Re: Resume problems

From: Rafael J. Wysocki
Date: Mon Oct 22 2007 - 19:15:26 EST

Next message: Ingo Molnar: "Re: [uml-devel] User Mode Linux still doesn't build in 2.6.23-final."
Previous message: Jeremy Fitzhardinge: "Re: [uml-devel] User Mode Linux still doesn't build in 2.6.23-final."
In reply to: Gabriel C: "Re: Resume problems"
Next in thread: Gabriel C: "Re: Resume problems"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Tuesday, 23 October 2007 01:00, Gabriel C wrote:
> Rafael J. Wysocki wrote:
> > On Monday, 22 October 2007 18:15, Gabriel C wrote:
> >> Hi all ,
> >>
> >> I'm running current git + aic7xxx suspend patch from http://bugzilla.kernel.org/show_bug.cgi?id=3062
> >> on a Dell Precision WorkStation 530 MT SMP box ( HT enabled ).
> >>
> >> Suspend works fine but on resume I have some problems.
> >> All CPU's but boot CPU won't come back , everything else seems fine.
> >
> > Can you please try to disable HT and suspend?
>
> So only 'Hibernation' is enabled in kernel and HT disabled in BIOS ?
>
> If you mean that , sure I can try doing so.

With suspend or hibernation enabled in the kernel, but with HT disabled in the
BIOS.

> I also could disable Suspend to RAM completly from BIOS as well if you want.

No, that rather won't work.

