From: Kerstin Jonsson <[email protected]>
When the SMP kernel decides to crash_kexec() the local APICs may have
pending interrupts in their vector tables.
The setup routine for the local APIC has a deficient mechanism for
clearing these interrupts, it only handles interrupts that has already
been dispatched to the local core for servicing (the ISR register)
safely, it doesn't consider lower prioritized queued interrupts stored
in the IRR register.
If you have more than one pending interrupt within the same 32 bit word
in the LAPIC vector table registers you may find yourself entering the
IO APIC setup with pending interrupts left in the LAPIC. This is a
situation for wich the IO APIC setup is not prepared. Depending of
what/which interrupt vector/vectors are stuck in the APIC tables your
system may show various degrees of malfunctioning.
That was the reason why the check_timer() failed in our system, the
timer interrupts was blocked by pending interrupts from the old kernel
when routed trough the IO APIC.
Additional comment from Jiri Bohac:
==============
If this should go into stable release,
I'd add some kind of limit on the number of iterations, just to be safe from
hard to debug lock-ups:
+if (loops++ > MAX_LOOPS) {
+ printk("LAPIC pending clean-up")
+ break;
+}
while (queued);
with MAX_LOOPS something like 1E9 this would leave plenty of time for the
pending IRQs to be cleared and would and still cause at most a second of delay
if the loop were to lock-up for whatever reason.
==============
>From [email protected]:
Merged Jiri suggestion into the patch.
Also made the max_loops depend on cpu_khz. Not sure how long an apic_read
takes, as it is on the CPU it may only be one cycle and we now wait 1 sec
in WARN_ON(..) case?
CC: [email protected]
CC: "Yinghai Lu" <[email protected]>
CC: [email protected]
CC: [email protected]
CC: "Kerstin Jonsson" <[email protected]>
Signed-off-by: Thomas Renninger <[email protected]>
---
arch/x86/kernel/apic/apic.c | 34 +++++++++++++++++++++++++---------
1 files changed, 25 insertions(+), 9 deletions(-)
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 3987e44..912dd59 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -51,6 +51,7 @@
#include <asm/smp.h>
#include <asm/mce.h>
#include <asm/kvm_para.h>
+#include <asm/tsc.h>
unsigned int num_processors;
@@ -1151,8 +1152,8 @@ static void __cpuinit lapic_setup_esr(void)
*/
void __cpuinit setup_local_APIC(void)
{
- unsigned int value;
- int i, j;
+ unsigned int value, queued;
+ int i, j, acked = 0, max_loops = cpu_khz * 1000;
if (disable_apic) {
arch_disable_smp_support();
@@ -1204,13 +1205,28 @@ void __cpuinit setup_local_APIC(void)
* the interrupt. Hence a vector might get locked. It was noticed
* for timer irq (vector 0x31). Issue an extra EOI to clear ISR.
*/
- for (i = APIC_ISR_NR - 1; i >= 0; i--) {
- value = apic_read(APIC_ISR + i*0x10);
- for (j = 31; j >= 0; j--) {
- if (value & (1<<j))
- ack_APIC_irq();
- }
- }
+ do {
+ queued = 0;
+ for (i = APIC_ISR_NR - 1; i >= 0; i--)
+ queued |= apic_read(APIC_IRR + i*0x10);
+
+ for (i = APIC_ISR_NR - 1; i >= 0; i--) {
+ value = apic_read(APIC_ISR + i*0x10);
+ for (j = 31; j >= 0; j--) {
+ if (value & (1<<j)) {
+ ack_APIC_irq();
+ acked++;
+ }
+ }
+ }
+ if (acked > 256) {
+ printk(KERN_ERR "LAPIC pending interrupts after %d EOI\n",
+ acked);
+ break;
+ }
+ max_loops--;
+ } while (queued && max_loops > 0);
+ WARN_ON(!max_loops);
/*
* Now that we are all set up, enable the APIC
--
1.6.3
On Tuesday 23 February 2010 12:51:25 Thomas Renninger wrote:
> From: Kerstin Jonsson <[email protected]>
...
> + int i, j, acked = 0, max_loops = cpu_khz * 1000;
Grmpfl, an unsigned long for max_loops, probably is a better idea...
What do you think, could this one get picked up then, anything else
to improve?
Thanks,
Thomas
On 02/23/2010 01:51 PM, Thomas Renninger wrote:
> From: Kerstin Jonsson<[email protected]>
>
> When the SMP kernel decides to crash_kexec() the local APICs may have
> pending interrupts in their vector tables.
