Hello.
On 05/19/2016 04:29 PM, Guenter Roeck wrote:
Commit da47b4572056 ("phy: add support for a reset-gpio specification")
causes the following xtensa qemu crash according to Guenter Roeck:
[ 9.366256] libphy: ethoc-mdio: probed
[ 9.367389] (null): could not attach to PHY
[ 9.368555] (null): failed to probe MDIO bus
[ 9.371540] Unable to handle kernel paging request at virtual address
0000001c
[ 9.371540] pc = d0320926, ra = 903209d1
[ 9.375358] Oops: sig: 11 [#1]
Of course, I'd like this patch to be reverted (as it was applied
erroneously instead of my patches)... but where's the backtrace? It's not
immediately clear how this code can cause kernel oops...
See http://www.spinics.net/lists/kernel/msg2258611.html.
OK, I'll try to comment on your analysis posted there:
GPIOLIB is not configured in my test case, meaning gpiod_get_optional()
returns -ENOSYS, and phy_probe() thus returns an error. Question here is if
it is really appropriate for the XXX_optional() gpiolib functions to return
an error if GPIOLIB is not configured. Either case, result is that pretty
much all phy registrations will now fail if GPIOLIB is not configured.
Yes, that's the primary issue. That's what needs to be fixed...
Also, I suspect that there may be a bug in the error handling path
of ethoc_probe(). No idea what exactly is wrong, though. Other drivers
use pretty much the same code sequence for mdio registration and associated
error handling.
I tried analyzing this code and it wasn't clear what's the issue...
I must mention that this code actually looked strange -- other drivers
call phy_connect_direct() etc. from their ndo_open() method, this driver does
this from the probe. One problem I'm seing is that iff register_netdev()
fails, it forgets to disconnect from the PHY.
Last but not least, something seems to be wrong with the use of dev_err()
with &netdev->dev if register_netdev() has not yet been called. Maybe someone
has some insight ?
Yes, this needs fixing as well.
Guenter
MBR, Sergei
