Hi Thor,
On Thu, Jul 25, 2019 at 10:16:45AM -0500, thor.thayer@xxxxxxxxxxxxxxx wrote:
> From: Thor Thayer <thor.thayer@xxxxxxxxxxxxxxx>
>
> Newer versions (V2) of Altera/Intel FPGAs CvP have different PCI
> Vendor Specific Capability offsets than the older (V1) Altera/FPGAs.
>
> Most of the CvP registers and their bitfields remain the same
> between both the older parts and the newer parts.
>
> This patchset implements changes to discover the Vendor Specific
> Capability offset and then add Stratix10 CvP support.
>
> V2 Changes:
> Remove inline designator from abstraction functions.
> Reverse Christmas Tree format for local variables
> Remove redundant mask from credit calculation
> Add commment for the delay(1) function in wait_for_credit()
>
> V3 Changes
> Return int instead of void for abstraction functions
> Check the return code from read in altera_cvp_chk_error()
> Move reset of current_credit_byte to clear_state().
> Check return codes of read/writes in added functions.
>
> Thor Thayer (3):
> fpga: altera-cvp: Discover Vendor Specific offset
> fpga: altera-cvp: Preparation for V2 parts.
> fpga: altera-cvp: Add Stratix10 (V2) Support
>
> drivers/fpga/altera-cvp.c | 348 ++++++++++++++++++++++++++++++++++++----------
> 1 file changed, 277 insertions(+), 71 deletions(-)
>
> --
> 2.7.4
>
this doesn't apply since I applied Carlos' patch 998c1de56dac7
("fpga: altera-cvp: Fix function definition argument") yesterday to
git://git.kernel.org/pub/scm/linux/kernel/git/mdf/linux-fpga.git
'for-next' branch.
Would you mind resending it on top of that and drop the extra dev_dbg()
for the offset in Patch [1/3]?
Sorry about that. Things are moving around a bit, but hopefully settle
down soon :)
- Moritz
