| 02:33.27 | *** join/#qi-hardware atommann (~atommann@58.251.2.94) |
| 03:26.58 | *** join/#qi-hardware jow_laptop (~jow@srv.mein.io) |
| 03:45.18 | *** join/#qi-hardware jow_laptop (~jow@srv.mein.io) |
| 05:33.40 | *** join/#qi-hardware nicksydney (~quassel@148.336.dsl.syd.iprimus.net.au) |
| 07:41.34 | *** join/#qi-hardware Nik05 (~Nik05@unaffiliated/nik05) |
| 07:56.06 | *** join/#qi-hardware xiangfu (~xiangfu@111.198.29.53) |
| 07:59.58 | *** join/#qi-hardware pcercuei (~paul@137.71.226.54) |
| 08:07.00 | *** join/#qi-hardware wolfspraul (~wolfsprau@xd9ba0822.dyn.telefonica.de) |
| 10:01.01 | *** join/#qi-hardware atommann (~atommann@58.251.2.94) |
| 11:33.03 | *** join/#qi-hardware jwhitmore (~jwhitmore@88.151.80.134) |
| 11:47.20 | *** join/#qi-hardware atommann (~atommann@58.251.2.94) |
| 12:46.25 | viric | does anybody understand the "Attention" warning in http://en.nanotec.com/products/1023-smci12-stepper-motor-controllers/ ? |
| 13:38.37 | whitequark | yes |
| 13:39.05 | whitequark | the power supply apparently has a virtual ground made using two capacitors between line and neutral |
| 13:41.55 | whitequark | usually done for EMI reasons which is likely a factor with a stepper controller |
| 13:44.12 | viric | what power supply? |
| 13:45.48 | whitequark | laptop |
| 13:48.24 | *** join/#qi-hardware dandon_ (~dandon@88.151.74.242) |
| 13:58.34 | viric | so what should I do? Power the board, unplug the laptop from the line, connect the board to the laptop with usb, ... |
| 13:58.40 | viric | and never connect the laptop to powerline again? |
| 14:01.43 | whitequark | well, it appears this is what they recommend |
| 14:01.56 | whitequark | alternatively you can use a line-isolated power supply for either of the laptop and the board |
| 14:02.19 | whitequark | e.g.: plug it into an isolating transformer, or use a supply that does /not/ have the above connection |
| 14:02.30 | whitequark | (poke your oscilloscope into GND of the supply and you'll see) |
| 14:06.35 | viric | what should I see in the DC GND of the supply? |
| 14:30.40 | whitequark | sine with a large amplitude |
| 14:39.19 | viric | between dc gnd and what? |
| 14:45.15 | *** join/#qi-hardware atommann (~atommann@222.248.72.156) |
| 14:53.13 | whitequark | and nothing. just put a probe there |
| 15:00.28 | viric | I don't have an oscilloscope here :) |
| 15:16.32 | *** join/#qi-hardware atommann (~atommann@155.94.207.116) |
| 15:28.45 | *** join/#qi-hardware Haswell (~broaderba@unaffiliated/haswell) |
| 15:34.06 | *** join/#qi-hardware atommann (~atommann@155.94.207.116) |
| 15:42.38 | *** join/#qi-hardware rjeffries (~rjeffries@pool-108-40-199-126.snloca.fios.verizon.net) |
| 15:42.50 | *** join/#qi-hardware paul_boddie (~paul_bodd@cm-84.215.167.166.getinternet.no) |
| 16:01.43 | paul_boddie | pcercuei: Looking at pinctrl for jz4740. Lots of repetition between the pinctrl driver and the DTS file. Is that normal? |
| 16:06.41 | pcercuei | paul_boddie: blame larsc D: |
| 16:06.49 | pcercuei | I think he's the author |
| 16:09.00 | pcercuei | paul_boddie: where is this driver? I can't find it |
| 16:09.21 | paul_boddie | It's the jz4770 driver I'm adapting. I haven't seen a jz4740 version. |
| 16:09.25 | pcercuei | oh |
| 16:09.27 | pcercuei | don't use htat |
| 16:09.48 | paul_boddie | No, I meant that in the pinctrl, there's a lot of GP?(?) stuff and then in the DTS the GPIOs are defined again. |
| 16:09.53 | pcercuei | yeah |
