Re: Long timeout when booting >= 2.6.38

From: Christian Hoffmann
Date: Thu May 05 2011 - 13:47:15 EST


On 05/05/2011 10:25 AM, Thomas Gleixner wrote:
On Wed, 4 May 2011, john stultz wrote:

On Wed, 2011-05-04 at 09:31 +0200, Christian Hoffmann wrote:
In fact, switching to acpi_pm doesn't seem to work:

$ dmesg | grep clock

[ 0.000000] Command line: BOOT_IMAGE=/vmlinuz-2.6.39-rc5-ch-broken+
root=/dev/mapper/vg-root ro quiet splash vt.handoff=7 clocksource=acpi_pm
[ 0.000000] Kernel command line:
BOOT_IMAGE=/vmlinuz-2.6.39-rc5-ch-broken+ root=/dev/mapper/vg-root ro
quiet splash vt.handoff=7 clocksource=acpi_pm
[ 0.000000] hpet clockevent registered
[ 1.413835] Switching to clocksource hpet
[ 1.420762] Override clocksource acpi_pm is not HRT compatible.
Cannot switch while in HRT/NOHZ mode
[ 147.940143] Refined TSC clocksource calibration: 2809.409 MHz.
[ 147.940147] Switching to clocksource tsc

So I think I've sorted this out.

The watchdog code is what enables CLOCK_SOURCE_VALID_FOR_HRES, but we
actually end up selecting the clocksource before we enqueue it into the
watchdog list, so that's why we see the warning.

I suspect the following will resolve this detail.

It does. Good catch!

Signed-off-by: John Stultz<johnstul@xxxxxxxxxx>

diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 6519cf6..0e17c10 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -685,8 +685,8 @@ int __clocksource_register_scale(struct clocksource *cs, u32 scale, u32 freq)
/* Add clocksource to the clcoksource list */
mutex_lock(&clocksource_mutex);
clocksource_enqueue(cs);
- clocksource_select();
clocksource_enqueue_watchdog(cs);
+ clocksource_select();
mutex_unlock(&clocksource_mutex);
return 0;
}
@@ -706,8 +706,8 @@ int clocksource_register(struct clocksource *cs)

mutex_lock(&clocksource_mutex);
clocksource_enqueue(cs);
- clocksource_select();
clocksource_enqueue_watchdog(cs);
+ clocksource_select();
mutex_unlock(&clocksource_mutex);
return 0;
}


I can now select clocksource=acpi_pm as kernel parameter. The boot still hangs after "MSI quirk detected...", but the timestamps in dmesg don't show the jump anymore.

[ 1.426107] NET: Registered protocol family 1
[ 1.426162] pci 0000:00:01.0: MSI quirk detected; subordinate MSI disabled
[ 2.300212] pci 0000:01:05.0: Boot video device
[ 2.300226] PCI: CLS 64 bytes, default 64


Full dmesg is here:
http://pastebin.com/raw.php?i=3Fctvsrp

Rgds,
Chris

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/