| *** tpb has joined #timvideos | 00:00 | |
| mithro | cr1901_modern: NFI | 00:06 |
|---|---|---|
| mithro | cr1901_modern: I think we probably want an MMU so we can run mainline kernel | 00:07 |
| shenki | mithro: which cpu are we going with? | 00:13 |
| mithro | or1k to start with | 00:14 |
| mithro | But the driver code should mostly be arch independent, right? | 00:15 |
| shenki | yep | 00:16 |
| shenki | do we have a project i can load onto my fpga? | 00:17 |
| mithro | The opsis-soc might be the best bet | 00:17 |
| mithro | Dunno what the state of the minispartan support is | 00:18 |
| shenki | ok | 00:18 |
| shenki | ok | 00:19 |
| mithro | You might be able to use the or1k from the hdmi2usb firmware repo | 00:19 |
| mithro | But without _florent_ 's dram rewrite, things seem to lock up if you are executing from the dram | 00:20 |
| mithro | (it's fine if you execute from bram) | 00:21 |
| *** ivodd has joined #timvideos | 00:21 | |
| mithro | Which is how we don't already run into the problem | 00:21 |
| mithro | Actually the base target might be fine | 00:22 |
| shenki | ok | 00:22 |
| shenki | i will hold off until we have some hardware to touch | 00:23 |
| mithro | If I understand the problem correctly it's caused by Multi master | 00:23 |
| mithro | I know the opsis-soc should work | 00:23 |
| cr1901_modern | mithro: Does NFI mean No ... Idea :P? | 00:24 |
| mithro | But an opsis is big to carry around | 00:24 |
| mithro | cr1901_modern: yes :-) | 00:24 |
| cr1901_modern | Just making sure. And yes, I really can't debug the lockup as is... it would be nice if the CPU would go to a specific vector on a bus error | 00:25 |
| cr1901_modern | but idk how that would be implemented | 00:25 |
| mithro | shenki: you can use https://gist.github.com/mithro/604da515edc1061a77a8ee6c1fe729e6 to get started | 00:26 |
| tpb | Title: Script to get all the enjoy-digital repos and set up a conda environment for using them in · GitHub (at gist.github.com) | 00:26 |
| mithro | shenki: the opsis-soc has a verilator simulation that you could get started with | 00:44 |
| mithro | No hardware needed for that | 00:45 |
| mithro | https://github.com/enjoy-digital/opsis-soc/blob/master/Makefile#L21 | 00:48 |
| tpb | Title: opsis-soc/Makefile at master · enjoy-digital/opsis-soc · GitHub (at github.com) | 00:48 |
| shenki | mithro: ok | 00:50 |
| *** sb0 has joined #timvideos | 01:11 | |
| *** puck has quit IRC | 01:27 | |
| *** puck has joined #timvideos | 01:30 | |
| *** ssk1328 has joined #timvideos | 01:42 | |
| *** nueces_ has joined #timvideos | 03:11 | |
| *** nueces has quit IRC | 03:14 | |
| *** rohitksingh_work has joined #timvideos | 04:00 | |
| mithro | morning ssk1328 | 04:35 |
| *** nueces_ has quit IRC | 04:47 | |
| *** ssk1328 has quit IRC | 05:15 | |
| *** rohitksingh_wor1 has joined #timvideos | 06:31 | |
| *** rohitksingh_work has quit IRC | 06:33 | |
| *** Bertl is now known as Bertl_zZ | 06:55 | |
| *** paradisaeidae has joined #timvideos | 07:03 | |
| *** sb0 has quit IRC | 07:14 | |
| *** paradisaeidae has quit IRC | 07:33 | |
| *** sb0 has joined #timvideos | 08:16 | |
| *** sb0 is now known as sb0____ | 08:17 | |
| *** sb0____ has quit IRC | 08:23 | |
| *** sb0 has joined #timvideos | 08:57 | |
| *** ssk1328 has joined #timvideos | 09:48 | |
| ssk1328 | mithro: ping? | 09:48 |
| ssk1328 | mithro: Any chace you saw the video | 09:48 |
| *** CARAM__ has quit IRC | 09:50 | |
| ssk1328 | mithro: Video of mixer with two DMA engines | 09:51 |
| *** CARAM__ has joined #timvideos | 09:52 | |
| mithro | ssk1328: no, I did not | 09:55 |
| ssk1328 | mithro: I added in the daily snippet | 09:55 |
| mithro | ssk1328: Great! I'll take a look in a bit | 09:55 |
| ssk1328 | mithro: https://photos.google.com/share/AF1QipMevEcGZvjoltkAkUoYSIHJHIdLSMyrzHIjgyAdXGAOSEN1RN085Y3moiTB9_spfA?key=bGZLX2FrdEozRnRyUllhRnJfMUtDTTdwWVkwLVN3 | 09:55 |
