From nobody Thu Apr 2 13:15:33 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id D85C6C04A95 for ; Sun, 25 Sep 2022 21:27:31 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233007AbiIYV13 (ORCPT ); Sun, 25 Sep 2022 17:27:29 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46810 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232716AbiIYV1U (ORCPT ); Sun, 25 Sep 2022 17:27:20 -0400 Received: from mail-qv1-xf41.google.com (mail-qv1-xf41.google.com [IPv6:2607:f8b0:4864:20::f41]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 18D952A26B for ; Sun, 25 Sep 2022 14:27:19 -0700 (PDT) Received: by mail-qv1-xf41.google.com with SMTP id c6so3315883qvn.6 for ; Sun, 25 Sep 2022 14:27:19 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date; bh=O3V0nPhtlK3TXA87oz1HqqVg77ploxiQ5BsPzqU7s6E=; b=PCIFTRmajDOtysXry44eI1u7u25E7Vbm6nwPh8iESWSBTaH8AsczQf+YWdDDfXoXBL HHwsT+B/8C3tvzMOEGqcnd4NJnmjteRgMJH3N6GHY8vCE00xeDE3nI2aTF/NRHcSh1u6 3KyKtEKlonOFHPS+506ttNgl7Y4hNzvfAq1Zt1DFyjH/b6OXgyzU+Wb4BXNJY+ne/j+f jWsOpYY7n4wfItI6f/u6WA8lsQM9kXyQVk7HLz/xgGAZwN4vw4CNpDM2wmvRn82YDjSL FfN0yblZwvwYk19Kx4MG8q/+QssygzxLd+YPTG8RJJn35TKOVpFthc/k7TACnk+yomZW 3Dpw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date; bh=O3V0nPhtlK3TXA87oz1HqqVg77ploxiQ5BsPzqU7s6E=; b=RyKR3F6t0aVIUrUNjJap5Zd5yib2w+i025PQkY3a1zj/16B92EYnQre9Sw+y/4nTBE ocv9RaKUr7vPQIwsOjfsgiiNkUhezdiuG+j78uKY7EVXTttDzYNPPPvNZwyazvCBq92B 9eiYs5vf0fYpOEugIK+1s7NVXYzK7o7psIPUVF7fdAv4kayao+U/19mQHzgq2ZXjatnU TLhA+ZiNpm8Ebj9JoXMZ2ao5Uv914HUQhRdQ+NwHVwZce4aMHLu90AK/1D/y81YIWE5I 55Bb8gju5oFe6ix/03Ms0R1UTCd7QsmS9R+jIOXOM9G1K57ii3JCcAyadAAHkERQhe8u GqWg== X-Gm-Message-State: ACrzQf3GFxphSVZ61n4kR4aGV2myYE0jrKv/Z+1uMv75YJhjNsnt5+XJ VU/arzT0MqC3rT83KG2j0NM= X-Google-Smtp-Source: AMsMyM4tIek80imNXPhVyo9v9D4fFIhgOmpttXfFHXSIXFvBM1ZdS7UJjmpujwj90DdN/liDVOPmxw== X-Received: by 2002:a05:6214:3011:b0:4ad:82d6:d579 with SMTP id ke17-20020a056214301100b004ad82d6d579mr14428920qvb.37.1664141238083; Sun, 25 Sep 2022 14:27:18 -0700 (PDT) Received: from sophie ([185.156.46.189]) by smtp.gmail.com with ESMTPSA id c10-20020a05620a11aa00b006bb83c2be40sm9921692qkk.59.2022.09.25.14.27.17 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sun, 25 Sep 2022 14:27:17 -0700 (PDT) From: Rebecca Mckeever To: Mike Rapoport , linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: David Hildenbrand , Rebecca Mckeever Subject: [PATCH 2/5] memblock tests: add top-down NUMA tests for memblock_alloc_exact_nid_raw Date: Sun, 25 Sep 2022 16:26:53 -0500 Message-Id: <2cc4c0972a472579da227007fc160d69e4f33727.1664138929.git.remckee0@gmail.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add tests for memblock_alloc_exact_nid_raw() where the simulated physical memory is set up with multiple NUMA nodes. Additionally, all of these tests set nid !=3D NUMA_NO_NODE. These tests are run with a top-down allocation direction. The tested scenarios are: Range unrestricted: - region can be allocated in the specific node requested: + there are no previously reserved regions + the requested node is partially reserved but has enough space Range restricted: - region can be allocated in the specific node requested after dropping min_addr: + range partially overlaps with two different nodes, where the first node is the requested node + range partially overlaps with two different nodes, where the requested node ends before min_addr + range overlaps with multiple nodes along node boundaries, and the requested node ends before min_addr Signed-off-by: Rebecca Mckeever --- .../memblock/tests/alloc_exact_nid_api.c | 344 ++++++++++++++++++ .../memblock/tests/alloc_exact_nid_api.h | 16 + 2 files changed, 360 insertions(+) diff --git a/tools/testing/memblock/tests/alloc_exact_nid_api.c