--- Log opened ma huhti 01 00:00:34 2024
2024-04-01T00:20:00 -!- jbo_ [~jbo@user/tct] has joined ##stm32
2024-04-01T00:21:20 -!- PaulFertser [paul@paulfertser.info] has quit [Ping timeout: 268 seconds]
2024-04-01T00:22:03 -!- jbo [~jbo@user/tct] has quit [Ping timeout: 255 seconds]
2024-04-01T00:27:43 -!- PaulFertser [paul@paulfertser.info] has joined ##stm32
2024-04-01T00:47:46 < veverak> anybody any tips for ADC+DMA example for stm32h503? I've got working setup on stm32g4: run ADC_Start_DMA(adc,data,length), let it read some values and than stop it. It combines adc + timer to trigger it + dma to store the values in buffer (normal buffer, non-circular)
2024-04-01T00:48:07 < veverak> I am migrating it to stm32h503, and all works fine as long as the DMA is not actually incrementing the read values
2024-04-01T00:48:30 < veverak> (        dma.Init.DestInc               = DMA_DINC_FIXED; on H5)
2024-04-01T00:48:48 < veverak> once I turn it on it breaks and also breaks stack so I can't debug stuff /o\
2024-04-01T00:49:14 < veverak> I am pretty sure I might be setting something wrong, hence why having example for "ADC+DMA" would be nice
2024-04-01T00:51:30 < zyp> where does the buffer live? stack?
2024-04-01T00:51:35 < veverak> The problem I am actually solving: I am PWMing a h-bridge for DC motor, and I would like to sample current shunt +-60 times per PWM period, so what currently happens is that I run ADC+DMA over one period
2024-04-01T00:51:48 < veverak> zyp: yeah, that would be too simple, it's in global variable
2024-04-01T00:52:08 < veverak> to be precise: the buffer is in class encapsulating the mechanism, and that class is instantiated as global
2024-04-01T00:52:16 < veverak> (C++)
2024-04-01T00:53:06 < zyp> haven't used h5, so I can't help you with any known working code, but I wouldn't expect any major differences from other platforms
2024-04-01T00:53:16 < veverak> so, I actually thought that there might be an issue, so I checked:
2024-04-01T00:53:39 < veverak> -         dma.Init.DestDataWidth         = DMA_DEST_DATAWIDTH_HALFWORD;      <<< this should set the data width to 16bits
2024-04-01T00:53:43 < zyp> you've set the size of the buffer right so it doesn't overwrite your entire memory including your stack?
2024-04-01T00:54:07 < veverak> this is the buffer definition: `        alignas( uint32_t ) uint16_t buffer[N];`
2024-04-01T00:54:20 < veverak> this is the call to start: `        alignas( uint32_t ) uint16_t buffer[N];`
2024-04-01T00:54:22 < veverak> waitu
2024-04-01T00:54:31 < veverak> this is the call to start: HAL_ADC_Start_DMA( &h, reinterpret_cast< uint32_t* >( &buffer ), N )
2024-04-01T00:54:38 < veverak> (`h` is proper handle)
2024-04-01T00:54:53 < veverak> so _I guess_ the types should be in order and the size of buffer is correct
2024-04-01T00:55:15 < zyp> the fact that h5 is v8m comes to mind, but this doesn't sound like a trustzone issue and I would guess you don't have trustzone enabled either way
2024-04-01T00:55:46 < veverak> I also checked internals of `ADC_Start_DMA` and it correctly scales the buffer length based on DataWidth as it wants to know bytes
2024-04-01T00:55:55 < veverak> zyp: well, I do have ICACHE, so there is that
2024-04-01T00:56:16 < zyp> I don't expect you're DMAing into the code area
2024-04-01T00:56:20 < veverak> thing is, the DMA peripheral is brand new in H5 relative to G4, the peripheral was redone entirely
2024-04-01T00:56:38 < zyp> ah
2024-04-01T00:57:25 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-01T00:57:26 < zyp> I don't presume they've added scatter-gather or anything like it to it?
2024-04-01T00:57:45 < veverak> not that I am aware of
2024-04-01T00:58:20 < qyx> is it different than h7?
2024-04-01T00:58:56 < veverak> dunno
2024-04-01T01:00:47 < veverak> btw: https://github.com/emsro/servio/blob/main/src/plt/stm32h5/adc.cpp here is the setup routine, maybe somebody can catch something in the setup that seems obviously wrong
2024-04-01T01:01:06 < veverak> note: this works, once I changed `dma.Init.DestInc` to increment it stops
2024-04-01T01:04:05 < zyp> what happens if you leave it on DMA_DINC_FIXED and try using &buffer[1]?
2024-04-01T01:04:16 < zyp> (i.e. could it be a memory alignment issue?)
2024-04-01T01:08:34 < zyp> hmm, what should I get from mouser?
2024-04-01T01:09:25 < zyp> I'm getting a STM32H7S78-DK, just wondering if there's anything else fun that I should add to the order
2024-04-01T01:15:20 < veverak> hmmm, I will try
2024-04-01T01:17:48 < veverak> what the hell
2024-04-01T01:18:02 < veverak> DINC_FIXED + HAL_ADC_Start_DMA( &h, reinterpret_cast< uint32_t* >( &buffer[0] ), N - 1 ) 
2024-04-01T01:18:06 < veverak> ^^ this does not work
2024-04-01T01:20:02 < veverak> yeah, I got a scent now
2024-04-01T01:20:52 < veverak> so N-1 does not have effect
2024-04-01T01:20:55 < veverak> basically, this:                 if ( HAL_ADC_Start_DMA( &h, reinterpret_cast< uint32_t* >( &buffer[0] ), N ) !=
2024-04-01T01:20:59 < veverak> causes hardfault reliably
2024-04-01T01:28:26 < veverak> LOL
2024-04-01T01:28:30 < veverak> zyp: fixed it
2024-04-01T01:28:37 < zyp> :)
2024-04-01T01:28:53 < veverak> so, what happens is that ADC works in _readings_ but DMA works in _bytes_
2024-04-01T01:29:10 < veverak> used = N - __HAL_DMA_GET_COUNTER( h.DMA_Handle ) / 2;
2024-04-01T01:29:35 -!- drzacek [~quassel@2a01:3d8:32a:fa00:f1a1:62d0:a6f2:c60d] has quit [Quit: https://quassel-irc.org - Czatuj komfortowo. Wszędzie.]
2024-04-01T01:29:36 < veverak> ^^ this hsould get me number of read values, `N` is size of buffer for _readings_, but GET_COUNTER returns how much bytes DMA used
2024-04-01T01:29:43 < veverak>  /2 was not there previously
2024-04-01T01:50:30 < veverak> 83,97,112,128,144,160,177,193,209,226,241,257,272,288,303,318,332,346,360,374,388,401,414,427,439,452,464,476,488,500,435,357,295,246,208,177,153,133,117,105,95,87,81,76,72,69,66,64,63,61,60,60,59,58,58,58,57,57,57,57,57,56,0,0
2024-04-01T01:50:34 < veverak> yaaay, data looks good
2024-04-01T02:03:37 < ColdKeyboard> Is there an easy way to test how much ram/flash a custom library would use other than setting up a full blown project?
2024-04-01T02:04:05 < ColdKeyboard> For example if you have a function pointer and then define a const array of function pointers
2024-04-01T02:05:58 < zyp> you can check the size of the symbols in a compiled library
2024-04-01T02:07:00 < zyp> but in practice it'll depend on which symbols will be used and which symbols will be thrown out
2024-04-01T02:07:16 < zyp> and for ram it only counts globals, not stack
2024-04-01T02:07:37 < zyp> and not context objects that you have to allocate and pass it yourself
2024-04-01T02:08:11 < zyp> so yes, you can estimate it, but you won't know the exact answer until you actually use it
2024-04-01T02:08:41 < zyp> how much the estimate and the actual numbers differ depends on the library and how much effort you put into estimating accurately
2024-04-01T02:09:40 < zyp> generally I wouldn't bother spending time on estimating unless you have a very good reason to
2024-04-01T02:10:49 < ColdKeyboard> Maybe you have a better suggestion for what I'm trying to achieve...
2024-04-01T02:11:04 < ColdKeyboard> Basically I will have a command that is 0x01->0xFF
2024-04-01T02:11:30 < ColdKeyboard> Instead of having a dumpster of if (cmd == ...) else if (cmd == ...) etc.
2024-04-01T02:11:52 < ColdKeyboard> I was thinking of having an array where at position [0x01] I have a function pointer that handles that command
2024-04-01T02:12:04 < ColdKeyboard> at [0x02] another function pointer that handles command 0x02 and so on
2024-04-01T02:12:31 < ColdKeyboard> So the check becomes just if fptr[cmd] != NULL and I should be good to go
2024-04-01T02:12:59 < ColdKeyboard> but then I have to have an array of 254 function pointers...
2024-04-01T02:13:07 < ColdKeyboard> Maybe a hash table or something else is better solution here?
2024-04-01T02:13:26 < zyp> so how many commands will you actually have?
2024-04-01T02:15:32 < zyp> there's two obvious reasonable ways of doing a dispatcher like you describe here, a function pointer table is one, a bunch of if/else statements are not
2024-04-01T02:15:40 < zyp> the other is a switch/case structure
2024-04-01T02:16:22 < zyp> depending on how dense/sparse the table is, they might result in more or less the same code
2024-04-01T02:17:40 < ColdKeyboard> I mean the worst case is I use all of the commands. More realistic scenario is I use some range, like from 1-20, or 20-50 etc
2024-04-01T02:18:07 < ColdKeyboard> I can enforce that for each device commands are within range. That's why I thought some hashing/shifting function + dispatcher 
2024-04-01T02:18:38 < ColdKeyboard> I'm not sure how fast the switch statement would be? Compared to if/else
2024-04-01T02:19:52 < ColdKeyboard> Or for example the absolute worst case is that I have only few commands like 1, 20, 22, 50
2024-04-01T02:24:26 < zyp> well, here's a comparison: https://godbolt.org/z/TKa8KaY9M
2024-04-01T02:25:49 < zyp> they pretty much do the same thing
2024-04-01T02:27:16 < zyp> https://godbolt.org/z/Y5aMdYTTr <- and here's with a sparse switch thrown into the mix
2024-04-01T02:29:53 < zyp> the sparse switch is effectively generating if/else structures because that saves flash, the dense switch uses tbb to effectively make a little table that each points to a branch instruction
2024-04-01T02:35:51 < zyp> so what this effectively means is that if you use an array of function pointers, it's a fixed time lookup no matter how many entries you've got or how sparse they are, but if the table is very sparse you'll spend a bunch of flash on empty entries
2024-04-01T02:36:39 < zyp> whereas if you use a switch, the compiler will generate compact code in any case, but it'll spend more cycles dealing with sparseness when you've got any
2024-04-01T02:36:58 < zyp> and for anything in between the extremes, it doesn't really matter
2024-04-01T02:37:31 < zyp> so pick the style you prefer, I guess
2024-04-01T02:37:40 < ColdKeyboard> Good point. Also thanks for the example. I forgot about the compiler explorer!
2024-04-01T02:38:37 < ColdKeyboard> I'm still not sure but maybe the switch statement is the easiest unless I know I'll have a sequential commands in which case I can do an array of function pointers
2024-04-01T02:38:54 < ColdKeyboard> Although switch might be easier to read...
2024-04-01T02:39:28 < zyp> switch is easier to read since it establishes the cmd -> function mapping more explicitly than the index of an array
2024-04-01T02:39:55 < zyp> except if you write C you could use designated array initializers
2024-04-01T03:13:03 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Remote host closed the connection]
2024-04-01T03:49:05 < englishman> https://libera.cat/
2024-04-01T04:21:22 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-01T04:42:57 -!- catphish [~quassel@user/catphish] has quit [Ping timeout: 256 seconds]
2024-04-01T06:30:27 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-01T06:59:05 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds]
2024-04-01T07:21:49 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-01T09:41:23 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-01T11:18:43 < qyx> purr meowy
2024-04-01T11:33:11 < qyx> experiencing my work desk mess runaway again
2024-04-01T11:43:32 < Steffanx> Someone played a good joke on you?
2024-04-01T11:45:51 -!- jtj [~jtj@212.66.207.170] has quit [Ping timeout: 260 seconds]
2024-04-01T11:52:17 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-01T11:57:21 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-01T12:00:48 -!- ilgrim [~ilgrim@xinu.me] has quit [Quit: WeeChat 4.1.2]
2024-04-01T12:31:38 < qyx> re github 2fa, I assume I can configure multiple options and use one of them?
2024-04-01T12:36:54 < qyx> and a subquestion, which oss android auth app is okish
2024-04-01T12:37:31 < qyx> https://f-droid.org/en/packages/com.beemdevelopment.aegis/ ?
2024-04-01T13:12:03 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-01T14:27:05 -!- ilgrim [~ilgrim@xinu.me] has joined ##stm32
2024-04-01T14:52:49 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 255 seconds]
2024-04-01T15:04:12 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-01T15:10:52 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 256 seconds]
2024-04-01T15:13:30 -!- drzacek [~quassel@2a01:3d8:425:d400:52b4:970f:baad:f68b] has joined ##stm32
2024-04-01T15:14:44 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-01T15:42:08 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-01T15:42:42 < Laurenceb_> I've taken a long hard look at my behaviour on Twitter and decided that I need therapy to deprogram myself from a number of extremist misogynistic, discriminatory and transphobic behaviours. I feel like this is a moment of clarity and I need to seize it and save myself before I fall deeper into extremism and hate.
2024-04-01T15:42:43 < Laurenceb_> Please get in touch if you can recommend a suitable therapist to give me the help I need as soon as possible.
2024-04-01T15:43:09 < Laurenceb_> oh wait
2024-04-01T15:43:19  * Laurenceb_ remembers he isn't in Scotland
2024-04-01T15:50:42 < specing> top kek
2024-04-01T16:04:19 < Steffanx> Have a look at the date Laurenceb_ . Assuming it was posted today or yesterday 
2024-04-01T16:05:05 < Steffanx> But you're always here to fool us, so what's new?
2024-04-01T16:20:08 -!- jbo_ is now known as jbo
2024-04-01T16:21:08 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-01T16:30:24 -!- jbo_ [~jbo@user/tct] has joined ##stm32
2024-04-01T16:31:46 -!- jbo [~jbo@user/tct] has quit [Ping timeout: 264 seconds]
2024-04-01T17:03:00 < machinehum> https://pastebin.com/raw/GEfKf76f
2024-04-01T17:03:04 < machinehum> That's cool
2024-04-01T17:11:11 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-01T17:15:08 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-01T18:14:50 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-01T19:10:00 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-01T19:39:26 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds]
2024-04-01T19:49:21 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 255 seconds]
2024-04-01T20:16:53 -!- boB_K7IQ [~boB_K7IQ@c-67-161-97-32.hsd1.wa.comcast.net] has joined ##stm32
2024-04-01T20:21:17 -!- boB_K7IQ [~boB_K7IQ@c-67-161-97-32.hsd1.wa.comcast.net] has quit [Read error: Connection reset by peer]
2024-04-01T20:27:17 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 240 seconds]
2024-04-01T20:33:19 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-01T20:34:45 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-01T20:36:47 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-01T20:43:27 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 256 seconds]
2024-04-01T20:51:54 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-01T20:54:03 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-01T21:08:51 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-01T21:10:34 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-01T21:18:47 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-01T21:36:24 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-01T21:57:12 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-01T21:58:14 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-01T21:59:05 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 252 seconds]
2024-04-01T22:03:05 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-01T22:07:34 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-01T22:08:16 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-01T22:10:25 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-01T22:12:45 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-01T22:14:51 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-01T22:18:55 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-01T22:23:50 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-01T22:27:42 -!- drzacek [~quassel@2a01:3d8:425:d400:52b4:970f:baad:f68b] has quit [Remote host closed the connection]
2024-04-01T22:39:45 -!- fenugrec_web [~fenugrec_@216.73.162.216] has joined ##stm32
2024-04-01T22:49:37 -!- fenugrec_web [~fenugrec_@216.73.162.216] has quit [Ping timeout: 250 seconds]
2024-04-01T23:04:15 -!- fenugrec_web [~fenugrec_@216.73.162.238] has joined ##stm32
2024-04-01T23:05:59 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-01T23:07:37 -!- BrainDamage is now known as NyanDawwmwage
2024-04-01T23:08:06 < qyx> lol
2024-04-01T23:14:43 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-01T23:30:48 -!- catphish [~quassel@user/catphish] has joined ##stm32
2024-04-01T23:33:02 < antto> Steffanx, i guess you meant those fancy 4-pin RGB LEDs which have digital input and output and can be chained in long strings?
2024-04-01T23:39:33 < Steffanx> Yes
2024-04-01T23:39:41 < Steffanx> Or 6 pin
2024-04-01T23:40:55 < qyx> I attempted this
2024-04-01T23:40:56 < qyx> 	const uint32_t invalid_instruction = 0xe0000000UL;
2024-04-01T23:40:56 < qyx> 	invalid_function_t invalid = (invalid_function_t)&invalid_instruction;
2024-04-01T23:41:23 < qyx> and I got INVSTATE usage fault
2024-04-01T23:41:39 < qyx> could anyone enlighten me?
2024-04-01T23:43:09 < zyp> you forgot to set LSB
2024-04-01T23:43:17 < zyp> thumb addresses always have LSB set
2024-04-01T23:43:37 < zyp> so you get INVSTATE because that's happens when you ask a cortex-m to change to arm mode
2024-04-01T23:43:54 < qyx> oh!
2024-04-01T23:44:16 < antto> Steffanx, on one hand that would eliminate the shift registers and/or drivers, cool, the question is where do you buy those LEDs from then
2024-04-01T23:44:52 < antto> i'm not too familiar, i looked into it, found a few different brands which seem like they try to mimic each other maybe
2024-04-01T23:45:28 < zyp> I added support for formatting spans:
2024-04-01T23:45:29 < zyp>     std::array<uint8_t, 6> mac_addr = {0x11, 0x22, 0x33, 0x44, 0x55, 0x66};
2024-04-01T23:45:29 < zyp>     logger.log<"MAC address: {:02x|:}", loc()>(std::span(mac_addr));
2024-04-01T23:45:40 < zyp> 22:42:25 example.cpp:73 MAC address: 11:22:33:44:55:66
2024-04-01T23:45:59 < zyp> part before | is element format, part after is separator
2024-04-01T23:46:02 < antto> i need LEDs that fit within 4 mm, because that's the "line size" of my 7seg segments, i can't have anything wider than that, including any solder joints
2024-04-01T23:46:53 < zyp> antto, you want adressable LEDs?
2024-04-01T23:47:05 < zyp> I use SK6805-EC15, they're like 1.5x1.5mm
2024-04-01T23:47:31 < antto> zyp, more like, i'm making this clock: https://i.imgur.com/VCGqcHb.png
2024-04-01T23:47:45 < antto> https://i.imgur.com/I0akyJs.png
2024-04-01T23:49:04 < antto> initially i was gonna put 14 74HC595 shift registers and a pile of resistors (one for each LED), then i found CAT4016 which is like a 16bit shift register for driving LEDs (current regulation), so i went for 7 of those and no resistors
2024-04-01T23:49:19 < antto> the price of that scheme is comparable
2024-04-01T23:49:58 < antto> recently the CAT4016 (in SOIC package) vanished, so i redid the layout with the QFN version of it
2024-04-01T23:50:51 < antto> i'm putting an ambient light sensor because i wanna adjust the LED brightness sanely so you don't go blind at night
2024-04-01T23:50:53 < sauce> are those segment frames off the shelf or is this a bespoke design for printing/laser
2024-04-01T23:51:15 < antto> sauce, the white shapes will have to be fabbed ;P~
2024-04-01T23:51:29 < sauce> cool
2024-04-01T23:51:43 < antto> their purpose is to block the light of each segment from bleeding into the neightbour segments
2024-04-01T23:52:12 < qyx> zyp: thanks now I get UNDEFINSTR correctly
2024-04-01T23:53:36 < zyp> antto, going by jlcpcb parts prices, SK6805-EC15 should be slightly cheaper than CAT4016
2024-04-01T23:54:07 < antto> where do i buy the "fancy" LEDs from tho
2024-04-01T23:54:21 < antto> don't tell me "adafruit, in packs of 10" pls
2024-04-01T23:54:52 < zyp> like I said, I checked jlcpcb parts prices, which means they've got them in stock
2024-04-01T23:55:14 < antto> okay
2024-04-01T23:55:14 < zyp> so assuming you're getting it assembled there, you just tell them to use them :p
2024-04-01T23:55:22 < antto> no
2024-04-01T23:55:47 < zyp> so who's assembling them?
2024-04-01T23:55:58 < antto> uh... moi
2024-04-01T23:56:08 < antto> who else assembles my own boards?
2024-04-01T23:56:31 < zyp> doesn't seem very worthwhile, but ok
2024-04-01T23:56:50 < antto> i need 2 or 3 of these for $home
2024-04-01T23:57:02 < antto> if all the sh*t works according to plan ;P~
2024-04-01T23:57:17 < zyp> in that case you might want to reconsider, they're small and fiddly, maybe go with something simpler to solder :p
2024-04-01T23:57:41 < zyp> I've reworked some, they're annoying to do by hand
2024-04-01T23:57:48 < antto> the first "fancy" LED i found was from a vendor i'm familiar with: https://www.optosupply.com/uppic/2023421763886.pdf
2024-04-01T23:58:01 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-01T23:58:07 < zyp> anyway, you can also just buy a reel from LCSC, been there done that too
2024-04-01T23:58:46 < antto> the 2.8mm width will fit within my 4mm segments
2024-04-01T23:58:52 < antto> with a tiny bit of air gap
2024-04-01T23:59:27 < Steffanx> I think mouser and co also sell similar LEDs. I think even wurth makes(or rebrands) then nowadays
2024-04-01T23:59:44 < antto> but... $$$wurth$$$
2024-04-01T23:59:45 < zyp> maybe get these: https://www.lcsc.com/product-detail/Light-Emitting-Diodes-LED_OPSCO-Optoelectronics-SK6805D-EC4228_C5440837.html
--- Day changed ti huhti 02 2024
2024-04-02T00:00:00 < Steffanx> Especially ws2812/13/whatever is hot these days
2024-04-02T00:00:15 < Steffanx> Protocol might vary though 
2024-04-02T00:00:18 < qyx> watch the Icc
2024-04-02T00:00:18 < zyp> there's like a ton of different packages with a SK6805 controller
2024-04-02T00:00:36 < qyx> some ws are nearly 1 mA when idle
2024-04-02T00:01:23 < Steffanx> Yeah that part is crazy qyx
2024-04-02T00:01:27 < qyx> imagine your clock drawing half a watt displaying nothing
2024-04-02T00:02:08 < Steffanx> Electricity is free in anttoland
2024-04-02T00:02:29 < antto> not when we gave away 1 or 2 nuclear reactors to .ua
2024-04-02T00:04:30 < antto> hm, these don't look very bright by specs
2024-04-02T00:07:37 -!- fenugrec_web [~fenugrec_@216.73.162.238] has quit [Ping timeout: 250 seconds]
2024-04-02T00:09:00 < Steffanx> qyx you want HD108, it claims a "Static power consumption" of 0 mA. 
