lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Tue, 28 Feb 2017 16:11:33 +0800
From:   Wanpeng Li <kernellwp@...il.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Mike Galbraith <efault@....de>,
        LKML <linux-kernel@...r.kernel.org>, Ingo Molnar <mingo@...e.hu>,
        Thomas Gleixner <tglx@...utronix.de>,
        Borislav Petkov <bp@...en8.de>
Subject: Re: tip.today - scheduler bam boom crash (cpu hotplug)

2017-02-28 16:08 GMT+08:00 Peter Zijlstra <peterz@...radead.org>:
> On Tue, Feb 28, 2017 at 09:51:07AM +0800, Wanpeng Li wrote:
>> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
>> index 4e95b2e..ed8eda4 100644
>> --- a/arch/x86/kernel/cpu/amd.c
>> +++ b/arch/x86/kernel/cpu/amd.c
>> @@ -557,7 +557,7 @@ static void early_init_amd(struct cpuinfo_x86 *c)
>>          set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
>>          if (check_tsc_unstable())
>>              clear_sched_clock_stable();
>> -    } else {
>> +    } else if (!boot_cpu_has(X86_FEATURE_HYPERVISOR)) {
>>          clear_sched_clock_stable();
>>      }
>
> That's wrong, you can have HYPERVISOR and still use
> native_sched_clock() (lguest does that for one).

My posting delay to check this in tsc_init().

