From patchwork Tue Sep 7 12:22:58 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Adhemerval Zanella Netto X-Patchwork-Id: 44883 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 3562C385E448 for ; Tue, 7 Sep 2021 12:24:50 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 3562C385E448 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1631017490; bh=b06zMraKWX8WzdRjbQYUxxTb/UDrjGJ1e6pICGIoH3E=; h=To:Subject:Date:In-Reply-To:References:List-Id:List-Unsubscribe: List-Archive:List-Post:List-Help:List-Subscribe:From:Reply-To:Cc: From; b=SLJ96r12JIesGmOze+LhZYKkIATP6JLoGiDAG2NyJwJWQZQq3ZHGhykgI8XIuW0Zi S6axKv4IHknfwp81JiBKM0UNNFjCuAeRRJjs7VLOx6oQHRbBu92m4+pkwHgYK+wB+N 0WJhNJVB2+GrrSoQfdbSYYrUCtO81gDp+Jac4DgM= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from mail-qt1-x836.google.com (mail-qt1-x836.google.com [IPv6:2607:f8b0:4864:20::836]) by sourceware.org (Postfix) with ESMTPS id 13475385842D for ; Tue, 7 Sep 2021 12:23:06 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 13475385842D Received: by mail-qt1-x836.google.com with SMTP id s32so7710739qtc.12 for ; Tue, 07 Sep 2021 05:23:06 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=b06zMraKWX8WzdRjbQYUxxTb/UDrjGJ1e6pICGIoH3E=; b=oc2ScZOVSfDBiL1iWs/wHsdxgNzS5l8sKXtQ/ZBo3fjoFC6sue5CG/3niXVlHJhBcx /1yObF6SZtCYfmSw35fZQO3SrH4Tnu2owVusfBq2EiNaa+j2JWozSVh7Jf8TtuuzQEwv cAkZ0/ejK6/zSjF0N/9jLgogrk29u7IVlyTfVlWEH1bD6RJ6RCzcgwz0H5ezlhl1m8ut EXr0y4fkQ0yDm7pbJleroLxWba8BHTIZppij20n9lzNnxUW39088NEOohFP62H0bCxe5 fQ6y67TeI5R3Ok+WnbfwmteNTQIGHzqOgVrYLDTAi9icD0EEuP1CJu1X8M7tNX90vfHX TYqg== X-Gm-Message-State: AOAM530lyWSMv/wtAUHqhGV64aBHqIeanl5BUROSnG5wEKARQlEuL2fT LI7QIxgvThtJWqm1NPUU/j/TggyeLpOpQQ== X-Google-Smtp-Source: ABdhPJyXaFbRGJpdC/O2zCCMBG6xoeFqO4QdTM0dwtxfoY9bpmcEzHMfmbjPB8ofW0qSprcNIoC7YQ== X-Received: by 2002:ac8:3e8a:: with SMTP id y10mr12340400qtf.31.1631017385492; Tue, 07 Sep 2021 05:23:05 -0700 (PDT) Received: from birita.. ([2804:431:c7cb:733d:f360:5bf3:9813:d8f5]) by smtp.gmail.com with ESMTPSA id r18sm6902148qtm.96.2021.09.07.05.23.04 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 07 Sep 2021 05:23:05 -0700 (PDT) To: libc-alpha@sourceware.org Subject: [PATCH v2 2/3] linux: Simplify get_nprocs Date: Tue, 7 Sep 2021 09:22:58 -0300 Message-Id: <20210907122259.79800-3-adhemerval.zanella@linaro.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20210907122259.79800-1-adhemerval.zanella@linaro.org> References: <20210907122259.79800-1-adhemerval.zanella@linaro.org> MIME-Version: 1.0 X-Spam-Status: No, score=-12.2 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_SHORT, RCVD_IN_DNSWL_NONE, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.4 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: Adhemerval Zanella via Libc-alpha From: Adhemerval Zanella Netto Reply-To: Adhemerval Zanella Cc: Florian Weimer , Nicolas Saenz Julienne Errors-To: libc-alpha-bounces+patchwork=sourceware.org@sourceware.org Sender: "Libc-alpha" This patch simplifies the memory allocation code and uses the sched routines instead of reimplement it. This still uses a stack allocation