Re: HPET regression in 2.6.26 versus 2.6.25 -- found another user with the same regression

[Yinghai Lu]

On Sat, Aug 23, 2008 at 1:33 PM, Rufus & Azrael
<rufus-azrael@xxxxxxxxxxxxxx> wrote:
> Yinghai Lu wrote:
>>
>> On Sat, Aug 23, 2008 at 1:15 PM, Rufus&  Azrael
>> <rufus-azrael@xxxxxxxxxxxxxx>  wrote:
>>
>>>
>>> Yinghai Lu wrote:
>>>
>>>>
>>>> On Sat, Aug 23, 2008 at 12:40 PM, Rufus&   Azrael
>>>> <rufus-azrael@xxxxxxxxxxxxxx>   wrote:
>>>>
>>>>
>>>>>
>>>>> Yinghai Lu wrote:
>>>>>
>>>>>
>>>>>>
>>>>>> On Sat, Aug 23, 2008 at 11:26 AM, Rufus&    Azrael
>>>>>> <rufus-azrael@xxxxxxxxxxxxxx>    wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>>
>>>>>>>> On Sat, Aug 23, 2008 at 4:58 AM, David
>>>>>>>> Witbrodt<dawitbro@xxxxxxxxxxxxx>
>>>>>>>>  wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>  please send out after booting with hpet=disable
>>>>>>>>>>  lspci -tv
>>>>>>>>>>  lspci -vvxxx
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>  Info attached:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> 00:14.0 SMBus: ATI Technologies Inc SBx00 SMBus Controller (rev 13)
>>>>>>>>        Subsystem: Elitegroup Computer Systems Device 2621
>>>>>>>>        Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop-
>>>>>>>> ParErr-
>>>>>>>> Stepping- SERR- FastB2B- DisINTx+
>>>>>>>>        Status: Cap+ 66MHz+ UDF- FastB2B- ParErr-
>>>>>>>> DEVSEL=medium>TAbort-
>>>>>>>> <TAbort-<MAbort->SERR-<PERR- INTx-
>>>>>>>>        Region 0: I/O ports at fa00 [size=16]
>>>>>>>>        Region 1: Memory at 80000000 (32-bit, non-prefetchable)
>>>>>>>> [size=1K]
>>>>>>>>        Capabilities: [b0] HyperTransport: MSI Mapping Enable- Fixed+
>>>>>>>>        Kernel driver in use: piix4_smbus
>>>>>>>> 00: 02 10 85 43 03 04 30 02 13 00 05 0c 00 00 80 00
>>>>>>>> 10: 01 fa 00 00 00 00 00 80 00 00 00 00 00 00 00 00
>>>>>>>> 20: 00 00 00 00 00 00 00 00 00 00 00 00 19 10 21 26
>>>>>>>> 30: 00 00 00 00 b0 00 00 00 00 00 00 00 00 00 00 00
>>>>>>>> 40: d4 3b 00 05 00 00 00 00 0f ff 00 00 00 00 00 00
>>>>>>>> 50: f0 01 f0 0e f0 0f f0 07 11 0b f0 0f 00 00 00 00
>>>>>>>> 60: 01 00 24 00 bf fd de 8f 3f 90 00 00 20 00 00 00
>>>>>>>> 70: 00 01 00 00 08 00 c0 fe ff 6e 00 00 00 00 f0 0f
>>>>>>>> 80: f0 0a f0 0f 00 00 00 00 00 00 00 00 8c 00 00 80
>>>>>>>> 90: 01 fa 00 00 fa de ff 00 00 00 00 00 00 00 00 00
>>>>>>>> a0: 00 00 ff ff ff ff f0 08 ff fd 12 02 16 7b 20 00
>>>>>>>> b0: 08 00 02 a8 00 00 00 00 00 00 00 00 f0 0f 08 1a
>>>>>>>> c0: ff ff ff ff 00 00 00 00 00 00 00 00 00 00 00 00
>>>>>>>> d0: 00 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00
>>>>>>>> e0: 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>>>>>>>> f0: d8 0c 00 00 00 00 44 00 00 00 00 00 aa 00 10 01
>>>>>>>> it seems kernel update BAR1 from 0xfed00000 to 0x80000000 after it
>>>>>>>> can
>>>>>>>> not give 0xfed00000 to that device.
>>>>>>>> please try attached patch, and send out lspci -vvxxxx again.
>>>>>>>> YH
>>>>>>>> [PATCH] x86: protect hpet in BAR for one ATI chipset
>>>>>>>>
>>>>>>>> so avoid kernel don't allocate nre resource for it because it can
>>>>>>>> not
>>>>>>>> allocate the old
>>>>>>>> address from BIOS.
>>>>>>>>
>>>>>>>> the same way like some IO APIC address in BAR handling
>>>>>>>>
>>>>>>>> Signed-off-by: Yinghai Lu<yhlu.kenrel@xxxxxxxxx>
>>>>>>>>
>>>>>>>> ---
>>>>>>>>  drivers/pci/quirks.c |   16 ++++++++++++++++
>>>>>>>>  1 file changed, 16 insertions(+)
>>>>>>>> Index: linux-2.6/drivers/pci/quirks.c
>>>>>>>> ===================================================================
>>>>>>>> --- linux-2.6.orig/drivers/pci/quirks.c
>>>>>>>> +++ linux-2.6/drivers/pci/quirks.c
>>>>>>>> @@ -1918,6 +1918,22 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_B
>>>>>>>>                         PCI_DEVICE_ID_NX2_5709S,
>>>>>>>>                         quirk_brcm_570x_limit_vpd);
>>>>>>>>
>>>>>>>> +static void __init quirk_hpet_in_bar(struct pci_dev *pdev)
>>>>>>>> +{
>>>>>>>> +       int i;
>>>>>>>> +       unsigned int base, size;
>>>>>>>> +
>>>>>>>> +       /* the BAR1 is the location of the HPET...we must