2024-04-02T00:10:21 < Steffanx> and uses are more spi like protocol
2024-04-02T00:17:02 < qyx> Using Cube/CubeMX means, that you constrain yourself to the "usual" usage cases envisaged by Cube/CubeMX authors. If you want to use any of the gazillion possible options and their combinations, ditch Cube and go for adult programming.
2024-04-02T00:17:06 < qyx> haha
2024-04-02T00:18:41 < antto> zyp, these seem brighter: https://www.lcsc.com/product-detail/Light-Emitting-Diodes-LED_XINGLIGHT-XL-3528RGBW-WS2812B_C2890364.html
2024-04-02T00:19:26 < Steffanx> Yeah there are a lot of variants, but different manufacturers
2024-04-02T00:19:42 < zyp> I've never had an issue with the LEDs not being bright enough, I typically run them at 8/255 for my use :p
2024-04-02T00:19:51 < zyp> but I've never built a clock so YMMV
2024-04-02T00:20:34 < Steffanx> *bt
2024-04-02T00:20:36 < Steffanx> *by
2024-04-02T00:20:37 < Steffanx> damnit
2024-04-02T00:20:37 < antto> ideally i would've blindly trusted the optosupply ones, but they aren't available anywhere, no clue about price too
2024-04-02T00:20:50 < antto> but they're very bright by specs
2024-04-02T00:21:31 < Steffanx> So buy a few and test them 
2024-04-02T00:21:45 < antto> nah
2024-04-02T00:21:54 < Steffanx> So they actually claim "No data input, zero power consumption when the light is off."  qyx 
2024-04-02T00:22:06 < qyx> sure
2024-04-02T00:22:18 < Steffanx> Yeah im going to buy a few
2024-04-02T00:22:45 < qyx> lmk if true then pls
2024-04-02T00:23:06 < Steffanx> They are rather old, so someone must have tried it by now.. 2019
2024-04-02T00:23:47 < Steffanx> oh 2020-9 for this one
2024-04-02T00:24:19 < antto> i need bright because i'll be putting a "decorative" PCB maybe 20mm away from this board, with the 7seg shapes exposed as nekkid FR4 which will act as diffuser
2024-04-02T00:24:29 < antto> a 1mm FR4 board that is
2024-04-02T00:24:33 < antto> with black soldermask
2024-04-02T00:24:56 < antto> (but i won't rely entirely on the soldermask to block light so i'll put copper too)
2024-04-02T00:25:39 < antto> did i mention this will be the most expensive clock? ;P~
2024-04-02T00:26:23 < Steffanx> Just have jlc3d print some fancy things
2024-04-02T00:28:13 < antto> i could maybe poke comet.bg to ask them to poke optosupply about their LEDs
2024-04-02T00:45:44 < antto> "Static power consumption 0.65mA typical" hmmmm...
2024-04-02T00:57:06 < Steffanx> Yeah .
2024-04-02T01:18:58 < nomorekaki> watchung a movie steff
2024-04-02T01:19:41 < nomorekaki> "Dual" filmed in funland
2024-04-02T01:29:23 < Steffanx> 5.8 ..
2024-04-02T01:30:00 < nomorekaki> so weird it's like hollywood movie cast but movie style is rather similar to minimalistic style of a famous finnish director
2024-04-02T01:30:39 < Steffanx> You mean people never come close except when in the sauna?
2024-04-02T01:33:42 < Steffanx> Are there even scenes in the sauna, nomorekaki ?
2024-04-02T01:33:57 < nomorekaki> I guess there was
2024-04-02T01:36:57 < nomorekaki> not saunaing but
2024-04-02T01:38:03 < nomorekaki> they got some grant from funland(film promotion fund) to film this movie so there must have been conditions..
2024-04-02T01:38:29 < nomorekaki> sauna visible for 4seconds minimum
2024-04-02T01:41:22 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T01:43:02 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T01:46:13 < qyx> zyp: what testing framework do you use (py)?
2024-04-02T01:46:19 < qyx> for HIL
2024-04-02T01:47:04 < qyx> basically flash a test fw, run tests using serial or gdb, get results, flash again a different suite,etc, etc
2024-04-02T01:47:16 < qyx> and generate some report
2024-04-02T01:47:56 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-02T01:52:51 < qyx> maybe pytest
2024-04-02T01:53:48 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-02T01:54:52 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T01:55:44 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T01:58:34 -!- NyanDawwmwage is now known as BrainDamage
2024-04-02T02:17:44 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T02:18:45 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T02:24:25 < ventYl> qyx: you basically need to create some test driver that will wrap the upload, execution and evaluation of test program on the target. then you can use whatever "test framework" you like.
2024-04-02T02:25:24 < ventYl> as long as framework can generate xUnit reports, you can represent them in variety of tools
2024-04-02T02:47:38 < nomorekaki> Steffanx: ngl that was the longest 1.5hours for a while
2024-04-02T03:25:18 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-02T04:11:53 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T04:12:43 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T07:46:32 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T07:47:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T08:35:12 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-02T08:52:51 < qyx> rip jbo_ 
2024-04-02T08:54:05 < qyx> https://www.theguardian.com/world/2024/apr/01/avalanche-at-zermatt-ski-resort-in-switzerland-kills-a-number-of-people
2024-04-02T09:18:21 < zyp> qyx, haven't done HIL unit tests
2024-04-02T09:18:57 < zyp> I figure part of the point of unit tests is that they don't have to be HIL
2024-04-02T09:22:40 < jpa-> i used to run nanopb unit tests on real hardware by having custom wrapper & semihosting, but nowadays i don't bother
2024-04-02T09:23:09 < jpa-> platformio test runner kind-of works, but i have found it more annoying than useful
2024-04-02T09:23:48 < zyp> if renode was less annoying to run, I would have written some unit tests that ran in renode the other day
2024-04-02T09:25:22 < zyp> but instead I ended up with some test cases that runs on the host, and some test cases that's not ran at all, just checking that they're building without error/warning
2024-04-02T09:25:31 < zyp> https://github.com/zyp/smolt/tree/main/tests/cpp
2024-04-02T09:27:35 < zyp> for the latter cases I'm gonna add validation that the resulting elfs contains the expected metadata
2024-04-02T09:28:54 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-02T09:58:54 -!- yoyofreeman [~yoyofreem@38.180.62.249] has joined ##stm32
2024-04-02T10:16:44 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has joined ##stm32
2024-04-02T10:22:24 < qyx> zyp: I am not saying unit tests
2024-04-02T10:23:25 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-02T10:27:52 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-02T10:44:49 < zyp> «flash again a different suite, etc» sounds like unit tests
2024-04-02T10:46:17 < zyp> the other kind of tests I've been doing are production tests, which by definition *have* to run on hardware
2024-04-02T10:49:30 < qyx> for example I am not dealing with simulating and catching various faults/exceptions, which are not unit tests and need to be run on the hardware itself, but they also need to be observed from the outside
2024-04-02T10:49:35 < qyx> *I am now
2024-04-02T10:49:59 < qyx> and of course cannot be run on the host
2024-04-02T10:50:15 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-02T10:53:35 < qyx> and I definitely will be running unit tests on the hardware itself, I already ran into too many problems with various crypto and other software running differently on the target
2024-04-02T10:58:25 < zyp> fair enough
2024-04-02T10:58:47 < zyp> still sounds like unit tests to me
2024-04-02T11:00:32 < zyp> and pytest should probably work well, with some reasonable fixture setup/reuse
2024-04-02T11:02:19 < zyp> e.g. you could have a fixture that takes care of controlling the target, and then if you've got multiple test cases sharing a firmware you could have another fixture that takes care of preparing that
2024-04-02T11:35:15 -!- splud [~noneya.bi@user/splud] has quit [Ping timeout: 255 seconds]
2024-04-02T11:48:57 -!- splud [~noneya.bi@user/splud] has joined ##stm32
2024-04-02T11:58:40 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-02T12:04:58 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has quit [Ping timeout: 255 seconds]
2024-04-02T12:20:13 -!- Laurenceb_ [~Laurenceb@cust226-dsl93-89-135.idnet.net] has joined ##stm32
2024-04-02T12:42:01 -!- Laurenceb_ [~Laurenceb@cust226-dsl93-89-135.idnet.net] has quit [Quit: Client closed]
2024-04-02T12:58:35 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has joined ##stm32
2024-04-02T13:43:24 -!- yoyofreeman [~yoyofreem@38.180.62.249] has quit [Ping timeout: 240 seconds]
2024-04-02T13:49:56 -!- yoyofreeman [~yoyofreem@38.180.62.249] has joined ##stm32
2024-04-02T13:54:40 < qyx> any idea why this thing sometimes returns 0? https://paste.jvnv.net/view/sdEqS
2024-04-02T13:55:16 < qyx> any barriers missing?
2024-04-02T13:55:48 < jpa-> which chip?
2024-04-02T13:59:44 < qyx> g491
2024-04-02T13:59:59 < zyp> qyx, that code doesn't make much sense
2024-04-02T14:00:10 < qyx> why?
2024-04-02T14:00:18 < jpa-> to me it seems like typical code to measure ratio between two clocks
2024-04-02T14:00:31 < qyx> yes, tim16 timebase is pclk which is known
2024-04-02T14:00:47 < qyx> I select LSE as TI1 to measure it is within bounds
2024-04-02T14:00:58 < zyp> what doesn't make sense is that you're not clearing CC1IF between line 16 and 19
2024-04-02T14:01:02 < qyx> then I enable CC1 input and wait for the first CC1 event to get the counter value
2024-04-02T14:01:07 < jpa-> CC1IF is cleared by CCR1 read
2024-04-02T14:01:08 < qyx> then I wait for another one
2024-04-02T14:01:15 < qyx> what jpa- says
2024-04-02T14:01:22 < qyx> I also tried explicitly, doesn't help
2024-04-02T14:01:42 < jpa-> so i guess the real question is whether some barrier is needed between lines 17 and 19
2024-04-02T14:01:43 < zyp> jpa-, wait, is that new?
2024-04-02T14:02:16 < jpa-> zyp: i don't think so, STM32F103 already has it
2024-04-02T14:02:24 -!- yoyofreeman [~yoyofreem@38.180.62.249] has quit [Ping timeout: 240 seconds]
2024-04-02T14:02:34 < jpa-> qyx: so you tried manual clear between lines 17 and 19?
2024-04-02T14:02:35 < qyx> if that helps, all ahb and apb busses are at 64 MHz, no prescalers
2024-04-02T14:02:42 < qyx> yep
2024-04-02T14:02:51 < jpa-> ok, then a barrier there is unlikely to help either
2024-04-02T14:03:06 < jpa-> and i think two reads to same peripheral should be strict ordered in any case
2024-04-02T14:03:46 < qyx> This flag is set by hardware. It is cleared by software (input capture or output compare
2024-04-02T14:03:49 < qyx> mode) or by reading the TIMx_CCR1 register (input capture mode only).
2024-04-02T14:03:56 < qyx> I wonder if it is cleared by writing 1 as usual
2024-04-02T14:04:01 < qyx> because I didn't try writing 0
2024-04-02T14:04:05 < qyx> and they are not mentioning it
2024-04-02T14:04:33 < jpa-> it is cleared by writing 0
2024-04-02T14:04:41 < qyx> how so
2024-04-02T14:04:42 < jpa-> the register bit diagram shows "rc_w0"
2024-04-02T14:05:14 < qyx> 	TIM_SR(TIM16) &= ~TIM_SR_CC1IF;
2024-04-02T14:05:19 < qyx> doesn't help either
2024-04-02T14:06:01 < jpa-> what is the expected LSE frequency? 32kHz?
2024-04-02T14:06:06 < qyx> yes
2024-04-02T14:06:38 < jpa-> i would put GPIO toggles around the while loops and scope it
2024-04-02T14:06:40 < qyx> when it works about 80% of time, ot measures correctly
2024-04-02T14:06:54 -!- yoyofreeman [~yoyofreem@38.180.62.249] has joined ##stm32
2024-04-02T14:07:51 < jpa-> and you could check whether CC1IF gets really cleared
2024-04-02T14:09:11 < qyx> also the first measure after enabling the timer shows about 16 kHz, but that is caused by a stale CC1IF (set between the timer is enabled and the function is entered)
2024-04-02T14:09:21 < qyx> which is solved by clearing it before the first while loop
2024-04-02T14:09:33 < zyp> wraparound doesn't look like it's sanely handled
2024-04-02T14:09:42 < zyp> i.e. when cc2 < cc1
2024-04-02T14:09:56 < qyx> both are uint16_t
2024-04-02T14:10:10 < zyp> yeah, but the subtraction promotes to int32_t
2024-04-02T14:10:29 < zyp> before the cast to uint32_t
2024-04-02T14:10:41 < qyx> really?
2024-04-02T14:11:21 < jpa-> yes
2024-04-02T14:12:08 < karlp> unsigned types were a mistake.
2024-04-02T14:12:16 < qyx> no
2024-04-02T14:12:54 < karlp> yes. treat everythign as signed, no surprises ever.
2024-04-02T14:14:10 < qyx> I though integer promotion only happens when different sizes and/or signed/unsigned are combined
2024-04-02T14:14:23 < qyx> I don't see a reason subtracting two uints should yield anything signed
2024-04-02T14:14:40 < jpa-> everything smaller gets promoted to int
2024-04-02T14:16:29 < qyx> I really hate computers
2024-04-02T14:17:35 < jpa-> that's fine, they are not fond of you either
2024-04-02T14:18:07 < qyx> k setting 	TIM_CNT(TIM16) = 0;
2024-04-02T14:18:09 < qyx> beforehand
2024-04-02T14:18:20 < qyx> instead of thinking how to handle overflows
2024-04-02T14:18:32 < zyp> just change the uint32_t in the cast to uint16_t
2024-04-02T14:19:11 < qyx> works too, thanks
2024-04-02T14:19:35  * qyx back to learn C in 21 days
2024-04-02T14:44:29 -!- Livio [~livio@user/livio] has quit [Ping timeout: 272 seconds]
2024-04-02T15:44:28 -!- BrainDamage_ [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-02T15:44:38 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Remote host closed the connection]
2024-04-02T15:49:20 -!- BrainDamage_ is now known as BrainDamage
2024-04-02T15:50:16 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-02T15:55:37 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-02T16:02:16 < karlp> I fucking love arm sometimes.
2024-04-02T16:02:19 < karlp> "The NVIC_IPR0-NVIC_IPR7 registers provide an 8-bit priority field for each interrupt. These
2024-04-02T16:02:21 < karlp> registers are only word-accessible.
2024-04-02T16:02:42 < karlp> so yeah, x /8x 0xE000E400 shows me some non-zero fields.
2024-04-02T16:03:02 < karlp> oh, no problem, x /32b 0xe000e400 instead, so i can count them out easily... => all zeros.
2024-04-02T16:03:30 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-02T16:17:17 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-02T16:35:50 < karlp> so.. freertos on cortex m0 just.. doesn't have any of this priority bit shit?
2024-04-02T16:35:58 < karlp> non of the examples have anything.
2024-04-02T16:36:10 < karlp> does m0 not have cpsid masking or something?
2024-04-02T16:49:44 < zyp> I don't think m0 does preemption
2024-04-02T16:50:20 < zyp> just interrupts on/off
2024-04-02T16:57:05 < karlp> yeah, primask is just a single bit.
2024-04-02T16:57:09 < karlp> man I hate this platform.
2024-04-02T16:57:24 < karlp> hurhur it's low power, hurhur
2024-04-02T17:10:51 < karlp> hrm, no it has priority based preemption, you can just cant' set a threshold to mask off, so none of freertos's configMAX_SYSCALL_INTERRUPT_PRIORITY stuff matters.
2024-04-02T17:11:04 < karlp> because you can only disable all interrupts or no-interrupts.
2024-04-02T17:33:57 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-02T17:50:41 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-02T17:50:56 < nomorekaki> Steffanx: musics https://www.youtube.com/watch?v=empLZMVWIP8
2024-04-02T17:55:20 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 260 seconds]
2024-04-02T18:14:03 < qyx> karlp: whuch hurhur low power, something specific?
2024-04-02T18:14:29 < zyp> m0
2024-04-02T18:15:12 < jpa-> m0, the core that is awesome if you come from AVR and that is terrible if you come from cortex-m3
2024-04-02T18:20:42 -!- Kerr [~quassel@174.31.48.154] has quit [Ping timeout: 255 seconds]
2024-04-02T18:23:08 < nomorekaki> would you use any peripherals for implementing 1wire slave?
2024-04-02T18:23:36 < nomorekaki> timer capture and timer match?
2024-04-02T18:39:34 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has quit [Ping timeout: 264 seconds]
2024-04-02T18:44:22 < nomorekaki> nice new attiny eventsys allows me to run timer for the bit being received from falling edge of RX
2024-04-02T18:44:36 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has joined ##stm32
2024-04-02T18:45:02 < nomorekaki> let's see if I could somehow make it also activate tx in case of tx bit
2024-04-02T18:45:36 < nomorekaki> just have interrupts when things are completed
2024-04-02T18:54:42 < nomorekaki> waveform out is not event source hmm
2024-04-02T18:59:13 < nomorekaki> okay I set TX timer to max and waveform match register per desired bit
2024-04-02T18:59:50 < nomorekaki> enable event from RX pin to run the timer if tx is desired
2024-04-02T19:00:50 < nomorekaki> timer stops running automatically as RX goes back high
2024-04-02T19:02:26 < nomorekaki> another timer has "timeout mode" literally made to count pulse width
2024-04-02T19:05:16 < nomorekaki> but there is pulse width measurement mode too looking rather similar
2024-04-02T19:10:13 < nomorekaki> another maybe has timeout and another doesnt
2024-04-02T19:12:08 < qyx> nomorekaki: nomoretimer, uart
2024-04-02T19:12:54 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 255 seconds]
2024-04-02T19:16:58 < nomorekaki> ye
2024-04-02T19:21:28 < nomorekaki> that would be the standard way i guess
2024-04-02T19:26:38 < nomorekaki> oh it has even single pin mode for that
2024-04-02T19:27:46 < nomorekaki> txd becomes open collector
2024-04-02T19:29:00 -!- flom84 [~flom84@user/flom84] has joined ##stm32
2024-04-02T19:29:09 -!- flom84 [~flom84@user/flom84] has quit [Remote host closed the connection]
2024-04-02T19:35:31 < nomorekaki> slave needs to synchronize to master start pulse qyx is that a problem?
2024-04-02T19:39:40 -!- boB_K7IQ [~boB_K7IQ@152.44.147.180] has joined ##stm32
2024-04-02T19:40:01 < nomorekaki> it needs to switch from receiving to transmitting during the start bit
2024-04-02T19:40:58 < nomorekaki> 1us is enough time to do it with interrupt though
2024-04-02T19:43:28 < nomorekaki> naked interrupt and 5 or 6 instructions from running state
2024-04-02T19:47:52 < qyx> nomorekaki: just send 0x80 or so and rx one byte
2024-04-02T19:48:25 < qyx> if it is 0x80, nothing happened, if 0xc0 or 0xe0 or 0xf0 or so, the slave extended the low period
2024-04-02T19:48:34 < nomorekaki> this is the slave
2024-04-02T19:48:45 < qyx> hm
2024-04-02T19:49:01 < qyx> ok that may be problematic
2024-04-02T19:49:42 < nomorekaki> but it really helps otherwise to know this uart stuff could be used and found maxims uart tool so I don't need to play with arduino
2024-04-02T19:51:31 < jpa-> 1wire is slow enough that if you don't have other load, you can just bitbang it
2024-04-02T19:51:40 < jpa-> (just in software)
2024-04-02T19:52:33 < nomorekaki> yes
2024-04-02T19:52:41 < nomorekaki> polls and loops
2024-04-02T19:52:52 < jpa-> lolls and poops
2024-04-02T19:53:26 < nomorekaki> I design it so I can utilize timers as bestest as possible
2024-04-02T19:54:23 < nomorekaki> if I feel like trying to keep the power minimal
2024-04-02T19:58:09 < nomorekaki> timers allow basically sleeping the DQ low periods
2024-04-02T19:59:46 < jpa-> which chip btw?
2024-04-02T20:02:46 < nomorekaki> attiny406
2024-04-02T20:04:29 -!- Netsplit *.net <-> *.split quits: yoyofreeman
2024-04-02T20:05:43 -!- Netsplit over, joins: yoyofreeman
2024-04-02T20:10:14 < jpa-> https://github.com/smurfix/owslave/blob/master/onewire.c  seems to use just interrupt + timer
2024-04-02T20:12:47 < jpa-> https://github.com/MarkusLange/OneWireSlave/blob/master/OneWireSlave.cpp  uses plain cpu-driven bitbang
2024-04-02T20:25:51 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 272 seconds]
2024-04-02T20:27:01 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-02T20:29:37 < ColdKeyboard> For some reason on my board ethernet gnd/shield is tied to board GND through 1M and 100nF in parallel
2024-04-02T20:30:15 < ColdKeyboard> Whenever the solenoid is actuated, the SSH session craps out but when I log back in, I can see that the device is still online and only my session dropped...
