From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pl1-f173.google.com (mail-pl1-f173.google.com [209.85.214.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3313530C371 for ; Fri, 31 Jul 2026 04:26:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.173 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785472010; cv=none; b=AWbSNFIRmxmaJ0m6ZnLHPFpBEvgPI6NJERQzjfkF2YDkUTvxSSnvVnHGcd+l7GX4hfEQTuK5XS1/osLJB1Oc+D2l32URWjv8p4TVVfXic6Zxm0oZ8bbRfjW3AchgVPDlagPxy6uU7X/SXxfw3jB74o33I98NeTezYNpVaprWY1s= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785472010; c=relaxed/simple; bh=jjcAxGJTKQOG/eellDIgMUp2SBdTJdIAPqwP7yBTyCk=; h=Date:From:To:cc:Subject:In-Reply-To:Message-ID:References: MIME-Version:Content-Type; b=VXKdFGu6OTMlOVHwT2t9YKEKq1sAcfrDWhTNjF7SsdT76O3TOGyeSySjCom0VBdxKh1C3iRZoEG/sEu5NsTq3lXVwYX3CcU+LryirskXhSRrHThP4jqA29meLaZYMh8WT02EjB2cfwuvNrEjqnfLmO6Wi/F9B696JhDtK079gEw= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=tb6W2F27; arc=none smtp.client-ip=209.85.214.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="tb6W2F27" Received: by mail-pl1-f173.google.com with SMTP id d9443c01a7336-2ccdf36f63dso74775ad.0 for ; Thu, 30 Jul 2026 21:26:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1785472007; x=1786076807; darn=vger.kernel.org; h=content-type:mime-version:references:message-id:in-reply-to:subject :cc:to:from:date:from:to:cc:subject:date:message-id:reply-to :content-type; bh=VtWsT75VcRdayV6UMHe+iQ7OyarIoZiTjxseMwp8AOM=; b=tb6W2F27u2ErAFpKMl/v/AS7XLnDSl2Mk0C6Y0errd1mBZAGzbV9os/24fwoXSiySJ bTfowY43tC0zRxgL9TzuCCkgculTlYla22Lv7oLGiRmtHLoqG09nG/f6cwF8p+zmfr9e 7JezzJvUWEf5YIQq5W/hIyM/DXJ1LPqYC0tqkY8Qd0sO/sxv6B0zwIFm1OdqcJABMoQ5 l77PxXjM/hUWmPorvPj/vtLg+sBEWa7Ct2KLz6D38IeRn10XeMcZWWC+/t7kfmSjgxGc w50/z61iJA5dBZfBHObQ6QF2Vk8qrDWwv7zh/rrgdmwFFb3YECuPGMrJGtNmVweMYisl VonQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1785472007; x=1786076807; h=content-type:mime-version:references:message-id:in-reply-to:subject :cc:to:from:date:x-gm-gg:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to:content-type; bh=VtWsT75VcRdayV6UMHe+iQ7OyarIoZiTjxseMwp8AOM=; b=bu1qpNGZTGen/f1IfgRKY6DfuALFoyiqP7dEwk7mogj7hj7qGXKg4JhRJFOHcFuEeD YzyA0S6Dv8UyAqWay8LgRJk3YLGuCk2gmrNlLfbcZFNrTus8rgdTv6rJGJG6S8uWRwNO aHpsIUC8LrgXpr9AItQ2Fe81EBT+I2uRLyTJNG7NygtW94JAlqfskikafpkIHb1tAEc9 eycvlhiUyqrf5GoC0hSO74y0x3cLSQvHKbb6Cc4vTgaxqRaDRVtTSZ3ujZHOVK8T94ih +ccSLvStA8L5G9D3pcHr5N1IeaDm4Izat3NqeIZMCPzUnRggaiov2xYcgFauuIu7Nej3 VK5A== X-Forwarded-Encrypted: i=1; AHgh+Rrllk5PJgJjByTO7jaZ5QVExB/wRUzyQmFdhKkdbiAJ9cTValOE9Lnyc1f1/r1GEbAZYiR5m05qEjTsEwc=@vger.kernel.org X-Gm-Message-State: AOJu0YzXY13pMYg8Oy169FiJOvcrGSfHrjBHVzf8OuWrwoRLBF3ReJfg tz32z3L/T4SAJMuu6Q6L7mxhhV9LQghPYLkW1vahECM5KAyt+bdqFQbOx5Hn2WJVfQ== X-Gm-Gg: AR+sD107FUIgO6egT4/75M71Yo0moBB3CuLwDSuAHw0ltHopbyOuNMZrdj6TzezY5Ia YNRO7YoEDO2guiaUhDGqkQAoS64ERdNGV83Zr0jS58yl+YoE1kjadkLIMH7CtFYUagSJm2N3B/w FUXQ1So0P7OWSxj5/8FWWYAJt9aUhmDREGXc8k8b3S9NVyO7F9HLHPufV62Ew0YVr7bEABzoBZu 1GSorZnn1j6/E9NxC+fxOt4EcRsasUAvZnPhbdWB9ZOL8WWvrZxXsRplpBXH1Nvu39cijPjDCCc NuRCB/PchJC2jzaIVXOUuoSBeXqKoi1A8P6C7ZmocGdpdwagOw3q7LwFU+Wcr4ESTLXESMJWe0R 9KKQxBLxsnbGY4zUBdzch+qVH0vXYxibEnHJROpGEYm+iOjyxoFTvg1+7G10q8Hmy5W9fnWQkv1 zIpkfiNGocxBplO4m4+FeF+22hb8T+Pd32Q+fJf7SR4Y6qycCEBDNNiDKvmHJ0TJid+X9sc3zTZ /jQyN1/Cjwoo09MsHYfhor2QnxoghGMe0p3qU+LN8gFuuav2CKVxVQJ28B/gXvg6iuOKcniMrYB dVc1Gxafks7/Jx9V2PB0sBq7Tw== X-Received: by 2002:a17:902:d04c:b0:2ca:be81:b469 with SMTP id d9443c01a7336-2d046b045femr923285ad.0.1785472006664; Thu, 30 Jul 2026 21:26:46 -0700 (PDT) Received: from [2a00:79e0:2eb4:9:a061:9e19:11a8:45b6] ([2a00:79e0:2eb4:9:a061:9e19:11a8:45b6]) by smtp.gmail.com with ESMTPSA id 98e67ed59e1d1-38fb2f65459sm97995a91.6.2026.07.30.21.26.45 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 30 Jul 2026 21:26:46 -0700 (PDT) Date: Thu, 30 Jul 2026 21:26:45 -0700 (PDT) From: David Rientjes To: Andrew Morton , David Hildenbrand , Christoph Lameter cc: Vlastimil Babka , Mathieu Desnoyers , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [patch 2/3] lib: test_slab: add synthetic benchmark for slab In-Reply-To: Message-ID: <430d42a4-6cbf-405c-8102-5ab5b930f906@google.com> References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Christoph Lameter Add a synthetic benchmark that can be used to measure performance of the slab allocator. This is used to analyze any improvements or regressions in slab allocation or freeing functions. The test is run by inserting the module: modprobe test_slab The module insertion will always fail so that it is automatically unloaded; the results are in the kernel log. Signed-off-by: Christoph Lameter Signed-off-by: David Rientjes --- lib/Kconfig.debug | 10 ++ lib/Makefile | 1 + lib/test_slab.c | 375 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 386 insertions(+) create mode 100644 lib/test_slab.c diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 6d743c8e27fa..a7e738e6afa8 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -3437,6 +3437,16 @@ config TEST_VMSTAT If unsure, say N. +config TEST_SLAB + tristate "Slab Allocator Benchmark" + default n + help + A synthetic benchmark that measures slab allocator performance, + specifically focusing on single thread testing by repeatedly + allocating then freeing memory and concurrent or remote allocations. + + If unsure, say N. + config RATELIMIT_KUNIT_TEST tristate "KUnit Test for correctness and stress of ratelimit" if !KUNIT_ALL_TESTS depends on KUNIT diff --git a/lib/Makefile b/lib/Makefile index 5f3ebabd0224..fe54690d130c 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -106,6 +106,7 @@ obj-$(CONFIG_TEST_REF_TRACKER) += test_ref_tracker.o obj-$(CONFIG_TEST_OBJPOOL) += test_objpool.o obj-$(CONFIG_TEST_KEXEC_HANDOVER) += test_kho.o obj-$(CONFIG_TEST_VMSTAT) += test_vmstat.o +obj-$(CONFIG_TEST_SLAB) += test_slab.o obj-$(CONFIG_TEST_FPU) += test_fpu.o test_fpu-y := test_fpu_glue.o test_fpu_impl.o diff --git a/lib/test_slab.c b/lib/test_slab.c new file mode 100644 index 000000000000..a43b0d73daab --- /dev/null +++ b/lib/test_slab.c @@ -0,0 +1,375 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Test module for synthetic in kernel slab allocator testing. + * + * The test is triggered by loading the module (which will fail). + * + * (C) 2009 Linux Foundation + */ + +#include +#include +#include +#include +#include +#include + +#define TEST_COUNT 10000 + +#ifdef CONFIG_SMP +#include +#include +#include +#include + +static struct test_struct { + struct task_struct *task; + int cpu; + int size; + int count; + void **v; + void (*test_p1)(struct test_struct *t); + void (*test_p2)(struct test_struct *t); + unsigned long start1; + unsigned long stop1; + unsigned long start2; + unsigned long stop2; +} test[NR_CPUS]; + +/* + * Allocate TEST_COUNT objects on cpus > 0 and then all the + * objects later on cpu 0 + */ +static void remote_free_test_p1(struct test_struct *t) +{ + int i; + + /* Perform no allocations on cpu 0 */ + for (i = 0; i < t->count; i++) { + u8 *p; + + if (smp_processor_id()) { + p = kmalloc(t->size, GFP_KERNEL); + /* Use