| mithro | ssk1328: What is happening in that video? | 10:00 |
| mithro | ssk1328: Seems to be flickering back and forth? | 10:01 |
| ssk1328 | pattern and input0 should mix? But only one of them is appearing at a time | 10:01 |
| ssk1328 | mithro: Yeah | 10:01 |
| ssk1328 | mithro: I suppose I can't read from 4 dma ports at a time? For bandwidth reasons? | 10:01 |
| mithro | ssk1328: How would that help? | 10:02 |
| ssk1328 | mithro: I am actually doing that right now | 10:02 |
| mithro | ssk1328: If you step through your video - I see a couple of mixed frames - but I don't know if that is the video or actually happening | 10:03 |
| mithro | ssk1328: It looks like you have connected two things to the output at the same time? | 10:04 |
| ssk1328 | mithro: I am not sure | 10:05 |
| ssk1328 | mithro: The phy object in Driver class | 10:05 |
| ssk1328 | mithro: Its for connectig Video to actual HDMI port of hardware correct? | 10:06 |
| mithro | ssk1328: Yes, that is my understanding - it takes a pixel stream and converts it to HDMI and sends it out the IO pins | 10:06 |
| ssk1328 | mithro: Okay I am updating the block diagram with exactly what I am doing | 10:07 |
| ssk1328 | mithro: Maybe I am making some incorrect assumptions/corrections | 10:07 |
| *** danielki has joined #timvideos | 10:14 | |
| *** CARAM__ has quit IRC | 10:15 | |
| *** CARAM__ has joined #timvideos | 10:17 | |
| mithro | ssk1328: ping when you want me to look at it | 10:17 |
| ssk1328 | mithro: Yes, see this https://docs.google.com/document/d/1g1c2IwCVxVzSHWdXbZ746HP-fnM4y1WqFuBZNkLi5mw/edit#heading=h.49gtwqs2n934 | 10:18 |
| tpb | Title: Mixer Block Design - Google Docs (at docs.google.com) | 10:18 |
| *** danielki has quit IRC | 10:18 | |
| ssk1328 | mithro: I have a TA meeting in 10 minutes, I will be back in an hour or so | 10:19 |
| ssk1328 | mithro: But you leave your comments here | 10:19 |
| mithro | ssk1328: that diagram looks like what I was thinking | 10:19 |
| *** olasd has quit IRC | 10:20 | |
| ssk1328 | But if I am instatntiating two DMA per HDMIOut and instantiate two HDMIOut, doesn't it mean, I am reading from four DMA ata time | 10:21 |
| mithro | ssk1328: Yes - which is why you need to do something slightly different for two outputs | 10:22 |
| *** olasd has joined #timvideos | 10:23 | |
| mithro | ssk1328: Updated what I think you need to do | 10:25 |
| ssk1328 | mithro: so I won't need two instantiations of HDMIOut module now | 10:33 |
| mithro | ssk1328: I'm suggesting getting rid of the HDMIOut module and splitting it into separate components | 10:35 |
| mithro | ssk1328: Tried to make it clearer what I was suggesting | 10:39 |
| *** Bertl_zZ is now known as Bertl | 12:42 | |
| *** rohitksingh_wor1 has quit IRC | 12:55 | |
| *** rohitksingh has joined #timvideos | 13:27 | |
| *** rohitksingh1 has joined #timvideos | 15:15 | |
| *** rohitksingh has quit IRC | 15:15 | |
| *** rohitksingh has joined #timvideos | 15:49 | |
| *** rohitksingh1 has quit IRC | 15:52 | |
| *** rohitksingh has quit IRC | 16:00 | |
| *** rohitksingh has joined #timvideos | 16:02 | |
| *** rohitksingh1 has joined #timvideos | 16:16 | |
| *** rohitksingh has quit IRC | 16:17 | |
| *** rohitksingh1 has quit IRC | 16:25 | |
| *** rohitksingh has joined #timvideos | 16:25 | |
| *** rohitksingh1 has joined #timvideos | 17:40 | |
| *** rohitksingh has quit IRC | 17:41 | |
| *** Bertl is now known as Bertl_oO | 18:03 | |
| *** rohitksingh1 has quit IRC | 18:25 | |
| *** rohitksingh has joined #timvideos | 18:29 | |
| *** rohitksingh has quit IRC | 18:36 | |
| *** rohitksingh has joined #timvideos | 19:01 | |
| *** danielki has joined #timvideos | 19:50 | |
| *** rohitksingh has quit IRC | 20:26 | |
| *** danielki has left #timvideos | 20:41 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!