2024-04-02T20:30:45 < jpa-> what does ssh -v or wireshark give as the reason for the session dropping?
2024-04-02T20:31:06 < ColdKeyboard> That's good point, I use Putty but maybe I can try ssh -v
2024-04-02T20:31:11 < jpa-> 1M + 100nF sounds normal enough, even though it is probably not standards compliant in the strictest sense
2024-04-02T20:32:02 < ColdKeyboard> I'll have to check a bit because everything else "seems normal". It's just when the lock solenoid is actuated, the SSH drops
2024-04-02T20:32:43 < jpa-> what kind of flyback voltage clamping arrangement do you have for the solenoid? does it happen when the solenoid is activated or when it is deactivated?
2024-04-02T20:55:35 < ColdKeyboard> SSH sends back -> client_loop: send disconnect: Connection reset
2024-04-02T20:55:50 < ColdKeyboard> The boards just has a diode connected across the solenoid
2024-04-02T20:56:34 < ColdKeyboard> And it happens only when the solenoid is activated/energized
2024-04-02T20:57:10 < ColdKeyboard> Or precisely only when it's going from not-energized to energized state
2024-04-02T20:57:11 < jpa-> are you sure the CPU doesn't reboot due to power glitch or something?
2024-04-02T20:57:49 < jpa-> a couple of corrupted packets shouldn't break TCP connection
2024-04-02T20:57:52 < ColdKeyboard> If I check the uptime after I re-login, it's hours or so
2024-04-02T20:58:26 < jpa-> perhaps it gets a "cable disconnected" notification from PHY and ends up bringing the interface down, causing the connections to close
2024-04-02T20:59:16 < ColdKeyboard> It's possible... I'll have to dig into it a bit more
2024-04-02T20:59:46 < jpa-> i'd scope the power rails to start with
2024-04-02T21:12:50 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-02T21:21:55 < ColdKeyboard> jpa- Weird question, but if there is a spike/ripple on the VCC line. Increasing the capacitor would help obiously, but would it be wise to also put an inductor inline to oppose current change or that would be a terrible idea?
2024-04-02T21:24:34 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T21:25:49 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T21:34:44 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-02T21:45:34 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 264 seconds]
2024-04-02T21:48:38 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-02T21:57:05 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-02T22:10:00 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-02T22:12:24 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has joined ##stm32
2024-04-02T22:13:24 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 260 seconds]
2024-04-02T22:40:05 < jpa-> ColdKeyboard: inductor can help or hurt depending on exact circuit.. it can also oscillate
2024-04-02T22:52:18 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-02T22:58:10 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-02T22:59:03 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-02T22:59:46 -!- qyx [~qyx@84.245.120.129] has quit [Ping timeout: 268 seconds]
2024-04-02T23:01:29 -!- qyx [~qyx@84.245.120.191] has joined ##stm32
2024-04-02T23:07:45 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-02T23:14:06 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-02T23:37:53 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-02T23:38:24 < nomorekaki> spice modeling says I can run attiny for hundreds of microseconds from 0402 1uF cap
2024-04-02T23:39:08 < nomorekaki> even switch cpu to "real clock" and consume a whole milliamp
2024-04-02T23:39:47 < nomorekaki> not hundreds of of microseconds but way longer I thought
2024-04-02T23:41:17 < nomorekaki> what do you think in practice does it really run from 0402 cap it feels so absurd that little grain of capacitor could power a mcu for any period of time
2024-04-02T23:42:55 < qyx> yes.
2024-04-02T23:43:12 < qyx> in other news, another qyx dementia
2024-04-02T23:43:26 < qyx> I can enable HSE, configure and enable PLL and switch SYSCLK to that PLL
2024-04-02T23:43:36 < qyx> but but I cannot switch it back to HSI16
2024-04-02T23:55:30 < qyx> thank you
2024-04-02T23:56:52 < nomorekaki> you counted on it?
2024-04-02T23:57:49 < nomorekaki> go through warm reset :o
--- Day changed ke huhti 03 2024
2024-04-03T00:07:03 < qyx> of course some retard disabled HSI16 in the bootloader
2024-04-03T00:07:09  * qyx hides
2024-04-03T00:10:49 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-03T00:15:37 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-03T00:40:00 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-03T01:00:01 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-03T01:02:25 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-03T01:18:45 -!- specing [~specing@user/specing] has quit [Ping timeout: 256 seconds]
2024-04-03T01:22:45 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-03T01:26:07 < qyx> I guess I can handle every possible clock mess now including HSE crystal detection
2024-04-03T01:27:04 < qyx> HSE bypass remains but I have to test a different HSE circuit for this and whether it is safe to try both bypass/no bypass
2024-04-03T01:42:44 < karlp> qyx: the hurhur low ower is becuase this fucking k32l2 m0+ was chosen for for "low power" but it's such a wildly small part of the power budget of the board.  an stm32f7 would have been just as suitable...
2024-04-03T01:46:11 < karlp> lol, fake it til you make it, then just stop doing the fake bit... https://gizmodo.com/amazon-reportedly-ditches-just-walk-out-grocery-stores-1851381116
2024-04-03T01:53:09 < qyx> noice
2024-04-03T01:53:34 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-03T02:00:40 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-03T02:02:38 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-03T02:16:10 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-03T02:21:09 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 272 seconds]
2024-04-03T02:25:42 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T02:27:21 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T03:08:38 -!- boB_K7IQ [~boB_K7IQ@152.44.147.180] has quit [Ping timeout: 252 seconds]
2024-04-03T03:14:25 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-03T03:31:51 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-03T03:32:21 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T03:33:04 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T04:07:32 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T04:07:58 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T04:11:16 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T04:11:59 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T04:20:40 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-03T04:22:33 -!- rkta_ [~rkta@user/rkta] has joined ##stm32
2024-04-03T04:23:32 -!- rkta [~rkta@user/rkta] has quit [Ping timeout: 256 seconds]
2024-04-03T04:58:53 -!- josuah [~josuah@46.23.94.12] has quit [Remote host closed the connection]
2024-04-03T05:27:46 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-03T05:39:14 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T06:28:35 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-03T06:32:52 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 255 seconds]
2024-04-03T07:34:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T07:36:16 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T07:40:20 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-03T07:40:53 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-03T07:49:20 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-03T08:02:01 -!- rkta_ is now known as rkta
2024-04-03T08:22:18 < jpa-> qyx: so did you figure out the CCR problem?
2024-04-03T08:24:39 < qyx> yes, wrong constant, bitten by locm3
2024-04-03T08:24:50 < qyx> I am curious how those examples worked
2024-04-03T08:24:58 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-03T08:25:54 < qyx> https://github.com/libopencm3/libopencm3-examples/blob/master/examples/stm32/l4/stm32l476g-disco/basics/basics.c#L81
2024-04-03T08:26:01 < qyx> apparently I am not the only one
2024-04-03T08:29:53 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-03T08:30:30 < qyx> but I was dealing with a ton of different issues yesterday so idk which one, there were also some "issues"
2024-04-03T08:33:39 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-03T08:41:36 -!- pwillard [sid136981@id-136981.hampstead.irccloud.com] has quit [Ping timeout: 268 seconds]
2024-04-03T08:45:04 -!- pwillard [sid136981@id-136981.hampstead.irccloud.com] has joined ##stm32
2024-04-03T08:50:04 -!- drkow [~k\o\w@2607:fea8:1d00:89f0:4662:11c8:f36e:f0ce] has joined ##stm32
2024-04-03T08:54:22 < qyx> rip tsmc
2024-04-03T08:54:27 -!- drfff [~k\o\w@2607:fea8:1d00:89f0:b0f9:e242:1752:8513] has quit [Ping timeout: 272 seconds]
2024-04-03T09:01:52 -!- yoyofreeman [~yoyofreem@38.180.62.249] has quit [Quit: Leaving]
2024-04-03T09:03:51 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-03T09:06:08 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-03T09:19:40 < qyx> if I wanted to skip the bootloader, is "reset halt; set_reg {pc 0x0802a82d sp 0x20014000}; resume" the right sequence?
2024-04-03T09:21:50 < jpa-> you may want to set NVIC->VTOR too unless your main app does it
2024-04-03T09:24:28 < qyx> true
2024-04-03T09:24:31 < qyx> thanks
2024-04-03T09:24:49 < qyx> yes I forgot to do that
2024-04-03T09:25:17 < qyx> when I start doing an actual work I feel like a pro debug consumer
2024-04-03T09:30:26 < jpa-> fortunately you have access to a channel full of rubber ducks
2024-04-03T09:31:19 < qyx> I'll pay it back when I start procrastinating again, no worries
2024-04-03T09:39:02 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-03T09:48:32 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-03T10:10:43 < ventYl> quack
2024-04-03T10:12:03 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has joined ##stm32
2024-04-03T10:23:50 -!- machinehum [~machinehu@xcpe-62-167-161-87.cgn.res.adslplus.ch] has left ##stm32 [WeeChat 4.2.1]
2024-04-03T10:25:42 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-03T11:12:44 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-03T11:32:53 -!- ventYl [~ventyl@adsl-dyn-213.95-102-75.t-com.sk] has quit [Ping timeout: 240 seconds]
2024-04-03T12:10:46 < karlp> qyx: yeah, that method doesn't have any doxygen hints either to what the set of viable values are.
2024-04-03T12:10:52 < karlp> sorry, full refund.
2024-04-03T12:12:40 < qyx> np
2024-04-03T12:28:14 < karlp> I love being able to do this from gbd: ` (gdb) call (void)memset(kmtb_buffer, 0, sizeof(kmtb_buffer))` 
2024-04-03T12:39:57 < jpa-> is gbd short for GNU buggered? that at least describes my state of mind after using gdb
2024-04-03T12:51:38 < karlp> well, maybe it's just taking what little pleasure there can be from the small things.
2024-04-03T13:45:01 -!- ventYl [~ventyl@adsl-dyn176.78-99-153.t-com.sk] has joined ##stm32
2024-04-03T13:51:20 < karlp> anyone have any idees ongdb's python api to get this to decode better? https://paste.jvnv.net/view/ElZOo
2024-04-03T13:53:44 < jpa-> maybe .format_address() ?
2024-04-03T13:54:13 < karlp> yes! that works
2024-04-03T13:54:14 < karlp> awesome
2024-04-03T13:54:21 < jpa-> happy to google
2024-04-03T13:54:44  * karlp foolishly trying to read the gdb docs on the python api
2024-04-03T13:58:57 < karlp> that's _way_ better now.
2024-04-03T14:00:11 < karlp> https://paste.jvnv.net/view/JUxGJ
2024-04-03T14:00:51 < karlp> I've given up on trying to stream MTB, but it can at least be the post mortem shit that seems to be all it's useful for
2024-04-03T14:06:19 < jpa-> https://sourceware.org/gdb/current/onlinedocs/gdb.html/Basic-Python.html   gdb docs is where i found it anyway :)
2024-04-03T14:06:29 < karlp> yeah, I was in the wrong section :)
2024-04-03T14:34:41 < qyx> ducks, mega giga integer literals, uint32_t freq = 16000000UL;
2024-04-03T14:34:49 -!- Laurenceb_ [~Laurenceb@cust226-dsl93-89-135.idnet.net] has joined ##stm32
2024-04-03T14:35:03 < qyx> it can be written freq = 16e6, but but double->uint
2024-04-03T14:35:26 < qyx> I am trying to find a way to macro it to be like 16_M, but I don't see any
2024-04-03T14:36:33 < karlp> c++...?
2024-04-03T14:36:34 < qyx> #define M * 1000000UL works but poses a risk
2024-04-03T14:37:11 < karlp> given they're literels, I just eat the work of doing "3 * 1024" or something
2024-04-03T14:37:55 < karlp> pretty sure I had some code using uint32-t = 16e6 or something too, and it sorts it out at compile time?
2024-04-03T14:37:59 < karlp> easy to godbolt...
2024-04-03T14:38:11 < qyx> hm yeah good idea
2024-04-03T14:40:29 < karlp> looks ok to me: https://godbolt.org/z/c75nEhKWY
2024-04-03T14:41:59 < qyx> 3e6 is ok too
2024-04-03T14:42:09 < qyx> k 3e6 looks good enough for me
2024-04-03T14:46:07 < karlp> 3e6 takes up more flash though :)
2024-04-03T14:46:16 < karlp> at least in my bullshit contrived example.
2024-04-03T14:46:54 < karlp> ah no, not when I do 3 * 1000 * 1000
2024-04-03T14:46:59 < karlp> wasn't being fair.
2024-04-03T14:47:29 -!- jbo_ is now known as jbo
2024-04-03T14:47:39 < karlp> somehow it can do "mov     r3, #3145728" but not "mov r3, #3000000"
2024-04-03T14:59:43 < jpa-> 0x300000 fits well with the ARM "N bits literal shifted by M bits" model
2024-04-03T15:03:29 -!- Laurenceb_ [~Laurenceb@cust226-dsl93-89-135.idnet.net] has quit [Quit: Client closed]
2024-04-03T15:20:46 -!- c10ud_ [~c10ud@95.236.174.95] has joined ##stm32
2024-04-03T15:24:35 -!- c10ud [~c10ud@user/c10ud] has quit [Ping timeout: 268 seconds]
2024-04-03T15:28:38 < karlp> you're telling me the assembly is a lie again right?
2024-04-03T15:28:47 < karlp> it's written taht way, but it's not actually?
2024-04-03T15:30:44 < karlp> godbolt at least is lyin well that it's ok: https://godbolt.org/z/GWTMGK6dn
2024-04-03T15:31:36 < jpa-> karlp: click "output -> compile to binary" and you'll get "offset out of range"
2024-04-03T15:31:59 < karlp> I think I'm done with computers
2024-04-03T15:32:23 < karlp> the burning never stops
2024-04-03T15:32:48 < jpa-> don't worry, you are on cortex-m0, it has only literals 0-255 anyway
2024-04-03T15:33:36 < jpa-> cortex-m3 can have 0-65535, or 0-255 shifted to any location, or 0-255 multiplied by 0x00010001, 0x01000100 or 0x01010101
2024-04-03T15:33:59 < karlp> ok, the first two i get, but .... those last ones? what are they from?
2024-04-03T15:34:03 < karlp> what are they for?
2024-04-03T15:34:15 < karlp> "lets' make up somethign to use the encoded options space"?
2024-04-03T15:35:46 < jpa-> they allow e.g. 0xFFFFFFFF to be encoded
2024-04-03T15:36:27 < jpa-> though i guess mvn #0 would do that too..
2024-04-03T15:41:10 < karlp> oh, multiplied by bits like thiat means you get to copy them into those cunks? 255 * 0x00010001 => 0x00ff00ff?
2024-04-03T15:41:26 < jpa-> sure
2024-04-03T15:41:27 < karlp> i can't do bitfield multiplication in my head
2024-04-03T15:41:43 < karlp> now I can see it a bit more clearly what those forms are useful for then...
2024-04-03T15:41:44 < jpa-> yeah, multiplying by 1 can be challenging
2024-04-03T15:41:52 < karlp> oh shush
2024-04-03T15:55:32 < qyx> https://paste.jvnv.net/view/TSpeB
2024-04-03T15:55:34 < qyx> half dun
2024-04-03T15:55:44 < qyx> I got some artifacts when dynamically reconfiguring clocks
2024-04-03T15:58:04 < qyx> so after a day of programming and adding 600 lines, I added exactly 0 features and solved 0 issues
2024-04-03T15:58:09 < qyx> and the code looks worse than before
2024-04-03T15:59:48 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-03T16:35:15 < karlp> meh, trying to do some hooks on critical sections in freertos, and someone clever has setup freertos to link as a static method
2024-04-03T17:34:55 < karlp> so, I thought a .a was just a pile of .o's.  so if I have "extern somefunc();" in a .file, and I call that function, and I expect it to be linked later, doesn't  it just get filled in later?
2024-04-03T17:35:07 < karlp> I seem to get an error creating the .a that the function doesn't exist?
2024-04-03T17:35:55 < karlp> no, ok, it's later, it does build the .a of freertos ok.
2024-04-03T17:35:58 < karlp> fucking linkers
2024-04-03T17:36:50 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-03T17:36:54 < karlp> fuckin lol, the function is static inline I guess that's why it can't "find it later"
2024-04-03T17:37:47 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds]
2024-04-03T17:46:07 < karlp> well, I've found some 500-600usec critical sections, which look like a bug, but none of them when it's overrrunning the uart.
2024-04-03T19:01:22 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-03T19:16:30 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 255 seconds]
2024-04-03T19:33:03 < qyx> were you hired to fix all problems?
2024-04-03T19:33:12 < karlp> effectively.
2024-04-03T19:33:43 < karlp> found a 6ms crit section too, but again, not interrupting my rx bytes...
2024-04-03T19:34:08 < qyx> I usually say I am not the saviour
2024-04-03T19:34:20 < karlp> well, sure, but sooner or later it has to get fixed...
2024-04-03T19:34:40 < karlp> and despite this being a "team" it's 6 teams of 1, not one team of 6.
2024-04-03T19:34:57 < karlp> went corpo and stood around someone else for an hour with a coffee for the afternoon instead.
2024-04-03T19:35:15 < karlp> hooked up some air shit and threw pretend chicken breasts at a machine to see if it worked.
2024-04-03T19:46:47 -!- Livio [~livio@user/livio] has quit [Ping timeout: 272 seconds]
2024-04-03T19:59:41 < qyx> yeah pair programming/debugging is a recognised technique too
2024-04-03T20:00:30 < qyx> anyway, I couldn't get the clock security system working
2024-04-03T20:01:01 < qyx> shorting out the oscillator output for a while, nothing happens, just freezes
2024-04-03T20:01:15 < qyx> nmi is enabled
2024-04-03T20:02:41 -!- scrts [~scrts2@23.28.144.38] has quit [Quit: The Lounge - https://thelounge.chat]
2024-04-03T20:03:04 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-03T20:18:58 < Ecco> Hi! Have you guys ever used the touch-sensing controller on an STM32?
2024-04-03T20:19:44 < jpa-> nope
2024-04-03T20:20:08 < jpa-> capacitive buttons suck
2024-04-03T20:22:32 < Ecco> how so?
2024-04-03T20:23:22 < Ecco> (my personal experience as a consumer has been a mixed bag. On some devices they never seem to work, but sometimes I think they're pretty neat)
2024-04-03T20:27:04 < specing> I have a portable induction cooker. When water leaves the pot due to too much heating, it sometimes spills over the touch controls
2024-04-03T20:27:25 < specing> this confuses the cooker, resulting in it "resetting" itself to the turn-on setting, which is maximum power
2024-04-03T20:29:20 < specing> I'll take physical buttons over touch any day. This is also the reason why we didn't buy a "proper" installable induction cooker.. I haven't found one that would have actual physical dials.
2024-04-03T20:30:47 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-03T20:40:08 < Ecco> specing: ha, that's a cool example :)
2024-04-03T20:40:14 < Ecco> (sucks, of course, but interesting nonetheless)
2024-04-03T20:40:32 < Ecco> I also have an induction stovetop
2024-04-03T20:40:37 < Ecco> (kind of exepensive)
2024-04-03T20:41:14 < Ecco> The design is better tho: when water spills, the thing detects it and enters an "error" state where the heat is turned off
2024-04-03T20:41:37 < Ecco> actually, it's funny you would mention this, because I kind of like the touch feature on the stovetop: it's *so* easy to clean
2024-04-03T20:41:58 < Ecco> The knobs on the one I had before were full of hard to remove gunk
2024-04-03T20:48:15 -!- drzacek [~quassel@2a01:3d8:460:5600:e8c3:941c:4d23:df75] has joined ##stm32
2024-04-03T20:48:15 -!- drzacek [~quassel@2a01:3d8:460:5600:e8c3:941c:4d23:df75] has quit [Client Quit]
2024-04-03T20:48:24 -!- drzacek [~quassel@2a01:3d8:460:5600:e8c3:941c:4d23:df75] has joined ##stm32
2024-04-03T21:00:39 < antto> unmount them and toss them in boiling water - gunk soup! \o/
2024-04-03T21:02:27 < Ecco> :-D
2024-04-03T21:40:49 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-03T21:51:06 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-03T21:51:12 < nomorekaki> https://twitter.com/nicolasburtey/status/1774457208492753020
2024-04-03T21:56:06 < stgl> I am still struggling with the STM32Cube USB middleware... as soon as I plug it in the USB port, systick interrupts stop for 35 ms (should be every ms) and then continues each 1.6ms, despite having highest priority
2024-04-03T21:56:36 < stgl> as soon as I unplug it, it crashes completely (this happens also with the example project, already made a forum post but no replies)
2024-04-03T21:56:42 < stgl> any ideas here?
2024-04-03T22:06:23 < nomorekaki> there are higher interrupts than systick
2024-04-03T22:07:40 < nomorekaki> do you have some dev board or how does the physical implementation look like?
2024-04-03T22:15:32 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-03T22:15:44 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-03T22:25:59 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-03T23:19:21 -!- josuah [~josuah@46.23.94.12] has joined ##stm32
2024-04-03T23:34:23 < Ecco> nomorekaki: hahaha good one
2024-04-03T23:35:00 -!- drzacek [~quassel@2a01:3d8:460:5600:e8c3:941c:4d23:df75] has quit [Quit: https://quassel-irc.org - Czatuj komfortowo. Wszędzie.]
2024-04-03T23:39:22 < Steffanx> It's not even the real story so.. Boring -_- the "autism" was much better
--- Day changed to huhti 04 2024
2024-04-04T01:31:44 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-04T01:33:55 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-04T02:13:11 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-04T03:52:18 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 268 seconds]
2024-04-04T05:14:46 -!- specing [~specing@user/specing] has quit [Read error: Connection reset by peer]
2024-04-04T05:49:50 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T05:51:01 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T07:01:37 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T07:02:10 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T07:05:47 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T07:06:58 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T07:14:44 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T07:15:11 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T08:11:37 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-04T08:58:22 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:db28:53b3:eb61:c6f] has joined ##stm32
2024-04-04T09:22:11 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-04T11:00:34 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Ping timeout: 264 seconds]
2024-04-04T11:46:00 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-04T11:57:14 < \dev\ice> is it possible count separate rising and falling edges on same gpio?
