Occasional mmc0 timeout on 2.6.25 master kernels
denver at ossguy.com
Wed Oct 8 12:34:49 EDT 2008
-----BEGIN PGP SIGNED MESSAGE-----
On Fri, May 2, 2008 at 12:24 PM, Denver Gingerich wrote:
> On Fri, May 2, 2008 at 12:12 PM, C. Scott Ananian wrote:
>> On Thu, May 1, 2008 at 10:52 PM, Denver Gingerich wrote:
>> > When booting 2.6.25 master kernels (specifically seen on 20080501.2
>> > and 20080501.3) from an SD card, the following error occurs every 5th
>> > boot or so:
>> > [ 3.512185] Waiting 2sec before mounting root device...
>> > [ 5.574603] VFS: Cannot open root device "mmcblk0p1" or
>> > [ 5.577210] Please append a correct "root=" boot option; here are the
>> > available partitions:
>> > ...
>> > [ 5.747213] Kernel panic - not syncing: VFS: Unable to mount root fs
>> > on unknown-block(0,0)
>> > [13.597220] mmc0: Timeout waiting for hardware interrupt.
>> > I have never encountered this error using 2.6.22-20071121.7
with over 50 boots.
>> Does increasing rootdelay help? ie, is the problem that setting up SD
>> is now taking (say) 1.5-2.5 seconds instead of 1-2 seconds, and so
>> sometimes the SD card is not present after the 2 second rootdelay?
> I can give it a shot, but I don't think it will help.
> 2.6.22-20071121.7 always ran fine with rootdelay=1 and when I first
> got the mounting problem on 2.6.25 I increased it to rootdelay=2. But
> the mounting problems continued after making that change. The type of
> error (mmc0 timeout) suggests that there are communication problems
> with the SD controller that would prevent the kernel from finding the
> partition table, even after waiting 10 seconds (the apparent length of
> the timeout given the timestamps).
This issue remains present on the 2.6.26-20080731.1 kernel (available
from http://dev.laptop.org/~dilinger/master/). However, it is
slightly different. With the new kernel, there is no list of
available partitions following the "here are the available partitions"
message. Additionally, there is no "mmc0: Timeout waiting for
hardware interrupt." message.
I tried the 2.6.26-20080731.1 kernel with rootdelay=1 and
rootdelay=30; both produced the same result.
Should I be reporting this to a bug tracking system? Or should I
refrain from doing so since this is a development kernel?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
-----END PGP SIGNATURE-----
More information about the Devel