| *** tpb has joined #timvideos | 00:00 | |
| mithro | tannewt: Afternoon | 00:40 |
|---|---|---|
| tannewt | hi hi mithro | 00:40 |
| mithro | tannewt: So, I need to do the USB descriptor dance... | 00:41 |
| tannewt | sounds like its time for tinyusb | 00:41 |
| tannewt | you could do just that part but you'll get higher level than you need to then | 00:42 |
| tannewt | tinyusb's api is mostly transactions | 00:42 |
| tannewt | and it handles all of the logic above that | 00:42 |
| mithro | So, how do we get started with tinyusb? | 00:43 |
| mithro | tannewt: Which tinyusb repo should I fork from? | 00:43 |
| tannewt | hathach's develop branch | 00:44 |
| tannewt | this example: https://github.com/hathach/tinyusb/tree/develop/examples/device/cdc_msc_hid | 00:44 |
| tpb | Title: tinyusb/examples/device/cdc_msc_hid at develop · hathach/tinyusb · GitHub (at github.com) | 00:44 |
| tannewt | we can hack it to turn msc and hid off | 00:45 |
| tannewt | https://github.com/hathach/tinyusb/blob/develop/examples/device/cdc_msc_hid/src/tusb_config.h#L97 is where to turn msc off | 00:45 |
| tpb | Title: tinyusb/tusb_config.h at develop · hathach/tinyusb · GitHub (at github.com) | 00:45 |
| tannewt | I'm working on porting instructions now. here is what I have so far: https://gist.github.com/tannewt/4ea0646d6847138ac7936eaae45ab5f6 | 00:47 |
| tpb | Title: start of tinyusb porting instructions · GitHub (at gist.github.com) | 00:47 |
| mithro | The first step to adding support is including the register definitions and startup code for the microcontroller in TinyUSB. | 00:49 |
| mithro | tannewt: There isn't any examples/device/cdc_msc_hid | 00:50 |
| mithro | Oh wait | 00:50 |
| mithro | my branch is 83 commits behind upstream | 00:50 |
| tannewt | we've been working hard on it if you forked earlier | 00:51 |
| mithro | Which is the best board to start from? | 00:53 |
| tannewt | whatever you are testing on I think | 00:55 |
| tannewt | ah, to start from? one of the samds | 00:55 |
| tannewt | they are the only ones setup with the makefile now | 00:55 |
| mithro | arm-none-eabi-gcc | 00:56 |
| mithro | Where do I change that? | 00:56 |
| tannewt | in the makefile | 00:57 |
| tannewt | it should be factored out :-) | 00:57 |
| tannewt | https://github.com/hathach/tinyusb/blob/develop/examples/device/cdc_msc_hid/Makefile#L15 | 00:58 |
| tpb | Title: tinyusb/Makefile at develop · hathach/tinyusb · GitHub (at github.com) | 00:58 |
| tannewt | looks like board.mk can override it | 00:58 |
| mithro | How does "BOARD_METRO_M4_EXPRESS" get defined? | 01:02 |
| tannewt | its based on the board given to make | 01:03 |
| tannewt | https://github.com/hathach/tinyusb/blob/develop/examples/device/cdc_msc_hid/Makefile#L60 | 01:03 |
| tpb | Title: tinyusb/Makefile at develop · hathach/tinyusb · GitHub (at github.com) | 01:03 |
| mithro | https://github.com/mithro/tinyusb/commit/820c92ac37e4f1521a9ff93fc80ca69ee30ada55 | 01:05 |
| tpb | Title: Starting supporting valentyusb. · mithro/tinyusb@820c92a · GitHub (at github.com) | 01:05 |
| tannewt | main shouldn't be provided by the board | 01:06 |
| tannewt | do init in board_init | 01:06 |
| mithro | https://www.irccloud.com/pastebin/e1EupdC5/ | 01:06 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 01:06 |
| tannewt | does the compiler handle packed? | 01:07 |
| mithro | $ lm32-unknown-elf-gcc --version | 01:08 |
| mithro | lm32-unknown-elf-gcc (8.2.0) 8.2.0 | 01:08 |
| tannewt | hrm, it should | 01:08 |
| tannewt | try commenting that check out | 01:10 |