>>>>>>>> +        * not touch this, so forcibly insert it into the resource
>>>>>>>> tree
>>>>>>>> */
>>>>>>>> +       base = pci_resource_start(pdev, 1);
>>>>>>>> +       size = pci_resource_len(pdev, 1);
>>>>>>>> +       if (base&&     size) {
>>>>>>>> +               insert_resource(&iomem_resource,&pdev->resource[1]);
>>>>>>>> +               dev_info(&dev->dev, "HPET at %04x-%04x\n", name,
>>>>>>>> base,
>>>>>>>> base + size - 1);
>>>>>>>> +       }
>>>>>>>> +}
>>>>>>>> +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, 0x3485,
>>>>>>>> quirk_hpet_in_bar);
>>>>>>>> +
>>>>>>>>  #ifdef CONFIG_PCI_MSI
>>>>>>>>  /* Some chipsets do not support MSI. We cannot easily rely on
>>>>>>>> setting
>>>>>>>>   * PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> Hi Yinghai,
>>>>>>>
>>>>>>>
>>>>>>> Your patch seems to be broken, see :
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>>  CC      drivers/pci/quirks.o
>>>>>>>> drivers/pci/quirks.c: In function 'quirk_hpet_in_bar':
>>>>>>>> drivers/pci/quirks.c:1804: erreur: 'dev' undeclared (first use in
>>>>>>>> this
>>>>>>>> function)
>>>>>>>> drivers/pci/quirks.c:1804: erreur: (Each undeclared identifier is
>>>>>>>> reported
>>>>>>>> only once
>>>>>>>> drivers/pci/quirks.c:1804: erreur: for each function it appears in.)
>>>>>>>> drivers/pci/quirks.c:1804: erreur: 'name' undeclared (first use in
>>>>>>>> this
>>>>>>>> function)
>>>>>>>> drivers/pci/quirks.c:1804: attention : trop d'arguments pour le
>>>>>>>> format
>>>>>>>> drivers/pci/quirks.c:1795: attention : unused variable 'i'
>>>>>>>> make[2]: *** [drivers/pci/quirks.o] Erreur 1
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>
>>>>>> please check v2.
>>>>>>
>>>>>> YH
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> Thanks Yinghai,
>>>>>
>>>>> Your patch applied, I attach lspci -vvxxxx.
>>>>>
>>>>> Regards.
>>>>>
>>>>>
>>>>> 00:14.0 SMBus: ATI Technologies Inc SBx00 SMBus Controller (rev 14)
>>>>>        Subsystem: Micro-Star International Co., Ltd. Device 7327
>>>>>        Control: I/O+ Mem- BusMaster- SpecCycle- MemWINV- VGASnoop-
>>>>> ParErr-
>>>>> Stepping- SERR- FastB2B- DisINTx+
>>>>>        Status: Cap+ 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium>TAbort-
>>>>> <TAbort-<MAbort->SERR-<PERR- INTx-
>>>>>        Region 0: I/O ports at 0b00 [size=16]
>>>>>        Capabilities: [b0] HyperTransport: MSI Mapping Enable- Fixed+
>>>>>        Kernel driver in use: piix4_smbus
>>>>> 00: 02 10 85 43 01 04 30 02 14 00 05 0c 00 00 80 00
>>>>> 10: 01 0b 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>>>>> 20: 00 00 00 00 00 00 00 00 00 00 00 00 62 14 27 73
>>>>> 30: 00 00 00 00 b0 00 00 00 00 00 00 00 00 00 00 00
>>>>> 40: d4 2b 00 05 00 00 00 00 0f ff 00 00 00 00 00 00
>>>>> 50: 10 01 00 00 00 00 80 08 11 0b f0 0f 00 00 00 00
>>>>> 60: 01 00 24 00 bf fd 9e 83 3f 90 00 00 20 00 00 00
>>>>> 70: 00 01 00 00 08 00 c0 fe ff 2e 00 00 00 00 f0 0f
>>>>> 80: f4 0a f0 0f 00 00 00 00 00 00 00 00 8c 00 00 80
>>>>> 90: 01 0b 00 00 3b de ff 00 00 00 00 00 00 00 00 00
>>>>> a0: 00 00 ff ff ff ff f0 08 02 f0 02 00 16 79 00 18
>>>>> b0: 08 00 02 a8 00 00 00 00 00 00 00 00 f0 0f 08 1a
>>>>> c0: ff ff ff ff 00 00 00 00 00 00 00 00 00 00 00 00
>>>>> d0: 00 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00
>>>>> e0: 20 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>>>>> f0: d8 0c 00 00 00 00 44 00 00 00 00 00 aa 00 10 01
>>>>>
>>>>>
>>>>
>>>> bios didn't set HPET in BAR1?
>>>>
>>>> YH
>>>>
>>>>
>>>>
>>>
>>> Yinghai,
>>>
>>> How do I check this ?
>>>
>>>
>>
>> it seems that you BIOS have HPET in acpi tables, but didn't set that
>> in BAR1 of 00:14.0
>>
>> or BIOS have set one magic bit to hide BAR1 to OS.
>>
>> please check if your hpet works. if so, we could have other quirks to
>> hide BAR1 for David's system.
>>
>> YH
>>
>>
>
> Yinghai,
>
> I found these lines in dmesg :
>
>> [    0.145977] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0, 0
>> [    0.145977] hpet0: 4 32-bit timers, 14318180 Hz
>
> do you think hpet works ?
>

cat /proc/timer_list

YH
--
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/