From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Alex Nixon <alex.nixon@citrix.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@elte.hu>
Subject: Re: [PATCH 3/5] x86: Unify x86_32 and x86_64 play_dead into one function
Date: Sat, 23 Aug 2008 22:57:10 -0700 [thread overview]
Message-ID: <48B0F836.4020506@goop.org> (raw)
In-Reply-To: <1219402335-29773-3-git-send-email-alex.nixon@citrix.com>
Alex Nixon wrote:
> Add the new play_dead into smpboot.c, as it fits more cleanly in there alongside other CONFIG_HOTPLUG functions.
> Separate out the common code into its own function.
>
Unfortunately this breaks with !CONFIG_SMP. The fix is a bit awkward,
because the right place to fix it is in asm-x86/smp.h, but that isn't
included by linux/smp.h when CONFIG_SMP isn't set.
J
> Signed-off-by: Alex Nixon <alex.nixon@citrix.com>
> Cc: Jeremy Fitzhardinge <jeremy@goop.org>
> Cc: Ingo Molnar <mingo@elte.hu>
> ---
> arch/x86/kernel/process_32.c | 32 --------------------------------
> arch/x86/kernel/process_64.c | 23 -----------------------
> arch/x86/kernel/smpboot.c | 29 +++++++++++++++++++++++++++++
> include/asm-x86/smp.h | 1 +
> 4 files changed, 30 insertions(+), 55 deletions(-)
>
> diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
> index c4abfab..aff137c 100644
> --- a/arch/x86/kernel/process_32.c
> +++ b/arch/x86/kernel/process_32.c
> @@ -74,38 +74,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
> return ((unsigned long *)tsk->thread.sp)[3];
> }
>
> -#ifdef CONFIG_HOTPLUG_CPU
> -#include <asm/nmi.h>
> -
> -/* We don't actually take CPU down, just spin without interrupts. */
> -void native_play_dead(void)
> -{
> - int cpu = raw_smp_processor_id();
> -
> - idle_task_exit();
> -
> - reset_lazy_tlbstate();
> -
> - irq_ctx_exit(cpu);
> -
> - mb();
> - /* Ack it */
> - __get_cpu_var(cpu_state) = CPU_DEAD;
> -
> - /*
> - * With physical CPU hotplug, we should halt the cpu
> - */
> - local_irq_disable();
> - /* mask all interrupts, flush any and all caches, and halt */
> - wbinvd_halt();
> -}
> -#else
> -void native_play_dead(void)
> -{
> - BUG();
> -}
> -#endif /* CONFIG_HOTPLUG_CPU */
> -
> /*
> * The idle thread. There's no useful work to be
> * done, so just try to conserve power and have a
> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
> index 1cca50c..b2bab8e 100644
> --- a/arch/x86/kernel/process_64.c
> +++ b/arch/x86/kernel/process_64.c
> @@ -87,29 +87,6 @@ void exit_idle(void)
> __exit_idle();
> }
>
> -#ifdef CONFIG_HOTPLUG_CPU
> -DECLARE_PER_CPU(int, cpu_state);
> -
> -#include <linux/nmi.h>
> -/* We halt the CPU with physical CPU hotplug */
> -void native_play_dead(void)
> -{
> - idle_task_exit();
> - mb();
> - /* Ack it */
> - __get_cpu_var(cpu_state) = CPU_DEAD;
> -
> - local_irq_disable();
> - /* mask all interrupts, flush any and all caches, and halt */
> - wbinvd_halt();
> -}
> -#else
> -void native_play_dead(void)
> -{
> - BUG();
> -}
> -#endif /* CONFIG_HOTPLUG_CPU */
> -
> /*
> * The idle thread. There's no useful work to be
> * done, so just try to conserve power and have a
> diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
> index c6832ca..58d8c6c 100644
> --- a/arch/x86/kernel/smpboot.c
> +++ b/arch/x86/kernel/smpboot.c
> @@ -1399,6 +1399,29 @@ void native_cpu_die(unsigned int cpu)
> }
> printk(KERN_ERR "CPU %u didn't die...\n", cpu);
> }
> +
> +void play_dead_common(void)
> +{
> + idle_task_exit();
> + reset_lazy_tlbstate();
> + irq_ctx_exit(raw_smp_processor_id());
> +
> + mb();
> + /* Ack it */
> + __get_cpu_var(cpu_state) = CPU_DEAD;
> +
> + /*
> + * With physical CPU hotplug, we should halt the cpu
> + */
> + local_irq_disable();
> +}
> +
> +void native_play_dead(void)
> +{
> + play_dead_common();
> + wbinvd_halt();
> +}
> +
> #else /* ... !CONFIG_HOTPLUG_CPU */
> int native_cpu_disable(void)
> {
> @@ -1410,4 +1433,10 @@ void native_cpu_die(unsigned int cpu)
> /* We said "no" in __cpu_disable */
> BUG();
> }
> +
> +void native_play_dead(void)
> +{
> + BUG();
> +}
> +
> #endif
> diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
> index dba14d3..95b28c1 100644
> --- a/include/asm-x86/smp.h
> +++ b/include/asm-x86/smp.h
> @@ -135,6 +135,7 @@ int native_cpu_up(unsigned int cpunum);
> int native_cpu_disable(void);
> void native_cpu_die(unsigned int cpu);
> void native_play_dead(void);
> +void play_dead_common(void);
>
> void native_send_call_func_ipi(cpumask_t mask);
> void native_send_call_func_single_ipi(int cpu);
>
next prev parent reply other threads:[~2008-08-24 5:57 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-22 10:52 [PATCH 1/5] x86: Add cpu hotplug hooks into smp_ops Alex Nixon
2008-08-22 10:52 ` [PATCH 2/5] x86_32: Clean up play_dead Alex Nixon
2008-08-22 10:52 ` [PATCH 3/5] x86: Unify x86_32 and x86_64 play_dead into one function Alex Nixon
2008-08-22 10:52 ` [PATCH 4/5] x86: Separate generic cpu disabling code from APIC writes in cpu_disable Alex Nixon
2008-08-22 10:52 ` [PATCH 5/5] Xen: Implement CPU hotplugging Alex Nixon
2008-08-25 9:25 ` Ingo Molnar
2008-08-25 9:41 ` Ingo Molnar
2008-08-24 5:57 ` Jeremy Fitzhardinge [this message]
2008-08-25 8:59 ` [PATCH 1/5] x86: Add cpu hotplug hooks into smp_ops Ingo Molnar
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=48B0F836.4020506@goop.org \
--to=jeremy@goop.org \
--cc=alex.nixon@citrix.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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