| tannewt | looks like it could be 10 bytes is the union isn't aligned | 01:11 |
| tannewt | https://github.com/mithro/tinyusb/blob/develop/src/device/dcd.h#L66 | 01:11 |
| tpb | Title: tinyusb/dcd.h at develop · mithro/tinyusb · GitHub (at github.com) | 01:11 |
| tannewt | does the lm32 handle unaligned access ok? | 01:11 |
| mithro | tannewt: Dunno, shouldn't the compiler handle that for you? | 01:28 |
| tannewt | it doesn't if you are dmaing data for samd | 01:29 |
| tannewt | thats why you fail that check I think. the lm32 compiler isn't aligning the union afaict | 01:29 |
| mithro | Oh wait | 01:29 |
| mithro | You *want* it unpacked | 01:30 |
| tannewt | right, the union should be aligned for that struct | 01:31 |
| tannewt | I was confusing myself | 01:31 |
| mithro | https://github.com/mithro/tinyusb/commit/7109ab065e0a4e025bd3154ef69ffec0edc87b1a | 01:31 |
| tpb | Title: Check offsetof of the two first int8. · mithro/tinyusb@7109ab0 · GitHub (at github.com) | 01:31 |
| tannewt | I think its setup_received that matters but I could be wrong | 01:32 |
| tannewt | I'm honestly not sure | 01:33 |
| tannewt | I'd just comment it out for now and move on | 01:33 |
| mithro | tannewt: https://github.com/mithro/tinyusb/commit/c88991d2dd82176459ccb6edf095dfa36fd1661d | 01:41 |
| tpb | Title: Check alignment of the required components (rather than structure size). · mithro/tinyusb@c88991d · GitHub (at github.com) | 01:41 |
| mithro | tannewt: I think that is what you are *actually* trying to check? | 01:41 |
| tannewt | ya I think so | 01:41 |
| mithro | https://github.com/mithro/tinyusb/commit/c74406a14b516e79de1a37b1322b9dd5c5523213 | 01:42 |
| tpb | Title: Check alignment of the required components (rather than structure size). · mithro/tinyusb@c74406a · GitHub (at github.com) | 01:42 |
| mithro | (Fixed the formatting) | 01:42 |
| mithro | board_led_control(BOARD_LED0, led_state); | 01:42 |
| tannewt | 👍 | 01:42 |
| mithro | I'm just going to comment that out for now? | 01:43 |
| tannewt | sure | 01:43 |
| tannewt | it just blinks to let you know its going | 01:43 |
| tannewt | you already know the clocks are going right | 01:43 |
| mithro | https://www.irccloud.com/pastebin/qdLsbfId/ | 01:43 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 01:43 |
| mithro | So I guess I need dcd_valentyusb.c and hal_valentyusb.c ? | 01:44 |
| tannewt | the juicy part is src/portable | 01:44 |
| tannewt | yup! | 01:44 |
| tannewt | in src/portable | 01:44 |
| mithro | What does dcd stand for? | 01:45 |
| tannewt | device class "device" I thiiink | 01:46 |
| tannewt | hal is for code thats shared regardless of host vs device | 01:47 |
| tannewt | and dcd is for device stuff | 01:47 |
| mithro | Looks like there is just tusb_hal_int_enable / tusb_hal_int_disable ? | 01:48 |
| tannewt | tusb_hal_init too | 01:50 |
| tannewt | which would turn on clocks to the usb peripheral for example | 01:51 |
| mithro | You have a very weird formatting style for your code... | 01:52 |
| tannewt | its not me, its hathach's style | 01:53 |
| tannewt | I mean to clang-format all of our stuff at some point | 01:53 |
| mithro | tannewt: https://github.com/mithro/tinyusb/commit/c23f4ce907cc86d8a54d70d4f4f6b8ec367a3dab | 02:01 |
| tpb | Title: Blank USB structures. · mithro/tinyusb@c23f4ce · GitHub (at github.com) | 02:01 |
| tannewt | ya, good start | 02:02 |
| mithro | What is "dcd_edpt_open" ? | 02:04 |
| tannewt | its called after the host picks a config | 02:05 |
| tannewt | for all non-control endpoints | 02:05 |
| mithro | So the only one I really care about is "dcd_edpt_xfer" ? | 02:11 |
