e1000e: Fix no connectivity when driver loaded with cable out
authorDavid Ertman <davidx.m.ertman@intel.com>
Tue, 25 Mar 2014 04:27:55 +0000 (04:27 +0000)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 7 Jun 2014 17:28:20 +0000 (10:28 -0700)
commita23e2fa7ab52dfec62dc40de5cea10aec7cacdda
treee8e1979c4103b4730eb346c1517a355e01242318
parent4305a2a568662fec9b05f4887cf1e0e37d02c127
e1000e: Fix no connectivity when driver loaded with cable out

commit b20a774495671f037e7160ea2ce8789af6b61533 upstream.

In commit da1e2046e5, the flow for enabling/disabling an Si errata
workaround (e1000_lv_jumbo_workaround_ich8lan) was changed to fix a problem
with iAMT connections dropping on interface down with jumbo frames set.
Part of this change was to move the function call disabling the workaround
to e1000e_down() from the e1000_setup_rctl() function.  The mechanic for
disabling of this workaround involves writing several MAC and PHY registers
back to hardware defaults.

After this commit, when the driver is loaded with the cable out, the PHY
registers are not programmed with the correct default values.  This causes
the device to be capable of transmitting packets, but is unable to recieve
them until this workaround is called.

The flow of e1000e's open code relies upon calling the above workaround to
expicitly program these registers either with jumbo frame appropriate settings
or h/w defaults on 82579 and newer hardware.

Fix this issue by adding logic to e1000_setup_rctl() that not only calls
e1000_lv_jumbo_workaround_ich8lan() when jumbo frames are set, to enable the
workaround, but also calls this function to explicitly disable the workaround
in the case that jumbo frames are not set.

Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/net/ethernet/intel/e1000e/netdev.c