diff mbox

ixgbe: Unsupported SFP+ modules on 10Gbit/s X520-DA2 NIC?

Message ID 1326925160.2795.45.camel@probook
State Changes Requested, archived
Delegated to: David Miller
Headers show

Commit Message

Jesper Dangaard Brouer Jan. 18, 2012, 10:19 p.m. UTC 
On Wed, 2012-01-18 at 22:45 +0100, Benny Amorsen wrote:
> Jesse Brandeburg <jesse.brandeburg@intel.com> writes:
> 
> > For X520 adapters, the documentation[1] states that which SFP+
> > adapters are/are not supported.  Direct attach cables are also
> > supported.
> >
> > [1] http://www.intel.com/support/network/adapter/pro100/sb/CS-030612.htm
> 
> I can't believe that locked optics have now arrived on commodity
> hardware. I have been trying to migrate to all-Intel networking at work;
> that effort is certainly on hold now.

I cannot understand why Intel are pulling a stunt like this! :-(

I have read the code, and the limitation comes from a EEPROM setting on
the NIC, see define "IXGBE_DEVICE_CAPS_ALLOW_ANY_SFP 0x1".

Here is a (untested) patch I believe removes the limitation in the
driver:


[PATCH] ixgbe: Always allow any SFP+ regardless of EEPROM setting.

Intel are trying to limit which SFP's we can use in our NICs.
We don't like this practices in the Linux Kernel.

Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Ben Greear Jan. 18, 2012, 10:43 p.m. UTC | #1 
On 01/18/2012 02:19 PM, Jesper Dangaard Brouer wrote:
> On Wed, 2012-01-18 at 22:45 +0100, Benny Amorsen wrote:
>> Jesse Brandeburg<jesse.brandeburg@intel.com>  writes:
>>
>>> For X520 adapters, the documentation[1] states that which SFP+
>>> adapters are/are not supported.  Direct attach cables are also
>>> supported.
>>>
>>> [1] http://www.intel.com/support/network/adapter/pro100/sb/CS-030612.htm
>>
>> I can't believe that locked optics have now arrived on commodity
>> hardware. I have been trying to migrate to all-Intel networking at work;
>> that effort is certainly on hold now.
>
> I cannot understand why Intel are pulling a stunt like this! :-(
>
> I have read the code, and the limitation comes from a EEPROM setting on
> the NIC, see define "IXGBE_DEVICE_CAPS_ALLOW_ANY_SFP 0x1".
>
> Here is a (untested) patch I believe removes the limitation in the
> driver:
>
>
> [PATCH] ixgbe: Always allow any SFP+ regardless of EEPROM setting.
>
> Intel are trying to limit which SFP's we can use in our NICs.
> We don't like this practices in the Linux Kernel.

I think that you should at least print some big warnings in
the kernel logs if you do this, as well as all the info you
can find on the non-supported SFP+ module in use so that folks can debug
things if the SFP+ doesn't properly work.

As previously mentioned, I found a case where some random SFP+
did NOT work with a similar hack in place...

Thanks,
Ben
diff mbox

Patch

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
index 7cf1e1f..2b13083 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
@@ -1061,6 +1061,8 @@  s32 ixgbe_identify_sfp_module_generic(struct ixgbe_hw *hw)
 		}
 
 		hw->mac.ops.get_device_caps(hw, &enforce_sfp);
+		/* Hack: Always allow any SFP regardless of EEPROM setting */
+		enforce_sfp |= IXGBE_DEVICE_CAPS_ALLOW_ANY_SFP;
 		if (!(enforce_sfp & IXGBE_DEVICE_CAPS_ALLOW_ANY_SFP) &&
 		    !((hw->phy.sfp_type == ixgbe_sfp_type_1g_cu_core0) ||
 		      (hw->phy.sfp_type == ixgbe_sfp_type_1g_cu_core1))) {