| tannewt | ya, that and the interrupt handler | 02:13 |
| tannewt | to queue events to tell the stack when things finish, reset and receive setup packets | 02:14 |
| mithro | What does "dcd_edpt_busy" mean? | 02:28 |
| tannewt | it checks to see if a transfer on an endpoint is pending | 02:31 |
| tannewt | aka you set the buffer but the host hasn't finished transferring the data | 02:31 |
| mithro | tannewt: https://github.com/mithro/tinyusb/blob/424b7245a82b5ec9873c8baa8e4a9cb7f4158269/src/portable/valentyusb/dcd_valentyusb.c | 02:35 |
| tpb | Title: tinyusb/dcd_valentyusb.c at 424b7245a82b5ec9873c8baa8e4a9cb7f4158269 · mithro/tinyusb · GitHub (at github.com) | 02:35 |
| mithro | tannewt: Could you clean up that file while I try and get things to link into the fpga image? | 02:38 |
| tannewt | I've actually gotta take a break for dinner and chores | 02:39 |
| tannewt | will be back on later though | 02:39 |
| mithro | cr1901_modern: ping? | 02:43 |
| *** futarisIRCcloud has joined #timvideos | 03:08 | |
| *** ivodd_ has quit IRC | 03:23 | |
| *** ivodd has joined #timvideos | 03:23 | |
| *** andi-m has quit IRC | 04:12 | |
| *** andi-m has joined #timvideos | 04:14 | |
| *** rohitksingh_work has joined #timvideos | 04:36 | |
| mithro | cr1901_modern: So, I've confirmed that disabling multiple+divide breaks the stub firmware | 04:43 |
| mithro | cr1901_modern: A quick look through the disassembly doesn't look like there is any div/mul instructions inside the firmware... | 04:51 |
| futarisIRCcloud | mithro: Probably some mul/div instructions in a library. | 05:00 |
| mithro | futarisIRCcloud: I mean the complete linked firmware | 05:00 |
| futarisIRCcloud | mithro: Hmm. Were there no exceptions, when the stub firmware crashed? What did you change the compiler args to? | 05:02 |
| mithro | The stub firmware acts strangely - it's hard to know exactly what is going on... | 05:03 |
| cr1901_modern | mithro: Which CPU (sorry not paying attention tonight lol) | 05:08 |
| mithro | LM32 | 05:08 |
| cr1901_modern | This matches _florent_'s experience. I suspect there's a nice bug in the core itself. | 05:09 |
| mithro | Seems like it | 05:18 |
| mithro | cr1901_modern: I'm currently running the stub firmware from rom on the 8k | 05:19 |
| mithro | With the usb core enabled.... | 05:19 |
| mithro | It's a bit tight :-P | 05:19 |
| mithro | Info: Device utilisation: | 05:20 |
| mithro | Info: ICESTORM_LC: 4572/ 7680 59% | 05:20 |
| mithro | Info: ICESTORM_RAM: 32/ 32 100% | 05:20 |
| mithro | Info: SB_IO: 12/ 256 4% | 05:20 |
| cr1901_modern | 32/32 block RAMs is prob fine | 05:21 |
| cr1901_modern | if it really bothers you, you could prob reduce the amount of RAM avail to 5.5kB (free 4 block RAMs) | 05:22 |
| mithro | kwargs['integrated_sram_size']=4*1024 | 05:28 |
| cr1901_modern | I do hope that's the changes you just made | 05:28 |
| mithro | firmware_rom_size = 8*1024 | 05:28 |
| cr1901_modern | 4kB is starting to _really_ cut it close for what you can do with micropython, b/c of the .data/.bss section | 05:29 |
| cr1901_modern | mithro: Is 32/32 used blocks _with_ "kwargs['integrated_sram_size']=4*1024"? | 05:31 |
| cr1901_modern | if so that's concerning | 05:31 |
| mithro | I'm not using micropython at the moment | 05:36 |
| tannewt | mithro, back from dinner. hows it going? | 05:42 |
| mithro | tannewt: I'm just about to start trying to link tinyusb into the firmware I have | 05:43 |
| tannewt | 👍 | 05:43 |
| mithro | gcc: error: build-valentyusb/obj/hw/bsp/valentyusb/board_valentyusb.o: No such file or directory | 05:49 |