Regards,
Wanpeng Li

>
>
> I suspect we can do something like the below. Since we fixed the
> clocksource watchdog to mark TSC unstable, and we're already fairly
> careful with using TSC for timekeeping anyway.
>
> ---
>  arch/x86/kernel/cpu/amd.c       |  4 ----
>  arch/x86/kernel/cpu/centaur.c   |  2 --
>  arch/x86/kernel/cpu/common.c    |  3 ---
>  arch/x86/kernel/cpu/cyrix.c     |  1 -
>  arch/x86/kernel/cpu/intel.c     |  4 ----
>  arch/x86/kernel/cpu/transmeta.c |  2 --
>  arch/x86/kernel/tsc.c           | 35 +++++++++++++++++++++++------------
>  7 files changed, 23 insertions(+), 28 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
> index 35a5d5dca2fa..c36140d788fe 100644
> --- a/arch/x86/kernel/cpu/amd.c
> +++ b/arch/x86/kernel/cpu/amd.c
> @@ -556,10 +556,6 @@ static void early_init_amd(struct cpuinfo_x86 *c)
>         if (c->x86_power & (1 << 8)) {
>                 set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
>                 set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
> -               if (check_tsc_unstable())
> -                       clear_sched_clock_stable();
> -       } else {
> -               clear_sched_clock_stable();
>         }
>
>         /* Bit 12 of 8000_0007 edx is accumulated power mechanism. */
> diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
> index 106bd3318121..44207b71fee1 100644
> --- a/arch/x86/kernel/cpu/centaur.c
> +++ b/arch/x86/kernel/cpu/centaur.c
> @@ -105,8 +105,6 @@ static void early_init_centaur(struct cpuinfo_x86 *c)
>  #ifdef CONFIG_X86_64
>         set_cpu_cap(c, X86_FEATURE_SYSENTER32);
>  #endif
> -
> -       clear_sched_clock_stable();
>  }
>
>  static void init_centaur(struct cpuinfo_x86 *c)
> diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
> index c188ae5a5d9f..0209907a63b1 100644
> --- a/arch/x86/kernel/cpu/common.c
> +++ b/arch/x86/kernel/cpu/common.c
> @@ -88,7 +88,6 @@ static void default_init(struct cpuinfo_x86 *c)
>                         strcpy(c->x86_model_id, "386");
>         }
>  #endif
> -       clear_sched_clock_stable();
>  }
>
>  static const struct cpu_dev default_cpu = {
> @@ -1077,8 +1076,6 @@ static void identify_cpu(struct cpuinfo_x86 *c)
>          */
>         if (this_cpu->c_init)
>                 this_cpu->c_init(c);
> -       else
> -               clear_sched_clock_stable();
>
>         /* Disable the PN if appropriate */
>         squash_the_stupid_serial_number(c);
> diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
> index 0a3bc19de017..a70fd61095f8 100644
> --- a/arch/x86/kernel/cpu/cyrix.c
> +++ b/arch/x86/kernel/cpu/cyrix.c
> @@ -185,7 +185,6 @@ static void early_init_cyrix(struct cpuinfo_x86 *c)
>                 set_cpu_cap(c, X86_FEATURE_CYRIX_ARR);
>                 break;
>         }
> -       clear_sched_clock_stable();
>  }
>
>  static void init_cyrix(struct cpuinfo_x86 *c)
> diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
> index fe0a615a051b..063197771b8d 100644
> --- a/arch/x86/kernel/cpu/intel.c
> +++ b/arch/x86/kernel/cpu/intel.c
> @@ -162,10 +162,6 @@ static void early_init_intel(struct cpuinfo_x86 *c)
>         if (c->x86_power & (1 << 8)) {
>                 set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
>                 set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
> -               if (check_tsc_unstable())
> -                       clear_sched_clock_stable();
> -       } else {
> -               clear_sched_clock_stable();
>         }
>
>         /* Penwell and Cloverview have the TSC which doesn't sleep on S3 */
> diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c
> index 8457b4978668..d77d07ab310b 100644
> --- a/arch/x86/kernel/cpu/transmeta.c
> +++ b/arch/x86/kernel/cpu/transmeta.c
> @@ -16,8 +16,6 @@ static void early_init_transmeta(struct cpuinfo_x86 *c)
>                 if (xlvl >= 0x80860001)
>                         c->x86_capability[CPUID_8086_0001_EDX] = cpuid_edx(0x80860001);
>         }
> -
> -       clear_sched_clock_stable();
>  }
>
>  static void init_transmeta(struct cpuinfo_x86 *c)
> diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
> index 46bcda4cb1c2..5ca0f52e1ba1 100644
> --- a/arch/x86/kernel/tsc.c
> +++ b/arch/x86/kernel/tsc.c
> @@ -327,9 +327,16 @@ unsigned long long sched_clock(void)
>  {
>         return paravirt_sched_clock();
>  }
> +
> +static inline bool using_native_sched_clock(void)
> +{
> +       return pv_time_ops.sched_clock == native_sched_clock;
> +}
>  #else
>  unsigned long long
>  sched_clock(void) __attribute__((alias("native_sched_clock")));
> +
> +static inline bool using_native_sched_clock(void) { return true; }
>  #endif
>
>  int check_tsc_unstable(void)
> @@ -1112,8 +1119,10 @@ static void tsc_cs_mark_unstable(struct clocksource *cs)
>  {
>         if (tsc_unstable)
>                 return;
> +
>         tsc_unstable = 1;
> -       clear_sched_clock_stable();
> +       if (using_native_sched_clock())
> +               clear_sched_clock_stable();
>         disable_sched_clock_irqtime();
>         pr_info("Marking TSC unstable due to clocksource watchdog\n");
>  }
> @@ -1135,18 +1144,20 @@ static struct clocksource clocksource_tsc = {
>
>  void mark_tsc_unstable(char *reason)
>  {
> -       if (!tsc_unstable) {
> -               tsc_unstable = 1;
> +       if (tsc_unstable)
> +               return;
> +
> +       tsc_unstable = 1;
> +       if (using_native_sched_clock())
>                 clear_sched_clock_stable();
> -               disable_sched_clock_irqtime();
> -               pr_info("Marking TSC unstable due to %s\n", reason);
> -               /* Change only the rating, when not registered */
> -               if (clocksource_tsc.mult)
> -                       clocksource_mark_unstable(&clocksource_tsc);
> -               else {
> -                       clocksource_tsc.flags |= CLOCK_SOURCE_UNSTABLE;
> -                       clocksource_tsc.rating = 0;
> -               }
> +       disable_sched_clock_irqtime();
> +       pr_info("Marking TSC unstable due to %s\n", reason);
> +       /* Change only the rating, when not registered */
> +       if (clocksource_tsc.mult)
> +               clocksource_mark_unstable(&clocksource_tsc);
> +       else {
> +               clocksource_tsc.flags |= CLOCK_SOURCE_UNSTABLE;
> +               clocksource_tsc.rating = 0;
>         }
>  }
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