object */ + *p = 17; + } else + p = NULL; + t->v[i] = p; + } +} + +static void remote_free_test_p2(struct test_struct *t) +{ + int i; + int cpu; + + /* All frees are completed on cpu zero */ + if (smp_processor_id()) + return; + + for_each_online_cpu(cpu) + for (i = 0; i < t->count; i++) { + u8 *p = test[cpu].v[i]; + + if (!p) + continue; + + *p = 16; + kfree(p); + } +} + +/* + * Allocate TEST_COUNT objects on cpu 0 and free them immediately on the + * other processors. + */ +static void alloc_n_free_test_p1(struct test_struct *t) +{ + int i; + int cpu; + char *p; + + if (smp_processor_id()) { + /* Consumer */ + for (i = 0; i < t->count / num_online_cpus(); i++) { + do { + p = t->v[i]; + if (!p) + cpu_relax(); + else + *p = 17; + } while (!p); + kfree(p); + t->v[i] = NULL; + } + return; + } + /* Producer */ + for (i = 0; i < t->count; i++) { + for_each_online_cpu(cpu) { + if (cpu) { + p = kmalloc(t->size, GFP_KERNEL); + /* Use object */ + *p = 17; + test[cpu].v[i] = p; + } + } + } +} + +/* + * Allocate TEST_COUNT objects and later free them all again + */ +static void kmalloc_alloc_then_free_test_p1(struct test_struct *t) +{ + int i; + + for (i = 0; i < t->count; i++) { + u8 *p = kmalloc(t->size, GFP_KERNEL); + + *p = 14; + t->v[i] = p; + } +} + +static void kmalloc_alloc_then_free_test_p2(struct test_struct *t) +{ + int i; + + for (i = 0; i < t->count; i++) { + u8 *p = t->v[i]; + + *p = 13; + kfree(p); + } +} + +/* + * Allocate TEST_COUNT objects. Free them immediately. + */ +static void kmalloc_alloc_free_test_p1(struct test_struct *t) +{ + int i; + + for (i = 0; i < TEST_COUNT; i++) { + u8 *p = kmalloc(t->size, GFP_KERNEL); + + *p = 12; + kfree(p); + } +} + +static atomic_t tests_running; +static atomic_t phase1_complete; +static DECLARE_COMPLETION(completion1); +static DECLARE_COMPLETION(completion2); +static DECLARE_COMPLETION(completion3); + +static int test_func(void *private) +{ + struct test_struct *t = private; + cpumask_t newmask = CPU_MASK_NONE; + + cpumask_set_cpu(t->cpu, &newmask); + set_cpus_allowed_ptr(current, &newmask); + t->v = kcalloc(t->count, sizeof(void *), GFP_KERNEL); + + atomic_inc(&tests_running); + wait_for_completion(&completion1); + t->start1 = get_cycles(); + t->test_p1(t); + t->stop1 = get_cycles(); + atomic_inc(&phase1_complete); + wait_for_completion(&completion2); + t->start2 = get_cycles(); + if (t->test_p2) + t->test_p2(t); + t->stop2 = get_cycles(); + atomic_dec(&tests_running); + wait_for_completion(&completion3); + kfree(t->v); + while (!kthread_should_stop()) { + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(10); + } + return 0; +} + +static void do_concurrent_test(void (*p1)(struct test_struct *), + void (*p2)(struct test_struct *), + int size, const char *name) +{ + int cpu; + unsigned long time1 = 0; + unsigned long time2 = 0; + unsigned long sum1 = 0; + unsigned long sum2 = 0; + + atomic_set(&tests_running, 0); + atomic_set(&phase1_complete, 0); + init_completion(&completion1); + init_completion(&completion2); + init_completion(&completion3); + + for_each_online_cpu(cpu) { + struct test_struct *t = &test[cpu]; + + t->cpu = cpu; + t->count = TEST_COUNT; + t->test_p1 = p1; + t->test_p2 = p2; + t->size = size; + t->task = kthread_run(test_func, t, "test%d", cpu); + if (IS_ERR(t->task)) { + pr_err("Failed to start test func\n"); + return; + } + } + + /* Wait till all processes are running */ + while (atomic_read(&tests_running) < num_online_cpus()) { + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(10); + } + complete_all(&completion1); + + /* Wait