b/tools/tes= ting/memblock/tests/alloc_exact_nid_api.c index 452aba1f19db..48c1b97a3f98 100644 --- a/tools/testing/memblock/tests/alloc_exact_nid_api.c +++ b/tools/testing/memblock/tests/alloc_exact_nid_api.c @@ -3,6 +3,21 @@ =20 #define FUNC_NAME "memblock_alloc_exact_nid_raw" =20 +/* + * contains the fraction of MEM_SIZE contained in each node in basis point + * units (one hundredth of 1% or 1/10000) + */ +static const unsigned int node_fractions[] =3D { + 2500, /* 1/4 */ + 625, /* 1/16 */ + 1250, /* 1/8 */ + 1250, /* 1/8 */ + 625, /* 1/16 */ + 625, /* 1/16 */ + 2500, /* 1/4 */ + 625, /* 1/16 */ +}; + /* * A simple test that tries to allocate a memory region within min_addr and * max_addr range: @@ -1190,6 +1205,334 @@ static int memblock_alloc_exact_nid_range_checks(vo= id) return 0; } =20 +/* + * A test that tries to allocate a memory region in a specific NUMA node t= hat + * has enough memory to allocate a region of the requested size. + * Expect to allocate an aligned region at the end of the requested node. + */ +static int alloc_exact_nid_top_down_numa_simple_check(void) +{ + int nid_req =3D 3; + struct memblock_region *new_rgn =3D &memblock.reserved.regions[0]; + struct memblock_region *req_node =3D &memblock.memory.regions[nid_req]; + void *allocated_ptr =3D NULL; + phys_addr_t size; + phys_addr_t min_addr; + phys_addr_t max_addr; + + PREFIX_PUSH(); + setup_numa_memblock(node_fractions); + + ASSERT_LE(SZ_4, req_node->size); + size =3D req_node->size / SZ_4; + min_addr =3D memblock_start_of_DRAM(); + max_addr =3D memblock_end_of_DRAM(); + + allocated_ptr =3D memblock_alloc_exact_nid_raw(size, SMP_CACHE_BYTES, + min_addr, max_addr, + nid_req); + + ASSERT_NE(allocated_ptr, NULL); + ASSERT_MEM_NE(allocated_ptr, 0, size); + + ASSERT_EQ(new_rgn->size, size); + ASSERT_EQ(new_rgn->base, region_end(req_node) - size); + ASSERT_LE(req_node->base, new_rgn->base); + + ASSERT_EQ(memblock.reserved.cnt, 1); + ASSERT_EQ(memblock.reserved.total_size, size); + + test_pass_pop(); + + return 0; +} + +/* + * A test that tries to allocate a memory region in a specific NUMA node t= hat + * is partially reserved but has enough memory for the allocated region: + * + * | +---------------------------------------+ | + * | | requested | | + * +-----------+---------------------------------------+----------+ + * + * | +------------------+ +-----+ | + * | | reserved | | new | | + * +-----------+------------------+--------------+-----+----------+ + * + * Expect to allocate an aligned region at the end of the requested node. = The + * region count and total size get updated. + */ +static int alloc_exact_nid_top_down_numa_part_reserved_check(void) +{ + int nid_req =3D 4; + struct memblock_region *new_rgn =3D &memblock.reserved.regions[1]; + struct memblock_region *req_node =3D &memblock.memory.regions[nid_req]; + void *allocated_ptr =3D NULL; + struct region r1; + phys_addr_t size; + phys_addr_t min_addr; + phys_addr_t max_addr; + + PREFIX_PUSH(); + setup_numa_memblock(node_fractions); + + ASSERT_LE(SZ_8, req_node->size); + r1.base =3D req_node->base; + r1.size =3D req_node->size / SZ_2; + size =3D r1.size / SZ_4; + min_addr =3D memblock_start_of_DRAM(); + max_addr =3D memblock_end_of_DRAM(); + + memblock_reserve(r1.base, r1.size); + allocated_ptr =3D memblock_alloc_exact_nid_raw(size, SMP_CACHE_BYTES, + min_addr, max_addr, + nid_req); + + ASSERT_NE(allocated_ptr, NULL); + ASSERT_MEM_NE(allocated_ptr, 0, size); + + ASSERT_EQ(new_rgn->size, size); + ASSERT_EQ(new_rgn->base, region_end(req_node) - size); + ASSERT_LE(req_node->base, new_rgn->base); + + ASSERT_EQ(memblock.reserved.cnt, 2); + ASSERT_EQ(memblock.reserved.total_size, size + r1.size); + + test_pass_pop(); + + return 