2024-04-04T12:02:53 < jpa-> timer encoder mode might be able to do it
2024-04-04T12:05:45 < qyx> I would say the count is gonna be the same?
2024-04-04T12:06:22 < jpa-> that's true too, you can just check GPIO state at end and add one
2024-04-04T12:15:21 < qyx> re. bosch trademark fee, does it mean I can use FDCAN instead of CAN-FD and everything is ok?
2024-04-04T12:17:51 < qyx> (there are no registrations found for FDCAN)
2024-04-04T12:19:07 < qyx> The CAN FD and CAN XL Protocol License is required for any implementation of the respective protocols, also for self-developed CAN FD/CAN XL modules or for CAN FD/CAN XL modules purchased from other vendors
2024-04-04T12:19:13 < qyx> huh? even for purchased modules?
2024-04-04T12:19:40 < qyx> so STM32 price doesn't include CAN-FD fees meaning I can't use it until I buy the protocol license?
2024-04-04T12:20:57 < ventYl> I guess that this is pretty standard patent licensing practice
2024-04-04T12:21:13 < ventYl> I would ask STM
2024-04-04T12:21:44 < ventYl> in some cases protocol license fee might be part of part price
2024-04-04T12:21:54 < qyx> indeed it is nonstandard
2024-04-04T12:23:14 < qyx> fee for IP is ok, fee for developing your own IP for the protocol is ok, but fee for using a third party made module which has already been paid for is not very common
2024-04-04T12:23:58 < ventYl> you have certainly not been working with h.264 'n' friends before :)
2024-04-04T12:24:04 < ventYl> there, this is pretty standard practice
2024-04-04T12:24:23 < qyx> robert, y u no bankrupt yet
2024-04-04T12:25:09 < qyx> fuk u robert then
2024-04-04T12:40:36 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 260 seconds]
2024-04-04T12:50:37 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T12:58:42 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T13:01:06 < c10ud_> is there any nice el-cheapo device with mcu+display+rs232 that one can use?
2024-04-04T13:04:30 < jpa-> something like STM32F429I-DISC1 ?
2024-04-04T13:05:29 < c10ud_> yup, nice one...i will look into the disco stuff
2024-04-04T13:08:06 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-04T13:08:18 < karlp> does it reallllly need rs232?
2024-04-04T13:08:36 < karlp> there's a few esp32 based platforms that have hard rs232 ports.
2024-04-04T13:09:15 < karlp> not seeing rs232 on that board jpa?
2024-04-04T13:12:03 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-04T13:13:09 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-04T13:13:57 < c10ud_> either 232 or can, i just need it to say hi to a motor which has both
2024-04-04T13:14:08 -!- duude__- is now known as duude__
2024-04-04T13:17:37 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has joined ##stm32
2024-04-04T13:29:34 -!- dkc [~dan@user/dkc] has quit [Quit: ZNC 1.8.2 - https://znc.in]
2024-04-04T13:29:42 < jpa-> karlp: yeah, but easy enough to add some breakout if real RS232 is needed
2024-04-04T13:30:26 -!- dkc [~dan@user/dkc] has joined ##stm32
2024-04-04T13:37:50 -!- dkc [~dan@user/dkc] has quit [Quit: ZNC 1.8.2 - https://znc.in]
2024-04-04T13:38:13 -!- dkc [~dan@user/dkc] has joined ##stm32
2024-04-04T13:54:06 < karlp> I wonder who works on a m0, and does enough dma traffic to need to be concerned abotu priorities vs round robin for channel arbitration.
2024-04-04T13:54:14 < karlp> you'd hav eto be really squeezing it all out.
2024-04-04T13:54:34 < qyx> hdmi output guy at rpi pico
2024-04-04T13:54:41 < qyx> *on
2024-04-04T13:55:03 < zyp> :)
2024-04-04T13:58:26 < zyp> while some people are shoehorning too much shit into too small microcontrollers, I picked up a stm32h7s7-dk for generic prototyping…
2024-04-04T13:59:19 < zyp> mostly because it's one of the few devboards that's got usb-c on everything and because it's got a proper 20-pin debug+trace header
2024-04-04T14:00:40 < qyx> hm it reminds me I have to do a board with u5
2024-04-04T14:01:15 < qyx> I must prepare to thorn-fail with hyperram beforehand
2024-04-04T14:04:49 < qyx> it is a myth water kills electronics
2024-04-04T14:05:04 < qyx> I just put stm32 with a HSE into water and it continues to work
2024-04-04T14:05:46 < qyx> oh LSE stopped
2024-04-04T14:05:47 < zyp> I think it's fairly common knowledge that the problem is not the water but impurities in the water
2024-04-04T14:06:16 < qyx> but condensed water doesn't have many impurities, does it?
2024-04-04T14:06:17 < karlp> I wouldn't say that's common at all.
2024-04-04T14:06:47 < zyp> no?
2024-04-04T14:06:55 < qyx> no, at least not here
2024-04-04T14:07:06 < qyx> everyone think water shorts electrical circuits
2024-04-04T14:07:16 < qyx> and everything is killed when you pour water on it instantly
2024-04-04T14:12:25 < qyx> ok so found the bug, the thing doesn't start because of the failed LSE
2024-04-04T14:12:43 < qyx> not even a boot loop because it happens before the watchdog is enabled
2024-04-04T14:14:02 < karlp> is it just me, or is this asserting on the alignment of the source pointer? even though it's copying things out field by field? https://github.com/nxp-mcuxpresso/mcux-sdk/blob/main/drivers/edma/fsl_edma.c#L92
2024-04-04T14:14:30 < jpa-> it's just trying to say "FU"
2024-04-04T14:15:11 < jpa-> compiler might optimize some of those to ldrd, so 4-byte alignment would make sense
2024-04-04T14:15:20 < jpa-> higher than that wouldn't matter no matter how it is copied
2024-04-04T14:16:03 < qyx> they are checking 32... bits?
2024-04-04T14:16:12 < jpa-> but considering they don't seem to know what "const" is for, i wouldn't expect them to be very knowledgeable on alignment either
2024-04-04T14:16:13 < qyx> (heh)
2024-04-04T14:16:30 < karlp> then there's this sort of magic weirdness: https://github.com/nxp-mcuxpresso/mcux-sdk-examples/blob/main/frdmk32l2a4s/driver_examples/lpuart/edma_rb_transfer/lpuart_edma_rb_transfer.c#L217-L220
2024-04-04T14:17:31 < zyp> NOPE();
2024-04-04T14:17:53 < jpa-> but hey, it was the initial commit, you can't expect much from initial commit..
2024-04-04T14:18:29 < zyp> pff
2024-04-04T14:18:30 < jpa-> i guess someone wanted to breakpoint there
2024-04-04T14:18:53 < qyx> planting a future backdoor
2024-04-04T14:18:58 < zyp> I've got a bad habit of waiting with the initial commit until I've got a MVP or something
2024-04-04T14:19:20 < zyp> ref. https://github.com/zyp/smolt/commit/5c05022
2024-04-04T14:19:56 < jpa-> me too
2024-04-04T14:20:19 < qyx> ok the device failed. HSE too
2024-04-04T14:22:20 < qyx> https://paste.jvnv.net/view/95QES
2024-04-04T14:22:21 < qyx> hah
2024-04-04T14:24:09 < zyp> «may be inaccurate», hah
2024-04-04T14:24:17 < zyp> more like *will* be inaccurate
2024-04-04T14:24:20 < jpa-> you could be really lucky!
2024-04-04T14:24:30 < zyp> LSI is ridiculously imprecise
2024-04-04T14:25:07 < zyp> good enough for watchdog, but not much else
2024-04-04T14:25:54 < zyp> if all you need RTC for is low power wakeups, then maybe good enough for that too
2024-04-04T14:25:56 < qyx> so the actual problem is I need to enable watchdog in the bootloader before any oscillator is started
2024-04-04T14:26:10 < zyp> if you need actual timekeeping, dividing down main clock is more precise than LSI
2024-04-04T14:26:19 < jpa-> doesn't iwdg have its own oscillator or something?
2024-04-04T14:26:27 < zyp> that's what LSI is
2024-04-04T14:26:40 < zyp> LSI is IWDG's own oscillator
2024-04-04T14:26:49 < qyx> yes I know it is but rtc is running in stop mode and rtc clock is advertised as available in stop mode
2024-04-04T14:27:06 < qyx> I also have a backup "system rtc" using a timer from sysclk
2024-04-04T14:28:04 < qyx> everything is far from complete
2024-04-04T14:47:05 < karlp> LSI for rtc is a joke.
2024-04-04T14:47:13 < karlp> not even sure it should be allowed.
2024-04-04T14:51:41 < qyx> rtc has many features other than "precise timekeeping" which are totally fine when run from LSI
2024-04-04T14:52:18 < qyx> periodic wakeups, alarms
2024-04-04T14:56:44 < karlp> fine with rtcv1, which has "long term counter"
2024-04-04T14:56:54 < karlp> but when it's telling you minutes and seconds in bcd....
2024-04-04T14:57:53 < qyx> oh yes I would kill for that, I need to handle TAI or GPS time and converting that is a nightmare
2024-04-04T15:37:16 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-04T15:53:52 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-04T15:55:13 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T15:57:02 -!- System_Error [~SystemErr@user/systemerror] has quit [Read error: Connection reset by peer]
2024-04-04T16:01:35 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-04T16:16:43 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has joined ##stm32
2024-04-04T16:33:12 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-04T17:00:31 -!- rpifan [~rpifan@p200300d267133a000dad49c73f9a9bfb.dip0.t-ipconnect.de] has joined ##stm32
2024-04-04T17:00:31 -!- rpifan [~rpifan@p200300d267133a000dad49c73f9a9bfb.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-04T17:00:31 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-04T18:01:12 -!- scrts [~scrts2@23.28.144.38] has quit [Ping timeout: 260 seconds]
2024-04-04T18:02:04 < karlp> https://twitter.com/mikesimonsen/status/1775683012598079639
2024-04-04T18:07:47 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-04T18:08:30 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-04T18:21:57 < ventYl> is there any reason for that?
2024-04-04T18:23:18 < BrainDamage> perhaps an eclipse
2024-04-04T18:28:10 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-04T18:28:38 < ventYl> that's one which came up to my mind, but I don't follow eclipse calendar
2024-04-04T18:50:10 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 246 seconds]
2024-04-04T18:59:01 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:db28:53b3:eb61:c6f] has quit [Ping timeout: 256 seconds]
2024-04-04T19:03:15 < karlp> yeah, total eclipse path over usa next week.
2024-04-04T19:54:19 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 255 seconds]
2024-04-04T20:05:29 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds]
2024-04-04T20:32:39 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-04T20:46:49 < Ecco> I'm (trying to) use the touch-sensing controller
2024-04-04T20:47:05 < Ecco> We're supposed to charge a sampling capacitor. It's a cap that's connected directly to a GPIO
2024-04-04T20:47:27 < Ecco> I'm trying to compute the RC constant, but… I've no idea what the resistor is. How much current can GPIOs provide?
2024-04-04T20:54:59 < Ecco> The datasheet says each GPIO can source 20mA. At 3V, I guess that means a 150 ohm ESR?
2024-04-04T21:03:27 < qyx> it is easy even without a touch controller
2024-04-04T21:03:42 < qyx> configure as opendrain outpit, pullup
2024-04-04T21:03:53 < qyx> don't bother with exact values
2024-04-04T21:04:21 < qyx> discharge with short pulse to ground, recharge with the internal pullup
2024-04-04T21:04:53 < qyx> it needs one tweak, there is a "routing interface" which allows you to short some analog gpio directly
2024-04-04T21:05:15 < qyx> so when measuring short to another timer channel configured as input capture
2024-04-04T21:05:43 < qyx> and measure the time it takes to switch the schmitt guy on the input to high
2024-04-04T21:06:16 < qyx> anyway, whatever medthod you are using, exact numbers are not important
2024-04-04T21:07:03 < qyx> there is also a flying cap discharge method where you don't sense the time but number of switchovers
2024-04-04T21:07:28 < jpa-> you can also measure against ADC sampling cap, but i'm not sure why you'd do any of these tricks if you have the touch controller
2024-04-04T21:07:56 < qyx> you need to calibrate the thing in runtime and adopt to enviromnent changes
2024-04-04T21:08:05 < qyx> capacity is temp dependent much
2024-04-04T21:08:19 < qyx> or high pass filter it
2024-04-04T21:09:05 < qyx> jpa-: becauseI like tricks
2024-04-04T21:12:26 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-04T21:12:34 < jpa-> tricksy slovakses
2024-04-04T21:21:48 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-04T21:29:53 < Ecco> Ha, interesting. The docs also just say that the sampling cap should just be "charged"
2024-04-04T21:30:14 < Ecco> so I don't need to be very accurate, just to make sure it's charged. But I feel like I kind of need to do the math
2024-04-04T21:31:27 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-04T21:31:37 < Ecco> Here's what I did: GPIO provides 20mA at 3V, so that's equivalent to a 150 ohms resistor. My capacitor is 47nF, so 5*RC is 32 us, which is roughly 1700 cycles at 48 MHz
2024-04-04T21:31:46 < Ecco> Does that seem reasonable? Or am I completely off?
2024-04-04T21:42:41 < srk> jpa-: relaymux <3 how did you make the silkscreen?
2024-04-04T22:12:15 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-04T22:19:45 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-04T23:04:14 -!- qyx [~qyx@84.245.120.191] has quit [Ping timeout: 268 seconds]
2024-04-04T23:05:49 -!- qyx [~qyx@84.245.120.146] has joined ##stm32
2024-04-04T23:18:32 < Steffanx> Whoa jpa- fell for the cube.
2024-04-04T23:18:43 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-04T23:19:06 < antto> They see me cuuubiiiin', they haaatiiin'
2024-04-04T23:20:57 < Ecco> :-D
2024-04-04T23:26:56 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-04T23:41:34 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-04T23:59:54 < qyx> so does anybody used the clock security system?
--- Day changed pe huhti 05 2024
2024-04-05T00:00:20 < qyx> or can anyone test? RCC_CR |= RCC_CR_CSSON; should enable it
2024-04-05T00:00:35 < qyx> and it switches to HSI when HSE is stopped
2024-04-05T00:00:42 < qyx> *it should
2024-04-05T00:00:51 < qyx> *has
2024-04-05T00:01:26 < qyx> idk if it should react to tweezers-stop but nothing happens
2024-04-05T00:02:08 < qyx> nothing in errata
2024-04-05T00:02:25 < qyx> the bit remains set when I check
2024-04-05T00:08:25 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T00:10:25 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T00:29:21 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T00:30:02 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T00:30:12 -!- System_Error [~SystemErr@user/systemerror] has quit [Read error: Connection reset by peer]
2024-04-05T00:34:07 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T00:34:54 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-05T00:43:55 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-05T00:57:00 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 260 seconds]
2024-04-05T01:34:09 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-05T01:46:27 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 256 seconds]
2024-04-05T01:49:07 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-05T01:58:29 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T01:59:17 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T01:59:28 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-05T02:09:52 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-05T02:14:52 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T02:15:25 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T02:21:23 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T02:22:01 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T02:54:56 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Killed (NickServ (GHOST command used by Spirit5321))]
2024-04-05T02:55:01 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-05T03:08:58 -!- DKordic [~DKordic@178.253.254.174] has quit [Ping timeout: 264 seconds]
2024-04-05T03:17:58 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 264 seconds]
2024-04-05T03:19:56 -!- Thorn [~Thorn@bl18-44-5.dsl.telepac.pt] has joined ##stm32
2024-04-05T03:19:56 -!- Thorn [~Thorn@bl18-44-5.dsl.telepac.pt] has quit [Changing host]
2024-04-05T03:19:56 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T03:34:03 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 260 seconds]
2024-04-05T03:36:00 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T04:35:17 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 240 seconds]
2024-04-05T04:37:36 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T04:42:03 -!- stgl [~stgl@2a03:b0c0:3:d0::cad:a001] has quit [Quit: ZNC 1.8.2 - https://znc.in]
2024-04-05T04:42:17 -!- stgl [~stgl@164.92.162.3] has joined ##stm32
2024-04-05T04:46:51 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 260 seconds]
2024-04-05T04:48:57 -!- Thorn [~Thorn@2001:8a0:dfe5:7b00:a906:f538:f11:3dfe] has joined ##stm32
2024-04-05T04:48:57 -!- Thorn [~Thorn@2001:8a0:dfe5:7b00:a906:f538:f11:3dfe] has quit [Changing host]
2024-04-05T04:48:57 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T05:13:11 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T05:13:53 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T05:25:22 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T05:25:56 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T06:07:36 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-05T06:48:10 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 255 seconds]
2024-04-05T06:50:20 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T07:53:15 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-05T08:10:55 < jpa-> srk: i made the silkscreen by drawing it in kicad and then ordering the pcb from jlcpcb
2024-04-05T08:11:19 < jpa-> Steffanx: have to use it to hate it
2024-04-05T08:11:52 -!- Kerr [~quassel@174.31.47.68] has joined ##stm32
2024-04-05T08:12:32 < jpa-> i have to admit the cube is not 100% terrible for simple stuff, though i didn't touch the "ide" part of it
2024-04-05T08:13:41 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 240 seconds]
2024-04-05T08:15:53 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T08:32:22 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 264 seconds]
2024-04-05T08:34:09 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T08:38:27 < qyx> isn't it too complex for simple stuff?
2024-04-05T08:39:33 < jpa-> not when you have 32kB to spare for flipping a few GPIOs
2024-04-05T08:46:52 < qyx> yesterday I removed rtc support from my bootloader (including mktime or whatever it was) and the binary shrinked by 13 kB
2024-04-05T08:47:58 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 256 seconds]
2024-04-05T08:50:11 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T09:02:46 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 246 seconds]
2024-04-05T09:05:13 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T09:18:55 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-05T10:16:42 -!- Thorn [~Thorn@user/thorn] has quit [Ping timeout: 268 seconds]
2024-04-05T10:18:48 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T10:18:50 -!- Thorn [~Thorn@user/thorn] has quit [Client Quit]
2024-04-05T10:19:52 < qyx> torture intensifies https://bin.jvnv.net/file/ywPWG/IMG_2024-04-05-09-17-10-543.jpg
2024-04-05T10:21:34 < c10ud_> nice balls
2024-04-05T10:23:35 < sauce> censored regions and suspicious white stains
2024-04-05T10:24:13 < qyx> sorry not public yet until end of 2024
2024-04-05T10:24:47 < sauce> what a tease
2024-04-05T10:25:01 < qyx> yeah I have a customer saying he is suspecting water condensation problems
2024-04-05T10:25:12 < qyx> so I am torturing the board with a water from a syringe a bit
2024-04-05T10:26:13 < c10ud_> coat the shit out of it and call it a day
2024-04-05T10:26:47 < qyx> kontakt 71 pur coating any good?
2024-04-05T10:27:12 < qyx> or kontakt 70 acrylic but apparently not suitable for wet conditions
2024-04-05T10:27:46 < zyp> what's the point of coating the board if you can't let it get wet afterwards?
2024-04-05T10:28:00 < c10ud_> honestly i dont remember what we used in *ahem* sensitive environment applications
2024-04-05T10:28:36 < c10ud_> if it can't get wet you rule out such issue playing a role in the malfunctions, i guess
2024-04-05T10:29:26 < c10ud_> oh i misread your statement
2024-04-05T10:29:50 < qyx> it is interesting it doesn't mess with the measuremnt, it is stable
2024-04-05T10:35:11 < qyx> Dear user as per our security policy your password has been expired.Please click below on Forgot Password to set a new one.
2024-04-05T10:35:14 < qyx> why st
2024-04-05T10:36:12 < zyp> ugh
2024-04-05T10:36:41 < zyp> flashbacks to when I worked for ST-E and was forced to change my password regularly
2024-04-05T10:37:06 < qyx> must not contain spaces
2024-04-05T10:37:42 < zyp> I forgot the exact rules, but the main issue was how often you had to change it
2024-04-05T10:38:04 < zyp> so you end up using a common base with the year/month as a suffix or some shit
2024-04-05T10:38:05 < qyx> I don't know because I need to generate a new one every time when I log into my.st
2024-04-05T10:38:16 < qyx> I just autogenerate it
2024-04-05T10:38:21 < qyx> but yeah
2024-04-05T10:41:09 < qyx> sign in, password bad, reset password, password ok, you are signed in, NO sign in again, ok
2024-04-05T10:41:24 < qyx> go to community, ok, create new post, NO sign in to create a post
2024-04-05T10:41:53 < qyx> You ranked up to
2024-04-05T10:41:54 < qyx> Associate II
2024-04-05T10:41:56 < qyx> thanks.
2024-04-05T11:11:51 < Steffanx> I feel very sorry for you qyx. Why do you have to endure all this?!
2024-04-05T12:29:27 -!- Posterdati [~Posterdat@user/Posterdati] has quit [Read error: Connection reset by peer]
2024-04-05T12:46:13 -!- Posterdati [~Posterdat@user/Posterdati] has joined ##stm32
2024-04-05T13:10:25 < qyx> so the CSS indeed works with a specific order of init
2024-04-05T13:10:43 < qyx> and the NMI is entered with some weird clock config which I have to find out
2024-04-05T13:37:22 < zyp> wtf, https://www.st.com/resource/en/application_note/an392-microcontrollers-and-triacbased-dimmers-stmicroelectronics.pdf suggests feeding a stm32c0 with 5V VDD
2024-04-05T13:38:20 < zyp> looks like they've just done a search and replace, s/ST6210/STM32C0/, because previous rev from 2009 said ST6210
2024-04-05T13:39:42 < zyp> (and yes, I checked, there's no 5V stm32c0, it'd be kinda neat if there were)
2024-04-05T13:40:27 < qyx> if c0 is marketed as an avr replacement, I would expect so
2024-04-05T13:42:09 -!- specing [~specing@user/specing] has quit [Ping timeout: 255 seconds]
2024-04-05T13:45:10 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-05T13:47:31 < qyx> I "solved" the issue by switching to LittleFS. According to the official docs, SPIFFS is deprecated, so this "solution" might actually be quite reasonable.
