mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Aithal, Srikanth" <Srikanth.Aithal@amd.com>
To: <mt4022@columbia.edu>, Thomas Gleixner <tglx@kernel.org>,
	John Stultz <jstultz@google.com>
Cc: Stephen Boyd <sboyd@kernel.org>,
	Miroslav Lichvar <mlichvar@redhat.com>,
	Borislav Petkov <bp@alien8.de>,
	K Prateek Nayak <kprateek.nayak@amd.com>,
	<linux-kernel@vger.kernel.org>, <tz2294@columbia.edu>,
	<stable@vger.kernel.org>
Subject: Re: [PATCH v2] clocksource: Use the clocksource under test for the skew check timeout
Date: Mon, 16 Nov 2026 13:00:50 +0530	[thread overview]
Message-ID: <a2854dd0-fd12-4a7b-81cf-4c9d823030b9@amd.com> (raw)
In-Reply-To: <20261113-clocksource-skew-timeout-v2-1-9361d9a18262@columbia.edu>


On 11/14/2026 9:42 AM, Mani Tofigh via B4 Relay wrote:
> [Some people who received this message don't often get email from devnull+mt4022.columbia.edu@kernel.org. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
>
> From: Mani Tofigh <mt4022@columbia.edu>
>
> watchdog_check_skew() waits for the remote CPU with interrupts disabled
> and gives up after a timeout measured with ktime_get_raw_fast_ns().
> While the clocksource is still jiffies during boot (or with
> clocksource=jiffies) that timeout can't expire, because only the boot
> CPU's tick advances jiffies and that's the CPU spinning with interrupts
> disabled.
>
> The boot then hangs silently when the remote CPU can't answer, e.g., in
> the stop_machine() of the switch away from jiffies:
>
>    CPU0                                  CPU1
>    multi_cpu_stop()                      multi_cpu_stop()
>                                            local_irq_disable()
>                                            /* waits for CPU0 */
>      <timer interrupt>
>        clocksource_watchdog()
>          /* sends CSD to CPU1 */
>          watchdog_check_skew()
>            /* waits for CPU1 forever */
>
> Measure the timeout with the clocksource under test instead, which keeps
> counting with interrupts disabled.  Also, when the remote CPU still
> hasn't answered by the next watchdog round, report it as stalled and
> mark the clocksource unstable instead of retrying forever.
>
> With the fix, the watchdog gives up after the timeout and the boot
> continues.
>
> Fixes: 763aacf86f1b ("clocksource: Rewrite watchdog code completely")
> Reported-by: Srikanth Aithal <Srikanth.Aithal@amd.com>
> Closes: https://lore.kernel.org/all/e4a41d43-4d60-4132-aa2d-70e038a3e9dd@amd.com/
> Suggested-by: Thomas Gleixner <tglx@kernel.org>
> Signed-off-by: Mani Tofigh <mt4022@columbia.edu>
> Tested-by: Srikanth Aithal <Srikanth.Aithal@amd.com>
> Cc: stable@vger.kernel.org
> ---
> The reproducer hits it within about 80s on a kernel booted with
> clocksource=jiffies, with both vCPUs of the guest on one host core:
> https://github.com/manitofigh/files/blob/main/linux-repros/wdboot.c
> ---
> Changes in v2:
> - Fold in the stall check (Thomas)
> - Drop the hard lockup backtrace (Thomas)
> - Add Srikanth's Reported-by, Closes and Tested-by
> - Link to v1: https://patch.msgid.link/20261102-clocksource-skew-timeout-v1-1-4ff412918591@columbia.edu
> ---
>   kernel/time/clocksource.c | 32 ++++++++++++++++++++++++--------
>   1 file changed, 24 insertions(+), 8 deletions(-)
>
> diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
> index 58c0ca11cdc5..5dfa72a2db68 100644
> --- a/kernel/time/clocksource.c
> +++ b/kernel/time/clocksource.c
> @@ -250,6 +250,7 @@ enum wd_result {
>          WD_FREQ_RESET,
>          WD_FREQ_SKEWED,
>          WD_CPU_TIMEOUT,
> +       WD_CPU_STALLED,
>          WD_CPU_SKEWED,
>   };
>
> @@ -305,11 +306,16 @@ static bool watchdog_wait_seq(struct watchdog_cpu_data *wd, u64 start, int seq)
>                  if (READ_ONCE(wd->result) != WD_SUCCESS)
>                          return false;
>
> -               /* Prevent endless loops if the other CPU does not react. */
> +               /*
> +                * Prevent endless loops if the other CPU does not react.  The
> +                * timeout uses the clocksource under test because the timekeeping
