15:11 <montjoie> sigmaris: could you retry my branch, it should handle both crypto now

15:51 <sigmaris> cool, I'll test again after work

16:21 <mort> Do y'all know what GPIO names like GPIO1_C5_d means? I assume it's a pin on GPIO1. Is it pin 0xC5 on GPIO1? Or does the C5 mean something else? And what does the _d mean?

16:26 <clarity> IIRC pins are named something like A0..7, B0..7, C0..7, D0..7

16:26 <clarity> Though some banks may have only 24 pins so D* are not present

16:26 <clarity> I don't know what _d is

16:26 <mort> so C5 would be 3*8+5

16:26 <clarity> I think so.

16:27 <clarity> mm wait

16:27 <clarity> Off by one

16:27 <clarity> A is 0*8+n so C must be 2*8+n

16:27 <mort> ah, right

19:58 <sigmaris> montjoie: I tried your branch (commit b69acda090) and got some self-tests failing when the driver loads: https://termbin.com/cp7n

20:20 <montjoie> sigmaris: found the bug, could you git pull && retry ?

20:21 <montjoie> I forgot to replace instance number in dma start...

20:39 <sigmaris> yep, seems better now - I don't see any errors when the driver probes: https://termbin.com/0hgt

20:50 <montjoie> sigmaris: thanks! do you want I put you as tested-by in my patch serie ?

20:52 <sigmaris> sure, you can put Tested-by: Hugh Cole-Baker <sigmaris@gmail.com>

20:58 <montjoie> and I found the TRM of rk3399, difference between crypto0 and 1 is that the 1 do not have RSA