| 16:09.55 | pcercuei | don't use that |
| 16:10.26 | pcercuei | we don't even use it anymore |
| 16:10.42 | paul_boddie | Do you use the jz4780 one, then? The MIPS/Imgtec one? |
| 16:10.55 | pcercuei | yes |
| 16:11.07 | pcercuei | the one you're looking at, won't read anything from devicetree |
| 16:11.23 | pcercuei | we're using this one now: https://github.com/gcwnow/linux/blob/jz-3.19/drivers/pinctrl/pinctrl-jz4780.c |
| 16:11.55 | paul_boddie | OK. I saw that, but wondered if it was appropriate. |
| 16:12.06 | pcercuei | with this devicetree: https://github.com/gcwnow/linux/blob/jz-3.19/arch/mips/boot/dts/jz4770.dtsi#L63 |
| 16:12.30 | pcercuei | I wrote pinctrl-jz4770, and one week after ImgTec released their set of drivers for the jz4780 |
| 16:12.37 | paul_boddie | OK. That might be a better starting point, I guess. |
| 16:13.17 | pcercuei | I think you could start from MIPS/CI20_Linux |
| 16:13.57 | pcercuei | they have some improvements to the jz4740 core that is not upstream (and not in the qi-kernel repo either) |
| 16:14.04 | paul_boddie | I was looking at gcwnow jz-3.18-dts and jz-3.19. The MIPS one didn't seem to be any different. |
| 16:14.43 | pcercuei | let's say it's getting more and more similar ;) |
| 16:14.56 | pcercuei | that wasn't really the case a few months ago when they first released the code |
| 16:16.30 | pcercuei | but we still have some crappy ingenic code in arch/mips/jz4770 |
| 16:16.44 | pcercuei | for instance, we don't have real support for DMA or timers |
| 16:17.13 | paul_boddie | What about the upstream jz4740 DMA support? |
| 16:17.24 | pcercuei | we use a MMC driver that would scare away any sane kernel developer |
| 16:17.44 | pcercuei | it's for jz4740; we would need to adapt it to jz4770 |
| 16:17.54 | pcercuei | the only difference AFAIK is that we have two DMA chips |
| 16:17.59 | pcercuei | so 2x 6 channels |
| 16:18.25 | paul_boddie | How does one find out the differences? I'm just hoping at the moment that most of the work has already been kind of done. ;-) |
| 16:18.49 | pcercuei | what do you mean? |
| 16:19.21 | paul_boddie | I'm just thinking about the details like registers, interrupts, and so on. |
| 16:19.54 | pcercuei | by reading the docs |
| 16:20.37 | paul_boddie | The programming manual, I guess. |
| 16:21.49 | pcercuei | yes |
| 16:23.05 | paul_boddie | I'll look at MIPS/CI20_Linux ci20-v3.18, then. |
| 16:24.27 | paul_boddie | (If any of that EOMA-68 hardware gets released with a jz4775 CPU, it'd be nice to see a kernel later than Ingenic's own code-drop version that actually supports it.) |
| 17:05.04 | paul_boddie | Stupid git! Is it normal that git fails when cloning and then leaves no trace of its CPU-thrashing antics? |
| 17:31.14 | *** join/#qi-hardware pcercuei (~pcercuei@2a02:810d:1b40:1118:1af4:6aff:fec5:1c8f) |
| 18:04.23 | *** join/#qi-hardware rjeffries (~rjeffries@216.4.56.164) |
| 18:56.27 | *** join/#qi-hardware wolfspraul (~wolfsprau@xd9ba0822.dyn.telefonica.de) |
| 19:17.34 | *** join/#qi-hardware rjeffries (~rjeffries@216.4.56.164) |
| 22:18.20 | *** join/#qi-hardware rjeffries (~rjeffries@97-86-138-47.dhcp.stls.mo.charter.com) |
| 22:42.19 | *** join/#qi-hardware freespace (foobar@167.160.169.163) |
| 22:51.27 | wpwrak | hmm, one for the gnu cpp wish list: #define foo(x) #ifdef x##foo ... |
| 23:09.57 | *** join/#qi-hardware atommann (~atommann@222.248.72.156) |
| 23:36.05 | *** join/#qi-hardware wolfspraul (~wolfsprau@x4d023443.dyn.telefonica.de) |