| *** tpb <[email protected]> has joined #litex | 00:00 | |
| *** somlo <[email protected]> has quit IRC (Ping timeout: 272 seconds) | 00:10 | |
| *** somlo <[email protected]> has joined #litex | 00:14 | |
| *** Degi <[email protected]> has quit IRC (Ping timeout: 256 seconds) | 01:34 | |
| *** Degi <[email protected]> has joined #litex | 01:34 | |
| *** bl0x <[email protected]> has joined #litex | 02:51 | |
| *** bl0x_ <[email protected]> has quit IRC (Ping timeout: 256 seconds) | 02:52 | |
| *** TMM_ <[email protected]> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) | 03:19 | |
| *** TMM_ <[email protected]> has joined #litex | 03:20 | |
| *** FabM <FabM!~FabM@2a03:d604:103:600:2e60:8c7c:e8fb:7990> has joined #litex | 07:24 | |
| *** lexano <[email protected]> has quit IRC (Ping timeout: 272 seconds) | 08:29 | |
| *** lexano <[email protected]> has joined #litex | 08:41 | |
| *** GNUmoon <GNUmoon!~GNUmoon@gateway/tor-sasl/gnumoon> has quit IRC (Remote host closed the connection) | 08:46 | |
| *** GNUmoon <GNUmoon!~GNUmoon@gateway/tor-sasl/gnumoon> has joined #litex | 08:47 | |
| *** feldim2425_ <feldim2425_!~feldim242@2001:871:25a:803d:1d38:f750:80f4:a20f> has joined #litex | 09:03 | |
| *** feldim2425 <[email protected]> has quit IRC (Ping timeout: 255 seconds) | 09:04 | |
| *** GNUmoon <GNUmoon!~GNUmoon@gateway/tor-sasl/gnumoon> has quit IRC (Remote host closed the connection) | 10:04 | |
| *** GNUmoon <GNUmoon!~GNUmoon@gateway/tor-sasl/gnumoon> has joined #litex | 10:05 | |
| *** knicklicht <knicklicht!~knicklich@2a02:908:1995:4220:61ab:5771:dcc6:2412> has joined #litex | 11:59 | |
| knicklicht | Can I configure my Lattice ECP5 so that the pins I use for I2C are configured to be pulled up internally? Is it safe to just attach a pull-up to pins that are configured as LVCMOS33?? | 12:00 |
|---|---|---|
| tnt | internal pull-ups tend to be rather weak while I2C like stronger ones ( like 4.7k ) | 12:05 |
| knicklicht | So it's fine to just attach external pull-ups to the FPGAs output. I was worried that it might be using some internal pull-down or something by default and combining that might damage the chip | 12:06 |
| tnt | nope | 12:09 |
| knicklicht | thanks :-) | 12:09 |
| knicklicht | Any Idea why the I2C core is pulling down the lines by default? | 13:19 |
| tnt | Not really, I never used the i2c core from litex, I have my own :) | 13:21 |
| *** knicklicht <knicklicht!~knicklich@2a02:908:1995:4220:61ab:5771:dcc6:2412> has quit IRC (Ping timeout: 260 seconds) | 17:47 | |
| *** wild <[email protected]> has joined #litex | 18:45 | |
| *** wild <[email protected]> has quit IRC (Remote host closed the connection) | 18:45 | |
| *** wild <[email protected]> has joined #litex | 18:45 | |
| *** wild <[email protected]> has quit IRC (Remote host closed the connection) | 18:46 | |
| *** WildCard <[email protected]> has joined #litex | 18:48 | |
| WildCard | Hey, I have a question about litePCIe. Can you DMA to the Host without the host setting up a region via BAR0? Or is this not allowed through the current stream implementation of DMA. | 18:50 |
| WildCard | Im trying to understand atm how I could DMA to any address, similar to what you can do with tools like pcileech. | 18:51 |
| zyp | yes, you can | 18:55 |
| zyp | the BAR0 is a target, i.e. answering read and write requests to a specific address range | 18:56 |
| zyp | the DMA block is an initiator, issuing read and write requests to arbitrary addresses | 18:57 |
| tnt | Although I'd expect any decent/recent OS to use the IOMMU so that you can't access anything that wasn't setup ... | 18:59 |
| *** WildCard <[email protected]> has quit IRC (Remote host closed the connection) | 19:09 | |
| *** WildCard <[email protected]> has joined #litex | 19:09 | |
| WildCard | Ok, but then the Host, needs to use BAR0 to allocate memory for DMA. What im asking about is arbitrary DMA, I guess all that is needed there is the BME bit set in the control register, and memory read TLP requests to come in. | 19:14 |
| WildCard | I am talking about without and IOMMU enabled, and very early boot for instance. | 19:15 |
| WildCard | like post enumeration. | 19:15 |
| WildCard | I didnt know if the stream implementation of the DMA in litepcie was restricted. maybe to certain sizes? | 19:16 |
| WildCard | ok, i see you saying it can do arbitrary messages. Sounds good enough then. And how do you interact with it? via the CSRs? So i could DMA via the bios.c in the RiscV software for example. Ive yet to see an example of some software using those registers. | 19:17 |
| WildCard | perhaps i missunderstood what was being described here. https://github.com/enjoy-digital/litepcie/issues/54 | 19:20 |
| WildCard | 'The core has been mostly used to create SDR or Video systems for now, and that's the reason we only have streaming interfaces.' | 19:21 |
| *** FabM <FabM!~FabM@armadeus/team/FabM> has quit IRC (Ping timeout: 260 seconds) | 20:14 | |
| *** WildCard <[email protected]> has quit IRC (Remote host closed the connection) | 21:58 | |
| *** WildCard <[email protected]> has joined #litex | 21:58 | |
| *** WildCard <[email protected]> has quit IRC (Remote host closed the connection) | 22:14 | |
| *** WildCard <[email protected]> has joined #litex | 22:14 | |
| *** TMM_ <[email protected]> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.) | 22:38 | |
| *** TMM_ <[email protected]> has joined #litex | 22:39 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!