2024-04-05T13:47:35 < qyx> https://github.com/espressif/arduino-esp32/issues/4334
2024-04-05T15:07:32 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-05T15:29:18 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-05T15:58:51 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-05T16:22:31 -!- dkc [~dan@user/dkc] has quit [Quit: ZNC 1.8.2 - https://znc.in]
2024-04-05T16:22:52 -!- dkc [~dan@user/dkc] has joined ##stm32
2024-04-05T16:53:12 -!- Xeroine [~Xeroine@user/xeroine] has quit [Remote host closed the connection]
2024-04-05T16:56:49 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-05T17:24:09 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-05T17:26:30 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-05T17:27:30 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-9081-64bb-61ac-ea42.fixed6.kpn.net] has joined ##stm32
2024-04-05T17:54:29 -!- rpifan [~rpifan@p200300d267133a008c87bb5dfd5ca765.dip0.t-ipconnect.de] has joined ##stm32
2024-04-05T17:54:29 -!- rpifan [~rpifan@p200300d267133a008c87bb5dfd5ca765.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-05T17:54:29 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-05T17:58:20 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 252 seconds]
2024-04-05T18:13:47 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-05T18:13:58 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-05T18:32:30 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds]
2024-04-05T18:34:54 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Quit: Konversation terminated!]
2024-04-05T19:02:54 -!- c10ud_ [~c10ud@95.236.174.95] has quit [Quit: Leaving]
2024-04-05T19:06:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T19:07:43 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T19:16:43 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-05T19:17:03 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-9081-64bb-61ac-ea42.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-05T19:22:14 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-05T19:23:45 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 272 seconds]
2024-04-05T19:28:41 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-05T19:31:37 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-05T19:37:55 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-05T20:13:14 -!- rpifan [~rpifan@p200300d267133a00416764fb3efb2727.dip0.t-ipconnect.de] has joined ##stm32
2024-04-05T20:13:14 -!- rpifan [~rpifan@p200300d267133a00416764fb3efb2727.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-05T20:13:14 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-05T20:15:48 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-05T20:18:01 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-05T20:37:50 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds]
2024-04-05T21:01:28 < jbo> zup
2024-04-05T21:01:30 < jbo> zyp
2024-04-05T21:01:39 < jpa-> whatzup
2024-04-05T21:02:01 < jbo> I have a basic USB-C question
2024-04-05T21:02:20 < jbo> if I use a USB-C connector but just need USB 2.0, I need to pull down both CC pins with one 5.1k resistor each, right?
2024-04-05T21:02:31 < jpa-> yes
2024-04-05T21:02:42 < jpa-> (for usb-c female)
2024-04-05T21:03:50 < jbo> well, it's sort of a bisexual thing, isn't it. I am working on the receptacle end. i.e. where you plug a USB-C cable in
2024-04-05T21:04:07 < jbo> it's overall female but has that poky bit in the center that penetrates into the cable connector
2024-04-05T21:04:27 < jpa-> sure sure, it feels a bit nasty but it's female enough
2024-04-05T21:04:33 < jbo> :D
2024-04-05T21:04:37 < jpa-> for usb-c plugs you only need one 5.1k
2024-04-05T21:04:56 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-05T21:04:59 < jbo> aye, so connect CC1 and CC2 together to a 5.1k to GND?
2024-04-05T21:05:26 < jpa-> for plug? only connect one CC1; for jack, separate 5.1k from each CC pin to GND
2024-04-05T21:06:01 < jbo> aye
2024-04-05T21:06:05 < jbo> what do I do with SBU1 and SBU2?
2024-04-05T21:06:29 < jpa-> put something fun onto them, like UART or SWD
2024-04-05T21:06:44 < jbo> hmpf, am I doing plug or jack in your terminology?
2024-04-05T21:06:57 < jpa-> how should i know? jack is female
2024-04-05T21:07:02 < jbo> "female"
2024-04-05T21:07:03 < jbo> :D
2024-04-05T21:09:42 < jbo> I'm just doing a simple FT2232 thingy
2024-04-05T21:09:44 < jbo> so no magic
2024-04-05T21:09:46 < zyp> leave SBU open unless you have a good reason to do otherwise
2024-04-05T21:10:36 < jbo> ack
2024-04-05T21:10:38 < jbo> thanks!
2024-04-05T21:10:41 < zyp> analog audio altmode uses SBU for analog audio, DP altmode uses SBU for DP AUX
2024-04-05T21:10:59 < zyp> those are the two main standard uses
2024-04-05T21:11:04 < jbo> yeah no... just FT2232H on the other end
2024-04-05T21:11:18 < zyp> and then you could use it for a custom altmode if you really wanted to
2024-04-05T21:11:30 < jpa-> JPA altmode uses SBU for programming, and doesn't negotiate about it
2024-04-05T21:11:43 < zyp> e.g. your ecpix-5 can do PCIe over USB-C with REFCLK on SBU
2024-04-05T21:13:03 < jbo> so single ended PCIe?
2024-04-05T21:13:20 < jbo> wait
2024-04-05T21:13:37 < jbo> never mind
2024-04-05T21:14:25 < zyp> no, SBU1 and SBU2 forms a balanced clock signal together
2024-04-05T21:18:29 < jbo> aye
2024-04-05T21:18:31 < jbo> jpa-, how's BQ?
2024-04-05T21:19:38 < jpa-> haven't touched it
2024-04-05T21:20:04 < jbo> :<
2024-04-05T21:27:23 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-05T21:32:07 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-3d1e-9dc1-eeac-4607.fixed6.kpn.net] has joined ##stm32
2024-04-05T21:33:16 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-05T21:48:50 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-05T21:50:11 < Ecco> I'm fighting (and losing) with openocd
2024-04-05T21:50:28 < jbo> Calling Dr. PaulFertser, Dr. PaulFertser to the reception desk please
2024-04-05T21:50:40 < jbo> hmm, I need to work on my PA skills
2024-04-05T21:51:16 < PaulFertser> jbo: hey, I do not have a degree
2024-04-05T21:53:43 < PaulFertser> Ecco: so what's the problem?
2024-04-05T21:55:59 < Ecco> I'm trying to attach and do a reset
2024-04-05T21:56:11 < Ecco> yet it always yields a hard fault:
2024-04-05T21:56:12 < Ecco> > [stm32h7x.cpu0] halted due to debug-request, current mode: Handler HardFault
2024-04-05T21:56:20 < Ecco> > xPSR: 0x01000003 pc: 0xfffffffe msp: 0xfffffffc
2024-04-05T21:56:42 < Ecco> (the MCU is properly probed and recognized tho)
2024-04-05T22:06:50 -!- Livio [~livio@user/livio] has quit [Quit: leaving]
2024-04-05T22:08:20 < PaulFertser> Ecco: it's expected for blank flash
2024-04-05T22:08:40 < PaulFertser> Ecco: the PC and SP match blank flash contents basically.
2024-04-05T22:12:56 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-05T22:33:04 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-05T22:43:36 < Steffanx> lol jbo i read that in this tune even though its not a perfect fit: https://www.youtube.com/watch?v=-1jPUB7gRyg
2024-04-05T22:49:09 < Steffanx> sorry i mean https://www.youtube.com/watch?v=MujkFIBP-ws ofcourse. not sure why i looked for that one. 
2024-04-05T22:55:20 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-05T22:55:51 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-05T22:56:31 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-05T23:11:30 -!- boB_K7IQ [~boB_K7IQ@152.44.147.180] has joined ##stm32
2024-04-05T23:14:06 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 255 seconds]
2024-04-05T23:15:54 -!- Thorn [~Thorn@user/thorn] has joined ##stm32
2024-04-05T23:30:45 -!- rpifan [~rpifan@p200300d267133a006af3f157f5d91740.dip0.t-ipconnect.de] has joined ##stm32
2024-04-05T23:30:45 -!- rpifan [~rpifan@p200300d267133a006af3f157f5d91740.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-05T23:30:45 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-05T23:33:51 < jbo> Steffanx, that was the reference I was trying to get to. but my memory failed.
2024-04-05T23:33:57 < jbo> glad that you picked up on it. :D
2024-04-05T23:34:18 < jbo> although there was another one, somewhere in Scrubs I think
2024-04-05T23:34:19 < jbo> but meh
2024-04-05T23:53:42 < PaulFertser> Scrubs was fun
--- Day changed la huhti 06 2024
2024-04-06T00:00:11 < ventYl> If this set of O-rings won't fit, I will set the bike on fire
2024-04-06T00:00:16 < jbo> hmpf... to QFN or to QFP
2024-04-06T00:00:19 < jbo> always a stupid choice
2024-04-06T00:00:56 < PaulFertser> ventYl: for is it for, the carbs?
2024-04-06T00:01:06 < ventYl> PaulFertser: yeah, last set lasted som 10 month
2024-04-06T00:01:30 < ventYl> then they told me: go and buy chineese repair kit, they fit
2024-04-06T00:01:34 < PaulFertser> So it was some counterfeit rubber?
2024-04-06T00:01:56 < zyp> jbo, doesn't matter unless it does, and in the latter case you know which you want
2024-04-06T00:02:13 < ventYl> well, nothing was fitting probably except of float bowl gasket, probably only because there is nothing you could mismatch it with
2024-04-06T00:02:32 < ventYl> PaulFertser: maybe, also, I selected O-rings based on what was available and seemed to fit
2024-04-06T00:02:53 < ventYl> as it turns out, I've been 0,5mm off, which meant that O-rings I selected were either too thick, or too thin
2024-04-06T00:02:54 < jbo> zyp <3
2024-04-06T00:03:22 < ventYl> thick ones made re-insterting parts back into carb one hell of a job, but were working for >5 years
2024-04-06T00:03:37 < ventYl> thin ones did not last even one full season
2024-04-06T00:03:54 < zyp> jbo, if you're going for compactness, you want qfn, otherwise qfp is typically easier to route since there's not a center pad in the way
2024-04-06T00:04:06 < PaulFertser> I only ride a push bike now...
2024-04-06T00:05:06 < zyp> easier to route in the sense that you've got the option to fan out pads both outwards and inwards
2024-04-06T00:05:14 < ventYl> I wouldn't like to push mine way too far
2024-04-06T00:05:24 < ventYl> Once I had to push it like one km away
2024-04-06T00:12:38 < jbo> is FTDI retarded or am I retarded?
2024-04-06T00:12:39 < jbo> https://ftdichip.com/wp-content/uploads/2024/03/DS_FT2232H.pdf
2024-04-06T00:12:43 < jbo> page 53
2024-04-06T00:13:02 < jbo> uses a 93C46 EEprom which according to it's datasheet has a Vcc range of 4.5V to 5V
2024-04-06T00:13:07 < jbo> and FTDI is powering it from 3.3V?
2024-04-06T00:13:38 < zyp> idk
2024-04-06T00:13:39 < zyp> https://www.st.com/resource/en/datasheet/m93c46-w.pdf
2024-04-06T00:14:03 < jbo> https://www.mouser.ch/datasheet/2/268/20001749K-277859.pdf
2024-04-06T00:14:56 < zyp> so what it means is that you're reading «93C46» too literally
2024-04-06T00:17:47 < jbo> hmpf
2024-04-06T00:18:28 < zyp> the ftdi doesn't care what brand or voltage your eeprom is, as long as it has a 93C46-compatible command set
2024-04-06T00:18:56 < jbo> I guess that makes sense
2024-04-06T00:22:13 < qyx> is it 2000 or why are we using 93c46
2024-04-06T00:22:33 < qyx> or I am a bit off
2024-04-06T00:22:48 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-06T00:24:32 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-06T00:24:44 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-06T00:33:46 < qyx> https://github.com/yaml/pyyaml/issues/724
2024-04-06T00:33:47 < qyx> really
2024-04-06T00:33:55 < qyx> I just segfaulted
2024-04-06T00:47:46 < qyx> another one https://github.com/actions/checkout/issues/217
2024-04-06T00:48:25 < qyx> that's because of deciding to do a bit of modern ci/cd
2024-04-06T00:51:36 < zyp> had to fix that pulling tags thing for orbtrace
2024-04-06T00:54:11 < qyx> did with: all_tags:true work?
2024-04-06T00:54:24 < zyp> https://github.com/orbcode/orbtrace/commit/7d5d3416d17ef8f06907566788b923d5ebccbea9 <- I ended up doing this so it can also figure out those tagged version + n commits tags for intermediate shit
2024-04-06T00:56:45 < qyx> hm yeah I need that too
2024-04-06T00:56:54 < qyx> if the command is legit I would steal it
2024-04-06T00:57:04 < zyp> it works for me, steal away
2024-04-06T00:58:37 < zyp> I don't remember what --prune is useful for, --unshallow is the one that grabs the rest of the history and I don't care about the submodules' history so I save time by skipping that
2024-04-06T01:14:07 < qyx> any idea how to setup a upstream tracking branch without checking it out?
2024-04-06T01:14:21 < qyx> that is git branch develop && git branch -u origin/develop develop
2024-04-06T01:14:22 < qyx> combined
2024-04-06T01:14:54 < specing> git branch --set-upstream-to=repo/branch develop
2024-04-06T01:15:23 < qyx> I was eyeballing this option and somewhat I didn't believe it is what I want
2024-04-06T01:17:14 < qyx> yeah it doesn't work
2024-04-06T01:17:23 < qyx> it complains the branch doesn't exist
2024-04-06T01:17:41 < qyx> --set-upstream-to is the same as -u I mentioned
2024-04-06T01:18:56 < zyp> why do you need that?
2024-04-06T01:20:02 < zyp> personally I don't tend to care much about what's tracking or not, most of the time it manages to set it up itself when I want it and the rest of the time I'm just explicit about what I want
2024-04-06T01:21:10 < qyx> my version script computes feature branch "distances" from the develop branch
2024-04-06T01:21:34 < qyx> and I am lazy to fix it the proper way, it needs a local develop branch
2024-04-06T01:23:27 < zyp> git branch develop origin/develop
2024-04-06T01:24:02 < zyp> that looks like it should set up tracking also
2024-04-06T01:24:29 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-3d1e-9dc1-eeac-4607.fixed6.kpn.net] has quit [Ping timeout: 240 seconds]
2024-04-06T01:41:14 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-06T01:42:49 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-06T01:44:16 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-06T01:45:38 < qyx> hm scons behaves differently when run from the github runner and locally on the runner
2024-04-06T01:50:45 < qyx> it misses a build step, two actually
2024-04-06T01:50:49 < qyx> also in logs
2024-04-06T01:51:45 < zyp> bikeshed question: in smolt, I'm gonna add a function to write a null token to the log transport, with the intended purpose of signalling the start of a logging session
2024-04-06T01:52:44 < zyp> for the ITM receiver, receiving a null token will act as a trigger to check if the ELF changed and reload the tokens if so
2024-04-06T01:53:16 < qyx> I would not call it a null token if it is a start token
2024-04-06T01:53:19 < zyp> and for a ringbuffer, a null token will effectively be interpreted as the beginning of the ringbuffer, i.e. it won't look further back in the buffer
2024-04-06T01:53:35 < zyp> by null, I mean literally a tag id value of 0
2024-04-06T01:54:01 < zyp> so the question is, what should I name this function?
2024-04-06T01:54:39 < zyp> my head suggested flush(), but it's not really what you'd normally expect a flush to do
2024-04-06T01:54:48 < qyx> reset?
2024-04-06T01:55:29 < zyp> don't think I like that either
2024-04-06T01:56:34 < zyp> I want something that sounds like a natural thing to do at the beginning of main() or whatever
2024-04-06T01:58:40 < zyp> also, the reason I'm solving elf reload this way instead of e.g. watching the elf with inotify or something is because I don't want it to reload as soon as you've rebuilt the elf, because the target doesn't start using the new tag ids before you've flashed and restarted it
2024-04-06T01:59:26 < zyp> so having the target explicitly tell me when it has restarted seems reasonable, and the zero id isn't used for anything else
2024-04-06T02:05:55 < qyx> this is insane, git safe_directory
2024-04-06T02:11:06 -!- rpifan [~rpifan@user/rpifan] has quit [Quit: Leaving]
2024-04-06T02:14:33 < BrainDamage> zyp: init?
2024-04-06T02:17:09 < qyx> scons is ignoring ALLLL my command line options
2024-04-06T02:17:19 < qyx> why?
2024-04-06T02:18:05 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 268 seconds]
2024-04-06T02:19:03 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T02:20:16 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T02:20:55 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T02:21:28 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T02:29:19 < qyx> if I run scons in a directory with a SConscript, it starts to do weird things and doesn't accept ANY arguments
2024-04-06T02:29:34 < qyx> on a different installation, it runs just fine
2024-04-06T02:29:46 < qyx> *SConstruct
2024-04-06T02:43:59 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-06T02:53:51 -!- boB_K7IQ [~boB_K7IQ@152.44.147.180] has quit [Ping timeout: 268 seconds]
2024-04-06T02:58:45 < zyp> hmm, the RTC in this thing supports both BCD calendar mode and free running binary counter mode
2024-04-06T02:58:51 < zyp> is that new?
2024-04-06T02:59:43 < zyp> yeah. how new?
2024-04-06T03:00:26 < zyp> g4 doesn't do that, so newer than g4
2024-04-06T03:02:36 < zyp> this thing = h7r/s, older h7 doesn't do that, h5 does
2024-04-06T03:03:31 < zyp> u5 also does
2024-04-06T03:08:13 < qyx> good to know
2024-04-06T03:08:41 < zyp> yeah, removes the issue you were complaining about the other day
2024-04-06T03:08:53 < qyx> omg dumb be I had .downloaded.stamp in my git repo..
2024-04-06T03:18:13 < qyx> I got it compiling on the 27th run
2024-04-06T03:19:09 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-06T03:33:09 < qyx> I really want u5/h5 board but I struggle with that
2024-04-06T05:31:27 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-06T05:47:28 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T05:47:58 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T05:49:52 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T05:50:25 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T07:31:38 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-06T07:40:26 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-06T09:32:57 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-5834-4170-91c8-9447.fixed6.kpn.net] has joined ##stm32
2024-04-06T10:11:33 < Steffanx> Uh why qyx? 
2024-04-06T10:29:55 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-06T11:07:02 < qyx> because I am lazy
2024-04-06T11:10:55 < Steffanx> Oh. 
2024-04-06T11:53:47 < qyx> this week I spent 35 hours fixing my broken code, doing code to fix the broken code, doing code to deal with the unfixed broken code and doing github's ci/cd in a blind faith to catch the rest of my broken code
2024-04-06T11:59:16 < Steffanx> Sounds like a good productive week
2024-04-06T12:12:04 < PaulFertser> Ecco: ping
2024-04-06T12:13:30 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-06T12:18:00 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-06T12:28:24 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has joined ##stm32
2024-04-06T12:32:52 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-06T12:59:08 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-06T13:22:58 < karlp> zyp: restart()?
2024-04-06T13:26:51 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T13:28:22 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T13:29:55 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-06T13:33:16 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T13:33:47 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-06T13:33:48 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T13:34:41 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T13:35:18 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T14:02:09 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T14:02:45 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T14:05:12 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-06T14:25:08 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-06T14:35:01 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-06T15:32:14 < Ecco> PaulFertser: pong? Sorry for the huge lag
2024-04-06T15:32:50 < Ecco> What you said made a lot of sense, I'm trying to figure out if that's indeed the problem
2024-04-06T15:46:51 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-06T16:04:17 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-5834-4170-91c8-9447.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-06T16:23:01 < PaulFertser> Ecco: at least there's no problem in your output. Probably there's no problem at all with anything?
2024-04-06T16:24:54 < Ecco> Yeah, I think you're right
2024-04-06T16:35:23 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-06T17:25:59 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-06T17:59:19 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-06T18:01:30 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 256 seconds]
2024-04-06T18:11:44 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-06T18:21:31 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-06T18:30:14 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-06T18:30:18 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-06T18:31:03 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 256 seconds]
2024-04-06T18:32:22 -!- duude__- is now known as duude__
2024-04-06T18:36:02 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-06T18:56:16 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-06T18:57:21 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-06T19:01:02 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-156e-cc1c-3f4e-dce6.fixed6.kpn.net] has joined ##stm32
2024-04-06T19:29:23 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit [Ping timeout: 264 seconds]
2024-04-06T19:33:49 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-06T19:53:19 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-06T20:09:05 < nomorekaki> https://youtu.be/QQhQ7yXogwo&t=398 using pcb as an indicator screen
2024-04-06T20:10:29 < nomorekaki> also some surface texture seems to be added to copper layer
2024-04-06T20:17:55 < sauce> i managed to avoid hearing that voice for at least a year and you had to go and ruin it
2024-04-06T20:21:22 -!- Livio [~livio@user/livio] has quit [Read error: Connection reset by peer]
2024-04-06T20:24:22 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-06T20:44:13 < Steffanx> You just lost the game.
2024-04-06T21:26:56 -!- boB_K7IQ [~boB_K7IQ@c-67-161-97-46.hsd1.wa.comcast.net] has joined ##stm32
2024-04-06T21:40:53 -!- boB_K7IQ [~boB_K7IQ@c-67-161-97-46.hsd1.wa.comcast.net] has quit [Ping timeout: 256 seconds]
2024-04-06T21:50:58 -!- boB_K7IQ [~boB_K7IQ@c-67-161-98-154.hsd1.wa.comcast.net] has joined ##stm32
2024-04-06T21:54:10 -!- F0olz [~F0olz@2a09:7c44:0:2f::1] has joined ##stm32
2024-04-06T21:55:18 < F0olz> Hi there! I've been reading the reference manual, and something is not quite clear to me: if I mess up the unlocking of the flash memory by writing an incorrect key, is there any way to recover from that without removing the power to my board?
2024-04-06T21:57:37 < zyp> I believe fucking up the unlock sequence locks it until the next reset
2024-04-06T21:57:55 < zyp> not power-on reset, just any reset, so you don't have to powercycle it
2024-04-06T21:58:40 < zyp> but I've never ended up in that situation that I can remember, so I haven't tested it in practice
2024-04-06T22:00:20 < F0olz> wait, there are multiple reset? Where can I learn about those?