> +                * clocksource can be jiffies, which doesn't advance while this CPU
> +                * waits with interrupts disabled.
> +                */
>                  if (cnt == 5000) {
> -                       u64 nsecs = ktime_get_raw_fast_ns();
> +                       struct clocksource *cs = wd->cs;
>
> -                       if (nsecs - start >=wd->timeout_ns) {
> +                       if (cycles_to_nsec_safe(cs, start, cs->read(cs)) >= wd->timeout_ns) {
>                                  watchdog_set_result(wd, WD_CPU_TIMEOUT);
>                                  return false;
>                          }
> @@ -322,12 +328,13 @@ static bool watchdog_wait_seq(struct watchdog_cpu_data *wd, u64 start, int seq)
>
>   static void watchdog_check_skew(struct watchdog_cpu_data *wd, int index)
>   {
> -       u64 prev, now, delta, start = ktime_get_raw_fast_ns();
> +       u64 prev, now, delta, start;
>          int local = index, remote = (index + 1) & 0x1;
>          struct clocksource *cs = wd->cs;
>
>          /* Set the local timestamp so that the first iteration works correctly */
> -       wd->cpu_ts[local] = cs->read(cs);
> +       start = cs->read(cs);
> +       wd->cpu_ts[local] = start;
>
>          /* Signal arrival */
>          atomic_inc(&wd->seq);
> @@ -392,7 +399,7 @@ static void __watchdog_check_cpu_skew(struct clocksource *cs, unsigned int cpu)
>
>          wd = per_cpu_ptr(&watchdog_cpu_data, cpu);
>          if (atomic_read(&wd->remote_inprogress) || wd_csd_locked(wd)) {
> -               watchdog_data.result = WD_CPU_TIMEOUT;
> +               watchdog_data.result = WD_CPU_STALLED;
>                  return;
>          }
>
> @@ -584,11 +591,16 @@ static void watchdog_print_freq_skew(struct clocksource *cs)
>          pr_warn("Clocksource %20s interval: %16lluns\n", cs->name, watchdog_data.cs_delta);
>   }
>
> -static void watchdog_handle_remote_timeout(struct clocksource *cs)
> +static void watchdog_print_remote_timeout(struct clocksource *cs)
>   {
>          pr_info_once("Watchdog remote CPU %u read timed out\n", watchdog_data.curr_cpu);
>   }
>
> +static void watchdog_print_remote_stall(struct clocksource *cs)
> +{
> +       pr_warn("Watchdog remote CPU %u stalled\n", watchdog_data.curr_cpu);
> +}
> +
>   static void watchdog_print_remote_skew(struct clocksource *cs)
>   {
>          pr_warn("Marking clocksource %s unstable due to inter CPU skew\n", cs->name);
> @@ -629,9 +641,13 @@ static void watchdog_check_result(struct clocksource *cs)
>
>          case WD_CPU_TIMEOUT:
>                  /* Remote check timed out. Try again next cycle. */
> -               watchdog_handle_remote_timeout(cs);
> +               watchdog_print_remote_timeout(cs);
>                  return;
>
> +       case WD_CPU_STALLED:
> +               watchdog_print_remote_stall(cs);
> +               break;
> +
>          case WD_CPU_SKEWED:
>                  watchdog_print_remote_skew(cs);
>                  break;
>
> ---
> base-commit: c1b96ba87e3caaa701d6dba254e58db7cc3c99be
> change-id: 20261102-clocksource-skew-timeout-f2a9e5fb3f75
>
> Best regards,
> --
> Mani Tofigh <mt4022@columbia.edu>


Retested v2 on Linus tip aebb727447640 (v7.4-rc3). The issue reported is 
fixed. Tested-by still holds for v2 as posted.


>
>

      reply	other threads:[~2026-11-16 11:24 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-14  4:12 Mani Tofigh via B4 Relay
2026-11-16  7:30 ` Aithal, Srikanth [this message]

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=a2854dd0-fd12-4a7b-81cf-4c9d823030b9@amd.com \
    --to=srikanth.aithal@amd.com \
    --cc=bp@alien8.de \
    --cc=jstultz@google.com \
    --cc=kprateek.nayak@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mlichvar@redhat.com \
    --cc=mt4022@columbia.edu \
    --cc=sboyd@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@kernel.org \
    --cc=tz2294@columbia.edu \
    /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

all inboxes | Powered by JetHome®