Saturday, 2019-05-11

*** tpb has joined #tomu00:00
xobsalexhw ( I just woke up, and I realized that the debug bridge has no trouble with this 2 byte offset, so it must be an issue in the epfifo. That's encouraging!00:18
*** futarisIRCcloud has quit IRC01:26
*** AmosSam has left #tomu01:57
*** AmosSam has joined #tomu01:57
*** emeb has quit IRC03:00
*** futarisIRCcloud has joined #tomu05:07
*** lathiat_ has joined #tomu05:24
*** Kamilion|ZNC has joined #tomu05:24
*** mifune_ has joined #tomu05:25
*** lathiat has quit IRC05:26
*** mifune has quit IRC05:26
*** Kamilion has quit IRC05:26
*** eater has quit IRC05:26
*** eater has joined #tomu05:26
*** Kamilion|ZNC is now known as Kamilion05:26
*** tpb` has joined #tomu05:39
*** tpb has quit IRC05:42
*** tpb` is now known as tpb05:42
*** rohitksingh has joined #tomu06:54
*** futarisIRCcloud has quit IRC07:17
*** rohitksingh has quit IRC07:46
*** rohitksingh has joined #tomu07:58
*** xobs has quit IRC08:33
*** shalzz has quit IRC08:33
*** alexhw[m] has quit IRC08:33
*** synaption[m] has quit IRC08:33
*** ptotter[m] has quit IRC08:33
*** nrossi has quit IRC08:33
*** alexhw[m] has joined #tomu08:44
*** shalzz has joined #tomu08:53
*** ptotter[m] has joined #tomu08:53
*** xobs has joined #tomu08:53
*** nrossi has joined #tomu08:53
*** synaption[m] has joined #tomu08:53
*** rohitksingh_ has joined #tomu09:01
*** rohitksingh has quit IRC09:02
*** AmosSam has left #tomu09:58
*** AmosSam has joined #tomu10:06
*** rohitksingh has joined #tomu11:24
*** rohitksingh_ has quit IRC11:24
*** AmosSam has left #tomu11:33
*** AmosSam has joined #tomu11:41
tntxobs: btw, not sure if you're doing it yet, but reading the flash unique id and using this to get a proper serial number in the USB descriptors would probably be nice for the production fw, so each device identifies itself properly.11:41
*** rohitksingh has quit IRC11:53
*** rohitksingh has joined #tomu11:53
ToxicFrogHrm. Ok, I can't get away with just cycle counting in main, because it looks like interrupt handlers in general are not working, and that means I can't speak USB.12:21
tntToxicFrog: what are you trying to do ?12:22
ToxicFrogtnt: ultimately, act as a USB HID keyboard and send a preprogrammed key sequence when one of the buttons is touched.12:24
ToxicFrogRight now, get the usb-hid-keyboard quickstart example working.12:24
ToxicFrogIf there were a precompiled usb-hid-quickstart.dfu I'd be seriously tempted to disassemble and edit that rather than continue down the rabbit hole of "why does my build environment produce binaries where everything works except interrupt handlers", honestly~12:26
tntToxicFrog: oh, wait are you on the tomu or fomu ?12:26
ToxicFrogI suspect that this is the SUSE version of the Ubuntu bug people linked me to last night, where it picks the wrong build of multilib to link against and ends up linking in code that the Cortex-M0 doesn't understand.12:33
ToxicFrogOk, yeah, this output from readelf looks suspicious:12:40
ToxicFrogTag_ARM_ISA_use: Yes12:40
ToxicFrogDoesn't the M0 only have THUMB?12:40
jerithHey it's a ToxicFrog.12:41
*** rohitksingh has quit IRC12:41
ToxicFrogjerith: hey it's a jerith! Fancy meeting you here.12:44
jerithIt's been a long time since I tinkered with Tomu.12:44
jerithOne day I want to put some Rust code on it.12:45
ToxicFrogjerith: I don't even particularly want a Tomu; I want something with the same form factor but with an AVR on it, since I already have this project working fine on the Digispark (an attiny85 board that plugs into a USB port, about the same size as a small flash drive)12:51
ToxicFrogBut that doesn't seem to exist, and someone at work suggested this instead12:51
ToxicFrogAnd it is pretty cool, but getting a working build environment set up is proving to be much more difficult than it was for the AVR.12:52
jerithTomu's the only programmable thing I've found with this form factor.12:54
jerith(But I haven't looked very hard.)12:54
ToxicFrogI have, and this is the only thing I've found.12:55
ToxicFrogOk. Completely discarding all installed ARM tools and replacing them with the version from appears to have sorted things out.13:03
tntwell there is the fomu but you're not going to have an easier time with that :p13:03
ToxicFrogYeah an FPGA is not actually solving the problem I have :P13:04
ToxicFrogOk, key injection working! Now I just need to modify usb-hid-keyboard to be able to send stuff like ESC and the F-keys13:07
ToxicFrogAnd since I already have working code for that for AVR it shouldn't be too hard to adapt13:07
xobsToxicFrog: thanks for verifying that Suse has the same issue...13:28
xobstnt: for the USB stuff, I'm going to try to get USB working again first. Then I'll add that!13:28
*** rohitksingh has joined #tomu14:03
ToxicFrogOh no, does the sample code in usb-hid-keyboard just busywait for a bit and then hope the host is ready to receive more packets, rather than waiting for the usb-ready interrupt? aaaaaaaaa14:26
*** emeb has joined #tomu14:33
*** emeb has quit IRC17:59
*** rohitksingh has quit IRC19:13
*** rohitksingh has joined #tomu19:25
*** rohitksingh has quit IRC20:10
alexhw[m]xobs: I found an issue in epfifo: the put signal (qualifying received data) is delayed by one clock cycle in order to allow interception by the debug interface, but the corresponding data is not delayed, so they get out of sync. (See the pull request)22:49
alexhw[m]Reducing the fifo depth in the RX pipeline to 2 should be fine by the way as it will neverhold more than one value.22:51

Generated by 2.13.1 by Marius Gedminas - find it at!