Re: ACPI regression in 2.6.25-rc6 (function keys stop working)

[Alexey Starikovskiy <astarikovskiy@suse.de>]

Tim Elliott wrote:
> I am seeing the same problem on my Lenovo 3000 v100. The function keys
> stopped working.
> 
> tim@tim-len:~$ cat /proc/interrupts
>            CPU0       CPU1
>   0:     164105          0   IO-APIC-edge      timer
>   1:      12560          0   IO-APIC-edge      i8042
>   9:          6          0   IO-APIC-fasteoi   acpi
>  12:     117534          0   IO-APIC-edge      i8042
>  14:         51          0   IO-APIC-edge      ide0
>  16:         31          0   IO-APIC-fasteoi   uhci_hcd:usb4,
> i915@pci:0000:00:02.0
>  18:     121892          0   IO-APIC-fasteoi   uhci_hcd:usb3, iwl3945
>  19:      36197          0   IO-APIC-fasteoi   uhci_hcd:usb2, ahci
>  20:       7931          0   IO-APIC-fasteoi   eth0
>  22:      11071          0   IO-APIC-fasteoi   sdhc0:slot0, HDA Intel
>  23:          5          0   IO-APIC-fasteoi   uhci_hcd:usb1, ehci_hcd:usb5
> NMI:          0          0   Non-maskable interrupts
> LOC:      11296      73879   Local timer interrupts
> RES:      23199      49164   Rescheduling interrupts
> CAL:        297      27033   function call interrupts
> TLB:        838       1286   TLB shootdowns
> TRM:          0          0   Thermal event interrupts
> THR:          0          0   Threshold APIC interrupts
> SPU:          0          0   Spurious interrupts
> ERR:          0
> 
> The problem goes away when reverting 2c81ce4c9c37b910210f2640c28e98a0c398dc26
> 
> Reverting the above and applying the patch in comment #38 of bug #9998
> also works (http://bugzilla.kernel.org/show_bug.cgi?id=9998).
reverting the above and patch in comment #38 of #9998 are the same thing.
I am really interested if after revert (or application of #38) you notice a
difference in interrupt count before and after applying #37 from #9998.

Thanks,
Alex.
> 
> Cheers,
> Tim
> 
> On 3/18/08, Alexey Starikovskiy <astarikovskiy@suse.de> wrote:
>> Hi Guillaume,
>>
>>  Please check if the patch to call _PSW methods, attached to bug #9998
>>  helps to reduce number of ACPI interrupts if you revert the 2c81ce4c9c3
>>
>>  Thanks,
>>  Alex.
>>
>>
>>  Guillaume Chazarain wrote:
>>  > Hi,
>>  >
>>  > There is an ACPI regression in 2.6.25-rc6, which causes the ACPI keys
>>  > to stop working on my laptop, an Asus V6VA. git bisect identified the
>>  > appended commit as guilty, and reverting it indeed fixes the problem
>>  > for me. Attached are my .config, a good dmesg and a bad one. The diff
>>  > between both dmesg reveals this interesting hunk:
>>  >
>>  > @@ -121,8 +121,9 @@
>>  >  ACPI: (supports S0 S1 S3 S4 S5)
>>  >  ACPI: Using IOAPIC for interrupt routing
>>  >  ACPI: EC: non-query interrupt received, switching to interrupt mode
>>  > +ACPI: EC: GPE storm detected, disabling EC GPE
>>  >  ACPI: EC: GPE = 0x1c, I/O: command/status = 0x66, data = 0x62
>>  > -ACPI: EC: driver started in interrupt mode
>>  > +ACPI: EC: driver started in poll mode
>>  >  ACPI: PCI Root Bridge [PCI0] (0000:00)
>>  >  pci 0000:00:1f.0: Enabled ICH6/i801 SMBus device
>>  >  pci 0000:00:1f.0: quirk: region 0800-087f claimed by ICH6 ACPI/GPIO/TCO
>>  >
>>  >
>>  > Thanks.
>>  >
>>  >
>>  > commit 2c81ce4c9c37b910210f2640c28e98a0c398dc26
>>  > Author: Alexey Starikovskiy <astarikovskiy@suse.de>
>>  > Date:   Tue Mar 11 13:30:00 2008 -0400
>>  >
>>  >     ACPI: EC: Handle IRQ storm on Acer laptops
>>  >
>>  >     On some Acer systems, the HW fails to clear the GPE source,
>>  >     causing an interrupt storm.
>>  >
>>  >     So in EC interrupt mode, we count how many interrupts we
>>  >     receive when waiting.  If we get more than 5, we give
>>  >     up on interrupt mode and revert to polling mode.
>>  >
>>  >     Also, for polling mode to work on Acers, we need
>>  >     to insert a delay.
>>  >
>>  >     Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
>>  >     Signed-off-by: Len Brown <len.brown@intel.com>
>>  >
>>  > diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
>>  > index caf873c..2c77359 100644
>>  > --- a/drivers/acpi/ec.c
>>  > +++ b/drivers/acpi/ec.c
>>  > @@ -129,6 +129,7 @@ static struct acpi_ec {
>>  >       struct mutex lock;
>>  >       wait_queue_head_t wait;
>>  >       struct list_head list;
>>  > +     atomic_t irq_count;
>>  >       u8 handlers_installed;
>>  >  } *boot_ec, *first_ec;
>>  >
>>  > @@ -181,6 +182,8 @@ static int acpi_ec_wait(struct acpi_ec *ec, enum
>>  > ec_event event, int force_poll)
>>  >  {
>>  >       int ret = 0;
>>  >
>>  > +     atomic_set(&ec->irq_count, 0);
>>  > +
>>  >       if (unlikely(event == ACPI_EC_EVENT_OBF_1 &&
>>  >                    test_bit(EC_FLAGS_NO_OBF1_GPE, &ec->flags)))
>>  >               force_poll = 1;
>>  > @@ -227,6 +230,7 @@ static int acpi_ec_wait(struct acpi_ec *ec, enum
>>  > ec_event event, int force_poll)
>>  >               while (time_before(jiffies, delay)) {
>>  >                       if (acpi_ec_check_status(ec, event))
>>  >                               goto end;
>>  > +                     msleep(5);
>>  >               }
>>  >       }
>>  >       pr_err(PREFIX "acpi_ec_wait timeout,"
>>  > @@ -529,6 +533,13 @@ static u32 acpi_ec_gpe_handler(void *data)
>>  >       struct acpi_ec *ec = data;
>>  >
>>  >       pr_debug(PREFIX "~~~> interrupt\n");
>>  > +     atomic_inc(&ec->irq_count);
>>  > +     if (atomic_read(&ec->irq_count) > 5) {
>>  > +             pr_err(PREFIX "GPE storm detected, disabling EC GPE\n");
>>  > +             acpi_disable_gpe(NULL, ec->gpe, ACPI_ISR);
>>  > +             clear_bit(EC_FLAGS_GPE_MODE, &ec->flags);
>>  > +             return ACPI_INTERRUPT_HANDLED;
>>  > +     }
>>  >       clear_bit(EC_FLAGS_WAIT_GPE, &ec->flags);
>>  >       if (test_bit(EC_FLAGS_GPE_MODE, &ec->flags))
>>  >               wake_up(&ec->wait);
>>  >
>>  >
>>
>>
>>  --
>>  To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>  the body of a message to majordomo@vger.kernel.org
>>  More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>  Please read the FAQ at  http://www.tux.org/lkml/
>>