Tyan S3992-E: hpet no longer working
Arno J. Klaassen
arno at heho.snv.jussieu.fr
Mon Jan 10 21:18:33 UTC 2011
Alexander Motin <mav at FreeBSD.org> writes:
> Arno J. Klaassen wrote:
>> John Baldwin <jhb at freebsd.org> writes:
>>
>>> On Thursday, January 06, 2011 5:32:08 pm Arno J. Klaassen wrote:
>>>> John Baldwin <jhb at freebsd.org> writes:
>>>>
>>>>> On Wednesday, January 05, 2011 4:39:24 pm Arno J. Klaassen wrote:
>>>>>> Hello,
>>>>>>
>>>>>> I have (a long-lasting) problem to get hpet attached to a Tyan S3992-E
>>>>>> MB. My last known working kernel is 7.1-PRERELEASE Sep 2 2008" , I
>>>>>> rarely cared about this board for a while...
>>>>>>
>>>>>> At that time the dmesg said :
>>>>>>
>>>>>>
>>>>>> acpi_hpet0: <High Precision Event Timer> iomem 0xfed00000-0xfed003ff
>>>>>> on acpi0
>>>>>> Timecounter "HPET" frequency 25000000 Hz quality 900
>>>>>>
>>>>>> now it says (debug.acpi.hpet_test="1", debug.acpi.layer="ACPI_TIMER",
>>>>>> debug.acpi.level="ACPI_LV_ALL_EXCEPTIONS" enabled) :
>>>>>>
>>>>>> hpet0: <High Precision Event Timer> iomem 0xfed00000-0xfed03fff on
>>>>>> acpi0
>>>>>> hpet0: vendor 0xffff, rev 0xff, 232831Hz 64bit, 32 timers, legacy route
>>>>>> hpet0: t0: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t1: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t2: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t3: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t4: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t5: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t6: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t7: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t8: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t9: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t10: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t11: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t12: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t13: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t14: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t15: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t16: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t17: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t18: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t19: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t20: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t21: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t22: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t23: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t24: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t25: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t26: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t27: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t28: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t29: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t30: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: t31: irqs 0xffffffff (31), MSI, 64bit, periodic
>>>>>> hpet0: 0.000000000: 4294967295 ... 4294967295 = 0
>>>>>> hpet0: time per call: 0 ns
>>>>>> hpet0: HPET never increments, disabling
>>>>>> device_attach: hpet0 attach returned 6
>>>>>>
>>>>>>
>>>>>> Some things strike me :
>>>>>>
>>>>>> 'vendor 0xffff, rev 0xf' and '4294967295 (== 0xffffffff)' as well
>>>>>> as 232831Hz
>>>>>>
>>>>>> the change in iomem range :
>>>>>>
>>>>>> OK : iomem 0xfed00000-0xfed003ff
>>>>>> KO : iomem 0xfed00000-0xfed03fff
>>>>>> ^^^^
>>>>>>
>>>>>> I can provide full dmesg and/or other extra needed info.
>>>>> Arno sent me his acpidump which includes this:
>>>>>
>>>>> Device (HPET)
>>>>> {
>>>>> Name (_HID, EisaId ("PNP0103"))
>>>>> Name (_UID, 0x34)
>>>>> Method (_STA, 0, NotSerialized)
>>>>> {
>>>>> Return (0x0F)
>>>>> }
>>>>>
>>>>> Method (_CRS, 0, NotSerialized)
>>>>> {
>>>>> Return (ResourceTemplate ()
>>>>> {
>>>>> Memory32Fixed (ReadWrite,
>>>>> 0xFED00000, // Address Base
>>>>> 0x00004000, // Address Length
>>>>> )
>>>>> })
>>>>> }
>>>>> }
>>>>>
>>>>> So it does look like we are doing what the DSDT tells us in terms
>>>>> of the memory address.
>>>> yop. That said, I made yet another copy-paste error: the last known
>>>> working kernel is 8.0-CURRENT Mar 1 2009 and the hpet says :
>>>>
>>>> acpi_hpet0: <High Precision Event Timer> iomem 0xfed00000-0xfed003ff
>>>> on acpi0
>>>> Timecounter "HPET" frequency 14318180 Hz quality 900
>>>>
>>>> [only the frequency differs, the memory range indeed then was reported as
>>>> 0x400 and not 0x4000 ]
>>>>
>>>>> Arno, are there any BIOS options that mention the HPET or have you updated
>>>>> your BIOS since you booted the 7.1 kernel?
>>>> yes .. I now use BIOS 1.06 released 06/09/09.
>>>> Can I somehow 'overide' the bios and force the driver to use 0X400 as
>>>> 'Address Length' in order to test if that makes the driver attach again?
>>> Changing the length wouldn't make a difference as we would still read the same
>>> registers since the start address is identical. I think the length is
>>> symptomatic of the BIOS doing something differently that has disabled the
>>> HPET.
>>
>> good point : this failure probably is not related to the FreeBSD-driver
>> : in the current BIOS under the submenu 'South Bridge Chipset
>> Configuration', the option to enable the HPET has disappeared (no
>> mention of that in the release-notes), whilst it was present in the
>> original BIOS, *and* disabled by default.
>>
>> Is it possible to write to some register during hpet_enable() and force
>> the timer to tick, regardless of the BIOS?
>
> Problem seems not about ticking, but about HPET registers working at
> all. Returning ffh values for everything more probably tells that HPET
> is just not in place where we look for it.
[forgive my ignorance, long time since I didn't do any of these hardware
architectural things ]
I googled around a little bit and found some issues with similar
chipsets, e.g. for xen 'enabl[ing] direct device assignment (PCI passthru)
on AMD IOMMU platforms.' :
http://www.mailinglistarchive.com/html/xen-devel@lists.xensource.com/2008-07/msg00282.html
Two remarks on that (quotes from the above thread) :
"Due to some BIOS issues, this patch has a workaround to enable ht1100
^^^^^^^^^^^ iommu."
(patch supposedly for ht1100 chipset, but the guy responding has
ht2000/ht1000, as I do)
[ from his linux dmesg (under xen I supposeà ] :
[ 57.300688] PCI-DMA: Disabling AGP.
[ 57.303516] PCI-DMA: aperture base @ c0000000 size 131072 KB
[ 57.303656] PCI-DMA: using GART IOMMU.
[ 57.303738] PCI-DMA: Reserving 128MB of IOMMU area in the AGP
aperture
These indeed are BIOS options : "Set GART size" with choice "AGP
Present/Disabled, 32MB, 64MB, 128MB [default], ... . (I tried all, no luck)
I have "device agp" in my kernel, but no trace of it in dmesg.
[ 57.307831] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
^^^^^^^^^^^^^^^^^^
Might this imply that it is not the same address-space as where FreeBSD
looks for it?
again, my $0.01 ...
For completenes, the two patches mentioned in the above thread can be
found at :
http://lists.xensource.com/archives/html/xen-devel/2007-11/txtJD2SnCs5Li.txt
http://lists.xensource.com/archives/html/xen-devel/2007-11/txtYFdQBW9AZT.txt
and both look rather 'hacky' ;-)
Thanx, Arno
More information about the freebsd-acpi
mailing list