2024-04-06T22:00:36 < zyp> the reference manual
2024-04-06T22:00:48 < zyp> which chip are you using?
2024-04-06T22:01:53 < F0olz> STM32H755
2024-04-06T22:02:25 < F0olz> Oh, indeed, there are even domain resets
2024-04-06T22:03:08 < zyp> https://www.st.com/resource/en/reference_manual/rm0399-stm32h745755-and-stm32h747757-advanced-armbased-32bit-mcus-stmicroelectronics.pdf <- page 356 has a decent overview over system reset sources
2024-04-06T22:06:13 < F0olz> sweet. So it looks like it be possible to somehow trigger a d1_rst to reset the Flash memory without resetting the entire CPU, right? or am I reading this wrong?
2024-04-06T22:07:09 < zyp> no
2024-04-06T22:07:30 < F0olz> oh, why?
2024-04-06T22:07:57 < zyp> hmm, hang on, this reference manual doesn't have the same description of the FLASH_KEYR register as the previous one I looked at
2024-04-06T22:08:20 < zyp> ah, see page 189
2024-04-06T22:08:34 < zyp> «Any wrong sequence locks up the corresponding register/bit until the next system reset»
2024-04-06T22:08:38 < zyp> *system* reset
2024-04-06T22:08:56 < zyp> that's the one that resets both cpus and more or less all peripherals
2024-04-06T22:09:59 < F0olz> hmm, but how would the flash know what kind of reset is happening?
2024-04-06T22:10:10 < F0olz> I mean, on the block diagram, they show d1_rst as an input, and that's it
2024-04-06T22:10:25 < F0olz> so *maybe* it's possible? (Even though the doc doesn't explicitely says so)?
2024-04-06T22:10:55 < zyp> which block diagram?
2024-04-06T22:11:49 < zyp> ah, figure 5
2024-04-06T22:12:33 < F0olz> yeah
2024-04-06T22:12:58 < F0olz> oh, wait, it also mentions po_rst
2024-04-06T22:13:07 < F0olz> so no, you're most likely right
2024-04-06T22:13:08 < zyp> hmm, yeah, flash appears to be in D1, so I guess a D1 reset should do it
2024-04-06T22:13:37 < zyp> unless the documentation is lying and a poweron reset is actually required
2024-04-06T22:14:09 < zyp> this isn't really an issue outside of the dualcore chips
2024-04-06T22:14:21 < F0olz> what's the difference between a system reset, a domain reset, and a power-on reset?
2024-04-06T22:15:32 < zyp> a power on reset can only be triggered by a power cycle, a system reset can also be triggered by the NRST pin or the SYSRESETREQ bit
2024-04-06T22:16:27 < zyp> and a domain reset is like a partial system reset, since in a dualcore system you might want to reset one cpu and not the other
2024-04-06T22:19:50 < F0olz> crystal clear. thanks!
2024-04-06T22:20:33 < zyp> watchdogs and stuff will also do a system reset
2024-04-06T22:22:46 -!- boB_K7IQ [~boB_K7IQ@c-67-161-98-154.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds]
2024-04-06T22:28:55 -!- qyx [~qyx@84.245.120.146] has quit [Ping timeout: 256 seconds]
2024-04-06T22:30:46 -!- qyx [~qyx@84.245.121.3] has joined ##stm32
2024-04-06T22:47:24 < nomorekaki> hello early
2024-04-06T22:57:37 < nomorekaki> is there ideal pcb material for constant thermal cycling
2024-04-06T22:58:46 < ventYl> hello kaki
2024-04-06T22:59:05 < nomorekaki> ventYl hi
2024-04-06T22:59:30 < nomorekaki> cyclic heating and cooling should I worry too much?
2024-04-06T23:00:24 < ventYl> I would probably worry more for solder than for the PCB
2024-04-06T23:00:38 < nomorekaki> yes
2024-04-06T23:00:52 < ventYl> like, it will suffer, but take a look on cars. interior can heat up to some 80 degrees if stationary on direct sun and go down to ~15 or even less during the night
2024-04-06T23:00:55 < nomorekaki> but is there anything to gain by changing pcb material?
2024-04-06T23:01:10 < ventYl> and they withstand more than 3000 such cycles before things go wrong
2024-04-06T23:01:24 < nomorekaki> how about SMD sizes?
2024-04-06T23:01:37 < nomorekaki> is that why cars use 1206 resistors and shit?
2024-04-06T23:02:19 < ventYl> they use a lot of "legs" rather than BGA or pads wherever possible
2024-04-06T23:02:27 < ventYl> to counter vibrations
2024-04-06T23:02:41 < nomorekaki> big QFPs
2024-04-06T23:02:52 < nomorekaki> no QFNs
2024-04-06T23:02:55 < ventYl> they are probably big because shit has a lot of in and out signals
2024-04-06T23:03:44 < nomorekaki> yes
2024-04-06T23:03:49 < nomorekaki> voltage and power rating
2024-04-06T23:03:50 -!- F0olz [~F0olz@2a09:7c44:0:2f::1] has quit [Quit: Client closed]
2024-04-06T23:04:04 < nomorekaki> surges, even continuos shorts
2024-04-06T23:04:30 < nomorekaki> but even for internal signaling there is no 0402
2024-04-06T23:05:16 < nomorekaki> that must have to do with mechanical reliability and stuff
2024-04-06T23:05:25 < ventYl> probably
2024-04-06T23:05:38 < nomorekaki> or potential oxidation not bridging it in a day or so
2024-04-06T23:06:40 < ventYl> oxidation is usually dealt with by not allowing moisture to enter the enclosure
2024-04-06T23:06:51 < ventYl> unless you are a complete moron, which some of people in automotive are
2024-04-06T23:14:00 < nomorekaki> why not make the electronics into a bucket where all the water pours once a drain gets clogged?
2024-04-06T23:14:31 < nomorekaki> actual thing in e60 iirc
2024-04-06T23:24:57 < ventYl> some of them do it. audi has a4 where circuit breakers are in poorly sealed comparment, so older cars have this "water cooled"
2024-04-06T23:25:10 < ventYl> peugeot also knew how to water cool their ECUs
2024-04-06T23:29:41 < nomorekaki> sometimes fire heated fuse box
2024-04-06T23:30:36 < nomorekaki> e60 fire heated battery main positive cable
2024-04-06T23:30:49 < nomorekaki> recalled and fixed though
2024-04-06T23:32:43 < ventYl> aaand some GMs also can have water cooled ECUs, because 0.02 silicone watertight sleeve is too expensive
2024-04-06T23:35:36 < nomorekaki> yeah never buying GM car ever I think
2024-04-06T23:38:38 < nomorekaki> don't know enough, but I know quality and reliability went into shit at some point when too much cost shaving occured in such things
2024-04-06T23:46:27 < nomorekaki> https://drive.google.com/file/d/1HRo4HtOCxDUPGCa6DhL0tiSejAhxLiBE/view?usp=drive_link making ds1821 emulator - or just a thermostat if I'm lazy. needs to fit in to-220 position in a heat sink
2024-04-06T23:48:41 -!- alan_o [~alan_o@2600:1700:1902:210f:48c0:1433:3ee:a7e7] has quit [Remote host closed the connection]
2024-04-06T23:49:00 -!- alan_o [~alan_o@2600:1700:1902:210f:6905:e17d:2044:5d0e] has joined ##stm32
2024-04-06T23:56:51 < Steffanx> Fancy. Why emulate it? 
2024-04-06T23:58:26 < Steffanx> And I bet that's an attiny..
2024-04-06T23:59:22 < nomorekaki> ds1821 is obsolete
2024-04-06T23:59:52 < nomorekaki> like buy them from ebay from only one seller type of obsolete
--- Day changed su huhti 07 2024
2024-04-07T00:04:50 < Steffanx> Oh I see.
2024-04-07T00:07:14 < Steffanx> Will you even go STM32 or will you wait till they're obsolete, nomorekaki?
2024-04-07T00:07:43 < nomorekaki> that's gonna take a while
2024-04-07T00:08:04 < nomorekaki> im not in hurry to use stm32
2024-04-07T00:08:32 < nomorekaki> to me it's just yet another cortex mcu
2024-04-07T00:09:23 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-07T00:09:37 < Steffanx> Yeah you should totally go with renesas.
2024-04-07T00:09:44 < Steffanx> Much more automotive 
2024-04-07T00:09:48 < nomorekaki> yes
2024-04-07T00:09:55 < nomorekaki> I have not tried that
2024-04-07T00:10:05 < nomorekaki> it's tempting
2024-04-07T00:10:41 < nomorekaki> but why use renesas cortex chip when they have the chips automotive clients actually use?
2024-04-07T00:10:49 < Steffanx> Oh dear are you sure?
2024-04-07T00:11:56 < nomorekaki> friend was reversing some toyota nissan ev driveline stuff and it was really exotic
2024-04-07T00:12:17 < nomorekaki> debugging probe and stuff was also really expensive
2024-04-07T00:12:22 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-07T00:13:43 < nomorekaki> didn't buy debug emulator development stuffs though had it reversed enough for his purposes to throw out the chip and replace it with arduino or something
2024-04-07T00:14:04 < nomorekaki> arduino with arm cortex
2024-04-07T00:26:20 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-07T00:30:33 -!- srk [~sorki@user/srk] has quit [Read error: Connection reset by peer]
2024-04-07T00:33:51 -!- srk [~sorki@user/srk] has joined ##stm32
2024-04-07T01:00:30 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-07T01:01:00 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-07T01:07:31 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-156e-cc1c-3f4e-dce6.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-07T01:27:25 < zyp> karlp, thanks for the suggestion, I ended up just making a metadata token named start_of_stream, used like logger.log<start_of_stream>();
2024-04-07T02:01:42 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has joined ##stm32
2024-04-07T02:03:49 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 268 seconds]
2024-04-07T02:11:43 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-07T02:15:34 -!- ventYl [~ventyl@adsl-dyn176.78-99-153.t-com.sk] has quit [Ping timeout: 264 seconds]
2024-04-07T02:23:08 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 252 seconds]
2024-04-07T02:27:33 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has quit [Quit: Bye]
2024-04-07T04:19:01 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Quit: Client closed]
2024-04-07T06:05:41 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Ping timeout: 240 seconds]
2024-04-07T07:17:52 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-07T07:26:05 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has quit [Ping timeout: 240 seconds]
2024-04-07T07:28:26 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has joined ##stm32
2024-04-07T08:05:29 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-07T08:06:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-07T09:00:08 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-448f-3301-62c7-faed.fixed6.kpn.net] has joined ##stm32
2024-04-07T09:34:54 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-07T10:56:40 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-07T11:17:11 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has joined ##stm32
2024-04-07T12:22:29 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-448f-3301-62c7-faed.fixed6.kpn.net] has quit [Ping timeout: 240 seconds]
2024-04-07T12:28:49 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-07T12:52:54 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-07T12:55:02 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-07T13:45:35 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-07T14:00:57 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-07T14:04:23 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-07T14:05:45 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 256 seconds]
2024-04-07T14:07:46 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-07T14:09:01 -!- duude__- [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-07T14:13:41 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 256 seconds]
2024-04-07T14:29:29 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-07T15:02:20 -!- ventYl [~ventyl@adsl-dyn36.78-98-110.t-com.sk] has joined ##stm32
2024-04-07T15:04:15 -!- ventYl [~ventyl@adsl-dyn36.78-98-110.t-com.sk] has quit [Client Quit]
2024-04-07T15:04:28 -!- ventYl [~ventyl@adsl-dyn36.78-98-110.t-com.sk] has joined ##stm32
2024-04-07T15:15:53 -!- con3 [~con3@164.90.228.156] has quit [Ping timeout: 256 seconds]
2024-04-07T16:40:27 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has quit [Quit: Bye]
2024-04-07T17:21:45 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-07T18:18:19 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-07T18:19:55 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-653b-d9bc-15c8-5576.fixed6.kpn.net] has joined ##stm32
2024-04-07T18:26:23 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-07T18:32:20 < antto> https://i.imgur.com/UrQeQ06.png
2024-04-07T18:48:52 < jbo> so dusty
2024-04-07T18:52:14 < Steffanx> Heh you switched to addressable LEDs 🎉
2024-04-07T18:52:34 < antto> yes x_x
2024-04-07T18:52:57 < antto> that shortened my mouse's life in half ;P~
2024-04-07T18:54:46 < jbo> I'm glad I'm not using mice anymore
2024-04-07T19:03:06 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-07T19:08:24 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Read error: Connection reset by peer]
2024-04-07T19:08:52 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-07T19:12:25 -!- BrainDamage [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-07T19:34:37 < Steffanx> What's your alternative mr jbo ? Trackballish thing?
2024-04-07T19:35:06 < Steffanx> neuralink?
2024-04-07T19:35:42 < jbo> trackball :)
2024-04-07T19:36:42 < Steffanx> which one is your favourite?
2024-04-07T19:39:22 < jbo> Logitech MX Ergo
2024-04-07T19:42:14 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-07T19:51:20 < Steffanx> I see
2024-04-07T19:51:38 < Steffanx> I expected something more fancy
2024-04-07T19:59:01 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has joined ##stm32
2024-04-07T20:00:28 < nomorekaki> Steffanx: https://drive.google.com/file/d/1HRo4HtOCxDUPGCa6DhL0tiSejAhxLiBE/view?usp=sharing
2024-04-07T20:02:12 < nomorekaki> I could make it way smaller
2024-04-07T20:03:02 < nomorekaki> keepouts are not even colliding
2024-04-07T20:03:21 < nomorekaki> *courtyards
2024-04-07T20:04:50 < Steffanx> is it actually temperature sensoring or is it just faking it?
2024-04-07T20:06:53 < nomorekaki> yeah attiny406 has internal temperature as adc channel
2024-04-07T20:10:11 < nomorekaki> for the application attiny + 100n + 2n700x would be sufficient, everything else is to support modeswitching to 1wire
2024-04-07T20:11:19 < nomorekaki> and slew rate control for 1wire operation has couple of components and mosfet reset state has one resistor
2024-04-07T20:14:33 < Steffanx> and the huge arse looking like a resistor part?
2024-04-07T20:16:02 < nomorekaki> modeswitching. you need to drive against whatever state the thermostat mode has
2024-04-07T20:16:36 < nomorekaki> also it has been specified in ds1821 datasheet to be 100ohm mosfet
2024-04-07T20:22:07 < nomorekaki> 100ohm and 5volts equals to 1206 package
2024-04-07T20:24:07 < nomorekaki> I'm designing smaller and smaller projects :O
2024-04-07T20:24:38 < nomorekaki> trying to find my level
2024-04-07T20:28:33 < nomorekaki> I doubt I get to actually implementing the 1wire mode will see. most likelly just program it with hardcoded temperature, polarity and hysteresis
2024-04-07T20:33:05 < nomorekaki> I wonder if I can space components just by copper to copper clearance
2024-04-07T20:34:33 < nomorekaki> never had hit the limit altough I have usually done assembly by shaking the board and tweezers with my hands until everything is in place
2024-04-07T20:36:37 < nomorekaki> for PCBA orders I have intentionally placed parts looselly in a sort of grid patterns
2024-04-07T20:45:54 < Steffanx> Are you going to solder headers to those pads? Why not just go with holes instead?
2024-04-07T20:48:30 < jbo> Steffanx, no fancyness required
2024-04-07T20:55:46 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 264 seconds]
2024-04-07T20:58:03 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-07T21:00:01 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-653b-d9bc-15c8-5576.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-07T21:00:18 < Steffanx> I just expected something different, jbo . Like those with the ball pretty much in the middle. Not the thumb controlled ball.
2024-04-07T21:10:48 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-07T21:12:51 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-07T21:12:55 < Laurenceb_> chatgpt knows
2024-04-07T21:13:01 < Laurenceb_> "Chris Chan, also known as Christian Weston Chandler, has had a colorful and often tumultuous love life. Let’s delve into some of the key figures in Chris’s romantic escapades:"
2024-04-07T21:14:36 < BrainDamage> unfortunately for us, we're stuck with LaurencebGPT
2024-04-07T21:18:08 < qyx> my sides
2024-04-07T21:43:49 -!- scrts [~scrts2@23.28.144.38] has quit [Ping timeout: 268 seconds]
2024-04-07T21:50:47 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-07T22:10:07 < Laurenceb_> >ur a virgin but CWC is a smelly, vile, repulsive literal motherfucker who nevertheless managed to lose his virginity and has a girlfriend
2024-04-07T22:10:48 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-07T22:12:59 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-07T22:39:17 -!- qyx [~qyx@84.245.121.3] has quit [Ping timeout: 240 seconds]
2024-04-07T22:40:14 -!- qyx [~qyx@84.245.120.233] has joined ##stm32
2024-04-07T22:47:47 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-653b-d9bc-15c8-5576.fixed6.kpn.net] has joined ##stm32
2024-04-07T22:48:56 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-07T22:49:21 -!- qyx [~qyx@84.245.120.233] has quit [Ping timeout: 255 seconds]
2024-04-07T22:51:03 -!- qyx [~qyx@84.245.121.38] has joined ##stm32
2024-04-07T22:51:11 -!- boB_K7IQ [~boB_K7IQ@c-76-135-53-111.hsd1.wa.comcast.net] has quit []
2024-04-07T23:03:10 < nomorekaki> Steffanx: same reason it has no vias
2024-04-07T23:03:39 < Steffanx> And why is that?
2024-04-07T23:04:05 < nomorekaki> its pressed to heat sink by clamp
2024-04-07T23:04:11 < Steffanx> Hm
2024-04-07T23:04:21 < nomorekaki> I just need to figure out how clamp from top surface
2024-04-07T23:04:31 < Steffanx> Was about to ask 
2024-04-07T23:05:02 < nomorekaki> epoxy maybe
2024-04-07T23:05:18 < nomorekaki> all the components are pretty much same height around 0.65mm
2024-04-07T23:05:23 < nomorekaki> or less
2024-04-07T23:06:50 < nomorekaki> no 0.85mm for mcu and everything else is less
2024-04-07T23:11:20 < nomorekaki> thought about placing components into islands and then have another pcb with holes for them
2024-04-07T23:12:52 -!- nuxil_ is now known as nuxil
2024-04-07T23:16:07 < nomorekaki> they really like to prob mill such pcb that is basically a frame with 3mm width edges
2024-04-07T23:17:59 < nomorekaki> maybe I could order a pcb with special routing steff
2024-04-07T23:18:24 < nomorekaki> kinda height map thing
2024-04-07T23:20:21 < nomorekaki> just 3d print a lid with indents for parts and fill with epoxy and press lid on
2024-04-07T23:25:01 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-07T23:28:06 < nomorekaki> I have 2comp polyurethane it might be good
2024-04-07T23:28:17 < Steffanx> That would probably work
2024-04-07T23:28:55 < nomorekaki> it's flexible and very good adhesion
2024-04-07T23:29:41 < nomorekaki> flexible means it wont break components
2024-04-07T23:30:52 < Steffanx> How are you going to program the thing?
2024-04-07T23:31:04 < nomorekaki> see the testpoint?
2024-04-07T23:31:27 < nomorekaki> thats reset/debug
2024-04-07T23:31:57 < Steffanx> Tinies do all over a single pin?
2024-04-07T23:32:09 < nomorekaki> nowdays yes
2024-04-07T23:32:15 < nomorekaki> including debugging
2024-04-07T23:33:10 < Steffanx> Also can't you make the pcb a tiny bit bigger, make the solder pads smaller. Rearrange shit and add the hole and use the gold old screw method? Use tented vias or otherwise some non electronically conductive pad.
2024-04-07T23:34:06 < Steffanx> Seems like your pcb is a tiny bit less wide than the to220
2024-04-07T23:36:38 < nomorekaki> true
2024-04-07T23:37:13 < ColdKeyboard> Anyone knows what kind of communication is going on between the "label maker" and the laser that is on the board?
2024-04-07T23:37:34 < ColdKeyboard> The one I have open has 10 pins and I can see two of them marked as TX and RX
2024-04-07T23:39:00 < Steffanx> Sniff it with your scope/logic analyzer and make a guess based on the signals :)
2024-04-07T23:39:41 < nomorekaki> what is "label maker" if not label maker?
2024-04-07T23:43:15 < ColdKeyboard> Good question; the control board of the label maker :)
2024-04-07T23:43:55 < nomorekaki> tape type?
2024-04-07T23:44:08 < ColdKeyboard> yeah
2024-04-07T23:44:49 < nomorekaki> which model and why you are trying to interface the laser?
2024-04-07T23:45:09 < ColdKeyboard> I have one where keyboard, screen and battery clip are busted but everything else works fine.
2024-04-07T23:45:37 < ColdKeyboard> I want to play around and maybe see if I can drive it from a micro and have some fun with it :)
2024-04-07T23:46:13 < ColdKeyboard> It's this one -> https://www.bhphotovideo.com/images/images1500x1500/brother_pth110_pt_h110_portable_label_maker_1305563.jpg
2024-04-07T23:46:22 < nomorekaki> https://www.youtube.com/watch?v=ag3opk4ae74 boy pranking folk on the street
2024-04-07T23:52:18 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-07T23:53:30 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-07T23:54:26 < nomorekaki> Steffanx: I'm going to make it narrower
2024-04-07T23:54:57 < Steffanx> Why?
2024-04-07T23:56:19 < nomorekaki> place for it is at best the width of to220
2024-04-07T23:57:17 < Steffanx> What's the orientation the heatsink flat on the pcb. Or.. ?
2024-04-07T23:58:45 < Steffanx> I mean you can extend your pcb down and use through hole pins right? I mean.. if the to220 was flat on the pcb/heat sink the pins have to go into the pcb as well.
