Dear Linux FPGA Maintainers,

I did not found a way to use the struct spi_transfer to manage the chip
select the way this fpga programming requires. Since the chip select is 
managed by this driver alone, it may interfere with the communication 
with other devices on the bus, which I tried to make clear in the 
documentation.

Patch 1:
- improved kconfig help: added warning on spi interference
- fixed: efinix_spi_probe() warn: passing zero to 'PTR_ERR'
- improved function naming: efinix_spi_*()
- fixed: fpga_manager_ops.write could be called multiple times during 
one time reprogramming
[attached: v3-0001-fpga-Add-Efinix-Trion-Titanium-serial-SPI-program.patch]

Patch 2:
- added warning on spi interference
- removed unused label on example
[attached: v3-0003-dt-bindings-fpga-Add-Efinix-serial-SPI-programmin.patch]

Patch 3:
- no changes
[attached: v3-0003-dt-bindings-vendor-prefix-Add-prefix-for-Efinix-I.patch]

Signed-off-by: Ian Dannapel <iansdannapel@gmail.com>

On 27/09/2024 16:14, iansdannapel@gmail.com wrote:
> Dear Linux FPGA Maintainers,
> 
> I did not found a way to use the struct spi_transfer to manage the chip
> select the way this fpga programming requires. Since the chip select is 
> managed by this driver alone, it may interfere with the communication 
> with other devices on the bus, which I tried to make clear in 

Start using b4... or use proper --cover-letter argument for git
format-patch. Your patchset is not correctly threaded.

Best regards,
Krzysztof