> > Vaibhav, > > I've now been through six different images and they all end up with the > same panic. I don't know what else to do except push on the RMA people for > a hardware change. But then I'd be afraid I'd create the same problem on a > new board... > > As for command timing, I've seen vastly different command sequences > leading up to the panic, so unless you mean the timing within cpsw_probe > itself it is hard to imagine how that could be a factor. And this is 100% > failure, with any image, on either memory, power cycled or reset in any > order. With or without console to USB or console to level converter > connection, with or without USB or ethernet connected. > > Obviously cpsw_probe needs to fail more gracefully. Hopefully with a clue > as to what is wrong. Could that be the "mdio failed with error -5" line? > > > Here's a 1 in 100 boots problem with that error: > http://permalink.gmane.org/gmane.linux.network/279586 > > But he gets a "random" MAC - I get the correct one: > | davinci_mdio: probe of 4a101000.mdio failed with error -5 > | Random MACID = 16:74:44:51:f1:0f > | gpio-keys volume_keys.6: Unable to claim irq 0; error -22 > | gpio-keys: probe of volume_keys.6 failed with error -22 > > > The guy mentioned way above with my same problem on only one of six BBB > boards shows the -5 error: > http://osdir.com/ml/beagleboard/2013-11/msg00230.html > > > Found (maybe?) the error doc: > > http://www.ks.uiuc.edu/Development/MDTools/namdlite/files/doc/html/file_8h.html#a48 > ----- > enum { MDIO_ERROR = -1, MDIO_LENGTH_ERRMSG = 240 } > ... > Error numbers. > Enumeration values: > MDIO_ERROR_NONE No error. > MDIO_ERROR_WARN A warning. > MDIO_ERROR_BADVAL An illegal value pertaining to file. > MDIO_ERROR_NOMEM Memory cannot be allocated. > MDIO_ERROR_OPEN Cannot open a file. > MDIO_ERROR_CLOSE Cannot close a file. > MDIO_ERROR_READ Unable to read from file. > MDIO_ERROR_WRITE Unable to write to file. > MDIO_ERROR_SEEK Unable to perform a byte seek within file. > MDIO_ERROR_SYNTAX Syntax error occurred with file. > MDIO_ERROR_UNXEOF The end-of-file marker occurred unexpectedly. > ----- > > But that doesn't make much sense to me... It is all about opening and > closing files, not probing devices. Yes, I know "everything in Unix is a > file", but those errors really do sound like file errors. > > Guess I'm in over my depth. > > Loren > > > On Monday, January 6, 2014 4:11:22 PM UTC-8, Vaibhav wrote: > > As per the commit logs of the mainline kernel a patch adding pm_runtime_* > calls in the driver went in around 3.7-rc3. I believe the images that you > have which are based on v3.8 will have them in place. > > I still can't think of a reason why it would worked earlier. Could be a > race > condition or could even be a hardware thing. Try changing the kernel build > to narrow it down. And yes, the error that you see would typically come up > when the init fails (typically clock). > >
-- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