buffer, so it can be used on malloc initialization code. Linux currently supports at maximum of 4096 cpus for most architectures: $ find -iname Kconfig | xargs git grep -A10 -w NR_CPUS | grep -w range arch/alpha/Kconfig- range 2 32 arch/arc/Kconfig- range 2 4096 arch/arm/Kconfig- range 2 16 if DEBUG_KMAP_LOCAL arch/arm/Kconfig- range 2 32 if !DEBUG_KMAP_LOCAL arch/arm64/Kconfig- range 2 4096 arch/csky/Kconfig- range 2 32 arch/hexagon/Kconfig- range 2 6 if SMP arch/ia64/Kconfig- range 2 4096 arch/mips/Kconfig- range 2 256 arch/openrisc/Kconfig- range 2 32 arch/parisc/Kconfig- range 2 32 arch/riscv/Kconfig- range 2 32 arch/s390/Kconfig- range 2 512 arch/sh/Kconfig- range 2 32 arch/sparc/Kconfig- range 2 32 if SPARC32 arch/sparc/Kconfig- range 2 4096 if SPARC64 arch/um/Kconfig- range 1 1 arch/x86/Kconfig-# [NR_CPUS_RANGE_BEGIN ... NR_CPUS_RANGE_END] range. arch/x86/Kconfig- range NR_CPUS_RANGE_BEGIN NR_CPUS_RANGE_END arch/xtensa/Kconfig- range 2 32 With x86 supporting 8192: arch/x86/Kconfig 976 config NR_CPUS_RANGE_END 977 int 978 depends on X86_64 979 default 8192 if SMP && CPUMASK_OFFSTACK 980 default 512 if SMP && !CPUMASK_OFFSTACK 981 default 1 if !SMP So using a maximum of 32k cpu should cover all cases (and I would expect once we start to have many more CPUs that Linux would provide a more straightforward way to query for such information). A test is added to check if sched_getaffinity can successfully return with large buffers. Checked on x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Florian Weimer --- posix/Makefile | 3 +- posix/tst-sched_getaffinity.c | 47 ++++++++++++++++++ sysdeps/unix/sysv/linux/getsysstats.c | 68 +++++++-------------------- 3 files changed, 67 insertions(+), 51 deletions(-) create mode 100644 posix/tst-sched_getaffinity.c diff --git a/posix/Makefile b/posix/Makefile index 059efb3cd2..09460a28e8 100644 --- a/posix/Makefile +++ b/posix/Makefile @@ -107,7 +107,8 @@ tests := test-errno tstgetopt testfnm runtests runptests \ tst-sysconf-empty-chroot tst-glob_symlinks tst-fexecve \ tst-glob-tilde test-ssize-max tst-spawn4 bug-regex37 \ bug-regex38 tst-regcomp-truncated tst-spawn-chdir \ - tst-wordexp-nocmd tst-execveat tst-spawn5 + tst-wordexp-nocmd tst-execveat tst-spawn5 \ + tst-sched_getaffinity # Test for the glob symbol version that was replaced in glibc 2.27. ifeq ($(have-GLIBC_2.26)$(build-shared),yesyes) diff --git a/posix/tst-sched_getaffinity.c b/posix/tst-sched_getaffinity.c new file mode 100644 index 0000000000..6d1fdcb05f --- /dev/null +++ b/posix/tst-sched_getaffinity.c @@ -0,0 +1,47 @@ +/* Tests for sched_getaffinity with large buffers. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + . */ + +#include +#include +#include + +static int +do_test (void) +{ + /* The values are larger than the default cpu_set_t. */ + const int bufsize[] = { 1<<11, 1<<12, 1<<13, 1<<14, 1<<15, 1<<16, 1<<17 }; + int cpucount[array_length(bufsize)]; + + for (int i = 0; i < array_length (bufsize); i++) + { + cpu_set_t *cpuset = CPU_ALLOC (bufsize[i]); + TEST_VERIFY (cpuset != NULL); + size_t size = CPU_ALLOC_SIZE (bufsize[i]); + TEST_COMPARE (sched_getaffinity (0, size, cpuset), 0); + cpucount[i] = CPU_COUNT_S (size, cpuset); + CPU_FREE (cpuset); + } + + for (int i = 0; i < array_length (cpucount) - 1; i++) + for (int j = 1; j < array_length (cpucount); j++) + TEST_COMPARE (cpucount[i], cpucount[j]); + + return 0; +} + +#include diff --git a/sysdeps/unix/sysv/linux/getsysstats.c b/sysdeps/unix/sysv/linux/getsysstats.c index e9c0dc4d83..1e3d88676d 100644 --- a/sysdeps/unix/sysv/linux/getsysstats.c +++ b/sysdeps/unix/sysv/linux/getsysstats.c @@ -28,61 +28,29 @@ #include #include -/* Compute the population count of the entire array. */ -static int -__get_nprocs_count (const unsigned long int *array, size_t length) -{ - int count = 0; - for (size_t i = 0; i < length; ++i) - if (__builtin_add_overflow (count, __builtin_popcountl (array[i]), - &count)) - return INT_MAX; - return count; -} - -/* __get_nprocs with a large buffer. */ -static int -__get_nprocs_large (void) -{ - /* This code cannot use scratch_buffer because it is used during - malloc initialization. */ - size_t pagesize = GLRO (dl_pagesize); - unsigned long int *page = __mmap (0, pagesize, PROT_READ | PROT_WRITE, - MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); - if (page == MAP_FAILED) - return 2; - int r = INTERNAL_SYSCALL_CALL (sched_getaffinity, 0, pagesize, page); - int count; - if (r > 0) - count = __get_nprocs_count (page, pagesize / sizeof (unsigned long int)); - else if (r == -EINVAL) - /* One page is still not enough to store the bits. A more-or-less - arbitrary value. This assumes t hat such large systems never - happen in practice. */ - count = GLRO (dl_pagesize) * CHAR_BIT; - else - count = 2; - __munmap (page, GLRO (dl_pagesize)); - return count; -} - int __get_nprocs (void) { - /* Fast path for most systems. The kernel expects a buffer size - that is a multiple of 8. */ - unsigned long int small_buffer[1024 / CHAR_BIT / sizeof (unsigned long int)]; - int r = INTERNAL_SYSCALL_CALL (sched_getaffinity, 0, - sizeof (small_buffer), small_buffer); + enum + { + max_num_cpus = 32768, + cpu_bits_size = CPU_ALLOC_SIZE (32768) + }; + + /* This cannot use malloc because it is used on malloc initialization. */ + __cpu_mask cpu_bits[cpu_bits_size / sizeof (__cpu_mask)]; + int r = INTERNAL_SYSCALL_CALL (sched_getaffinity, 0, cpu_bits_size, + cpu_bits); if (r > 0) - return __get_nprocs_count (small_buffer, r / sizeof (unsigned long int)); + return CPU_COUNT_S (cpu_bits_size, (cpu_set_t*) cpu_bits); else if (r == -EINVAL) - /* The kernel requests a larger buffer to store the data. */ - return __get_nprocs_large (); - else - /* Some other error. 2 is conservative (not a uniprocessor - system, so atomics are needed). */ - return 2; + /* The input buffer is still not enough to store the number of cpus. This + is an arbitrary values assuming such systems should be rare and there + is no offline cpus. */ + return max_num_cpus; + /* Some other error. 2 is conservative (not a uniprocessor system, so + atomics are needed). */ + return 2; } libc_hidden_def (__get_nprocs) weak_alias (__get_nprocs, get_nprocs)