| mithro | gcc: error: nosys.specs: No such file or directory | 05:49 |
| tannewt | try removing the nosys.specs thing | 05:49 |
| mithro | What is nosys.specs? | 05:50 |
| tannewt | do you have hw/bsp/valentyusb/board_valentyusb.c ? | 05:50 |
| mithro | tannewt: I just touched an empty file to make the error go away... | 05:50 |
| tannewt | its to reduce the number of c internals included I think | 05:50 |
| tannewt | there are a couple functions need in there to init the board | 05:51 |
| mithro | tannewt: You really should obey "V=1" in your makefile... | 05:51 |
| tannewt | go ahead and add it | 05:52 |
| tannewt | I was trying to keep it as small as I could | 05:52 |
| tannewt | but thats impossible with make | 05:52 |
| mithro | tannewt: Found a bug -> https://github.com/mithro/tinyusb/blob/develop/hw/bsp/metro_m0_express/board.mk#L14 | 06:01 |
| tpb | Title: tinyusb/board.mk at develop · mithro/tinyusb · GitHub (at github.com) | 06:01 |
| mithro | tannewt: should be LDFLAGS not LD_FLAGS | 06:01 |
| tannewt | ah, its probably redundant with the CFLAGS | 06:02 |
| tannewt | fix is welcome | 06:02 |
| mithro | tannewt: no make clean? | 06:03 |
| tannewt | that should work | 06:03 |
| tannewt | https://github.com/mithro/tinyusb/blob/424b7245a82b5ec9873c8baa8e4a9cb7f4158269/examples/device/cdc_msc_hid/Makefile#L121 | 06:04 |
| tpb | Title: tinyusb/Makefile at 424b7245a82b5ec9873c8baa8e4a9cb7f4158269 · mithro/tinyusb · GitHub (at github.com) | 06:04 |
| mithro | Oh, still need the -C | 06:04 |
| tannewt | you can cd to the directory too and it'll work without -C | 06:05 |
| mithro | What is __be2n ? | 06:10 |
| tannewt | https://github.com/hathach/tinyusb/blob/master/tinyusb/common/compiler/compiler_gcc.h#L125 | 06:11 |
| tpb | Title: tinyusb/compiler_gcc.h at master · hathach/tinyusb · GitHub (at github.com) | 06:11 |
| tannewt | I didn't write most of this code | 06:11 |
| mithro | tannewt: it also seems to be somewhat defined in tusb_common.h ? | 06:12 |
| tannewt | that is just what google found for me. maybe it moved | 06:13 |
| *** Kripton has quit IRC | 06:13 | |
| mithro | I'm a bit confused why it isn't picking up the definitions from compiler_gcc.h ... | 06:16 |
| mithro | oh | 06:17 |
| mithro | You don't have bigendian definitions.... | 06:17 |
| *** Kripton has joined #timvideos | 06:26 | |
| mithro | tannewt: well fooy... | 06:29 |
| mithro | https://www.irccloud.com/pastebin/yrefRo2K/ | 06:30 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 06:30 |
| tannewt | are you doing the debug build? | 06:30 |
| mithro | not deliberately | 06:30 |
| tannewt | did you disable msc? | 06:30 |
| tannewt | kk you'd have to give it DEBUG=1 | 06:30 |
| mithro | Probably not? how do I do that? | 06:32 |
| tannewt | https://github.com/mithro/tinyusb/blob/develop/examples/device/cdc_msc_hid/src/tusb_config.h#L99 | 06:33 |
| tpb | Title: tinyusb/tusb_config.h at develop · mithro/tinyusb · GitHub (at github.com) | 06:33 |
| mithro | Okay, that fixed the sram - but still to big for user flash.... | 06:34 |
| mithro | /home/tansell/github/timvideos/litex-buildenv/build/conda/bin/../lib/gcc/lm32-elf/8.2.0/../../../../lm32-elf/bin/ld: region `user_flash' overflowed by 5264 bytes | 06:35 |
| tannewt | how much do you have to work with? | 06:35 |
| tannewt | can you post the map somewhere? | 06:35 |
| mithro | On the BX we have have 128 KBit of block ram for ROM+RAM+USB FIFO (== 32kbyte) | 06:37 |
| mithro | Sorry | 06:38 |
| mithro | that should be 16kilobytes.... | 06:38 |
| tannewt | that isn't alot | 06:38 |
