| *** tpb has joined #litex | 00:00 | |
| *** freemint has quit IRC | 01:03 | |
| *** freemint has joined #litex | 01:05 | |
| *** _whitelogger has quit IRC | 03:15 | |
| *** _whitelogger has joined #litex | 03:18 | |
| *** _whitelogger has quit IRC | 05:12 | |
| *** _whitelogger has joined #litex | 05:15 | |
| *** _whitelogger has quit IRC | 05:36 | |
| *** _whitelogger has joined #litex | 05:39 | |
| *** _whitelogger has quit IRC | 06:06 | |
| *** _whitelogger has joined #litex | 06:09 | |
| *** _whitelogger has quit IRC | 06:54 | |
| *** _whitelogger has joined #litex | 06:57 | |
| *** _whitelogger has quit IRC | 07:30 | |
| *** _whitelogger has joined #litex | 07:33 | |
| *** freemint has quit IRC | 10:10 | |
| *** freemint has joined #litex | 10:16 | |
| *** freemint has quit IRC | 10:35 | |
| *** rohitksingh has quit IRC | 11:18 | |
| *** rohitksingh has joined #litex | 11:24 | |
| *** rohitksingh has quit IRC | 11:51 | |
| *** freemint has joined #litex | 14:00 | |
| *** freemint has quit IRC | 14:03 | |
| *** freemint has joined #litex | 14:16 | |
| *** freemint has quit IRC | 14:45 | |
| *** freemint has joined #litex | 15:28 | |
| *** rohitksingh has joined #litex | 17:04 | |
| *** rohitksingh has quit IRC | 17:51 | |
| somlo | the default Rocket Linux variant is set up with (nSets=4, nWays=1, nTLBs=4) for both the L1D and L1I caches, and per my earlier experiments connecting LiteDRAM directly to the cached-RAM axi port, we get a 7% performace boost going from 64-bit to 128-bit data width, and a 1% boost going 128->256. | 17:55 |
|---|---|---|
| somlo | it was nagging at me that some of that is muddied by the presence and size of the cache, so I generated low-cache Rocket variants with (nsets=4, nways=1, ntlbs=4) | 17:56 |
| *** rohitksingh has joined #litex | 17:56 | |
| somlo | MUCH worse performance overall, but going from 64-128 mem_axi data width got me 18% better perormance, and going from 128->256 data width got me a 3% improvement | 17:56 |
| somlo | so, more improvement from mem_axi width doubling in the absence of a large L1 cache | 17:57 |
| somlo | _florent_, daveshah: ^^ | 17:57 |
| somlo | just wanted to make sure it's actually worth adding "wide" rocket variants as a first choice, and keeping data-width FSM based conversion as a backup plan only | 17:58 |
| *** ambro718 has joined #litex | 18:07 | |
| somlo | but with small caches, performance was really BAD in absolute terms, so it took a few days to get all the results :) | 18:07 |
| *** keesj has quit IRC | 19:00 | |
| *** freemint has quit IRC | 20:08 | |
| scanakci | _florent_: I am a bit confused about --rom-init in terms of provided file format. My understanding is that mem.init file in generated gateware folder contains rom content. If I put my program into mem.init (each line includes 32 bit number in hexa format), I successfully simulate the program. If I use the --rom-init with the same file, It does not work. | 20:15 |
| scanakci | For a reason, mem.init file does not contain what I want to have in ROM. | 20:16 |
| *** freemint has joined #litex | 20:22 | |
| _florent_ | somlo: thanks for the results, so do you think it's useful to use 128-bit data-width with the L1 cache? | 20:34 |
| _florent_ | scanakci: --rom-init should point to a binary file, mem_x.init are the memory initialization files for the generated verilog | 20:40 |
| _florent_ | scanakci: but if you pass a binary file to --rom-init, you should have a mem_x.init file with the same content (but in hexa format) | 20:41 |
| scanakci | thanks _florent:. I will try it soon. | 20:41 |
| scanakci | I could generate a bitstream of Litex+Blackparrot. As far as I see, there is not --rom-init option in genesys2.py file. | 20:42 |
| scanakci | How am I supposed to load Litex-ROM? I thought that it is again reading from file during systhesis with readmemh. | 20:43 |
| *** CarlFK has quit IRC | 20:46 | |
| _florent_ | indeed, it's not available from the command line, but i could add that | 20:46 |
| *** rohitksingh has quit IRC | 21:02 | |
| *** rohitksingh has joined #litex | 21:10 | |
| scanakci | okay. using binary file with --rom-init still did not give me my expected mem.init file. For now, I will manually update mem.init in gateware for both simulation and FPGA. | 21:54 |
| *** ambro718 has quit IRC | 22:08 | |
| *** rohitksingh has quit IRC | 22:15 | |
| *** CarlFK has joined #litex | 22:15 | |
| *** somlo has quit IRC | 23:44 | |
| *** somlo has joined #litex | 23:46 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!