From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3603F449B1B for ; Tue, 1 Sep 2026 23:24:27 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788305069; cv=none; b=C3YJxtRZOCmfkteQP7T5TWLfHdRSshpLcejCQsSIAZeWudMzyJdETxEf3eK6QqWmAe/izsiwukrj1JkwSrVIjunsMLZzJ/2I8zI4ttcrsxkfzgtP31Sd5VY5zWwijEUT+Aoy/G41PnKyXI+ugw9EV7fjfsS4peyK8rZVvqz6pCM= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788305069; c=relaxed/simple; bh=a/Gew3qvJ/AdKgGy7/W9Q3AQuqozlFsBijJTj+NAobY=; h=From:To:Cc:Subject:Date:Message-Id:MIME-Version:Content-Type; b=Hr6WheXXZjLMaQDCBMHnMkK0tRGL4vZafK2t90ElQkCnaj4pxaLH+HdfnoFGBc5Fc07prK7OGtR8AXSO/NTXBophMgShMred0oXNGYxkWruZ5krYb/dkxHDeD3lybXqDJ3UqJiqRaqvx3yqz50CjoDyUtC5hVS5y+stobRmyEaM= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=C4gEmpXX; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="C4gEmpXX" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 374AC1F000E9; Tue, 1 Sep 2026 23:24:24 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1788305067; bh=0IfPqucy8DfnTKp/2gWMiMi8WWSj25F+64egVqUdKRg=; h=From:To:Cc:Subject:Date; b=C4gEmpXXqmYDM1Pm09UQVGKITgrhJdjZsiAnmp3HJa8cNs89U2PzZhLUasWcH1i8e iarqtP0CcSSnLhbKbZoobAandpBpJJjcIV6qQc1fYcapPBO3sY6L/uOWEUmzJjmQSm NjZVpT1WPZqplcg6HcbpAoJ0KF4vnb/A6+tifct6qKqpq71xCFoGRsFnRCi1Iqol+n jq3r5Iz1PqE3q/Sa/xZ3sIEY3QNidmiAh6LSQLTiI4GDrnWKjo6NJsTeBEpCFdRCBV PaSz4OFOkKT/xAojq2Nv+LutVg18X6/iNdXFnWK0dgVKR/3xK6wXXoYScYAkBGzKZK hFNNUhD22iu5A== From: "Barry Song (Xiaomi)" To: akpm@linux-foundation.org, linux-mm@kvack.org Cc: axelrasmussen@google.com, baolin.wang@linux.alibaba.com, baoquan.he@linux.dev, chenridong@xiaomi.com, david@kernel.org, hannes@cmpxchg.org, kasong@tencent.com, lianux.mm@gmail.com, linux-kernel@vger.kernel.org, ljs@kernel.org, lyugaofei@xiaomi.com, mhocko@kernel.org, qi.zheng@linux.dev, shakeel.butt@linux.dev, stevensd@chromium.org, wangzicheng@honor.com, weixugc@google.com, yuanchu@google.com, zhangbo56@xiaomi.com, "Barry Song (Xiaomi)" Subject: [PATCH v3 0/7] mm/mglru: speed up inc_min_seq() and fix cold/hot inversions Date: Wed, 2 Sep 2026 07:24:14 +0800 Message-Id: <20260901232421.40157-1-baohua@kernel.org> X-Mailer: git-send-email 2.39.3 (Apple Git-146) 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=UTF-8 Content-Transfer-Encoding: 8bit This is an aging speedup series split out from the MGLRU swappiness series [1], with the inc_min_seq changes separated to make them easier to review. Currently, inc_min_seq() has two primary issues that affect both performance and folio hotness assessment: 1. It processes each folio one by one, while many operations can be batched or skipped. For example, a batch of folios can be moved together from the oldest generation to the second-oldest generation, and the associated counting can also be done in batches. 2. It may cause potential cold/hot inversion by placing promoted folios (which have been scanned and found to have young PTEs) behind non-promoted folios. A similar inversion can also occur among non-promoted folios, as tail folios from the oldest generation are placed before head folios when moving them to the second-oldest generation. This series tries to batch operations as much as possible and fix the potential cold/hot inversion by keeping promoted folios ahead of non-promoted folios, while also preserving the order of non-promoted folios when moving them from the oldest generation to the second-oldest generation. Minor issue: inc_min_seq() also counts protected folios improperly, as promoted folios should be skipped, as in sort_folio(). We need a stable workload with a stable number of folios to measure aging and evaluate the speedup in inc_min_seq(). So I asked ChatGPT to generate the microbenchmark below. It ages an LRU vec containing 512 MB of memory 100 times: #define _GNU_SOURCE #include #include #include #include #include #include #include #include #include #define SIZE (512UL * 1024 * 1024) #define LRU_GEN "/sys/kernel/debug/lru_gen" #define TARGET_CGROUP "/system.slice/agetest.scope" #define START_GEN 3 #define END_GEN 103 static long long nsec_diff(const struct timespec *start, const struct timespec *end) { return (end->tv_sec - start->tv_sec) * 1000000000LL + (end->tv_nsec - start->tv_nsec); } static int find_memcg_id(void) { FILE *fp; char line[4096]; int memcg_id; fp = fopen(LRU_GEN, "r"); if (!fp) { perror("fopen lru_gen"); return -1; } while (fgets(line, sizeof(line), fp)) { char *p; if (strncmp(line, "memcg ", 6)) continue; p = line + 6; if (sscanf(p, "%d", &memcg_id) != 1) continue; /* * The memcg path follows the numeric ID. */ p = strchr(p, ' '); if (!p) continue; if (strstr(p, TARGET_CGROUP)) { fclose(fp); return memcg_id; } } fclose(fp); fprintf(stderr, "Cannot find %s\n", TARGET_CGROUP); return -1; } int main(void) { void *addr; int memcg_id; int fd; long long total_ns = 0; /* * mmap 512 MB and touch every page. */ addr = mmap(NULL, SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); if (addr == MAP_FAILED) { perror("mmap"); return 1; } memset(addr, 0x55, SIZE); printf("mmap: %p, size: %lu MB\n", addr, SIZE / 1024 / 1024); /* * Find the memcg ID automatically. */ memcg_id = find_memcg_id(); if (memcg_id < 0) return 1; printf("memcg: %d (%s)\n", memcg_id, TARGET_CGROUP); printf("aging generation %d -> %d\n", START_GEN, END_GEN); fd = open(LRU_GEN, O_WRONLY); if (fd < 0) { perror("open lru_gen"); return 1; } for (int gen = START_GEN; gen <= END_GEN; gen++) { char buf[128]; int len; struct timespec start, end; long long ns; len = snprintf(buf, sizeof(buf), "+ %d 0 %d\n", memcg_id, gen); clock_gettime(CLOCK_MONOTONIC, &start); if (write(fd, buf, len) != len) { perror("write lru_gen"); close(fd); return 1; } clock_gettime(CLOCK_MONOTONIC, &end); ns = nsec_diff(&start, &end); total_ns += ns; printf("gen %3d: %8.3f ms\n", gen, ns / 1000000.0); fflush(stdout); } close(fd); printf("\nTotal: %.3f ms\n", total_ns / 1000000.0); printf("Average: %.3f ms\n", total_ns / (double)(END_GEN - START_GEN + 1) / 1000000.0); while (1) sleep(1); return 0; } Run the above microbenchmark with: systemd-run --scope --unit=agetest -p MemoryMax=1024M ./agetest I’m seeing a significant speedup in inc_min_seq() on my x86 PC: W/o patch: Running scope as unit: agetest.scope mmap: 0x72c1b5a00000, size: 512 MB memcg: 12673 (/system.slice/agetest.scope) aging generation 3 -> 103 gen 3: 7.433 ms gen 4: 0.949 ms gen 5: 2.535 ms gen 6: 5.043 ms gen 7: 5.041 ms gen 8: 5.027 ms ... gen 100: 5.035 ms gen 101: 5.011 ms gen 102: 5.029 ms gen 103: 5.056 ms Total: 503.946 ms Average: 4.990 ms W/ patch: Running scope as unit: agetest.scope mmap: 0x775ec5200000, size: 512 MB memcg: 12717 (/system.slice/agetest.scope) aging generation 3 -> 103 gen 3: 7.558 ms gen 4: 0.916 ms gen 5: 2.277 ms gen 6: 2.328 ms … gen 100: 2.303 ms gen 101: 2.297 ms gen 102: 2.305 ms gen 103: 2.312 ms Total: 236.635 ms Average: 2.343 ms The average aging time drops from 4.990 ms to 2.343 ms! Thanks, Xueyuan, for testing this on ARM[2]. It actually shows an even larger improvement. Xueyuan tested this series on his arm64 machine (24 cores, 4K base pages) and reproduced the improvement: THP=never (PTE): baseline: 7.644 ms patched: 2.964 ms (-61.2%) THP=always (PMD): baseline: 0.0373 ms patched: 0.0292 ms (-21.8%) [1] https://lore.kernel.org/linux-mm/20260812121658.69965-1-baohua@kernel.org/ [2] https://lore.kernel.org/linux-mm/20260827035416.3012015-1-xueyuan.chen21@gmail.com/ -v3: * Collect tags from Ridong, Lian and Kunwu, thanks! * Add some comments as Ridong's suggestion in "enhance cold/hot inversion handling", thanks! -v2: * Collect tags from Kairui, Baoquan, and Xueyuan, thanks! * Split `__folio_inc_gen()` into a separate patch and add `VM_WARN_ON` checks for the generation and active state, per Baoquan and Kairui, thanks! * Make `delta` a `long` instead of `unsigned long`, per Lian Wang, thanks! * Fix checkpatch issues by converting macros to `static inline` functions, per Kairui, thanks! * Drop the batched protected update since it doesn't show a significant performance improvement, following Baoquan's comments. https://lore.kernel.org/linux-mm/20260827234704.63163-1-baohua@kernel.org/ -v1: https://lore.kernel.org/linux-mm/20260821102538.22642-1-baohua@kernel.org/ Barry Song (Xiaomi) (7): mm/mglru: separate folio generation update from LRU accounting mm/mglru: batch update lrugen->nr_pages in inc_min_seq() mm/mglru: enhance cold/hot inversion handling in inc_min_seq() mm/mglru: exclude folios promoted by aging from protected in inc_min_seq() mm/mglru: make LRU folio prefetch helper an inline function mm/mglru: move folios from oldest gen to second-oldest gen from head to tail mm/mglru: batch move folios to the second-oldest gen's LRU mm/vmscan.c | 136 ++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 106 insertions(+), 30 deletions(-) -- 2.34.1