I just upgraded the kernel on my System76 Gazelle laptop (basically an
ASUS Z62FP without the Microsoft tax) and the hard drive began
freezing every few minutes. I blacklisted ata_piix and the problem
went away, so that really narrows down where the bug has to be. Here's
the relevant part of the dmesg:
[ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
[ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
cdb 0x25 data 8 in
[ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
0x4 (timeout)
[ 931.004000] ata1: port is slow to respond, please be patient (Status 0xd0)
[ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
[ 954.028000] ata1: soft resetting port
[ 954.380000] ata1.00: configured for UDMA/100
[ 954.560000] ata1.01: configured for UDMA/33
[ 954.560000] ata1: EH complete
Complete dmesg and lspci -vvx are attached; please CC me because I'm
not subscribed. My sincere apologies if this is already fixed in the
main kernel.
Keenan Pepper
On Sun, 4 Feb 2007 00:32:12 -0500 "Keenan Pepper" <[email protected]> wrote:
> I just upgraded the kernel on my System76 Gazelle laptop (basically an
> ASUS Z62FP without the Microsoft tax) and the hard drive began
> freezing every few minutes. I blacklisted ata_piix and the problem
> went away, so that really narrows down where the bug has to be. Here's
> the relevant part of the dmesg:
>
> [ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> [ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
> cdb 0x25 data 8 in
> [ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
> 0x4 (timeout)
> [ 931.004000] ata1: port is slow to respond, please be patient (Status 0xd0)
> [ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
> [ 954.028000] ata1: soft resetting port
> [ 954.380000] ata1.00: configured for UDMA/100
> [ 954.560000] ata1.01: configured for UDMA/33
> [ 954.560000] ata1: EH complete
>From what version of the kernel did you upgrade?
It is very helpful for us to knwo the latest version which worked OK.
> Complete dmesg and lspci -vvx are attached; please CC me because I'm
> not subscribed. My sincere apologies if this is already fixed in the
> main kernel.
You're running 2.6.20-rc6. This won't have been fixed.
Keenan Pepper wrote:
> I just upgraded the kernel on my System76 Gazelle laptop (basically an
> ASUS Z62FP without the Microsoft tax) and the hard drive began
> freezing every few minutes. I blacklisted ata_piix and the problem
> went away, so that really narrows down where the bug has to be. Here's
> the relevant part of the dmesg:
>
> [ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2
> frozen
> [ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
> cdb 0x25 data 8 in
> [ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
> 0x4 (timeout)
> [ 931.004000] ata1: port is slow to respond, please be patient (Status
> 0xd0)
> [ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
It seems that something is issuing READ_CAPACITY to your drive or trying
to open the device triggering media revalidation periodically and your
drive doesn't like it. Your harddisk is sharing the channel with the
ODD so it can't be accessed till the command times out and the ODD gets
recovered.
* Does putting in a readable CD or DVD make any difference?
* Does the problem occur in single mode where no HAL or whatever is running?
PS: It *really* helps if you include full dmesg including the detection
messages and whole error messages when reporting bugs. If you need
snippet of it to explain, copy that part but keep the whole dmesg at the
end. Kernel developers are usually pretty good at filtering useful info
from tons of kernel messages.
--
tejun
Tejun Heo wrote:
> Keenan Pepper wrote:
>> I just upgraded the kernel on my System76 Gazelle laptop (basically an
>> ASUS Z62FP without the Microsoft tax) and the hard drive began
>> freezing every few minutes. I blacklisted ata_piix and the problem
>> went away, so that really narrows down where the bug has to be. Here's
>> the relevant part of the dmesg:
>>
>> [ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2
>> frozen
>> [ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
>> cdb 0x25 data 8 in
>> [ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
>> 0x4 (timeout)
>> [ 931.004000] ata1: port is slow to respond, please be patient (Status
>> 0xd0)
>> [ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
>
> It seems that something is issuing READ_CAPACITY to your drive or trying
> to open the device triggering media revalidation periodically and your
> drive doesn't like it. Your harddisk is sharing the channel with the
> ODD so it can't be accessed till the command times out and the ODD gets
> recovered.
>
> * Does putting in a readable CD or DVD make any difference?
>
> * Does the problem occur in single mode where no HAL or whatever is running?
>
> PS: It *really* helps if you include full dmesg including the detection
> messages and whole error messages when reporting bugs. If you need
> snippet of it to explain, copy that part but keep the whole dmesg at the
> end. Kernel developers are usually pretty good at filtering useful info
> from tons of kernel messages.
