From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751801AbdKEQyv (ORCPT ); Sun, 5 Nov 2017 11:54:51 -0500 Received: from mx1.redhat.com ([209.132.183.28]:54190 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751527AbdKEQyn (ORCPT ); Sun, 5 Nov 2017 11:54:43 -0500 DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com D5B08C04AC5A Authentication-Results: ext-mx07.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx07.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=prarit@redhat.com From: Prarit Bhargava To: linux-kernel@vger.kernel.org Cc: Andi Kleen , Prarit Bhargava , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, Peter Zijlstra , Dave Hansen , Piotr Luc , Kan Liang , Borislav Petkov , Stephane Eranian , Arvind Yadav , Andy Lutomirski , Christian Borntraeger , "Kirill A. Shutemov" , Tom Lendacky , He Chen , Mathias Krause , Tim Chen , Vitaly Kuznetsov Subject: [PATCH v5 2/3] x86/topology: Avoid wasting 128k for package id array Date: Sun, 5 Nov 2017 11:54:27 -0500 Message-Id: <20171105165428.32108-3-prarit@redhat.com> In-Reply-To: <20171105165428.32108-1-prarit@redhat.com> References: <20171105165428.32108-1-prarit@redhat.com> X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.31]); Sun, 05 Nov 2017 16:54:43 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Andi Kleen I was looking at large early boot allocations and noticed that since (1f12e32f x86/topology: Create logical package id) every 64bit system allocates a 128k array to convert logical package ids. This happens because the array is sized to (MAX_LOCAL_APIC * u16) = 128k. This is a lot of waste especially for most systems which have one or two sockets. Use a dynamically allocated array of size logical_packages to map the logical and physical packages. [v2]: Decrease logical_packages when the last thread in a socket is removed. [v3]: Add more logic to keep logical and physical package IDs in synch. [v4]: Keep logical mapping static by using hybrid approach of a small logical to physical array and keeping logical cpu information in cpu_data. [v5]: Change kmalloc to GFP_ATOMIC to fix "sleeping function" warning on virtual machines. Remove phys_pkg_id. Add spinlock to avoid concurrency issues. Signed-off-by: Andi Kleen Signed-off-by: Prarit Bhargava Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org Cc: Peter Zijlstra Cc: Andi Kleen Cc: Dave Hansen Cc: Piotr Luc Cc: Kan Liang Cc: Borislav Petkov Cc: Stephane Eranian Cc: Prarit Bhargava Cc: Arvind Yadav Cc: Andy Lutomirski Cc: Christian Borntraeger Cc: "Kirill A. Shutemov" Cc: Tom Lendacky Cc: He Chen Cc: Mathias Krause Cc: Tim Chen Cc: Vitaly Kuznetsov --- arch/x86/kernel/smpboot.c | 97 +++++++++++++++++++++++++---------------------- 1 file changed, 51 insertions(+), 46 deletions(-) diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index ad59edd84de7..580261f8a2bf 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -99,13 +99,12 @@ DEFINE_PER_CPU_READ_MOSTLY(cpumask_var_t, cpu_llc_shared_map); DEFINE_PER_CPU_READ_MOSTLY(struct cpuinfo_x86, cpu_info); EXPORT_PER_CPU_SYMBOL(cpu_info); -/* Logical package management. We might want to allocate that dynamically */ -static int *physical_to_logical_pkg __read_mostly; -static unsigned long *physical_package_map __read_mostly;; -static unsigned int max_physical_pkg_id __read_mostly; +/* Logical package management.