> >
> >> ...
> >>
> >> Oct 22 15:02:28 lara [ 49.618795] Enabling non-boot CPUs ...
> >> Oct 22 15:02:28 lara [ 49.622211] PM: Adding info for No Bus:msr1
> >> Oct 22 15:02:28 lara [ 49.622259] PM: Adding info for No Bus:cpu1
> >> Oct 22 15:02:28 lara [ 49.622302] SMP alternatives: switching to SMP code
> >> Oct 22 15:02:28 lara [ 49.623536] Booting processor 1/1 eip 3000
> >> Oct 22 15:02:28 lara [ 54.638093] Not responding.
> >> Oct 22 15:02:28 lara [ 54.638096] Inquiring remote APIC #1...
> >> Oct 22 15:02:28 lara [ 54.638099] ... APIC #1 ID: failed
> >> Oct 22 15:02:28 lara [ 54.638204] ... APIC #1 VERSION: failed
> >> Oct 22 15:02:28 lara [ 54.638307] ... APIC #1 SPIV: failed
> >> Oct 22 15:02:28 lara [ 54.638427] skipping cpu1, didn't come online
> >> Oct 22 15:02:28 lara [ 54.638602] PM: Removing info for No Bus:msr1
> >> Oct 22 15:02:28 lara [ 54.638643] PM: Removing info for No Bus:cpu1
> >> Oct 22 15:02:28 lara [ 54.638678] Error taking CPU1 up: -5
> >> Oct 22 15:02:28 lara [ 54.640908] PM: Adding info for No Bus:msr2
> >> Oct 22 15:02:28 lara [ 54.640939] PM: Adding info for No Bus:cpu2
> >> Oct 22 15:02:28 lara [ 54.640976] SMP alternatives: switching to SMP code
> >> Oct 22 15:02:28 lara [ 54.641961] Booting processor 2/2 eip 3000
> >> Oct 22 15:02:28 lara [ 59.656795] Not responding.
> >> Oct 22 15:02:28 lara [ 59.656799] Inquiring remote APIC #2...
> >> Oct 22 15:02:28 lara [ 59.656803] ... APIC #2 ID: failed
> >> Oct 22 15:02:28 lara [ 59.656907] ... APIC #2 VERSION: failed
> >> Oct 22 15:02:28 lara [ 59.657011] ... APIC #2 SPIV: failed
> >> Oct 22 15:02:28 lara [ 59.657131] skipping cpu2, didn't come online
> >> Oct 22 15:02:28 lara [ 59.657300] PM: Removing info for No Bus:msr2
> >> Oct 22 15:02:28 lara [ 59.657343] PM: Removing info for No Bus:cpu2
> >> Oct 22 15:02:28 lara [ 59.657379] Error taking CPU2 up: -5
> >> Oct 22 15:02:28 lara [ 59.659605] PM: Adding info for No Bus:msr3
> >> Oct 22 15:02:28 lara [ 59.659637] PM: Adding info for No Bus:cpu3
> >> Oct 22 15:02:28 lara [ 59.659673] SMP alternatives: switching to SMP code
> >> Oct 22 15:02:28 lara [ 59.660725] Booting processor 3/3 eip 3000
> >> Oct 22 15:02:28 lara [ 64.675517] Not responding.
> >> Oct 22 15:02:28 lara [ 64.675520] Inquiring remote APIC #3...
> >> Oct 22 15:02:28 lara [ 64.675524] ... APIC #3 ID: failed
> >> Oct 22 15:02:28 lara [ 64.675628] ... APIC #3 VERSION: failed
> >> Oct 22 15:02:28 lara [ 64.675731] ... APIC #3 SPIV: failed
> >> Oct 22 15:02:28 lara [ 64.675859] skipping cpu3, didn't come online
> >> Oct 22 15:02:28 lara [ 64.676017] PM: Removing info for No Bus:msr3
> >> Oct 22 15:02:28 lara [ 64.676059] PM: Removing info for No Bus:cpu3
> >> Oct 22 15:02:28 lara [ 64.676092] Error taking CPU3 up: -5
> >> Oct 22 15:02:28 lara [ 64.676326] evxfevnt-0079 [00] enable : System is already in ACPI mode
> >>
> >> ...
> >>
> >> After I've played with a lot boot options I found out booting with ' acpi=ht ' will make the CPU's work again but now
> >> I have a problem on Suspend. Everything seems to just go down disks etc but the box itself is for some reason still on.
> >> So I've tested reboot=<> options with no luck.
> >> ( after waiting 5 minutes to be sure everything is really off I can just hit power button). On resume now everything is fine.
> >>
> >> I'm not really sure what is wrong here acpi/hibernation/cpu-hotplug or a mix of all so I'm CC'ing linux-acpi as well.
> >> The only thing I noticed is the 'Breaking affinity for irq XX' on suspend without acpi=ht messages.
> >>
> >> I can't even tell whatever other kernel versions are working because aic7xxx driver didn't got suspend support till now
> >> ( or at least never worked here ). I know suspend worked fine on windows with that box.
> >>
> >> There is my config and dmesg ( good and bad one ) :
> >>
> >>
> >> http://194.231.229.228/suspend/acpi=ht_working_dmesg.txt
