From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752740AbeFDKB2 (ORCPT ); Mon, 4 Jun 2018 06:01:28 -0400 Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:43080 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752680AbeFDKBZ (ORCPT ); Mon, 4 Jun 2018 06:01:25 -0400 From: Srikar Dronamraju To: Ingo Molnar , Peter Zijlstra Cc: LKML , Mel Gorman , Rik van Riel , Srikar Dronamraju , Thomas Gleixner Subject: [PATCH 12/19] sched:numa Remove numa_has_capacity Date: Mon, 4 Jun 2018 15:30:21 +0530 X-Mailer: git-send-email 2.7.4 In-Reply-To: <1528106428-19992-1-git-send-email-srikar@linux.vnet.ibm.com> References: <1528106428-19992-1-git-send-email-srikar@linux.vnet.ibm.com> X-TM-AS-GCONF: 00 x-cbid: 18060410-0016-0000-0000-000001D7739B X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 18060410-0017-0000-0000-0000322A0352 Message-Id: <1528106428-19992-13-git-send-email-srikar@linux.vnet.ibm.com> X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2018-06-04_07:,, signatures=0 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 priorityscore=1501 malwarescore=0 suspectscore=2 phishscore=0 bulkscore=0 spamscore=0 clxscore=1015 lowpriorityscore=0 mlxscore=0 impostorscore=0 mlxlogscore=999 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1805220000 definitions=main-1806040123 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 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 --- 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