[PATCH net-2.6.25] New driver "sfc" for Solarstorm SFC4000 controller - 5th attempt

[PATCH net-2.6.25] New driver "sfc" for Solarstorm SFC4000 controller - 5th attempt

[Robert Stonehouse]

This is a resubmission of a new driver for Solarflare network controllers.

The driver supports several types of PHY (10Gbase-T, XFP, CX4) on six
different 10G and 1G boards. There is a driver for the network controller
and an MTD driver that allows access to the flash/EEPROM.

NICs based on this controller are now available from SMC as part numbers
SMC10GPCIe-XFP and SMC10GPCIe-10BT.

FYI the previous thread was:
  http://marc.info/?l=linux-netdev&m=119999015817920&w=2

In previous postings we were asked to have the MTD part of this patch
reviewed and we have responded to the comments received so far:
  http://lists.infradead.org/pipermail/linux-mtd/2008-January/020355.html

Things that have changed since the last patch:
 - Removed residual support for more than one port per controller.
 - Moved LED control from PHY operations to board operations.
 - Added support for controlling LRO through ethtool {get,set}_flags().
 - Removed obsolete workaround for reservation of MSI-X tables by PCI core.
 - Improved reliability of PHY loopback setting.
 - Improved reliability of I2C operations.
 - Fixed on-load self-test when using RSS.
 - Simplified MTD driver following review on linux-mtd list.
 - Removed some redundant information from struct efx_nic.
 - hardware monitoring (for PHY link polling) is now a workqueue item
rather than using a timer
 - mac_lock becomes a mutex to support the previous item.

We would welcome more review of this code and would like to know if there
are any objections to this being merged for 2.6.25.


The patch (against net-2.6.25) is at:
  https://support.solarflare.com/netdev/5/net-2.6.25-sfc-2.2.0045.patch

The new files may also be downloaded as a tarball:
  https://support.solarflare.com/netdev/5/net-2.6.25-sfc-2.2.0045.tgz

And for verification there is:
  https://support.solarflare.com/netdev/5/MD5SUMS

Regards

--
Rob Stonehouse

Re: [PATCH net-2.6.25] New driver "sfc" for Solarstorm SFC4000 controller - 5th attempt

[Ben Hutchings]

Robert Stonehouse wrote:
<snip>
>  - Improved reliability of I2C operations.
<snip>

This reminds me that Randy Dunlap raised the question of whether we
could integrate with the kernel I2C and SPI systems.  I did some
investigatory work on this but it looked like it would add complexity
and increase our code size.

Ben.

-- 
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.

Re: [PATCH net-2.6.25] New driver "sfc" for Solarstorm SFC4000 controller - 5th attempt

[Jeff Garzik]

Robert Stonehouse wrote:
> This is a resubmission of a new driver for Solarflare network controllers.
> 
> The driver supports several types of PHY (10Gbase-T, XFP, CX4) on six
> different 10G and 1G boards. There is a driver for the network controller
> and an MTD driver that allows access to the flash/EEPROM.
> 
> NICs based on this controller are now available from SMC as part numbers
> SMC10GPCIe-XFP and SMC10GPCIe-10BT.
> 
> FYI the previous thread was:
>   http://marc.info/?l=linux-netdev&m=119999015817920&w=2
> 
> In previous postings we were asked to have the MTD part of this patch
> reviewed and we have responded to the comments received so far:
>   http://lists.infradead.org/pipermail/linux-mtd/2008-January/020355.html
> 
> Things that have changed since the last patch:
>  - Removed residual support for more than one port per controller.
>  - Moved LED control from PHY operations to board operations.
>  - Added support for controlling LRO through ethtool {get,set}_flags().
>  - Removed obsolete workaround for reservation of MSI-X tables by PCI core.
>  - Improved reliability of PHY loopback setting.
>  - Improved reliability of I2C operations.
>  - Fixed on-load self-test when using RSS.
>  - Simplified MTD driver following review on linux-mtd list.
>  - Removed some redundant information from struct efx_nic.
>  - hardware monitoring (for PHY link polling) is now a workqueue item
> rather than using a timer
>  - mac_lock becomes a mutex to support the previous item.
> 
> We would welcome more review of this code and would like to know if there
> are any objections to this being merged for 2.6.25.
> 
> 
> The patch (against net-2.6.25) is at:
>   https://support.solarflare.com/netdev/5/net-2.6.25-sfc-2.2.0045.patch
> 
> The new files may also be downloaded as a tarball:
>   https://support.solarflare.com/netdev/5/net-2.6.25-sfc-2.2.0045.tgz
> 
> And for verification there is:
>   https://support.solarflare.com/netdev/5/MD5SUMS

Please email to me in patch format...