*/ unsigned int __max_logical_packages __read_mostly; EXPORT_SYMBOL(__max_logical_packages); +static u16 *logical_to_physical_pkg_map; static unsigned int logical_packages __read_mostly; +static DEFINE_SPINLOCK(logical_pkg_lock); /* Maximum number of SMT threads on any online core */ int __max_smt_threads __read_mostly; @@ -278,6 +277,31 @@ static void notrace start_secondary(void *unused) cpu_startup_entry(CPUHP_AP_ONLINE_IDLE); } +/** + * topology_phys_to_logical_pkg - Map a physical package id to a logical + * + * Returns logical package id or -1 if not found + */ +int topology_phys_to_logical_pkg(unsigned int phys_pkg) +{ + int log_pkg; + int found = 0; + unsigned long flags; + + spin_lock_irqsave(&logical_pkg_lock, flags); + for (log_pkg = 0; log_pkg < logical_packages; log_pkg++) + if (logical_to_physical_pkg_map[log_pkg] == phys_pkg) { + found = 1; + break; + } + spin_unlock_irqrestore(&logical_pkg_lock, flags); + + if (found) + return log_pkg; + return -1; +} +EXPORT_SYMBOL(topology_phys_to_logical_pkg); + /** * topology_update_package_map - Update the physical to logical package map * @pkg: The physical package id as retrieved via CPUID @@ -285,17 +309,12 @@ static void notrace start_secondary(void *unused) */ int topology_update_package_map(unsigned int pkg, unsigned int cpu) { - unsigned int new; - - /* Called from early boot ? */ - if (!physical_package_map) - return 0; - - if (pkg >= max_physical_pkg_id) - return -EINVAL; + int new; + u16 *ltp_pkg_map_new; + unsigned long flags; - /* Set the logical package id */ - if (test_and_set_bit(pkg, physical_package_map)) + new = topology_phys_to_logical_pkg(pkg); + if (new >= 0) goto found; if (logical_packages >= __max_logical_packages) { @@ -305,34 +324,31 @@ int topology_update_package_map(unsigned int pkg, unsigned int cpu) } new = logical_packages++; - if (new != pkg) { - pr_info("CPU %u Converting physical %u to logical package %u\n", - cpu, pkg, new); + + /* Allocate and copy a new array */ + ltp_pkg_map_new = kmalloc(logical_packages * sizeof(u16), GFP_ATOMIC); + BUG_ON(!ltp_pkg_map_new); + spin_lock_irqsave(&logical_pkg_lock, flags); + if (logical_to_physical_pkg_map) { + memcpy(ltp_pkg_map_new, logical_to_physical_pkg_map, + logical_packages * sizeof(u16)); + kfree(logical_to_physical_pkg_map); } - physical_to_logical_pkg[pkg] = new; + logical_to_physical_pkg_map = ltp_pkg_map_new; + logical_to_physical_pkg_map[new] = pkg; + spin_unlock_irqrestore(&logical_pkg_lock, flags); + if (pkg != new) + pr_info("CPU %u Converting physical %u to logical package %u\n", + cpu, pkg, new); found: - cpu_data(cpu).logical_proc_id = physical_to_logical_pkg[pkg]; + cpu_data(cpu).logical_proc_id = new; return 0; } -/** - * topology_phys_to_logical_pkg - Map a physical package id to a logical - * - * Returns logical package id or -1 if not found - */ -int topology_phys_to_logical_pkg(unsigned int phys_pkg) -{ - if (phys_pkg >= max_physical_pkg_id) - return -1; - return physical_to_logical_pkg[phys_pkg]; -} -EXPORT_SYMBOL(topology_phys_to_logical_pkg); - static void __init smp_init_package_map(struct cpuinfo_x86 *c, unsigned int cpu) { unsigned int ncpus; - size_t size; /* * Today neither Intel nor AMD support heterogenous systems. That @@ -363,21 +379,10 @@ static void __init smp_init_package_map(struct cpuinfo_x86 *c, unsigned int cpu) } __max_logical_packages = DIV_ROUND_UP(total_cpus, ncpus); - logical_packages = 0; - - /* - * Possibly larger than what we need as the number of apic ids per - * package can be smaller than the actual used apic ids. - */ - max_physical_pkg_id = DIV_ROUND_UP(MAX_LOCAL_APIC, ncpus); - size = max_physical_pkg_id * sizeof(unsigned int); - physical_to_logical_pkg = kmalloc(size, GFP_KERNEL); - memset(physical_to_logical_pkg, 0xff, size); - size = BITS_TO_LONGS(max_physical_pkg_id) * sizeof(unsigned long); - physical_package_map = kzalloc(size, GFP_KERNEL); - pr_info("Max logical packages: %u\n", __max_logical_packages); + logical_packages = 0; + topology_update_package_map(c->phys_proc_id, cpu); } -- 2.15.0.rc0.39.g2f0e14e64