[PATCH] SPI: bcm2835: move to the transfer_one driver model

[Stephen Warren]

On 03/28/2015 12:11 PM, Martin Sperl wrote:
>> On 28.03.2015, at 05:09, Stephen Warren <swarren at wwwdotorg.org> wrote:
...
>>> As for testing: I have also tried to test with mmc_spi, but I have not
>>> been able to make that driver work reliably in any recent kernel
>>> versions.
>>> Most of the time I see timeouts - and with lots of different SD-cards...
>>>
>>> IIRC the last time I tested it successfully was with 3.12.
>>
>> It'd be great if you could use "git bisect" to track down the change
>> that broke this.
> 
> The problem is that it is a lot of kernel-versions I would have to test
> to figure out why and with the rpi used for compiling the kernel it 
> becomes a prohibitive long exercise...

Why not just cross-compile from a faster machine? It probably only takes
a few minutes per commit to test (and git bisect is pretty optimal
w.r.t. the number of commits you need to test).

> Also I may have a slightly different setup now compared to when I did
> the test initially, so this may be the trigger as well. 

That's also pretty easy to test; just go back and retest the old kernel
version that you believe was working. If it's still working but the
latest kernel is broken, there's been a regression. If the old kernel
also doesn't work, then something has changed in your setup.