2024-04-07T23:58:45 < nomorekaki> vertical
2024-04-07T23:59:00 < Steffanx> Oh 
2024-04-07T23:59:04 < nomorekaki> vertical made for to220s in row
--- Day changed ma huhti 08 2024
2024-04-08T00:01:44 < Steffanx> Would still try something like this, but I might be missing some important detail https://usercontent.irccloud-cdn.com/file/yMmvqUNA/1000018571.webp
2024-04-08T00:03:49 < nomorekaki> heat sink has no holes
2024-04-08T00:03:56 < nomorekaki> it has spring clamps
2024-04-08T00:05:34 < Steffanx> Oh yes. That's one detail i forgot already 😅
2024-04-08T00:13:58 < nomorekaki> minimum pcb size for jlc is 5x5mm
2024-04-08T00:20:53 < nomorekaki> steff I left you nice laughts in the link above
2024-04-08T00:22:00 < nomorekaki> lil dude is a menace
2024-04-08T00:44:20 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 260 seconds]
2024-04-08T00:46:45 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-08T00:52:03 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-08T01:07:55 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-08T01:18:08 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 252 seconds]
2024-04-08T01:19:47 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-08T01:24:04 < qyx> k I githubbed enough for today
2024-04-08T01:24:05 < ColdKeyboard> This is quite fun. So the "laser" module in the label maker has TX and RX lines. RX is low for ~200ns for every X mm of tape. And then TX sends some binary/line data to actuate laser (I assume)
2024-04-08T01:25:30 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-08T01:34:26 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-08T02:07:41 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-653b-d9bc-15c8-5576.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-08T02:50:39 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-08T04:47:47 -!- nomorekaki [~nomorekak@37-136-135-97.rev.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-08T06:03:51 -!- qyx [~qyx@84.245.121.38] has quit [Ping timeout: 256 seconds]
2024-04-08T06:05:49 -!- qyx [~qyx@84.245.120.220] has joined ##stm32
2024-04-08T07:10:09 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-08T09:15:37 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-08T09:23:26 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-08T09:24:00 < qyx> In 2020, Mbed TLS joined the TrustedFirmware project.[7] 
2024-04-08T09:24:01 < qyx> noooo
2024-04-08T09:24:30 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T09:32:31 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-08T09:34:22 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T09:41:41 -!- Livio [~livio@user/livio] has quit [Ping timeout: 240 seconds]
2024-04-08T09:51:42 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-08T09:52:38 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-08T10:08:50 -!- Thorn [~Thorn@user/thorn] has quit [Quit: Depression is merely anger without enthusiasm]
2024-04-08T10:37:19 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-08T11:46:44 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Ping timeout: 256 seconds]
2024-04-08T11:51:15 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Remote host closed the connection]
2024-04-08T12:39:43 < karlp> did githubbing successfully help you forget about elections?
2024-04-08T12:40:37 < qyx> fuk elections
2024-04-08T12:40:52 < qyx> and the whole country
2024-04-08T12:43:10 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-08T12:45:08 < Steffanx> Lol such rage
2024-04-08T12:46:13 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-08T12:47:26 < karlp> it's rage or apathy.
2024-04-08T12:47:39 < karlp> and apathy makes me angry at myself
2024-04-08T12:47:44 < karlp> so... rage or rage.
2024-04-08T13:27:29 < karlp> fuck convenntionalcommits.org
2024-04-08T13:30:04 < ventYl> why? it looks reasonable
2024-04-08T13:30:11 < karlp> mostly.
2024-04-08T13:30:17 < karlp> but the fix prefix is bullshit
2024-04-08T13:30:33 < karlp> I want to have components, not just fix spewed allover the first line.
2024-04-08T13:30:51 < ventYl> I always wondered who it is possible that my chineese coworker was OK with compiling changelog for half a day
2024-04-08T13:30:52 < karlp> giving up human parsing to nominally make machine parsing work better, with shitty machine parsing heuristics scripts
2024-04-08T13:31:23 < karlp> if you think that autoextracting fix() lines from gitc ommits magically gives you a useful changelog, you're out of your goddamn mind.
2024-04-08T13:31:54 < karlp> fucking hooks also just broke my commit
2024-04-08T13:32:10 < karlp> try to edit the text for them, git commit --amend, it fails, and throws it away
2024-04-08T13:32:16 < karlp> so git commit --amend again is blank slate again.
2024-04-08T13:32:40 < karlp> I abhor projects taht foist this level of commit police onto end users
2024-04-08T13:32:51 < karlp> I can't even commit locally without following upstream rules on things.
2024-04-08T13:34:44 < qyx> til abhor, til foist, thanks
2024-04-08T13:35:37 < qyx> also, I like the idea of forcing the commit message style
2024-04-08T13:35:50 < qyx> whether this particular one is good or not, idk yet
2024-04-08T13:38:23 < karlp> i can accept it on upstream, that's theirs
2024-04-08T13:38:45 < karlp> but installing 50000 python packages and making my local commits fail because I didn't format it the same way? fuck off.
2024-04-08T13:39:29 < karlp> I'm very much unconcinved that using the first 5-6 chars of every single commit message with "fix, feat, chore, doc, test" is good use of that prime realestate.
2024-04-08T13:40:11 < karlp> anyway, two patches sent upstream with marel.com emails, yay, finally.
2024-04-08T13:40:37 < karlp> got permission to not wait on work ever getting their own org set up properly oto push from, as lonng it goes up with work emails on them.
2024-04-08T13:40:43 < qyx> where do you see the convention must be forced?
2024-04-08T13:40:50 < karlp> esp-idf.
2024-04-08T13:40:53 < qyx> they explicitly say you don't have to adhere
2024-04-08T13:40:55 < qyx> oh
2024-04-08T13:41:10 < karlp> the conventionalcommits people might say "none of this is required"
2024-04-08T13:41:21 < karlp> but that didn't stop people building commit hook plugins that follow it to the letter.
2024-04-08T13:41:51 < karlp> it's like that meme about scifi author with the terror machine, and techbros building the terror machine from the book!
2024-04-08T13:42:46 < karlp> these are all the checks taht are getting run on any commit in esp-idf: https://paste.jvnv.net/view/QJGTw
2024-04-08T13:43:11 < jpa-> i find the boundaries between fix, feat and chore rather arbitrary; if e.g. something is broken by a new version of dependency, is that a fix (it fixes a problem), a feat (it now works with a new version), or a chore (stuff happens, you need to update things)
2024-04-08T13:43:12 < qyx> to be fair, I am using most of the ideas already
2024-04-08T13:43:13 < karlp> and yeah, getting it right locally makes it way easier to merge it all later.
2024-04-08T13:43:38 < karlp> oh, the general concept is fine,
2024-04-08T13:43:54 < karlp> but still, thiking you can use this to make useful machine generated release notes is still wild.
2024-04-08T13:43:57 < qyx> that's easy in the sense of semver, if a fix doesn't change how the thing looks from the outside, it is a fix
2024-04-08T13:44:06 < qyx> but fix may also be a feature or a breaking change
2024-04-08T13:44:18 < qyx> in this case it is feature or a breaking change
2024-04-08T13:48:03 < qyx> also, cloudflare pages for documentation builds, much approved
2024-04-08T13:51:24 < ventYl> the only aspect of github actions for page generation I am not entirely OK with is to use orphan branch to hold the content
2024-04-08T13:51:49 < qyx> same here, that's the reason I looked for something else
2024-04-08T13:59:23 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Ping timeout: 264 seconds]
2024-04-08T14:16:09 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-08T14:20:16 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-08T14:51:00 < qyx> re addressable LEDs
2024-04-08T14:51:09 < qyx> do we have angled versions?
2024-04-08T14:52:02 < qyx> https://www.sparkfun.com/products/23649
2024-04-08T14:52:52 < zyp> yes
2024-04-08T14:53:47 < qyx> idk if they are their own https://www.sparkfun.com/products/12986
2024-04-08T14:53:51 < zyp> e.g. https://jlcpcb.com/partdetail/OpscoOptoelectronics-SK6805SIDES/C2909057 and https://jlcpcb.com/partdetail/3185288-SK6805SIDE_G001/C2909060
2024-04-08T14:54:58 < qyx> hm I need to put them out of the case in some kind of 3d printed holder
2024-04-08T14:55:02 < qyx> like those spakrksuns
2024-04-08T14:55:04 < zyp> that said, I like angled lightpipes, see e.g. https://bin.jvnv.net/file/aizay.jpg
2024-04-08T14:55:06 < qyx> *sparkfuns
2024-04-08T14:55:13 < zyp> rightmost board
2024-04-08T14:55:19 < qyx> me too but for this application they are too dim
2024-04-08T14:55:29 < zyp> too dim?
2024-04-08T14:55:48 < zyp> I doubt these would be
2024-04-08T14:55:52 < qyx> I need something blasting in the user direction to be noticed from ~5 m
2024-04-08T14:56:20 < qyx> actually I don't need RGB, I need pink
2024-04-08T14:56:32 < qyx> but find me a pink LED.. there is ONE on TME
2024-04-08T14:56:44 < qyx> I mean, available
2024-04-08T14:59:04 < zyp> is this an indicator or some sort of illumination?
2024-04-08T15:00:27 < qyx> it is indicating by illumination
2024-04-08T15:01:01 < zyp> like an alarm to grab attention without it being looked at?
2024-04-08T15:01:06 < nomorekaki> Steffanx: https://drive.google.com/file/d/1HRo4HtOCxDUPGCa6DhL0tiSejAhxLiBE/view?usp=sharing
2024-04-08T15:01:57 < qyx> kinda, it is for those big locker box panels and I need to get user attention where to go
2024-04-08T15:02:09 < qyx> hence diffused and bright
2024-04-08T15:02:43 < qyx> because they are not of the "self opening" type
2024-04-08T15:03:20 < zyp> I'd still give the domed top variant of the lightpipes I use a try
2024-04-08T15:04:55 < zyp> I've got chinaclones of https://www.bivar.com/parts_content/Datasheets/SLP3-150-XXX-X.pdf
2024-04-08T15:06:10 < zyp> does it have to be pink? InGaN green is crazy bright and should make up for the loss in the pipe
2024-04-08T15:06:27 < qyx> yeah, marketing requirement :>
2024-04-08T15:07:06 < zyp> how about the adressable LEDs that includes a white die? run the white and red together at 100%
2024-04-08T15:07:31 < zyp> although the offset between the dies might not work out with a lightpipe
2024-04-08T15:07:38 < qyx> that was my original plan
2024-04-08T15:08:05 < qyx> then I considered exactly this issue and tried to find RGBW or at least RGB in a 5 mm diffused THT package
2024-04-08T15:08:19 < qyx> and found those sparkfun LEDs
2024-04-08T15:10:28 < zyp> I think the problem with white is that LEDs don't make white directly, but couple a die with a phosphor that turns the light white
2024-04-08T15:10:41 < zyp> which creates mechanical limitations on how close you can put it to other dies
2024-04-08T15:11:43 < zyp> «Most "white" LEDs in production today use a 450nm – 470nm blue GaN (gallium nitride) LED covered by a yellowish phosphor coating usually made of cerium doped yttrium aluminium garnet (YAG:Ce) crystals which have been powdered and bound in a type of viscous adhesive. The LED chip emits blue light, part of which is converted to yellow by the YAG:Ce. The single crystal form of YAG:Ce is actually considered a 
2024-04-08T15:11:49 < zyp> scintillator rather than a phosphor. Since yellow light stimulates the red and green receptors of the eye, the resulting mix of blue and yellow light gives the appearance of white.»
2024-04-08T15:13:24 < qyx> The Hyper Red source color devices are made with
2024-04-08T15:13:24 < qyx> AlGaInP on GaAs substrate Light Emitting Diode
2024-04-08T15:14:50 < qyx> so AlGaInP red, InGaN blue, InGaN green
2024-04-08T15:15:00 < zyp> yeah
2024-04-08T15:15:19 < qyx> https://www.mouser.sk/ProductDetail/Kingbright/WP154A4SUREQBFZGW?qs=w3%252Bcv6FlocrIOROaa2rdEg%3D%3D
2024-04-08T15:15:33 < qyx> I am actually considering a right angled PCB to mount them
2024-04-08T15:15:54 < zyp> is that easier than just bending the legs? :)
2024-04-08T15:16:17 < qyx> for assembly, maybe
2024-04-08T15:16:26 < qyx> idk
2024-04-08T15:31:36 < nomorekaki> do you guys have some sort of architecture for production tester?
2024-04-08T15:32:18 < nomorekaki> or just ad-hoc all the way?
2024-04-08T15:35:05 < zyp> in what sense?
2024-04-08T15:35:50 < nomorekaki> have some sort of modularity?
2024-04-08T15:35:51 < zyp> I've done a bunch of work on production testing over the last year
2024-04-08T15:36:15 < zyp> effectively built a whole reusable production test infrastructure
2024-04-08T15:39:15 < zyp> test specs are yaml, describes test sequences to run, each step calls a python function, yaml can spec arguments, e.g. «check_dut_analog» or «check_jig_analog», takes pin name and thresholds as arguments
2024-04-08T15:39:43 < nomorekaki> so it's linux box
2024-04-08T15:39:56 < nomorekaki> does it have comfuser as a master?
2024-04-08T15:40:13 < zyp> no, it's python, it's designed to be able to run on anything including windows
2024-04-08T15:40:18 < nomorekaki> does it run the sequence from pc?
2024-04-08T15:40:23 < zyp> yeah
2024-04-08T15:40:32 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Ping timeout: 252 seconds]
2024-04-08T15:40:40 < nomorekaki> what is pc interfacing with?
2024-04-08T15:41:27 < zyp> on the jig, there's a controller board that can both do SWD and has a bunch of gpios it can poke the target with and ADC channels to read values with
2024-04-08T15:41:46 < zyp> it has a RPC over TCP over USB that the python software talks to
2024-04-08T15:42:10 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-08T15:42:34 < zyp> and then in the DUT firmware there's a test stub that can be invoked over RPC, kinda giving you another RPC that lets you run snippets of code on the DUT
2024-04-08T15:43:10 < zyp> so the python functions can either invoke commands on the jig controller, or on the DUT itself via the jig controller
2024-04-08T15:44:15 < nomorekaki> rpc over tpc hmm
2024-04-08T15:44:20 < nomorekaki> never heard of that
2024-04-08T15:44:38 < nomorekaki> *tcp
2024-04-08T15:45:19 < zyp> it's a custom RPC protocol, very simple
2024-04-08T15:46:43 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T15:46:45 < nomorekaki> my testers use uart
2024-04-08T15:46:49 < nomorekaki> one way
2024-04-08T15:47:15 < nomorekaki> it just vommits text
2024-04-08T15:47:38 < zyp> so here's one section of a test spec: https://paste.jvnv.net/view/dwnY3
2024-04-08T15:48:11 < zyp> that check_analog looks like this: https://paste.jvnv.net/view/nprxm
2024-04-08T16:24:52 -!- Livio [~livio@user/livio] has quit [Ping timeout: 246 seconds]
2024-04-08T16:52:32 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T17:20:21 < qyx> hm I can set window background to red, blue, gray, whatever.. but not white
2024-04-08T17:26:28 < qyx> in Qt6
2024-04-08T17:41:47 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has joined ##stm32
2024-04-08T17:41:57 -!- Linux_Kerio [~Linux_Ker@chello085216198152.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-08T18:16:07 < jbo> qyx, what?
2024-04-08T18:34:47 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 264 seconds]
2024-04-08T18:36:31 -!- ventYl [~ventyl@adsl-dyn36.78-98-110.t-com.sk] has quit [Ping timeout: 260 seconds]
2024-04-08T18:50:17 < qyx> jbo: yeah
2024-04-08T18:55:03 < qyx> it is greyish even if I set white, either with a palette or setstyle
2024-04-08T18:55:29 < qyx> even the yellow looks abit weirdas if there was some half opaque overlay
2024-04-08T19:18:20 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-08T19:33:15 -!- ventYl [~ventyl@adsl-dyn58.78-98-154.t-com.sk] has joined ##stm32
2024-04-08T19:37:37 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-08T19:42:22 -!- josuah [~josuah@46.23.94.12] has quit [Ping timeout: 268 seconds]
2024-04-08T19:42:23 -!- Alexer [~alexer@alexer.net] has quit [Ping timeout: 268 seconds]
2024-04-08T19:42:23 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Ping timeout: 268 seconds]
2024-04-08T19:42:53 -!- Alexer [~alexer@alexer.net] has joined ##stm32
2024-04-08T19:43:33 -!- BrainDamage [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-08T19:58:39 < Steffanx> I already received the 0 mA addressable leds qyx .. didnt even have time to make a pcb yet -_-
2024-04-08T20:37:24 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-08T20:39:14 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-08T21:03:37 < BrainDamage> time to fly wire them
2024-04-08T21:27:37 < qyx> Steffanx: what for, just measure vss/vdd
2024-04-08T21:28:09 < Steffanx> I already did and its actually < 1 uA. (datasheet of this one claims 1 uA)
2024-04-08T21:55:12 < Steffanx> Sadly after turning it on fully and off again.. its 62 uA
2024-04-08T21:55:23 < Steffanx> but that can just me doing something wrong
2024-04-08T22:06:53 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-6594-e3d3-49f7-f504.fixed6.kpn.net] has joined ##stm32
2024-04-08T22:16:57 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T22:38:59 -!- Livio [~livio@user/livio] has quit [Ping timeout: 264 seconds]
2024-04-08T22:47:11 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-08T22:48:21 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-08T23:17:41 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a803-16c9-7bb-d97.fixed6.kpn.net] has joined ##stm32
2024-04-08T23:21:46 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-6594-e3d3-49f7-f504.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-08T23:36:22 < Steffanx> Like driving the thing with a spi @ 3.3V instead of the required 5V. 
2024-04-08T23:36:44 < Steffanx> Anyway, it seems the claim is true qyx
2024-04-08T23:42:19 < zyp> nice
2024-04-08T23:42:31 < zyp> do they come in small packages as well?
2024-04-08T23:44:55 < Steffanx> I recall seeing 2020
2024-04-08T23:45:48 < Steffanx> But it's a bit of a mess as well. The protocol descriptions differ here and there or are just entirely wrong. 
2024-04-08T23:47:31 < zyp> oh, shame, other leds never have that issue
2024-04-08T23:47:33 < zyp> (https://github.com/orbcode/orbtrace/commit/483545a)
2024-04-08T23:47:48 < Steffanx> 😜
2024-04-08T23:48:49 < zyp> also https://github.com/orbcode/orbtrace/commit/598b4c9
2024-04-08T23:49:56 < zyp> half the production test failures on the first orbtrace series was leds that were just unhappy about the reset period
2024-04-08T23:55:24 < Steffanx> Anyway, time to put a few on a PCB add a level shifter and test with a few more.
--- Day changed ti huhti 09 2024
2024-04-09T00:21:03 < zyp> what were they called again?
2024-04-09T00:24:29 < qyx> they must be using some trickery, like relay on the input
2024-04-09T00:27:46 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-09T00:30:31 < Steffanx> Hd108
2024-04-09T00:39:54 < zyp> hmm, yeah 2020 exists, but availability doesn't seem impressive
2024-04-09T00:50:00 < Steffanx> True same applies to the 5050.
2024-04-09T00:51:40 < zyp> it's actually listed at jlcpcb, but says unavailable, and LCSC doesn't list it at all
2024-04-09T00:52:28 < zyp> wonder if somebody got it listed by sourcing it themselves and having it shipped to jlc
2024-04-09T01:00:12 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-09T01:01:24 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-09T01:15:37 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 255 seconds]
2024-04-09T01:32:53 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a803-16c9-7bb-d97.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-09T01:46:15 -!- c10ud_ [~c10ud@95.236.174.95] has joined ##stm32
2024-04-09T01:49:22 -!- c10ud [~c10ud@user/c10ud] has quit [Ping timeout: 255 seconds]
2024-04-09T02:40:39 -!- fenugrec [~f@192.214.232.39] has quit [Ping timeout: 255 seconds]
2024-04-09T02:41:47 -!- fenugrec [~f@192.214.232.39] has joined ##stm32
2024-04-09T02:47:10 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-09T02:59:06 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-09T02:59:07 -!- ferdna [~ferdna@user/ferdna] has quit [Remote host closed the connection]
2024-04-09T03:17:51 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-09T03:18:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-09T03:29:46 -!- sugarbeet [~barbas@81.4.123.134] has quit [Ping timeout: 252 seconds]
2024-04-09T03:31:00 -!- sugarbeet [~barbas@81.4.123.134] has joined ##stm32
2024-04-09T05:31:01 -!- nuxil [~nuxil@141.195.51.41] has quit [Read error: Connection reset by peer]
2024-04-09T07:50:43 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-c515-b802-47f3-3b19.fixed6.kpn.net] has joined ##stm32
2024-04-09T08:00:55 < jpa-> the WS2812B-B that I had jlcpcb put on boards for me last year were a total disappointment; supposedly work down to 2.7 Vin - nope, the voltage drop on GND tends to ruin that; several were broken right away, and two more have broken from vibration or temperature since then (which is especially annoying as I put epoxy over them to protect.. so difficult to change)
2024-04-09T08:11:25 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-c515-b802-47f3-3b19.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-09T08:40:18 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-09T08:47:20 < antto> huh
2024-04-09T08:48:09 < antto> jpa-, i am currently planning to put 104 of these on my clock: https://www.lcsc.com/datasheet/lcsc_datasheet_2402181504_XINGLIGHT-XL-3528RGBW-WS2812B_C2890364.pdf
2024-04-09T08:48:31 < antto> i'll be putting them myself of course
2024-04-09T08:48:44 < antto> the datasheet makes me so unconfident
2024-04-09T08:52:56 < antto> if i could, i would've used these ones: https://www.optosupply.com/uppic/2023421763886.pdf
2024-04-09T08:53:43 < antto> they appear physically compatible, i might try to see if $distributor could poke the $manufacturer for a few hundred
2024-04-09T09:04:51 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-c515-b802-47f3-3b19.fixed6.kpn.net] has joined ##stm32
2024-04-09T09:18:34 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-09T09:23:30 < jpa-> todays obvious yet amazing discovery: ifixit magnetic project mat holds stainless steel stencils nicely in place
2024-04-09T09:29:48 < ventYl> isn't stainless steel non-magnetic?