Auch, scrap the last paragraph. You actually did. Sorry about that. ;-)
I see your controller is ICH7 and the errors are over ten minutes apart.
It really seems that HAL, smartd, hddtemp or whatever is poking the dvd
writer and it isn't happy about it. In addition to the above questions,
you can read cd/dvd's using the drive, right?
Thanks.
--
tejun
>From the dmesg, I found a problem NOT relative to the PATA driver.
I would like to know if cpu freq driver works? If not, please enter a
acpi bug on bugzilla.kernel.org.
"[ 17.368000] ACPI Exception (exoparg2-0442): AE_AML_PACKAGE_LIMIT,
Index (0FFFFFFFD) is beyond end of object [20060707]
[ 17.368000] ACPI Error (psparse-0537): Method parse/execution
failed [\_PR_.CPU1._PSS] (Node c1468ba8), AE_AML_PACKAGE_LIMIT
[ 17.368000] ACPI Exception (acpi_processor-0235):
AE_AML_PACKAGE_LIMIT, Evaluating _PSS [20060707]
[ 17.368000] ACPI Exception (exoparg2-0442): AE_AML_PACKAGE_LIMIT,
Index (0FFFFFFFD) is beyond end of object [20060707]
[ 17.368000] ACPI Error (psparse-0537): Method parse/execution
failed [\_PR_.CPU2._PSS] (Node c14689f0), AE_AML_PACKAGE_LIMIT
[ 17.368000] ACPI Exception (acpi_processor-0235):
AE_AML_PACKAGE_LIMIT, Evaluating _PSS [20060707]
"
On 2/4/07, Keenan Pepper <[email protected]> wrote:
> I just upgraded the kernel on my System76 Gazelle laptop (basically an
> ASUS Z62FP without the Microsoft tax) and the hard drive began
> freezing every few minutes. I blacklisted ata_piix and the problem
> went away, so that really narrows down where the bug has to be. Here's
> the relevant part of the dmesg:
>
> [ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> [ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
> cdb 0x25 data 8 in
> [ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
> 0x4 (timeout)
> [ 931.004000] ata1: port is slow to respond, please be patient (Status 0xd0)
> [ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
> [ 954.028000] ata1: soft resetting port
> [ 954.380000] ata1.00: configured for UDMA/100
> [ 954.560000] ata1.01: configured for UDMA/33
> [ 954.560000] ata1: EH complete
>
> Complete dmesg and lspci -vvx are attached; please CC me because I'm
> not subscribed. My sincere apologies if this is already fixed in the
> main kernel.
>
> Keenan Pepper
>
>
Luming Yu wrote:
> From the dmesg, I found a problem NOT relative to the PATA driver.
> I would like to know if cpu freq driver works? If not, please enter a
> acpi bug on bugzilla.kernel.org.
The acpi-cpufreq driver does work, but there are some other
ACPI-related problems. Most importantly for me, S3 sleep does not
work, and although S1 sleep works, the Fn key combinations stop
working afterwards. Should I file a bug about this?
Keenan
On Sun, 4 Feb 2007, Keenan Pepper wrote:
> I just upgraded the kernel on my System76 Gazelle laptop (basically an
> ASUS Z62FP without the Microsoft tax) and the hard drive began
> freezing every few minutes. I blacklisted ata_piix and the problem
> went away, so that really narrows down where the bug has to be. Here's
> the relevant part of the dmesg:
>
> [ 924.000000] ata1.01: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2
> frozen
> [ 924.000000] ata1.01: cmd a0/01:00:00:00:00/00:00:00:00:00/b0 tag 0
> cdb 0x25 data 8 in
> [ 924.000000] res 40/00:03:00:00:00/00:00:00:00:00/b0 Emask
> 0x4 (timeout)
> [ 931.004000] ata1: port is slow to respond, please be patient (Status 0xd0)
> [ 954.028000] ata1: port failed to respond (30 secs, Status 0xd0)
> [ 954.028000] ata1: soft resetting port
> [ 954.380000] ata1.00: configured for UDMA/100
> [ 954.560000] ata1.01: configured for UDMA/33
> [ 954.560000] ata1: EH complete
This seems kind of similar to my problems with ICH7 and other SATA on the
same comp. See <http://bugzilla.kernel.org/show_bug.cgi?id=7883>. I'm
getting very similar messages.
Martin