> >> http://194.231.229.228/suspend/dmesg_broken_cpus_on_resume.txt
> >> http://194.231.229.228/suspend/config
> >
> > Well, I think we have a problem with the CPU hotplug.
> >
> > Can you try to offline-online CPUs (without suspending) and see if that works?
>
> Yes does work when I do it manually :
>
> [ 6687.595842] CPU 1 is now offline
> [ 6687.711425] CPU 2 is now offline
> [ 6687.819330] CPU 3 is now offline
> [ 6687.819337] SMP alternatives: switching to UP code
> [ 6702.109605] SMP alternatives: switching to SMP code
> [ 6702.110634] Booting processor 1/1 eip 3000
> [ 6702.122140] Initializing CPU#1
> [ 6702.182045] Calibrating delay using timer specific routine.. 3989.26 BogoMIPS (lpj=1994633)
> [ 6702.182063] CPU: After generic identify, caps: bfebfbff 00000000 00000000 00000000 00004400 00000000 00000000 00000000
> [ 6702.182085] CPU: Trace cache: 12K uops, L1 D cache: 8K
> [ 6702.182091] CPU: L2 cache: 512K
> [ 6702.182096] CPU: Physical Processor ID: 0
> [ 6702.182102] CPU: After all inits, caps: bfebfbff 00000000 00000000 0000b080 00004400 00000000 00000000 00000000
> [ 6702.182118] Intel machine check architecture supported.
> [ 6702.182130] Intel machine check reporting enabled on CPU#1.
> [ 6702.182137] CPU1: Intel P4/Xeon Extended MCE MSRs (12) available
> [ 6702.182143] CPU1: Thermal monitoring enabled
> [ 6702.183563] CPU1: Intel(R) Xeon(TM) CPU 2.00GHz stepping 07
> [ 6702.184488] checking TSC synchronization [CPU#0 -> CPU#1]: passed.
> [ 6702.205500] Switched to high resolution mode on CPU 1
> [ 6702.210400] SMP alternatives: switching to SMP code
> [ 6702.212196] Booting processor 2/2 eip 3000
> [ 6702.222693] Initializing CPU#2
> [ 6702.282950] Calibrating delay using timer specific routine.. 3988.88 BogoMIPS (lpj=1994443)
> [ 6702.282962] CPU: After generic identify, caps: 3febfbff 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> [ 6702.282974] CPU: Trace cache: 12K uops, L1 D cache: 8K
> [ 6702.282977] CPU: L2 cache: 512K
> [ 6702.282980] CPU: Physical Processor ID: 3
> [ 6702.282983] CPU: After all inits, caps: 3febfbff 00000000 00000000 0000b080 00000000 00000000 00000000 00000000
> [ 6702.282991] Intel machine check architecture supported.
> [ 6702.282998] Intel machine check reporting enabled on CPU#2.
> [ 6702.283001] CPU2: Intel P4/Xeon Extended MCE MSRs (12) available
> [ 6702.283005] CPU2: Thermal monitoring enabled
> [ 6702.283300] CPU2: Intel(R) XEON(TM) CPU 2.00GHz stepping 04
> [ 6702.284296] checking TSC synchronization [CPU#1 -> CPU#2]: passed.
> [ 6702.305317] Switched to high resolution mode on CPU 2
> [ 6702.312356] SMP alternatives: switching to SMP code
> [ 6702.313995] Booting processor 3/3 eip 3000
> [ 6702.324511] Initializing CPU#3
> [ 6702.384864] Calibrating delay using timer specific routine.. 3988.87 BogoMIPS (lpj=1994438)
> [ 6702.384875] CPU: After generic identify, caps: 3febfbff 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> [ 6702.384888] CPU: Trace cache: 12K uops, L1 D cache: 8K
> [ 6702.384891] CPU: L2 cache: 512K
> [ 6702.384894] CPU: Physical Processor ID: 3
> [ 6702.384897] CPU: After all inits, caps: 3febfbff 00000000 00000000 0000b080 00000000 00000000 00000000 00000000
> [ 6702.384905] Intel machine check architecture supported.
> [ 6702.384912] Intel machine check reporting enabled on CPU#3.
> [ 6702.384915] CPU3: Intel P4/Xeon Extended MCE MSRs (12) available
> [ 6702.384919] CPU3: Thermal monitoring enabled
> [ 6702.385146] CPU3: Intel(R) XEON(TM) CPU 2.00GHz stepping 04
> [ 6702.386252] checking TSC synchronization [CPU#1 -> CPU#3]: passed.
> [ 6702.407259] Switched to high resolution mode on CPU 3
>
> ...
>
> done with :
> for i in cpu1 cpu2 cpu3; do echo 0 >/sys/devices/system/cpu/$i/online; done
>
> for i in cpu1 cpu2 cpu3; do echo 1 >/sys/devices/system/cpu/$i/online; done