| mithro | We can use XIP spiflash for data in the future | 06:39 |
| mithro | s/data/ro stuff/ | 06:39 |
| mithro | tannewt: But I was hoping to get the descriptor stuff going with everything embedded in the gateware.... | 06:40 |
| tannewt | well lets take a look at the map to see what we're still including | 06:40 |
| tannewt | and figure out how big user_flash is | 06:41 |
| mithro | tannewt: I'm just trying to bump user_flash up to 10kbytes | 06:41 |
| tannewt | I have no idea how tiny tinyusb is | 06:42 |
| tannewt | I know circuitpython's footprint was reduced | 06:42 |
| mithro | Okay, I shifted 2kbytes from sram to user_flash | 06:47 |
| mithro | /usr/local/google/home/tansell/github/timvideos/litex-buildenv/build/conda/bin/../lib/gcc/lm32-elf/8.2.0/../../../../lm32-elf/bin/ld: region `user_flash' overflowed by 3216 bytes | 06:48 |
| tannewt | a map file (from the build output) will help debug whats in the flash | 06:52 |
| mithro | so, I just pushed everything.... | 06:54 |
| mithro | tannewt: https://github.com/mithro/litex-buildenv - branch tinyusb-work | 06:55 |
| tpb | Title: GitHub - mithro/litex-buildenv: An environment for building LiteX based FPGA designs. Makes it easy to get everything you need! (at github.com) | 06:55 |
| tannewt | can you post the map file? I don't have the toolchain setup and am debugging something else at the moment | 06:55 |
| mithro | tannewt: https://paste.ubuntu.com/p/yFxDzgFq84/ | 06:57 |
| tpb | Title: Ubuntu Pastebin (at paste.ubuntu.com) | 06:57 |
| tannewt | thanks looking | 06:59 |
| *** thaytan has quit IRC | 07:01 | |
| *** thaytan has joined #timvideos | 07:03 | |
| *** ChanServ sets mode: +v thaytan | 07:03 | |
| tannewt | its weird to me that it has rand and srand | 07:04 |
| tannewt | I don't think the linker can be smart with libbase-nofloat.a because it wasn't compiled with lto | 07:07 |
| tannewt | the strings from tinyusb seem pretty large to me too | 07:09 |
| tannewt | https://www.irccloud.com/pastebin/2wyA2K6Q/ | 07:10 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 07:10 |
| tannewt | if you have a bin I'll look at whats in there | 07:10 |
| mithro | I need to run away | 07:21 |
| tannewt | k, goodnight! sorry I can't repro here atm. will try to do that later this week | 07:27 |
| tannewt | currently on the critical path for a feather nrf52840 | 07:27 |
| *** CarlFK has quit IRC | 07:35 | |
| mithro | hrm... | 07:36 |
| mithro | lm32-elf-ld: main.o: plugin needed to handle lto object | 07:36 |
| tannewt | weird, haven't seen that | 07:38 |
| mithro | hrm, now it can't find puts/printf/memcpy.... | 08:24 |
| tannewt | after lto? | 08:25 |
| tannewt | maybe its dropped when the library is created? | 08:25 |
| tannewt | I usually compile it all at once | 08:25 |
| mithro | grep still says that libbase-nofloat.a matches the name "memset" | 08:26 |
| tannewt | though I'm not sure what the link map looks like for the tinyusb example | 08:26 |
| tannewt | mithro, I've checked out your branch if you want me to poke at it with you | 08:34 |
| mithro | tannewt: So, it doesn't look like it's loading the libbase-float contents.... | 08:35 |
| tannewt | into the link command? | 08:36 |
| mithro | https://www.irccloud.com/pastebin/ZxRs35io/ | 08:37 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 08:37 |
| mithro | Looking at the .map file - it seems these things are in a "/tmp/cc....ltrans.o" file? | 08:39 |
| tannewt | ya, thats how the lto works | 08:40 |
| mithro | Were as the stub firmware has a section at the top which says | 08:40 |