till all processes have completed phase 1 */ + while (atomic_read(&phase1_complete) < num_online_cpus()) { + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(10); + } + complete_all(&completion2); + + /* Wait till all processes have completed phase 2 */ + while (atomic_read(&tests_running)) { + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(10); + } + complete_all(&completion3); + + for_each_online_cpu(cpu) + kthread_stop(test[cpu].task); + + pr_alert("%s(%d):", name, size); + for_each_online_cpu(cpu) { + struct test_struct *t = &test[cpu]; + + time1 = t->stop1 - t->start1; + time2 = t->stop2 - t->start2; + sum1 += time1; + sum2 += time2; + pr_cont(" %d=%lu", cpu, time1 / TEST_COUNT); + if (p2) + pr_cont("/%lu", time2 / TEST_COUNT); + } + pr_cont(" Average=%lu", sum1 / num_online_cpus() / TEST_COUNT); + if (p2) + pr_cont("/%lu", sum2 / num_online_cpus() / TEST_COUNT); + pr_cont("\n"); + schedule_timeout(200); +} +#endif + +static int slab_test_init(void) +{ + void **v = kmalloc_array(TEST_COUNT, sizeof(void *), GFP_KERNEL); + unsigned int i; + cycles_t time1, time2, time; + int rem; + int size; + + pr_alert("test init\n"); + + pr_alert("Single thread testing\n"); + pr_alert("=====================\n"); + pr_alert("1. Kmalloc: Repeatedly allocate then free test\n"); + for (size = 8; size <= PAGE_SIZE << 2; size <<= 1) { + time1 = get_cycles(); + for (i = 0; i < TEST_COUNT; i++) { + u8 *p = kmalloc(size, GFP_KERNEL); + + *p = 22; + v[i] = p; + } + time2 = get_cycles(); + time = time2 - time1; + + pr_alert("%i times kmalloc(%d) ", i, size); + time = div_u64_rem(time, TEST_COUNT, &rem); + pr_cont("-> %llu cycles ", (unsigned long long) time); + + time1 = get_cycles(); + for (i = 0; i < TEST_COUNT; i++) { + u8 *p = v[i]; + + *p = 23; + kfree(p); + } + time2 = get_cycles(); + time = time2 - time1; + + pr_cont("kfree "); + time = div_u64_rem(time, TEST_COUNT, &rem); + pr_cont("-> %llu cycles\n", (unsigned long long) time); + } + + pr_alert("2. Kmalloc: alloc/free test\n"); + for (size = 8; size <= PAGE_SIZE << 2; size <<= 1) { + time1 = get_cycles(); + for (i = 0; i < TEST_COUNT; i++) { + u8 *p = kmalloc(size, GFP_KERNEL); + + kfree(p); + } + time2 = get_cycles(); + time = time2 - time1; + + pr_alert("%i times kmalloc(%d)/kfree ", i, size); + time = div_u64_rem(time, TEST_COUNT, &rem); + pr_cont("-> %llu cycles\n", (unsigned long long) time); + } + kfree(v); +#ifdef CONFIG_SMP + pr_info("Concurrent allocs\n"); + pr_info("=================\n"); + for (i = 3; i <= PAGE_SHIFT; i++) { + do_concurrent_test(kmalloc_alloc_then_free_test_p1, + kmalloc_alloc_then_free_test_p2, + 1 << i, "Kmalloc N*alloc N*free"); + } + for (i = 3; i <= PAGE_SHIFT; i++) { + do_concurrent_test(kmalloc_alloc_free_test_p1, NULL, + 1 << i, "Kmalloc N*(alloc free)"); + } + + pr_info("Remote free test\n"); + pr_info("================\n"); + for (i = 3; i <= PAGE_SHIFT; i++) { + do_concurrent_test(remote_free_test_p1, + remote_free_test_p2, + 1 << i, "N*remote free"); + } + + pr_info("1 alloc N free test\n"); + pr_info("===================\n"); + for (i = 3; i <= PAGE_SHIFT; i++) { + do_concurrent_test(alloc_n_free_test_p1, + NULL, + 1 << i, "1 alloc N free"); + } + +#endif + return -EAGAIN; /* Fail will directly unload the module */ +} + +static void slab_test_exit(void) +{ + pr_alert("test exit\n"); +} + +module_init(slab_test_init) +module_exit(slab_test_exit) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christoph Lameter and Mathieu Desnoyers"); +MODULE_DESCRIPTION("SLAB test");