| *** tpb has joined #litex | 00:00 | |
| *** _franck_ has quit IRC | 00:07 | |
| *** _whitelogger has quit IRC | 01:43 | |
| *** _whitelogger has joined #litex | 01:46 | |
| *** shenki has quit IRC | 02:11 | |
| keesj | I found is quite usefull for debugging | 07:41 |
|---|---|---|
| *** fjullien has joined #litex | 08:42 | |
| *** fjullien is now known as _franck_ | 08:43 | |
| bunnie[m] | <somlo "bunnie: "sdcore_issue_cmd_write""> yes, it works in the version of litesdcard i was using, but it's a really old version, at commit 067299dac3b8d7e5fe9dc3f28a71841be1b24227 | 09:09 |
| bunnie[m] | i checked and sdcore_issue_cmd_write is in my csr.h for that version. | 09:09 |
| _florent_ | bunnie[m]: i don't think litesdcard changed too much since this, but i'm going to do some tests and will post the results here | 09:32 |
| bunnie[m] | ok cool. if memory serves correct, this change is because we use an 8-bit width CSR space on the netv2 image, and the original sdcard implementation assumed a 32-bit CSR? so there were a few little hacks i had to patch in to make the 8-bit width work. it's been a while, tho, my memory is fuzzy. | 09:48 |
| bunnie[m] | to wit, the only thing i do with the SD card interface is run a factory test, which is reading the card ID registers and blasting in a block of data and checking it comes out ok to make sure the hardware is soldered together correctly. it's not used in the production NeTV2 image. | 09:49 |
| _florent_ | bunnie[m]: yes that's also what i remember | 09:54 |
| _florent_ | somlo: i just updated the litesdcard example designs, and switch to the Nexys4DDR for the Artix7 example | 11:35 |
| _florent_ | somlo: i also switched the CPU from lm32 to vexriscv that is now the default in LiteX | 11:36 |
| _florent_ | somlo: i think you should first try to reproduce my results on the Nexys4DDR: https://gist.github.com/enjoy-digital/d68e70f526ee8bcae325651dfbc12542 | 11:37 |
| tpb | Title: gist:d68e70f526ee8bcae325651dfbc12542 · GitHub (at gist.github.com) | 11:37 |
| _florent_ | Here we initialize the clock to 10MHz, then initialize the SDCard, do a write/read test at 10MHz and then increase the frequency up to 40MHz | 11:38 |
| somlo | _florent_: thanks! How long shoud "sdinit" take? I think it's hanging when I try to replicate your results, as I don't get anything after a couple of minutes: https://pastebin.com/QrYiTMJr | 15:45 |
| tpb | Title: __ _ __ _ __ / / (_) /____ | |/_/ / /__/ / __/ - - Pastebin.com (at pastebin.com) | 15:45 |
| somlo | _florent_: my migen git sha1 is really "4c00f5b", but since it's packaged as an RPM there's no way to grab it using a git command :) | 15:49 |
| *** _franck_ has quit IRC | 17:08 | |
| *** _franck_ has joined #litex | 17:41 | |
| *** somlo has quit IRC | 18:29 | |
| *** somlo has joined #litex | 18:46 | |
| _florent_ | somlo: sdinit should be 2-3 seconds max | 20:17 |
| _florent_ | somlo: can you activate the debug in the firmware? | 20:18 |
| _florent_ | btw, without driving explicitly rst to 0, i was having the same result you have (https://github.com/enjoy-digital/litesdcard/blob/master/examples/nexys4ddr.py#L86) | 20:20 |
| tpb | Title: litesdcard/nexys4ddr.py at master · enjoy-digital/litesdcard · GitHub (at github.com) | 20:20 |
| _florent_ | Also, it could be useful to test with another SDCard if you have one | 20:21 |
| _florent_ | i could send the you a working bitstream/firmware tomorrow if you want to test that on your setup | 20:23 |
| futarisIRCcloud | mithro / _florent_ : What's the best way to run hdmi2usb to get a litex scope trace? Also, are the sync pulses regenerated or passed through from the input? | 21:30 |
| futarisIRCcloud | https://github.com/timvideos/HDMI2USB-litex-firmware/issues/478#issuecomment-572493368 | 21:30 |
| tpb | Title: HDMI2USB/Opsis not working with Decimator MD-CROSS v2 · Issue #478 · timvideos/HDMI2USB-litex-firmware · GitHub (at github.com) | 21:30 |
| futarisIRCcloud | xobs: is there a simple gateware that will output a hdmi test pattern? Just want to try hooking up a NeTV2 to a decimator. | 21:33 |
| *** _franck_ has quit IRC | 22:35 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!