0; +} + +/* + * A test that tries to allocate a memory region that spans over the min_a= ddr + * and max_addr range and overlaps with two different nodes, where the fir= st + * node is the requested node: + * + * min_addr + * | max_addr + * | | + * v v + * | +-----------------------+-----------+ | + * | | requested | node3 | | + * +-----------+-----------------------+-----------+--------------+ + * + + + * | +-----------+ | + * | | rgn | | + * +-----------------------+-----------+--------------------------+ + * + * Expect to drop the lower limit and allocate a memory region that ends at + * the end of the requested node. + */ +static int alloc_exact_nid_top_down_numa_split_range_low_check(void) +{ + int nid_req =3D 2; + struct memblock_region *new_rgn =3D &memblock.reserved.regions[0]; + struct memblock_region *req_node =3D &memblock.memory.regions[nid_req]; + void *allocated_ptr =3D NULL; + phys_addr_t size =3D SZ_512; + phys_addr_t min_addr; + phys_addr_t max_addr; + phys_addr_t req_node_end; + + PREFIX_PUSH(); + setup_numa_memblock(node_fractions); + + req_node_end =3D region_end(req_node); + min_addr =3D req_node_end - SZ_256; + max_addr =3D min_addr + size; + + allocated_ptr =3D memblock_alloc_exact_nid_raw(size, SMP_CACHE_BYTES, + min_addr, max_addr, + nid_req); + + ASSERT_NE(allocated_ptr, NULL); + ASSERT_MEM_NE(allocated_ptr, 0, size); + + ASSERT_EQ(new_rgn->size, size); + ASSERT_EQ(new_rgn->base, req_node_end - size); + ASSERT_LE(req_node->base, new_rgn->base); + + ASSERT_EQ(memblock.reserved.cnt, 1); + ASSERT_EQ(memblock.reserved.total_size, size); + + test_pass_pop(); + + return 0; +} + +/* + * A test that tries to allocate a memory region that spans over the min_a= ddr + * and max_addr range and overlaps with two different nodes, where the req= uested + * node ends before min_addr: + * + * min_addr + * | max_addr + * | | + * v v + * | +---------------+ +-------------+---------+ | + * | | requested | | node1 | node2 | | + * +----+---------------+--------+-------------+---------+----------+ + * + + + * | +---------+ | + * | | rgn | | + * +----------+---------+-------------------------------------------+ + * + * Expect to drop the lower limit and allocate a memory region that ends at + * the end of the requested node. + */ +static int alloc_exact_nid_top_down_numa_no_overlap_split_check(void) +{ + int nid_req =3D 2; + struct memblock_region *new_rgn =3D &memblock.reserved.regions[0]; + struct memblock_region *req_node =3D &memblock.memory.regions[nid_req]; + struct memblock_region *node2 =3D &memblock.memory.regions[6]; + void *allocated_ptr =3D NULL; + phys_addr_t size; + phys_addr_t min_addr; + phys_addr_t max_addr; + + PREFIX_PUSH(); + setup_numa_memblock(node_fractions); + + size =3D SZ_512; + min_addr =3D node2->base - SZ_256; + max_addr =3D min_addr + size; + + allocated_ptr =3D memblock_alloc_exact_nid_raw(size, SMP_CACHE_BYTES, + min_addr, max_addr, + nid_req); + + ASSERT_NE(allocated_ptr, NULL); + ASSERT_MEM_NE(allocated_ptr, 0, size); + + ASSERT_EQ(new_rgn->size, size); + ASSERT_EQ(new_rgn->base, region_end(req_node) - size); + ASSERT_LE(req_node->base, new_rgn->base); + + ASSERT_EQ(memblock.reserved.cnt, 1); + ASSERT_EQ(memblock.reserved.total_size, size); + + test_pass_pop(); + + return 0; +} + +/* + * A test that tries to allocate memory within min_addr and max_add range = when + * the requested node and the range do not overlap, and requested node ends + * before min_addr. The range overlaps with multiple nodes along node + * boundaries: + * + * min_addr + * | max_addr + * | | + * v v + * |-----------+ +----------+----...----+----------+ | + * | requested | | min node | ... | max node | | + * +-----------+-----------+----------+----...