From patchwork Mon Sep 6 18:11:32 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: 44870 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 ABF73385AC31 for ; Mon, 6 Sep 2021 18:13:30 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org ABF73385AC31 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1630952010; bh=M5jjQNMX9a2IyeWj6skfl4WfVx/SMtSw8VY6NQ3cqN0=; 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=cHQ8YL1UgxLqeIH8XRqYhk4cZWhi2KnFqc6DzbktwxqCLGDc5FgVIfxXSgJQXylHf 2VpRXbcz5WEDg54ntcwZwH4V6g7yeYtgSrnA1BD4P95rYRoJSDpYRq4UkvfpC1YHu+ QcKEOBFTZAFAzIrmxYtprzL22fSXeyapRCRC31Mk= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from mail-qv1-xf34.google.com (mail-qv1-xf34.google.com [IPv6:2607:f8b0:4864:20::f34]) by sourceware.org (Postfix) with ESMTPS id E57253857022 for ; Mon, 6 Sep 2021 18:11:40 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org E57253857022 Received: by mail-qv1-xf34.google.com with SMTP id p17so4378506qvo.8 for ; Mon, 06 Sep 2021 11:11:40 -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=M5jjQNMX9a2IyeWj6skfl4WfVx/SMtSw8VY6NQ3cqN0=; b=N9J6rkUJs12m9E7njQJjCfQRR3agzLGFwC49SFQ5zpXgrYRYr63yBgjh+2Rleqh7xo 9xLR+ZHR3VgUuPwv88mVb369oMGpKstepAwtkx1cBy/siBKIDFpGz+VN+J7j+UV1A0Kx tKbpFmvPRq2jXLpDZ70o8kQfSXgoKfZnwq1dqBbLPFE/kv5kueh38Kemu0Wi8UNaA57K YqJSc0kRxFHH3EsHWDFyOdjfFMvdKgIvbh+tKtnDfoZRhF57hLB1xE6zrznQLPG2NQCX BauOSRLBRNnksIYqGAM1QXqw5wuqkZ/qgJlVjqJUfMDa3AKv+DpEzeSNzQ0tkxvYiXrp e6cg== X-Gm-Message-State: AOAM533fn4hzKksDTsC+PrtljNZsHQgKFd1kk4i5urVpK6bjDYCit+Wm hJxQ8UI9x/92Mk3m4V/dcg3ga8LdpsE6/w== X-Google-Smtp-Source: ABdhPJxa0/1hWzYE0RqbBi3mtFVnaOUoOurNl5gHUXF7r6+O0V7pRa8jI33e7+XxCDidP+HOUWiTtg== X-Received: by 2002:ad4:4b14:: with SMTP id r20mr13389749qvw.61.1630951900388; Mon, 06 Sep 2021 11:11:40 -0700 (PDT) Received: from birita.. ([2804:431:c7cb:733d:210:e147:8627:2d6]) by smtp.gmail.com with ESMTPSA id j23sm7175647qkl.65.2021.09.06.11.11.39 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 06 Sep 2021 11:11:40 -0700 (PDT) To: libc-alpha@sourceware.org Subject: [PATCH 2/3] linux: Simplify get_nprocs Date: Mon, 6 Sep 2021 15:11:32 -0300 Message-Id: <20210906181133.1401140-3-adhemerval.zanella@linaro.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20210906181133.1401140-1-adhemerval.zanella@linaro.org> References: <20210906181133.1401140-1-adhemerval.zanella@linaro.org> MIME-Version: 1.0 X-Spam-Status: No, score=-12.4 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, 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. --- posix/Makefile | 3 +- sysdeps/unix/sysv/linux/getsysstats.c | 64 ++++++--------------------- 2 files changed, 16 insertions(+), 51 deletions(-) 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/sysdeps/unix/sysv/linux/getsysstats.c b/sysdeps/unix/sysv/linux/getsysstats.c index e9c0dc4d83..b9710f8319 100644 --- a/sysdeps/unix/sysv/linux/getsysstats.c +++ b/sysdeps/unix/sysv/linux/getsysstats.c @@ -28,61 +28,25 @@ #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); + /* This cannot use malloc because it is used on malloc initialization. */ + enum { max_num_cpus = 32768 }; + size_t cpu_bits_size = CPU_ALLOC_SIZE (max_num_cpus); + __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)