From: David Woodhouse <dwmw2@infradead.org>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
Dave Hansen <dave.hansen@linux.intel.com>,
x86@kernel.org, "H. Peter Anvin" <hpa@zytor.com>,
Paolo Bonzini <pbonzini@redhat.com>,
"Paul E. McKenney" <paulmck@kernel.org>,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
rcu@vger.kernel.org, mimoja@mimoja.de, hewenliang4@huawei.com,
hushiyuan@huawei.com, luolongjun@huawei.com,
hejingxian@huawei.com
Subject: [PATCH 00/11] Parallel CPU bringup for x86_64
Date: Thu, 9 Dec 2021 15:09:27 +0000 [thread overview]
Message-ID: <20211209150938.3518-1-dwmw2@infradead.org> (raw)
Dusting off this patch series from February, in which we reduce the time
taken for bringing up CPUs on a 96-way Skylake box from 500ms to about
34ms.
There is more parallelism to be had here, including a 1:many TSC sync
(or just *no* TSC sync, in the kexec case), and letting the APs all run
through their own states from CPUHP_BRINGUP_CPU to CPUHP_AP_ONLINE_IDLE
in parallel too. But I'll take a mere factor of 15 for the time being.
We can also have a careful look at the remaining time spent in the
initial INIT/SIPI phase and see what we can shave off it.
David Woodhouse (10):
x86/apic/x2apic: Fix parallel handling of cluster_mask
rcu: Kill rnp->ofl_seq and use only rcu_state.ofl_lock for exclusion
rcu: Add mutex for rcu boost kthread spawning and affinity setting
cpu/hotplug: Add dynamic parallel bringup states before CPUHP_BRINGUP_CPU
x86/smpboot: Reference count on smpboot_setup_warm_reset_vector()
x86/smpboot: Split up native_cpu_up into separate phases
cpu/hotplug: Move idle_thread_get() to <linux/smpboot.h>
x86/tsc: Avoid synchronizing TSCs with multiple CPUs in parallel
x86/smp: Bring up secondary CPUs in parallel
x86/kvm: Silence per-cpu pr_info noise about KVM clocks and steal time
Thomas Gleixner (1):
x86/boot: Support parallel startup of secondary CPUs
arch/x86/include/asm/realmode.h | 3 +++
arch/x86/include/asm/smp.h | 9 ++++++-
arch/x86/kernel/acpi/sleep.c | 1 +
arch/x86/kernel/apic/apic.c | 2 +-
arch/x86/kernel/apic/x2apic_cluster.c | 82 +++++++++++++++++++++++++++++++++++++-------------------------
arch/x86/kernel/head_64.S | 71 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
arch/x86/kernel/kvm.c | 6 ++---
arch/x86/kernel/kvmclock.c | 2 +-
arch/x86/kernel/smpboot.c | 243 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-------------------------------------------------
arch/x86/kernel/tsc_sync.c | 7 ++++++
arch/x86/realmode/init.c | 3 +++
arch/x86/realmode/rm/trampoline_64.S | 14 +++++++++++
include/linux/cpuhotplug.h | 2 ++
include/linux/smpboot.h | 7 ++++++
kernel/cpu.c | 27 +++++++++++++++++++--
kernel/rcu/tree.c | 65 ++++++++++++++++++++++++-------------------------
kernel/rcu/tree.h | 7 +++---
kernel/rcu/tree_plugin.h | 10 ++++++--
kernel/smpboot.c | 2 +-
kernel/smpboot.h | 2 --
20 files changed, 418 insertions(+), 147 deletions(-)
next reply other threads:[~2021-12-09 15:10 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-09 15:09 David Woodhouse [this message]
2021-12-09 15:09 ` [PATCH 01/11] x86/apic/x2apic: Fix parallel handling of cluster_mask David Woodhouse
2021-12-09 15:09 ` [PATCH 02/11] rcu: Kill rnp->ofl_seq and use only rcu_state.ofl_lock for exclusion David Woodhouse
2021-12-09 17:18 ` Paul E. McKenney
2021-12-09 18:52 ` David Woodhouse
2021-12-09 18:31 ` Neeraj Upadhyay
2021-12-09 18:43 ` David Woodhouse
2021-12-09 19:21 ` [PATCH v1.1 " David Woodhouse
2021-12-10 4:26 ` Neeraj Upadhyay
2021-12-13 8:57 ` David Woodhouse
2021-12-13 9:11 ` Neeraj Upadhyay
2021-12-09 15:09 ` [PATCH 03/11] rcu: Add mutex for rcu boost kthread spawning and affinity setting David Woodhouse
2021-12-09 17:20 ` Paul E. McKenney
2021-12-09 15:09 ` [PATCH 04/11] cpu/hotplug: Add dynamic parallel bringup states before CPUHP_BRINGUP_CPU David Woodhouse
2021-12-09 15:09 ` [PATCH 05/11] x86/smpboot: Reference count on smpboot_setup_warm_reset_vector() David Woodhouse
2021-12-09 15:09 ` [PATCH 06/11] x86/smpboot: Split up native_cpu_up into separate phases David Woodhouse
2021-12-09 15:09 ` [PATCH 07/11] cpu/hotplug: Move idle_thread_get() to <linux/smpboot.h> David Woodhouse
2021-12-09 15:09 ` [PATCH 08/11] x86/tsc: Avoid synchronizing TSCs with multiple CPUs in parallel David Woodhouse
2021-12-09 15:43 ` Peter Zijlstra
2021-12-09 15:50 ` David Woodhouse
2021-12-09 15:09 ` [PATCH 09/11] x86/boot: Support parallel startup of secondary CPUs David Woodhouse
2021-12-09 15:50 ` Peter Zijlstra
2021-12-14 11:33 ` David Woodhouse
2021-12-09 15:09 ` [PATCH 10/11] x86/smp: Bring up secondary CPUs in parallel David Woodhouse
2021-12-09 15:09 ` [PATCH 11/11] x86/kvm: Silence per-cpu pr_info noise about KVM clocks and steal time David Woodhouse
2021-12-09 17:39 ` Paolo Bonzini
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211209150938.3518-1-dwmw2@infradead.org \
--to=dwmw2@infradead.org \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=hejingxian@huawei.com \
--cc=hewenliang4@huawei.com \
--cc=hpa@zytor.com \
--cc=hushiyuan@huawei.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luolongjun@huawei.com \
--cc=mimoja@mimoja.de \
--cc=mingo@redhat.com \
--cc=paulmck@kernel.org \
--cc=pbonzini@redhat.com \
--cc=rcu@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).