From patchwork Sun Aug 27 00:37:27 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: SeongJae Park X-Patchwork-Id: 136965 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a59:a7d1:0:b0:3f2:4152:657d with SMTP id p17csp2612209vqm; Sat, 26 Aug 2023 18:02:17 -0700 (PDT) X-Google-Smtp-Source: AGHT+IFphxVTw8DdFN0hPOjJzjv7+JGzfiayKoYc9/ecEr6pZ1qEpTU2qMDc3VYc5mE51ZlE9K7H X-Received: by 2002:a17:907:7845:b0:9a1:8f6f:6873 with SMTP id lb5-20020a170907784500b009a18f6f6873mr14542941ejc.33.1693098137512; Sat, 26 Aug 2023 18:02:17 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1693098137; cv=none; d=google.com; s=arc-20160816; b=Sn4tHtIv9jYTpTwpmeLq/Wc2NCna0fQRpkc0jzQITWCgfwCB4MWlR/LpPRCT0e+B/N mcQJTgrZI1HckJO3RNBXxDbA9sk2zTdmece5tpdEyW8AHxH22bx4VJUeYfgI11JFYHUW 3dX9I+z71ywaUq1oUZ562UAL1L0ZQAEM4YhAF60zZLUEV3HMGi9MONMG5sN5cTvHX3Gp 9O01ofWwIK9YVaQpBat+PRPcwIlzR0OVAL6sh5ay3VcYAsFWO/JDHJRcthzfDUlONX3H uhBzY3Fp1ODYojkvQsJCUVHCbYQsRp3DD3QY6+sZaARqSw+3XqWLKAKflZIFh5ovkUgu rJKQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :message-id:date:subject:cc:to:from:dkim-signature; bh=TLz5pVGZjQKqBabfGtEv5P2bas6p9XJHMe7z2lyAl2w=; fh=HTgu+GxaUjOE8uwz5yYew3I5e0tZp8EF0T7tetbenMs=; b=zAu3o5WJ49z5F6k9Chn19BKIZa21WSzMHMViS1yu0OuDvF3cBaSFv8OmlHrbscCas9 ag8+Bd5TeawQdbKAL3DLOA7YL75AbYrlHKmqnwcUM4Wg6G/k+d5uwS2E20I+dl64Jpsg CM+BbeRddHfhaHwbmEKKTMWOky0Bz8pVVNxV0v/pPUKVHflULwDrsI3UdrpIrMMd+o0D Znrx1u4BWlIdmtZOFwz+PZEf4Al8ntH8vAh5FjpCRChPAj6KeB1TTtEaaH7gvqZQr6MC k8uyN81f9vxx/dWLVM4tA28vBvN71xbDpRqqSI3H6qjWrNTlZ/wSmjX4wjZlct2e8bxJ k3/A== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=Lo1DIY8g; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id qt16-20020a170906ecf000b00993a37b5e5csi2732379ejb.394.2023.08.26.18.01.10; Sat, 26 Aug 2023 18:02:17 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; dkim=pass header.i=@kernel.org header.s=k20201202 header.b=Lo1DIY8g; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229723AbjH0AiE (ORCPT + 99 others); Sat, 26 Aug 2023 20:38:04 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:59554 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229475AbjH0Ahh (ORCPT ); Sat, 26 Aug 2023 20:37:37 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 055EF109 for ; Sat, 26 Aug 2023 17:37:35 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id 7C7CF61DD3 for ; Sun, 27 Aug 2023 00:37:34 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 67961C433C8; Sun, 27 Aug 2023 00:37:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1693096653; bh=I6+EjuMhnMqA0A4SsaSLYkN9eETY2McANvNftJkI0dM=; h=From:To:Cc:Subject:Date:From; b=Lo1DIY8gtf7Th1wUHVHnXZJ3Je8HOVS7taKpYUE+ksqp1o3qqQAi8al7eP48M5cN+ GmG3Vqx/Bw8Cdmp8aIippxj1Or6nMYiNT/Eb2pC5XUgj011KZd4sLeKySRbWdSGTxG CA50E5u+hRNcxigWAd2byVptPZbNQElXZ9f7fNIJzGjx7qq0FjqfSA2ahVPDnkysmH gTrM485E4xuUdgtBCU8us98C6IYD424xWrIMR17/kc2ZQFETWgJ+IuQc67NHRGH7Zt WG8jgH85rjBPuhISdj4mqrlUtJ7aCQa++ElPL23Q0QJ63OGyc4tKTX5ZTNq/OPHaic ureLkSb8a+7LQ== From: SeongJae Park To: damon@lists.linux.dev Cc: SeongJae Park , Andrew Morton , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [RFC PATCH] mm/damon/core: use number of passed access sampling as a timer Date: Sun, 27 Aug 2023 00:37:27 +0000 Message-Id: <20230827003727.49369-1-sj@kernel.org> X-Mailer: git-send-email 2.25.1 MIME-Version: 1.0 X-Spam-Status: No, score=-2.1 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF, RCVD_IN_DNSWL_BLOCKED,SPF_HELO_NONE,SPF_PASS autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-getmail-retrieved-from-mailbox: INBOX X-GMAIL-THRID: 1775342072202721866 X-GMAIL-MSGID: 1775342072202721866 DAMON sleeps for sampling interval after each sampling, and check if it's time for further doing aggregation and ops updating using ktime_get_coarse_ts64() and baseline timestamps for the two periodic operations. That's for making the operations occur at deterministic timing. However, it turned out it could still result in indeterministic and even not-that-intuitive results. After all, timer functions, and especially sleep functions that DAMON uses to wait for specific timing, could contain some errors. Those errors are legal, so no problem. However, depending on such legal timing errors, the nr_accesses can be larger than aggregation interval divided by sampling interval. For example, with the default setting (5 ms sampling interval and 100 ms aggregation interval) we frequently show regions having nr_accesses larger than 20. Also, if the execution of a DAMOS scheme takes a long time, next aggregation could happen before enough number of samples are collected. Since access check sampling is the smallest unit work of DAMON, using the number of passed sampling intervals as the DAMON-internal timer can easily avoid these problems. That is, convert aggregation and ops update intervals to numbers of sampling intervals that need to be passed before those operations be executed, count the number of passed sampling intervals, and invoke the operations as soon as the specific amount of sampling intervals passed. Make the change. Signed-off-by: SeongJae Park --- include/linux/damon.h | 14 ++++++-- mm/damon/core.c | 84 +++++++++++++++++++------------------------ 2 files changed, 48 insertions(+), 50 deletions(-) diff --git a/include/linux/damon.h b/include/linux/damon.h index ab3089de1478..9a32b8fd0bd3 100644 --- a/include/linux/damon.h +++ b/include/linux/damon.h @@ -524,8 +524,18 @@ struct damon_ctx { struct damon_attrs attrs; /* private: internal use only */ - struct timespec64 last_aggregation; - struct timespec64 last_ops_update; + /* number of sample intervals that passed since this context started */ + unsigned long passed_sample_intervals; + /* + * number of sample intervals that should be passed before next + * aggregation + */ + unsigned long next_aggregation_sis; + /* + * number of sample intervals that should be passed before next ops + * update + */ + unsigned long next_ops_update_sis; /* public: */ struct task_struct *kdamond; diff --git a/mm/damon/core.c b/mm/damon/core.c index 988dc39e44b1..83af336bb0e6 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -456,8 +456,11 @@ struct damon_ctx *damon_new_ctx(void) ctx->attrs.aggr_interval = 100 * 1000; ctx->attrs.ops_update_interval = 60 * 1000 * 1000; - ktime_get_coarse_ts64(&ctx->last_aggregation); - ctx->last_ops_update = ctx->last_aggregation; + ctx->passed_sample_intervals = 0; + ctx->next_aggregation_sis = ctx->attrs.aggr_interval / + ctx->attrs.sample_interval; + ctx->next_ops_update_sis = ctx->attrs.ops_update_interval / + ctx->attrs.sample_interval; mutex_init(&ctx->kdamond_lock); @@ -577,6 +580,9 @@ static void damon_update_monitoring_results(struct damon_ctx *ctx, */ int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs) { + unsigned long sample_interval; + unsigned long remaining_interval_us; + if (attrs->min_nr_regions < 3) return -EINVAL; if (attrs->min_nr_regions > attrs->max_nr_regions) @@ -584,6 +590,20 @@ int damon_set_attrs(struct damon_ctx *ctx, struct damon_attrs *attrs) if (attrs->sample_interval > attrs->aggr_interval) return -EINVAL; + sample_interval = attrs->sample_interval ? attrs->sample_interval : 1; + + /* adjust next_aggregation_sis */ + remaining_interval_us = ctx->attrs.sample_interval * + (ctx->next_aggregation_sis - ctx->passed_sample_intervals); + ctx->next_aggregation_sis = ctx->passed_sample_intervals + + remaining_interval_us / sample_interval; + + /* adjust next_ops_update_sis */ + remaining_interval_us = ctx->attrs.sample_interval * + (ctx->next_ops_update_sis - ctx->passed_sample_intervals); + ctx->next_ops_update_sis = ctx->passed_sample_intervals + + remaining_interval_us / sample_interval; + damon_update_monitoring_results(ctx, attrs); ctx->attrs = *attrs; return 0; @@ -757,38 +777,6 @@ int damon_stop(struct damon_ctx **ctxs, int nr_ctxs) return err; } -/* - * damon_check_reset_time_interval() - Check if a time interval is elapsed. - * @baseline: the time to check whether the interval has elapsed since - * @interval: the time interval (microseconds) - * - * See whether the given time interval has passed since the given baseline - * time. If so, it also updates the baseline to current time for next check. - * - * Return: true if the time interval has passed, or false otherwise. - */ -static bool damon_check_reset_time_interval(struct timespec64 *baseline, - unsigned long interval) -{ - struct timespec64 now; - - ktime_get_coarse_ts64(&now); - if ((timespec64_to_ns(&now) - timespec64_to_ns(baseline)) < - interval * 1000) - return false; - *baseline = now; - return true; -} - -/* - * Check whether it is time to flush the aggregated information - */ -static bool kdamond_aggregate_interval_passed(struct damon_ctx *ctx) -{ - return damon_check_reset_time_interval(&ctx->last_aggregation, - ctx->attrs.aggr_interval); -} - /* * Reset the aggregated monitoring results ('nr_accesses' of each region). */ @@ -1292,18 +1280,6 @@ static void kdamond_split_regions(struct damon_ctx *ctx) last_nr_regions = nr_regions; } -/* - * Check whether it is time to check and apply the operations-related data - * structures. - * - * Returns true if it is. - */ -static bool kdamond_need_update_operations(struct damon_ctx *ctx) -{ - return damon_check_reset_time_interval(&ctx->last_ops_update, - ctx->attrs.ops_update_interval); -} - /* * Check whether current monitoring should be stopped * @@ -1436,6 +1412,8 @@ static int kdamond_fn(void *data) sz_limit = damon_region_sz_limit(ctx); while (!kdamond_need_stop(ctx)) { + unsigned long sample_interval; + if (kdamond_wait_activation(ctx)) break; @@ -1446,11 +1424,17 @@ static int kdamond_fn(void *data) break; kdamond_usleep(ctx->attrs.sample_interval); + ctx->passed_sample_intervals++; if (ctx->ops.check_accesses) max_nr_accesses = ctx->ops.check_accesses(ctx); - if (kdamond_aggregate_interval_passed(ctx)) { + sample_interval = ctx->attrs.sample_interval ? + ctx->attrs.sample_interval : 1; + if (ctx->passed_sample_intervals == + ctx->next_aggregation_sis) { + ctx->next_aggregation_sis += + ctx->attrs.aggr_interval / sample_interval; kdamond_merge_regions(ctx, max_nr_accesses / 10, sz_limit); @@ -1465,7 +1449,11 @@ static int kdamond_fn(void *data) ctx->ops.reset_aggregated(ctx); } - if (kdamond_need_update_operations(ctx)) { + if (ctx->passed_sample_intervals == + ctx->next_ops_update_sis) { + ctx->next_ops_update_sis += + ctx->attrs.ops_update_interval / + sample_interval; if (ctx->ops.update) ctx->ops.update(ctx); sz_limit = damon_region_sz_limit(ctx);