| mithro | "Archive member included to satisfy reference by file (symbol)" | 08:41 |
| tannewt | but it shouldn't need the whole archive | 08:41 |
| mithro | Oh... | 08:42 |
| mithro | The archives are being passed in as compile time objects... | 08:42 |
| tannewt | right, and may need to be compiled differently to work with lto | 08:43 |
| tannewt | or we figure out how to remove the reference :-) | 08:43 |
| tannewt | how are you building it from within the litex-env? | 08:44 |
| tannewt | I think I'm close to having it setup right | 08:45 |
| mithro | Yes | 08:45 |
| mithro | I think the issue is that it doesn't think anything in the archive is compatible? | 08:45 |
| tannewt | not sure | 08:48 |
| mithro | Ahh ha! | 08:48 |
| mithro | /usr/local/google/home/tansell/github/timvideos/litex-buildenv/build/conda/bin/../lib/gcc/lm32-elf/8.2.0/../../../../lm32-elf/bin/ld: region `user_flash' overflowed by 468 bytes | 08:49 |
| tannewt | so close! | 08:49 |
| tannewt | whats the build command from within the litex env? | 08:49 |
| mithro | https://www.irccloud.com/pastebin/ouq1jR7V/ | 08:49 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 08:49 |
| tannewt | do you have a tick timer? | 08:49 |
| tannewt | or a cycle count? | 08:50 |
| tannewt | I'm in `(LX P=tinyfpga_bx T=usb.minimal F=stub R=tinyusb-work) [tannewt@dalinar litex-buildenv]$` | 08:50 |
| tannewt | that was my best guess | 08:50 |
| mithro | looks pretty good | 08:53 |
| mithro | Let me commit and push what I have | 08:54 |
| tannewt | kk | 08:55 |
| mithro | tannewt: pushed | 08:58 |
| mithro | tannewt: process is currently | 08:58 |
| mithro | `make firmware` in the top level | 08:58 |
| mithro | cd third_party/valentyusb/third_party/tinyusb | 08:58 |
| mithro | make -C examples/device/cdc_msc_hid BOARD=valentyusb clean && make -C examples/device/cdc_msc_hid BOARD=valentyusb | 08:59 |
| mithro | Make sure you update all the submodules... | 08:59 |
| tannewt | need to push litex? | 09:00 |
| tannewt | `Fetched in submodule path 'third_party/litex', but it did not contain efb0801c20aa397b3c425299aa6835aad2a364c1.` | 09:00 |
| mithro | tannewt: You'll probably need to add my litex repo | 09:03 |
| tannewt | kk | 09:04 |
| tannewt | k, got it | 09:05 |
| mithro | I really must head home now | 09:05 |
| tannewt | yes for sure! | 09:06 |
| tannewt | I'll poke a little then go to bed | 09:06 |
| tannewt | I think I'm where you are | 09:07 |
| tannewt | missing refs to memcpy, memset and abort | 09:07 |
| mithro | tannewt: hrm, I was sure I pushed the fix to that? | 09:07 |
| tannewt | hrm | 09:08 |
| tannewt | 7ac239195078735922a9f399a0526f9e7073da6b is my latest in tinyusb | 09:08 |
| mithro | That is what I'm on.... | 09:08 |
| tannewt | ah, I needed to clean | 09:09 |
| tannewt | eesh its late | 09:09 |
| tannewt | https://www.irccloud.com/pastebin/KfefH8X5/ | 09:09 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 09:09 |
| mithro | tannewt: I wonder why yours is 1208 bytes? | 09:09 |
| tannewt | whats yours? | 09:10 |
| mithro | Oh it linked! | 09:10 |
| tannewt | whoa! | 09:11 |
| mithro | 11k ROM, 1k RAM :-P | 09:11 |
| tannewt | 🎉 | 09:11 |
| mithro | Pushed that change.... | 09:12 |
| mithro | tannewt: Did you do a make clean in the top level ? | 09:12 |
| tannewt | no | 09:13 |
| mithro | tannewt: you'll probably want to do that... | 09:13 |
| tannewt | yup | 09:14 |
| tannewt | got it linked here too | 09:14 |
| mithro | Well, I'm not going to try this firmware tonight.... | 09:14 |
| mithro | tannewt: In theory, cp xxx.bin file over top of the stub firmware, then "make gateware && make flash" | 09:15 |