> The setup routine for the local APIC has a deficient mechanism for
> clearing these interrupts, it only handles interrupts that has already
> been dispatched to the local core for servicing (the ISR register)
> safely, it doesn't consider lower prioritized queued interrupts stored
> in the IRR register.
>
> If you have more than one pending interrupt within the same 32 bit word
> in the LAPIC vector table registers you may find yourself entering the
> IO APIC setup with pending interrupts left in the LAPIC. This is a
> situation for wich the IO APIC setup is not prepared. Depending of
> what/which interrupt vector/vectors are stuck in the APIC tables your
> system may show various degrees of malfunctioning.
> That was the reason why the check_timer() failed in our system, the
> timer interrupts was blocked by pending interrupts from the old kernel
> when routed trough the IO APIC.
>
> Additional comment from Jiri Bohac:
> ==============
> If this should go into stable release,
> I'd add some kind of limit on the number of iterations, just to be safe from
> hard to debug lock-ups:
>
> +if (loops++> MAX_LOOPS) {
> + printk("LAPIC pending clean-up")
> + break;
> +}
> while (queued);
>
> with MAX_LOOPS something like 1E9 this would leave plenty of time for the
> pending IRQs to be cleared and would and still cause at most a second of delay
> if the loop were to lock-up for whatever reason.
> ==============
>
> From [email protected]:
> Merged Jiri suggestion into the patch.
> Also made the max_loops depend on cpu_khz. Not sure how long an apic_read
> takes, as it is on the CPU it may only be one cycle and we now wait 1 sec
> in WARN_ON(..) case?
>
>
An apic_read() can take a couple of microseconds when running
virtualized, so this loop may run for hours. On the other hand,
virtualized hardware is unlikely to misbehave.
Still I recommend using a clocksource (tsc would do) and not a loop count.
--
error compiling committee.c: too many arguments to function
> ________________________________________
> From: Avi Kivity [[email protected]]
> Sent: Tuesday, February 23, 2010 1:03 PM
> To: Thomas Renninger
> Cc: [email protected]; Kerstin Jonsson; [email protected]; Yinghai Lu; [email protected]; [email protected]
> Subject: Re: [PATCH] x86 apic: Ack all pending irqs when crashed/on kexec
>
> On 02/23/2010 01:51 PM, Thomas Renninger wrote:
>
>> From: Kerstin Jonsson<[email protected]>
>>
>> When the SMP kernel decides to crash_kexec() the local APICs may have
>> pending interrupts in their vector tables.
>> The setup routine for the local APIC has a deficient mechanism for
>> clearing these interrupts, it only handles interrupts that has already
>> been dispatched to the local core for servicing (the ISR register)
>> safely, it doesn't consider lower prioritized queued interrupts stored
>> in the IRR register.
>>
>> If you have more than one pending interrupt within the same 32 bit word
>> in the LAPIC vector table registers you may find yourself entering the
>> IO APIC setup with pending interrupts left in the LAPIC. This is a
>> situation for wich the IO APIC setup is not prepared. Depending of
>> what/which interrupt vector/vectors are stuck in the APIC tables your
>> system may show various degrees of malfunctioning.
>> That was the reason why the check_timer() failed in our system, the
>> timer interrupts was blocked by pending interrupts from the old kernel
>> when routed trough the IO APIC.
>>
>> Additional comment from Jiri Bohac:
>> ==============
>> If this should go into stable release,
>> I'd add some kind of limit on the number of iterations, just to be safe from
>> hard to debug lock-ups:
>>
>> +if (loops++> MAX_LOOPS) {
>> + printk("LAPIC pending clean-up")
>> + break;
>> +}
>> while (queued);
>>
>> with MAX_LOOPS something like 1E9 this would leave plenty of time for the
>> pending IRQs to be cleared and would and still cause at most a second of delay
>> if the loop were to lock-up for whatever reason.
>> ==============
>>
>> From [email protected]:
>> Merged Jiri suggestion into the patch.
>> Also made the max_loops depend on cpu_khz. Not sure how long an apic_read
>> takes, as it is on the CPU it may only be one cycle and we now wait 1 sec
>> in WARN_ON(..) case?
>>
>>
>>
>
> An apic_read() can take a couple of microseconds when running
> virtualized, so this loop may run for hours. On the other hand,
> virtualized hardware is unlikely to misbehave.
>
> Still I recommend using a clocksource (tsc would do) and not a loop count.
>
> --
> error compiling committee.c: too many arguments to function
>
>
>
Is it possible/thinkable to distinguish between real and virtual targets?
I.e. to somehow detect that the target is a virtual machine and adapt accordingly.
There may be other cases as well, in which one would benefit from taking
target type into consideration when e.g. estimating the reasonable number of cycles
for a specific operation.