Hm, well.

Please apply the appended patch and then try:

# echo 8 > /proc/sys/kernel/printk
# echo 5 > /sys/power/pm_test_level
# echo mem > /sys/power/state
(should wait for approx. 3 sec. and return to the boot prompt)
# echo 4 > /sys/power/pm_test_level
# echo mem > /sys/power/state
(should wait for approx. 3 sec. and return to the boot prompt)
...
# echo 1 > /sys/power/pm_test_level
# echo mem > /sys/power/state
(should wait for approx. 3 sec. and return to the boot prompt)

and see if you can reproduce the problem and for which test level.

[Echoing 0 to /sys/power/pm_test_level restores the normal behavior.]

Greetings,
Rafael

---
kernel/power/main.c | 75 ++++++++++++++++++++++++++++++++++++++++++++-------
kernel/power/power.h | 10 ++++++
2 files changed, 76 insertions(+), 9 deletions(-)

Index: linux-2.6/kernel/power/main.c
===================================================================
--- linux-2.6.orig/kernel/power/main.c
+++ linux-2.6/kernel/power/main.c
@@ -28,6 +28,46 @@ BLOCKING_NOTIFIER_HEAD(pm_chain_head);

DEFINE_MUTEX(pm_mutex);

+#ifdef CONFIG_PM_DEBUG
+int pm_test_level = TEST_NONE;
+
+static int suspend_test(int level)
+{
+ if (pm_test_level == level) {
+ printk(KERN_INFO "suspend debug: Waiting for 3 seconds.\n");
+ mdelay(3000);
+ return 1;
+ }
+ return 0;
+}
+
+static ssize_t pm_test_level_show(struct kset *kset, char *buf)
+{
+ return sprintf(buf, "%d\n", pm_test_level);
+}
+
+static ssize_t
+pm_test_level_store(struct kset *kset, const char *buf, size_t n)
+{
+ int val;
+
+ if (sscanf(buf, "%d", &val) != 1)
+ return -EINVAL;
+
+ if (val < TEST_NONE || val > TEST_FREEZER)
+ return -EINVAL;
+
+ pm_test_level = val;
+
+ return n;
+}
+
+power_attr(pm_test_level);
+#else /* !CONFIG_PM_DEBUG */
+static inline int suspend_test(int level) { return 0; }
+#endif /* !CONFIG_PM_DEBUG */
+
+
#ifdef CONFIG_SUSPEND

/* This is just an arbitrary number */
@@ -133,7 +173,10 @@ static int suspend_enter(suspend_state_t
printk(KERN_ERR "Some devices failed to power down\n");
goto Done;
}
- error = suspend_ops->enter(state);
+
+ if (!suspend_test(TEST_CORE))
+ error = suspend_ops->enter(state);
+
device_power_up();
Done:
arch_suspend_enable_irqs();
@@ -164,16 +207,25 @@ int suspend_devices_and_enter(suspend_st
printk(KERN_ERR "Some devices failed to suspend\n");
goto Resume_console;
}
+
+ if (suspend_test(TEST_DEVICES))
+ goto Resume_devices;
+
if (suspend_ops->prepare) {
error = suspend_ops->prepare();
if (error)
goto Resume_devices;
}
+
+ if (suspend_test(TEST_PLATFORM))
+ goto Finish;
+
error = disable_nonboot_cpus();
- if (!error)
+ if (!error && !suspend_test(TEST_CPUS))
suspend_enter(state);

enable_nonboot_cpus();
+ Finish:
if (suspend_ops->finish)
suspend_ops->finish();
Resume_devices:
@@ -240,12 +292,17 @@ static int enter_state(suspend_state_t s
printk("done.\n");

pr_debug("PM: Preparing system for %s sleep\n", pm_states[state]);
- if ((error = suspend_prepare()))
+ error = suspend_prepare();
+ if (error)
goto Unlock;

+ if (suspend_test(TEST_FREEZER))
+ goto Finish;
+
pr_debug("PM: Entering %s sleep\n", pm_states[state]);
error = suspend_devices_and_enter(state);

+ Finish:
pr_debug("PM: Finishing wakeup.\n");
suspend_finish();
Unlock:
@@ -363,18 +420,18 @@ pm_trace_store(struct kset *kset, const
}

power_attr(pm_trace);
+#endif /* CONFIG_PM_TRACE */

static struct attribute * g[] = {
&state_attr.attr,
+#ifdef CONFIG_PM_TRACE
&pm_trace_attr.attr,
+#endif
+#ifdef CONFIG_PM_DEBUG
+ &pm_test_level_attr.attr,
+#endif
NULL,
};
-#else
-static struct attribute * g[] = {
- &state_attr.attr,
- NULL,
-};
-#endif /* CONFIG_PM_TRACE */

static struct attribute_group attr_group = {
.attrs = g,
Index: linux-2.6/kernel/power/power.h
===================================================================
--- linux-2.6.orig/kernel/power/power.h
+++ linux-2.6/kernel/power/power.h
@@ -211,3 +211,13 @@ static inline int pm_notifier_call_chain
return (blocking_notifier_call_chain(&pm_chain_head, val, NULL)
== NOTIFY_BAD) ? -EINVAL : 0;
}
+
+/* Suspend test levels */
+enum {
+ TEST_NONE,
+ TEST_CORE,
+ TEST_CPUS,
+ TEST_PLATFORM,
+ TEST_DEVICES,
+ TEST_FREEZER
+};
-
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/

Next message: Ingo Molnar: "Re: [uml-devel] User Mode Linux still doesn't build in 2.6.23-final."
Previous message: Jeremy Fitzhardinge: "Re: [uml-devel] User Mode Linux still doesn't build in 2.6.23-final."
In reply to: Gabriel C: "Re: Resume problems"
Next in thread: Gabriel C: "Re: Resume problems"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]