| tannewt | you know I'm tempted :-) | 09:15 |
| tannewt | is it loaded via usb or uart? | 09:16 |
| mithro | or modify targets/tinyfpga_bx/usb.py "firmware_filename" to point to the new location | 09:16 |
| mithro | tannewt: flash will load via the original tinyusb bootloader | 09:16 |
| mithro | tannewt: Need to wire up a UART like shown here -> https://github.com/timvideos/litex-buildenv/wiki/HowTo-FuPy-on-iCE40-Boards#hardware-modification-needed | 09:17 |
| tpb | Title: HowTo FuPy on iCE40 Boards · timvideos/litex-buildenv Wiki · GitHub (at github.com) | 09:17 |
| mithro | tannewt: To get console output | 09:18 |
| mithro | tannewt: I'm pretty sure I haven't gotten the pullup connected correctly yet | 09:18 |
| tannewt | making gateware... | 09:19 |
| mithro | Also, the USB core doesn't seem to work with nextpnr | 09:19 |
| tannewt | meaning it just increases iterations? | 09:20 |
| tannewt | nvm, still going | 09:20 |
| mithro | Doesn't respond to USB packets | 09:21 |
| mithro | Even though timing still looks pretty good | 09:22 |
| daveshah | Is it on latest master where the pll issue on the tinyfpga was fixed? | 09:22 |
| tannewt | https://www.irccloud.com/pastebin/DiNmgKNe/ | 09:22 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 09:22 |
| mithro | daveshah: I believe so | 09:22 |
| mithro | tannewt: just touch that file for now | 09:23 |
| tannewt | k | 09:23 |
| mithro | I would also recommend flashing the stub firmware first | 09:23 |
| tannewt | https://www.irccloud.com/pastebin/EoNoZGCg/ | 09:24 |
| tpb | Title: Snippet | IRCCloud (at www.irccloud.com) | 09:24 |
| mithro | And checking you get a prompt from that | 09:24 |
| mithro | Need to upgrade tinyprog from git | 09:24 |
| tannewt | k, I think its time to call it a night | 09:25 |
| tannewt | so close but there are likely more bugs past this one | 09:25 |
| daveshah | mithro: does it work with arachne? | 09:25 |
| mithro | I'm already walking home | 09:26 |
| mithro | daveshah: haven't tested with arachne | 09:26 |
| mithro | daveshah: it has two clock domains | 09:26 |
| daveshah | That would rule out an issue in Yosys | 09:26 |
| daveshah | arachne-pnr supports 0 clock domains | 09:27 |
| daveshah | Yet plenty of people have used it for designs with 2 or 3 clock domains | 09:27 |
| *** futarisIRCcloud has quit IRC | 09:27 | |
| mithro | I can test tomorrow | 09:28 |
| daveshah | Thanks | 09:28 |
| tannewt | ya, looks like its stuck in reset | 09:34 |
| tannewt | the latest tinyprog pypi releases are missing py3 support | 09:35 |
| tannewt | https://github.com/tinyfpga/TinyFPGA-Bootloader/issues/25 | 09:35 |
| tpb | Title: tinyprog 1.0.23 has no py3 wheel · Issue #25 · tinyfpga/TinyFPGA-Bootloader · GitHub (at github.com) | 09:35 |
| tannewt | anyway, good work! | 09:35 |
| tannewt | I'm off to bed | 09:35 |
| *** futarisIRCcloud has joined #timvideos | 10:37 | |
| futarisIRCcloud | https://twitter.com/q3k/status/1067237002385518593 | 10:37 |
| *** futarisIRCcloud has quit IRC | 12:47 | |
| *** rohitksingh_work has quit IRC | 13:05 | |
| *** rohitksingh has joined #timvideos | 14:56 | |
| *** CarlFK has joined #timvideos | 15:55 | |
| *** ChanServ sets mode: +v CarlFK | 15:55 | |
| *** rohitksingh has quit IRC | 16:05 | |
| *** rohitksingh has joined #timvideos | 16:14 | |
| *** CarlFK has quit IRC | 16:37 | |
| *** froztbyte has quit IRC | 17:49 | |
| *** froztbyte has joined #timvideos | 17:50 | |
| *** froztbyte has joined #timvideos | 17:50 | |
| *** tsglove has quit IRC | 18:15 | |
