[dpdk-dev] [PULL REQUEST] i40e base driver update

[Thomas Monjalon]

Hi Helin,

> Helin Zhang (33):
>       i40e: copyright update
>       i40e: disable setting of phy configuration
>       i40e: adjustment of register definitions and relevant
>       i40e/base: rename 'err' to 'perrno'
>       i40e/base: support of building both PF and VF driver together
>       i40e/base: support of CEE DCBX on recent firmware versions
>       i40e: replacement of 'i40e_debug_read_register()'
>       i40e/base: rework of 'i40e_hmc_get_object_va'
>       i40e/base: update of shadow RAM read/write functions
>       i40e/base: catch NVM write semaphore timeout and retry
>       i40e/base: check for AQ timeout in aq_rc decode
>       i40e/base: fix up NVM update sm error handling
>       i40e/base: enhancement of polling NVM semaphore
>       i40e/base: enhancements of NVM checksum calculation
>       i40e/base: replacement of DEBUGOUT() with i40e_debug()
>       i40e/base: add fw build info to AQ data
>       i40e/base: define and use i40e_is_vf()
>       i40e/base: grab NVM devstarter version not image version
>       i40e/base: enhancements on adminq init and sending asq command
>       i40e/base: i40e_aq_get_link_info() should be used directly
>       i40e/base: add new interfaces for future use
>       i40e/base: update of get/set LED functions
>       i40e/base: clean up sparse complaint in i40e_debug_aq
>       i40e/base: get pf_id from HW rather than PCI function
>       i40e/base: find partition_id in npar mode, and disable FCOE by default
>       i40e/base: Reassign incorrect PHY type as a workaround for a FW issue
>       i40e/base: add AOC phy types to case statement in get_media_type
>       i40e/base: support for iSCSI capability
>       i40e/base: set FLAG_RD when sending driver version to FW
>       i40e/base: future proof some sizeof calls
>       i40e/base: add more virtual channel operations for future use
>       i40e/base: rework of structures and macros for future use
>       i40e/base: modifications for future use

There was a build break due log rework in the middle of this series.
Don't forget to check series with "git rebase -x make".
I adjusted the patches order to fix it.

Some titles and commit messages were a bit too vague and didn't describe
why changes are done. I tried to improve titles a bit.
In general, giving the name of a function in a commit title doesn't give
the right level of information. Titles starting with "update" or "enhance"
won't be descriptive enough and would probably be split.
I'm not completely satisfied with it but most of things are well described
so I guess it's enough for a base driver update of a young PMD.
Though, it would be nice to have more precise descriptions of why things
are done for next updates.

Applied, thanks