----+----------+------+ + * + + + * | +-----+ | + * | | rgn | | + * +-----+-----+----------------------------------------------------+ + * + * Expect to drop the lower limit and allocate a memory region that ends at + * the end of the requested node. + */ +static int alloc_exact_nid_top_down_numa_no_overlap_low_check(void) +{ + int nid_req =3D 0; + struct memblock_region *new_rgn =3D &memblock.reserved.regions[0]; + struct memblock_region *req_node =3D &memblock.memory.regions[nid_req]; + struct memblock_region *min_node =3D &memblock.memory.regions[2]; + struct memblock_region *max_node =3D &memblock.memory.regions[5]; + void *allocated_ptr =3D NULL; + phys_addr_t size =3D SZ_64; + phys_addr_t max_addr; + phys_addr_t min_addr; + + PREFIX_PUSH(); + setup_numa_memblock(node_fractions); + + min_addr =3D min_node->base; + max_addr =3D region_end(max_node); + + allocated_ptr =3D memblock_alloc_exact_nid_raw(size, SMP_CACHE_BYTES, + min_addr, max_addr, + nid_req); + + ASSERT_NE(allocated_ptr, NULL); + ASSERT_MEM_NE(allocated_ptr, 0, size); + + ASSERT_EQ(new_rgn->size, size); + ASSERT_EQ(new_rgn->base, region_end(req_node) - size); + + ASSERT_EQ(memblock.reserved.cnt, 1); + ASSERT_EQ(memblock.reserved.total_size, size); + + test_pass_pop(); + + return 0; +} + +/* Test case wrappers for NUMA tests */ +static int alloc_exact_nid_numa_simple_check(void) +{ + test_print("\tRunning %s...\n", __func__); + memblock_set_bottom_up(false); + alloc_exact_nid_top_down_numa_simple_check(); + + return 0; +} + +static int alloc_exact_nid_numa_part_reserved_check(void) +{ + test_print("\tRunning %s...\n", __func__); + memblock_set_bottom_up(false); + alloc_exact_nid_top_down_numa_part_reserved_check(); + + return 0; +} + +static int alloc_exact_nid_numa_split_range_low_check(void) +{ + test_print("\tRunning %s...\n", __func__); + memblock_set_bottom_up(false); + alloc_exact_nid_top_down_numa_split_range_low_check(); + + return 0; +} + +static int alloc_exact_nid_numa_no_overlap_split_check(void) +{ + test_print("\tRunning %s...\n", __func__); + memblock_set_bottom_up(false); + alloc_exact_nid_top_down_numa_no_overlap_split_check(); + + return 0; +} + +static int alloc_exact_nid_numa_no_overlap_low_check(void) +{ + test_print("\tRunning %s...\n", __func__); + memblock_set_bottom_up(false); + alloc_exact_nid_top_down_numa_no_overlap_low_check(); + + return 0; +} + +int __memblock_alloc_exact_nid_numa_checks(void) +{ + test_print("Running %s NUMA tests...\n", FUNC_NAME); + + alloc_exact_nid_numa_simple_check(); + alloc_exact_nid_numa_part_reserved_check(); + alloc_exact_nid_numa_split_range_low_check(); + alloc_exact_nid_numa_no_overlap_split_check(); + alloc_exact_nid_numa_no_overlap_low_check(); + + return 0; +} + int memblock_alloc_exact_nid_checks(void) { prefix_reset(); @@ -1199,6 +1542,7 @@ int memblock_alloc_exact_nid_checks(void) dummy_physical_memory_init(); =20 memblock_alloc_exact_nid_range_checks(); + memblock_alloc_exact_nid_numa_checks(); =20 dummy_physical_memory_cleanup(); =20 diff --git a/tools/testing/memblock/tests/alloc_exact_nid_api.h b/tools/tes= ting/memblock/tests/alloc_exact_nid_api.h index 4408719de3b9..cef419d55d2a 100644 --- a/tools/testing/memblock/tests/alloc_exact_nid_api.h +++ b/tools/testing/memblock/tests/alloc_exact_nid_api.h @@ -5,5 +5,21 @@ #include "common.h" =20 int memblock_alloc_exact_nid_checks(void); +int __memblock_alloc_exact_nid_numa_checks(void); + +#ifdef CONFIG_NUMA +static inline int memblock_alloc_exact_nid_numa_checks(void) +{ + __memblock_alloc_exact_nid_numa_checks(); + return 0; +} + +#else +static inline int memblock_alloc_exact_nid_numa_checks(void) +{ + return 0; +} + +#endif /* CONFIG_NUMA */ =20 #endif --=20 2.25.1