Real:      752.04      919.89      836.81       63.29     9.281%
numa03.sh       Sys:      115.10      133.35      125.46        7.78     26.30%
numa03.sh      User:    58736.44    70084.26    65103.67     4416.10     8.758%
numa04.sh      Real:      418.43      709.69      512.53      104.17     1.679%
numa04.sh       Sys:      242.99      370.47      297.39       42.20     18.76%
numa04.sh      User:    34916.14    48429.54    38955.65     4928.05     -0.38%
numa05.sh      Real:      379.27      434.05      403.70       17.79     1.530%
numa05.sh       Sys:      145.94      344.50      268.72       68.53     35.20%
numa05.sh      User:    32679.32    35449.75    33989.10      913.19     0.406%

Signed-off-by: Srikar Dronamraju <sri...@linux.vnet.ibm.com>
---
 kernel/sched/fair.c | 37 +++----------------------------------
 1 file changed, 3 insertions(+), 34 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 259c343..709c77c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1417,7 +1417,6 @@ struct numa_stats {
        unsigned long compute_capacity;
 
        unsigned int nr_running;
-       int has_free_capacity;
 };
 
 /*
@@ -1444,8 +1443,7 @@ static void update_numa_stats(struct numa_stats *ns, int 
nid)
         * the @ns structure is NULL'ed and task_numa_compare() will
         * not find this node attractive.
         *
-        * We'll either bail at !has_free_capacity, or we'll detect a huge
-        * imbalance and bail there.
+        * We'll detect a huge imbalance and bail there.
         */
        if (!cpus)
                return;
@@ -1456,7 +1454,6 @@ static void update_numa_stats(struct numa_stats *ns, int 
nid)
 
        capacity = min_t(unsigned, capacity,
                DIV_ROUND_CLOSEST(ns->compute_capacity, SCHED_CAPACITY_SCALE));
-       ns->has_free_capacity = (ns->nr_running < capacity);
 }
 
 struct task_numa_env {
@@ -1672,7 +1669,6 @@ static void task_numa_find_cpu(struct task_numa_env *env,
         * than swapping tasks around, check if a move is possible.
         */
        move = !load_too_imbalanced(src_load, dst_load, env);
-
        for_each_cpu(cpu, cpumask_of_node(env->dst_nid)) {
                /* Skip this CPU if the source task cannot migrate */
                if (!cpumask_test_cpu(cpu, &env->p->cpus_allowed))
@@ -1683,31 +1679,6 @@ static void task_numa_find_cpu(struct task_numa_env *env,
        }
 }
 
-/* Only move tasks to a NUMA node less busy than the current node. */
-static bool numa_has_capacity(struct task_numa_env *env)
-{
-       struct numa_stats *src = &env->src_stats;
-       struct numa_stats *dst = &env->dst_stats;
-
-       if (src->has_free_capacity && !dst->has_free_capacity)
-               return false;
-
-       /*
-        * Only consider a task move if the source has a higher load
-        * than the destination, corrected for CPU capacity on each node.
-        *
-        *      src->load                dst->load
-        * --------------------- vs ---------------------
-        * src->compute_capacity    dst->compute_capacity
-        */
-       if (src->load * dst->compute_capacity * env->imbalance_pct >
-
-           dst->load * src->compute_capacity * 100)
-               return true;
-
-       return false;
-}
-
 static int task_numa_migrate(struct task_struct *p)
 {
        struct task_numa_env env = {
@@ -1764,8 +1735,7 @@ static int task_numa_migrate(struct task_struct *p)
        update_numa_stats(&env.dst_stats, env.dst_nid);
 
        /* Try to find a spot on the preferred nid. */
-       if (numa_has_capacity(&env))
-               task_numa_find_cpu(&env, taskimp, groupimp);
+       task_numa_find_cpu(&env, taskimp, groupimp);
 
        /*
         * Look at other nodes in these cases:
@@ -1795,8 +1765,7 @@ static int task_numa_migrate(struct task_struct *p)
                        env.dist = dist;
                        env.dst_nid = nid;
                        update_numa_stats(&env.dst_stats, env.dst_nid);
-                       if (numa_has_capacity(&env))
-                               task_numa_find_cpu(&env, taskimp, groupimp);
+                       task_numa_find_cpu(&env, taskimp, groupimp);
                }
        }
 
-- 
1.8.3.1

Reply via email to