| *** tsglove has joined #timvideos | 18:22 | |
| *** rohitksingh has quit IRC | 18:35 | |
| *** rohitksingh has joined #timvideos | 18:53 | |
| *** nrossi has quit IRC | 19:00 | |
| *** nancy[m] has quit IRC | 19:00 | |
| *** shivm28[m] has quit IRC | 19:00 | |
| *** felix[m]2 has quit IRC | 19:00 | |
| *** ducky[m] has quit IRC | 19:00 | |
| *** micolous[m] has quit IRC | 19:00 | |
| *** synaption[m] has quit IRC | 19:00 | |
| *** master1588[m] has quit IRC | 19:00 | |
| *** pzieba[m] has quit IRC | 19:00 | |
| *** futaris[m] has quit IRC | 19:00 | |
| *** jea[m] has quit IRC | 19:00 | |
| *** CarlFK[m] has quit IRC | 19:00 | |
| *** nbags[m] has quit IRC | 19:00 | |
| *** xobs has quit IRC | 19:00 | |
| *** jfng has quit IRC | 19:00 | |
| *** CarlFK[m] has joined #timvideos | 19:19 | |
| *** ChanServ sets mode: +v CarlFK[m] | 19:19 | |
| *** rohitksingh has quit IRC | 19:36 | |
| *** jea[m] has joined #timvideos | 19:42 | |
| *** felix[m]2 has joined #timvideos | 19:42 | |
| *** nrossi has joined #timvideos | 19:42 | |
| *** jfng has joined #timvideos | 19:42 | |
| *** xobs has joined #timvideos | 19:42 | |
| *** master1588[m] has joined #timvideos | 19:42 | |
| *** futaris[m] has joined #timvideos | 19:42 | |
| *** nbags[m] has joined #timvideos | 19:42 | |
| *** shivm28[m] has joined #timvideos | 19:42 | |
| *** micolous[m] has joined #timvideos | 19:42 | |
| *** pzieba[m] has joined #timvideos | 19:42 | |
| *** nancy[m] has joined #timvideos | 19:42 | |
| *** ducky[m] has joined #timvideos | 19:42 | |
| *** synaption[m] has joined #timvideos | 19:42 | |
| *** Kripton has quit IRC | 19:44 | |
| *** Kripton has joined #timvideos | 19:48 | |
| *** rohitksingh has joined #timvideos | 19:50 | |
| *** pzieba[m] has quit IRC | 20:02 | |
| *** felix[m]2 has quit IRC | 20:02 | |
| *** jfng has quit IRC | 20:02 | |
| *** ducky[m] has quit IRC | 20:02 | |
| *** nancy[m] has quit IRC | 20:02 | |
| *** futaris[m] has quit IRC | 20:02 | |
| *** nbags[m] has quit IRC | 20:02 | |
| *** nrossi has quit IRC | 20:02 | |
| *** jea[m] has quit IRC | 20:03 | |
| *** synaption[m] has quit IRC | 20:03 | |
| *** shivm28[m] has quit IRC | 20:03 | |
| *** micolous[m] has quit IRC | 20:03 | |
| *** xobs has quit IRC | 20:03 | |
| *** master1588[m] has quit IRC | 20:03 | |
| *** CarlFK[m] has quit IRC | 20:03 | |
| *** CarlFK has joined #timvideos | 20:05 | |
| *** ChanServ sets mode: +v CarlFK | 20:05 | |
| *** CarlFK[m] has joined #timvideos | 20:11 | |
| *** ChanServ sets mode: +v CarlFK[m] | 20:11 | |
| *** rohitksingh has quit IRC | 20:15 | |
| *** jea[m] has joined #timvideos | 20:33 | |
| *** felix[m]2 has joined #timvideos | 20:33 | |
| *** xobs has joined #timvideos | 20:33 | |
| *** jfng has joined #timvideos | 20:33 | |
| *** nrossi has joined #timvideos | 20:33 | |
| *** ducky[m] has joined #timvideos | 20:33 | |
| *** nbags[m] has joined #timvideos | 20:33 | |
| *** futaris[m] has joined #timvideos | 20:33 | |
| *** shivm28[m] has joined #timvideos | 20:33 | |
| *** micolous[m] has joined #timvideos | 20:33 | |
| *** master1588[m] has joined #timvideos | 20:33 | |
| *** pzieba[m] has joined #timvideos | 20:33 | |
| *** nancy[m] has joined #timvideos | 20:33 | |
| *** synaption[m] has joined #timvideos | 20:33 | |
| *** cr1901_modern1 has joined #timvideos | 21:08 | |
| *** cr1901_modern has quit IRC | 21:10 | |
| *** cr1901_modern1 has quit IRC | 21:14 | |
| *** cr1901_modern has joined #timvideos | 21:14 | |
| *** CarlFK has quit IRC | 22:33 | |
| *** futarisIRCcloud has joined #timvideos | 23:53 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!