Skip to contentSkip to conversation

Bug in HPET RTC emulation.

DISCUSSION Alok Kataria 2 participants 3 messages Feb 18, 2010 – Feb 22, 2010 1 message not in the archive

Subsystems (2)
lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

[Copying LKML]

<ping>

Not sure if you guys got a chance to look at this patch, the fix is pretty trivial too.

Thanks,
Alok

Quoted 63 lines · Alok Kataria

Hi Venki/ Vojtech,

We think there exists a bug in the HPET code that emulates the RTC.

In the normal case, when the RTC frequency is set, the rtc driver tells the hpet code about it here:

int hpet_set_periodic_freq(unsigned long freq)

{
	uint64_t clc;

	if (!is_hpet_enabled())
		return 0;

	if (freq <= DEFAULT_RTC_INT_FREQ)
		hpet_pie_limit = DEFAULT_RTC_INT_FREQ / freq;
	else {
		clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC;
		do_div(clc, freq);
		clc >>= hpet_clockevent.shift;
		hpet_pie_delta = (unsigned long) clc;
	}
	return 1;
}

If freq is set to 64Hz (DEFAULT_RTC_INT_FREQ) or lower, then hpet_pie_limit (a static) is set to non-zero.
Then, on every one-shot HPET interrupt, hpet_rtc_timer_reinit is called to compute the next timeout. Well, that function has this logic:

	if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit)
		delta = hpet_default_delta;
	else
		delta = hpet_pie_delta;

Since hpet_pie_limit is not 0, hpet_default_delta is used. That corresponds to 64Hz.

Now, if you set a different rtc frequency, you'll take the else path through hpet_set_periodic_freq, but unfortunately no one resets hpet_pie_limit back to 0.

Boom....now you are stuck with 64Hz HPET interrupts forever.

The patch below just resets the hpet_pie_limit value when requested freq is greater than DEFAULT_RTC_INT_FREQ, which we think fixes this problem.

Thanks,
Alok

Index: linux-2.6/arch/x86/kernel/hpet.c
===================================================================

arch/x86/kernel/hpet.c +1 -0
@@ -1135,6 +1135,7 @@ int hpet_set_periodic_freq(unsigned long
 		do_div(clc, freq);
 		clc >>= hpet_clockevent.shift;
 		hpet_pie_delta = clc;
+		hpet_pie_limit = 0;
 	}
 	return 1;
 }
#2 lore raw
Quoted 1 line · On Thu, 18 Feb 2010 14:30:08 -0800 Alok Kataria <[email protected]> wrote:

[Copying LKML]

Please do that all the time!

Quoted 5 lines

<ping>

Not sure if you guys got a chance to look at this patch, the fix is pretty trivial too.

Please resend everything, including a Signed-off-by: as per Documentation/SubmittingPatches.

Thanks.

#3 lore raw
Quoted 6 lines · Andrew Morton
Quoted 1 line · On Thu, 18 Feb 2010 14:30:08 -0800 Alok Kataria <[email protected]> wrote:

[Copying LKML]

Please do that all the time!

Quoted 8 lines
Quoted 5 lines

<ping>

Not sure if you guys got a chance to look at this patch, the fix is pretty trivial too.

Please resend everything, including a Signed-off-by: as per Documentation/SubmittingPatches.

Hi Andrew,

Here it is.

--

We think there exists a bug in the HPET code that emulates the RTC.

In the normal case, when the RTC frequency is set, the rtc driver tells the hpet code about it here:

int hpet_set_periodic_freq(unsigned long freq)

{
        uint64_t clc;

        if (!is_hpet_enabled())
                return 0;

        if (freq <= DEFAULT_RTC_INT_FREQ)
                hpet_pie_limit = DEFAULT_RTC_INT_FREQ / freq;
        else {
                clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC;
                do_div(clc, freq);
                clc >>= hpet_clockevent.shift;
                hpet_pie_delta = (unsigned long) clc;
        }
        return 1;
}

If freq is set to 64Hz (DEFAULT_RTC_INT_FREQ) or lower, then hpet_pie_limit (a static) is set to non-zero.
Then, on every one-shot HPET interrupt, hpet_rtc_timer_reinit is called to compute the next timeout. Well, that function has this logic:

        if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit)
                delta = hpet_default_delta;
        else
                delta = hpet_pie_delta;

Since hpet_pie_limit is not 0, hpet_default_delta is used. That corresponds to 64Hz.

Now, if you set a different rtc frequency, you'll take the else path through hpet_set_periodic_freq, but unfortunately no one resets hpet_pie_limit back to 0.

Boom....now you are stuck with 64Hz RTC interrupts forever.

The patch below just resets the hpet_pie_limit value when requested freq is greater than DEFAULT_RTC_INT_FREQ, which we think fixes this problem.

---

 arch/x86/kernel/hpet.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
arch/x86/kernel/hpet.c +1 -0
index ad80a1c..a79a358 100644
@@ -1143,6 +1143,7 @@ int hpet_set_periodic_freq(unsigned long freq)
 		do_div(clc, freq);
 		clc >>= hpet_clockevent.shift;
 		hpet_pie_delta = clc;
+		hpet_pie_limit = 0;
 	}
 	return 1;
 }