2024-04-09T09:32:28 < jpa-> apparently not all alloys
2024-04-09T09:32:29 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-09T09:32:55 < jpa-> the stencils are not as magnetic as normal steel, but over the large area there is still significant force
2024-04-09T09:33:31 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-09T09:34:33 -!- duude__- is now known as duude__
2024-04-09T09:35:56 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-09T09:37:21 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Ping timeout: 255 seconds]
2024-04-09T09:42:29 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-b96d-d51e-2ae7-ab85.fixed6.kpn.net] has joined ##stm32
2024-04-09T09:46:22 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-c515-b802-47f3-3b19.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-09T10:43:31 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-09T10:46:20 < zyp> I keep noticing an «embedded shit» directory in this one codebase I'm doing some work on
2024-04-09T10:46:29 < zyp> except they've bundled this thing: https://github.com/Sensirion/embedded-sht
2024-04-09T10:46:59 < zyp> it's distracting
2024-04-09T10:49:48 < jpa-> i always have embedded shit in my projects but i'm not as outright about it
2024-04-09T10:55:54 < zyp> the people who wrote this code also have an interesting idea of what a task is
2024-04-09T10:56:36 < zyp> it appears that to them it means «function», because pretty much every function is prefixed with task_
2024-04-09T11:16:52 -!- nuxil [~nuxil@141.195.51.41] has joined ##stm32
2024-04-09T11:19:56 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-09T11:20:05 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 256 seconds]
2024-04-09T11:52:30 < zyp> this codebase is also full of headers that relies on stdint.h and stdbool.h, but doesn't include them, and so relies on them already being included first
2024-04-09T11:53:11 < zyp> which breaks when I'm attempting to build them separately
2024-04-09T11:53:25 < zyp> (I'm tasked with adding unit tests to this project)
2024-04-09T12:07:25 < karlp> heh, I've seen/used task_ prefix for the top level functions that are called repeatedly in a superloop style.
2024-04-09T12:07:36 < karlp> it's akin to "main" for each module.
2024-04-09T12:12:35 < Steffanx> It's not really weird to call that a task I think? 
2024-04-09T12:17:21 < qyx> I should eat something as a task
2024-04-09T12:18:52 < zyp> karlp, yeah, that's understandable
2024-04-09T12:30:20 < zyp> wtf is this bullshit: https://github.com/pololu/vl53l1x-st-api-arduino/blob/master/vl53l1x-st-api/vl53l1_types.h#L88
2024-04-09T12:36:56 < qyx> haha
2024-04-09T12:37:21 < zyp> just told the compiler to define NULL=0 and that made it happy
2024-04-09T12:37:48 < qyx> oh it continues
2024-04-09T12:38:06 < qyx> the rest of the file is the same
2024-04-09T12:38:43 < qyx> why on earth should there be no stdint.h
2024-04-09T12:38:58 < qyx> are we back in 89?
2024-04-09T12:39:33 < karlp> this is what happens when people work without supervision and don't question the reasons.
2024-04-09T12:46:47 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-09T12:48:04 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-09T12:48:44 -!- duude__- is now known as duude__
2024-04-09T13:01:18 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-09T13:21:37 < jpa-> zyp: but why is NULL not defined if stddef.h is included?
2024-04-09T14:18:19 < zyp> I have no idea
2024-04-09T14:20:33 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-09T14:21:36 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-09T15:04:40 < ventYl> jpa-: NULL is a POSIX thing, not ISO C
2024-04-09T15:09:20 < zyp> I checked https://en.cppreference.com/w/c/types/NULL and came to the same conclusion as jpa- 
2024-04-09T15:22:54 < mawk> at least say it's ((void *)0) and not just 0
2024-04-09T15:26:31 < zyp> meh, who cares
2024-04-09T15:28:47 -!- Livio [~livio@user/livio] has quit [Ping timeout: 264 seconds]
2024-04-09T15:59:55 < ventYl> "aux" is a reserved file name in Windows
2024-04-09T16:00:02 < ventYl> my creepy RTOS cannot be checked out there
2024-04-09T16:04:30 < mawk> CON too ventYl 
2024-04-09T16:04:43 < mawk> it means cunt in french
2024-04-09T16:04:54 < mawk> both meanings of the word
2024-04-09T16:05:05 < ventYl> yeah, after I realized that "AUX" is a reserved name, I recalled that there are several others
2024-04-09T16:05:13 < ventYl> back from the days of DOS 2.x probably
2024-04-09T16:05:20 < mawk> yeah
2024-04-09T16:05:36 < mawk> now the special names are GUIDs so it's impossible to stumble upon by accident
2024-04-09T16:05:50 < mawk> in NTFS
2024-04-09T16:05:54 < mawk> and with NT
2024-04-09T16:06:02 < ventYl> yet, for quite some time I wondered why Git for Windows can't extract something as trivial as include/aux/systick.h
2024-04-09T16:29:37 < qyx> zyp: what conclusion is that
2024-04-09T16:30:25 < qyx> oh that it is posix?
2024-04-09T16:43:40 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-09T17:07:33 -!- rpifan [~rpifan@p200300d267133a007194969f071e01d8.dip0.t-ipconnect.de] has joined ##stm32
2024-04-09T17:07:33 -!- rpifan [~rpifan@p200300d267133a007194969f071e01d8.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-09T17:07:33 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T17:21:14 < karlp> so, ios just... blocks bluetooth classic SPP devices? like, they won't even present as existing?
2024-04-09T17:21:31 < karlp> it's hard to get straigth answers out of apple, but, really?
2024-04-09T17:21:40 < karlp> just zero legacy hardware for you, ever...
2024-04-09T17:33:21 < PaulFertser> Apple knows better what iphone users desire.
2024-04-09T17:35:52 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-09T17:37:30 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-09T17:57:18 < karlp> eh, it works on windows, linux and android phones, I've done my part I think...
2024-04-09T18:00:07 < PaulFertser> How about doing it in BLE and allowing web browser apps to talk to it via Bluetooth API or whatever it's called?
2024-04-09T18:02:32 < karlp> yeah, it appears that adding BLE "SPP like" is the general workaround for iphones.
2024-04-09T18:02:41 < karlp> can do that I guess. 
2024-04-09T18:13:48 < PaulFertser> On the plus side you do not need to bother with any "stores" on any platform if you go that way.
2024-04-09T18:14:00 < PaulFertser> Because https://developer.mozilla.org/en-US/docs/Web/API/Web_Bluetooth_API
2024-04-09T18:17:02 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-09T18:26:31 < karlp> oh, that's not my problem.
2024-04-09T18:26:43 < karlp> web bluetooth is chrome only still, which is pretty garbage, I wouldn't rely on taht.
2024-04-09T18:27:13 < karlp> well, edge too apparently, if anyone uses that.
2024-04-09T18:32:51 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 255 seconds]
2024-04-09T18:38:20 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 268 seconds]
2024-04-09T18:45:09 < PaulFertser> Yeah, it sucks in its own ways.
2024-04-09T18:45:21 < PaulFertser> Feels like most things in modern IT suck big time one way or another :/
2024-04-09T18:59:52 < karlp> oh boi, I love string processing in C.
2024-04-09T19:03:06 < karlp> imagine this ever going wrong, super robust... https://paste.jvnv.net/view/GjpgS
2024-04-09T19:03:39 < BrainDamage> I like that function, it has a pleasing aesthetic symmetry
2024-04-09T19:44:59 -!- rpifan [~rpifan@p200300d267133a008f7b949ecb26b45a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-09T19:44:59 -!- rpifan [~rpifan@p200300d267133a008f7b949ecb26b45a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-09T19:44:59 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T19:52:14 < karlp> ah, manual json constructing too.
2024-04-09T19:52:26 < karlp> returning broken incomplete json on the slightest mis configuration.
2024-04-09T19:58:29 < jpa-> i have bunch of that
2024-04-09T19:58:43 < jpa-> it even has snprintf() that still ends up overrunning the buffer
2024-04-09T20:00:26 < jpa-> i wish there was JSONv2 that would be roughly the same, but fix the most stupid things like not allowing trailing comma or NaN or comments
2024-04-09T20:03:53 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 252 seconds]
2024-04-09T20:11:41 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 240 seconds]
2024-04-09T20:49:56 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T20:53:44 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-09T20:55:12 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-09T20:55:47 < zyp> jpa-, it exists, it's called yaml
2024-04-09T21:00:52 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-09T21:15:10 < qyx> no, yaml bans tabs which is a regression
2024-04-09T21:15:37 < BrainDamage> yaml is a bit crazy
2024-04-09T21:16:26 < BrainDamage> 80% of this is "you have to quote strings" https://ruudvanasseldonk.com/2023/01/11/the-yaml-document-from-hell but the remaining 20% is even crazier
2024-04-09T21:43:04 < Steffanx> heh they actually removed comments from json?!
2024-04-09T21:45:44 < qyx> I don't think they were even allowed
2024-04-09T21:46:39 < Steffanx> Yes they were apparently. 
2024-04-09T21:48:01 < qyx> wat "Comments were removed from the JSON specification to prevent developers from using them to specify parser directives"
2024-04-09T21:48:09 < qyx> https://sentry.io/answers/comments-in-json/
2024-04-09T21:52:18 < qyx> BrainDamage: lol the doc
2024-04-09T21:57:39 < qyx> yes it is a bit BrainDamaged
2024-04-09T22:13:25 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-09T22:13:43 < Laurenceb_> >Chris-chan was listening to a 'binaural beats' YouTube video meant to hypnotize someone until their penis turns into a vagina
2024-04-09T22:13:54 < Laurenceb_> the more u know
2024-04-09T22:14:29 < Steffanx> Hi laurenceGPT
2024-04-09T22:27:59 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-09T22:36:49 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-09T22:37:08 < nomorekaki> hyperlurence
2024-04-09T22:37:37 < jpa-> zyp: yaml is way too different and no interoperability
2024-04-09T22:38:02 < qyx> no, json is a valid subset of yaml
2024-04-09T22:38:18 < qyx> so it offers full interoperability
2024-04-09T22:38:32 < jpa-> it is? huh, i never realized based on how they look
2024-04-09T22:38:37 < qyx> keep in mind yaml has a ton of different grammars
2024-04-09T22:39:05 < qyx> https://john-millikin.com/json-is-not-a-yaml-subset
2024-04-09T22:39:06 < qyx> haha
2024-04-09T22:39:57 < jpa-> great
2024-04-09T22:40:27 < qyx> ok but the guy is clickbaiting, fuk him
2024-04-09T22:41:40 < qyx> because neither of his points is valid for parsing json with a yaml parser
2024-04-09T22:41:54 < qyx> (in yaml1.2)
2024-04-09T22:52:17 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 272 seconds]
2024-04-09T22:58:20 -!- rpifan [~rpifan@p200300d267133a009e657f6a79edc38c.dip0.t-ipconnect.de] has joined ##stm32
2024-04-09T22:58:20 -!- rpifan [~rpifan@p200300d267133a009e657f6a79edc38c.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-09T22:58:20 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T23:02:07 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-09T23:02:28 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T23:10:07 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-09T23:10:20 -!- rpifan [~rpifan@p200300d267133a00f3f0d9ff02ab1747.dip0.t-ipconnect.de] has joined ##stm32
2024-04-09T23:10:20 -!- rpifan [~rpifan@p200300d267133a00f3f0d9ff02ab1747.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-09T23:10:20 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T23:12:45 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-09T23:14:07 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-09T23:14:23 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-09T23:41:03 -!- nuxil [~nuxil@141.195.51.41] has quit [Read error: Connection reset by peer]
2024-04-09T23:43:02 -!- nuxil [~nuxil@141.195.51.41] has joined ##stm32
2024-04-09T23:52:03 -!- nuxil [~nuxil@141.195.51.41] has quit [Read error: Connection reset by peer]
2024-04-09T23:53:43 -!- nuxil [~nuxil@141.195.51.41] has joined ##stm32
--- Day changed ke huhti 10 2024
2024-04-10T00:23:40 -!- Alexer [~alexer@alexer.net] has quit [Ping timeout: 268 seconds]
2024-04-10T00:24:00 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-10T00:24:39  * Laurenceb_ is wondering if VESC could be made to run on nucleo-f429zi
2024-04-10T00:24:48 < Laurenceb_> maybe I should try this, sounds fun...
2024-04-10T00:25:57 < karlp> jpa-: this is... worse. there's a json library in tree, it' sused to start constructing the messages, then it hands to some raw string shit that fails, and it just... splats back half a json string to the browser and pretends everything's ok...
2024-04-10T00:27:06 < karlp> I mean... just keep using the json library ffs.
2024-04-10T00:27:23 < zyp> a few months ago, I worked on a project that got a http server with a json API
2024-04-10T00:27:53 < karlp> that sounds sane. this has the json api only via websockets....
2024-04-10T00:28:19 < karlp> bleh. it's so uninspiring to work on it
2024-04-10T00:28:32 < zyp> the request handler is made with a parser-generator with a grammar that both handles the HTTP protocol and a JSON POST payload on the fly
2024-04-10T00:29:26 < qyx> sounds fun
2024-04-10T00:29:53 < zyp> I was adding bearer token auth to it, originally only did http digest auth
2024-04-10T00:30:57 < zyp> since bearer tokens are opaque, I cheated by making the tokens look like digests so I didn't have to modify the parser
2024-04-10T00:32:14 < specing> Laurenceb_: f429 is probably a supercomputer wrt. what is needed of it
2024-04-10T00:33:44 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Ping timeout (120 seconds)]
2024-04-10T00:36:46 < Laurenceb_> specing: not really if I need fast control and sensorless operation
2024-04-10T00:37:03 < Laurenceb_> atm we have tms320 at 100Mhz brogrammed with simulink and its too slow
2024-04-10T00:37:38 < Laurenceb_> officially only 405 and 407 is supported but I see some forum posters claim to have used 429
2024-04-10T00:37:42 < zyp> well duh, it's simulink
2024-04-10T00:37:46 < Laurenceb_> lol yeah
2024-04-10T00:38:01 < zyp> I'm not overly impressed by the code simulink generates
2024-04-10T00:38:06 < Laurenceb_> I looked as some of the c code and it was horrific, worse than n00b arduino
2024-04-10T00:38:20 < Laurenceb_> the matlab generated code can be very good, but simulink blocks are often poor
2024-04-10T00:38:24 < zyp> and yeah, you can run f407 code unmodified on f429, it's a direct superset
2024-04-10T00:38:36 < Laurenceb_> ok
2024-04-10T00:39:07 < zyp> f429 is just f427 with lcd periph and f427 is just f407 with dram support and dualbank flash
2024-04-10T00:39:08 < Laurenceb_> ideally I want to test ethernet comms - chibios has support for 429 ethernet so it might be feasible, but extra cpu load
2024-04-10T00:39:48 < Laurenceb_> hardware - https://www.wolfspeed.com/crd600da12e-xm3/
2024-04-10T00:39:54 < zyp> I dropped f427 into a f407 design once because client came back and wanted more flash
2024-04-10T00:41:06 < Laurenceb_> internally the wolfspeed uses a ti launchpad plugged into the main pcb O_o
2024-04-10T00:42:03 < Laurenceb_> I could probably literally bodge the nucleo on using some veroboard lol
2024-04-10T00:42:27 < qyx> looks like a good plan when you are dealing with 600 kW of power
2024-04-10T00:42:34 < Laurenceb_> heh
2024-04-10T00:42:54 < Laurenceb_> we have been to 1MW peak
2024-04-10T00:43:11 < qyx> how did that happen you were given such a dual use good to handle
2024-04-10T00:43:55 < Laurenceb_> dual use good to handle?
2024-04-10T00:44:34 < qyx> "dual use goods" capable of killing people when not handled carefully or intentionally misused by terorrists etc.
2024-04-10T00:45:08 < qyx> sorry I don't english
2024-04-10T00:46:19 < Laurenceb_> oh we just got it from a chinese seller for $8k, not digikey or similar
2024-04-10T00:47:10 < Laurenceb_> today hyperboss asked me to make a lighter motor with magnetic aluminium
2024-04-10T00:50:20 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-b96d-d51e-2ae7-ab85.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-10T00:51:36 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-10T01:00:05 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T01:00:59 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T01:36:45 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-10T01:44:05 -!- c10ud_ [~c10ud@95.236.174.95] has quit [Read error: Connection reset by peer]
2024-04-10T01:45:09 < zyp> the fuck do you need to motorize a lighter for? what are you gonna burn?
2024-04-10T01:46:59 < qyx> lold
2024-04-10T02:10:53 < nuxil> zyp, thats what you took out of that ? im more concerned about where he will find this magnetic aluminium he needs to use 😛
2024-04-10T02:11:38 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-10T02:11:58 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 264 seconds]
2024-04-10T02:21:57 < qyx> squirrel cage motors have been using it for ages
2024-04-10T02:35:33 -!- Alexer [~alexer@alexer.net] has joined ##stm32
2024-04-10T02:49:34 < nuxil> yes ofc. aluminium is subject to lenz law, and can be used in motors. but its still not a magnetic metall 🙂
2024-04-10T02:55:03 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Killed (NickServ (GHOST command used by Spirit5325))]
2024-04-10T02:55:08 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-10T02:57:22 < nuxil>  eddy can tell more about how it works. im not sure if hes around tho 😛
2024-04-10T03:17:41 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 240 seconds]
2024-04-10T03:25:46 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-10T03:26:30 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T03:27:33 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T04:03:54 -!- veverak [~veverak@ip-89-103-173-67.bb.vodafone.cz] has quit [Ping timeout: 255 seconds]
2024-04-10T04:31:43 -!- Flea86 [~maomao@user/Flea86] has joined ##stm32
2024-04-10T04:36:18 -!- veverak [~veverak@ip-89-103-173-67.bb.vodafone.cz] has joined ##stm32
2024-04-10T04:37:14 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-10T04:46:53 < NEYi> Hey, people! Does anybody here have older CubeMX firmware packs? 
2024-04-10T04:46:53 < NEYi> I've been playing with STM32s some long time, 5 years ago or more.
2024-04-10T04:46:53 < NEYi> Decided to remember the glory days..!
2024-04-10T04:46:53 < NEYi> Downloaded the latest CubeMX v6...
2024-04-10T04:46:53 < NEYi> And it turned out to be so ugly - only a mother could love it. ( o_o)
2024-04-10T04:46:54 < NEYi> I don't get how could someone design a program with such a "progressive" interface.
2024-04-10T04:46:56 < NEYi> And it also takes 1 gig on the hard drive.
2024-04-10T04:46:58 < NEYi> EMPTY! :O
2024-04-10T04:47:00 < NEYi> Like I also gotta download firmware packages and so on.
2024-04-10T04:47:02 < NEYi> And it directly says it will spy on me. ( O_o)
2024-04-10T04:47:04 < NEYi> Well, I decided to download older cubes...
2024-04-10T04:47:08 < NEYi> And the further down I went - the better they get.
2024-04-10T04:47:10 < NEYi> I like the latest release of version v4 most... The program is only 50 megabytes! :D
2024-04-10T04:47:12 < NEYi> But I can't download the FW packages for it.
2024-04-10T04:47:14 < NEYi> They aren't on ST website anymore.
2024-04-10T04:47:16 < NEYi> Only the latest ones that are unsupported by the classic CubeMX.
2024-04-10T04:47:18 < NEYi> Does anybody have older MCU firmwares? Files like "stm32cube_fw_f1_v161.zip", and such?
2024-04-10T04:47:20 < NEYi> Or knows where I can download those?
2024-04-10T04:47:22 < NEYi> ST's server is down, the program won't download anything.
2024-04-10T04:47:24 < NEYi> And I can't find this stuff anywhere on the web.
2024-04-10T04:47:26 < NEYi> As if it was deleted from the existence.
2024-04-10T04:55:29 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T04:56:35 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T04:57:02 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T04:57:40 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T05:14:15 < NEYi> ...
2024-04-10T05:14:16 < NEYi> !!! False alarm... I've installed Cube v5, and it both connects to the server, and there's an option to download even very old packs, that are nowhere on the web, and not even on ST's website. I better grab all that stuff quick, and treasure it on my HDD. :D
2024-04-10T06:04:41 -!- qyx [~qyx@84.245.120.220] has quit [Ping timeout: 268 seconds]
2024-04-10T06:05:57 -!- qyx [~qyx@84.245.120.207] has joined ##stm32
2024-04-10T06:32:29 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-10T06:40:03 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T06:41:31 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T07:51:55 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-10T08:21:00 < nuxil> hu? it should download what is needed when you select the mcu.
2024-04-10T08:21:10 < nuxil> nm... 
2024-04-10T08:21:44 < nuxil> good idea to scroll down to the end 😛
2024-04-10T08:39:01 < octorian> So I go ahead and add USB HID Keyboard host-side support to my device, and it works all well and good with the assumption that USB Keyboards will only ever return simple 8-byte reports for keypresses under normal circumstances. Works fine with every keyboard I can find. Until tonight, when I found my first exception...
2024-04-10T08:39:49 < octorian> A stupid el-cheapo USB foot pedal (that pretends to be a keyboard).  It sends 9-byte reports prefixed with a Report ID, where I might actually have to parse the USB report descriptor to handle it in a "generic" way.
2024-04-10T08:48:36 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-10T09:09:39 < jpa-> octorian: i have done similar, i just always used last 8 bytes of packet
2024-04-10T09:10:31 < jpa-> the only two formats i ever saw were the 8-byte plain or 9-byte with prefix
2024-04-10T09:10:35 < octorian> I mean that's the easy kludge.  Or simply assume that if 9 bytes, and if report ID is 0x01, then offset by 1.
2024-04-10T09:11:09 < octorian> I guess the question is whether 9-byte /w prefix will ever have a prefix other than 0x01, or contain different data.
2024-04-10T09:11:41 < jpa-> i have seen prefix 0x03
2024-04-10T09:11:54 < jpa-> and probably if you'd plug in a mouse, it would have some other stuff
2024-04-10T09:11:55 < zyp> you know, this is a solved problem
2024-04-10T09:11:55 < octorian> Though one idea is a very quick-and-dirty parser that doesn't fully parse the data, just tries to find the report IDs that match the right type.
2024-04-10T09:12:14 < zyp> HID spec have what they call «boot protocol» for keyboards and mice
2024-04-10T09:12:22 < octorian> jpa-, what devices have you seen 0x03 on?
2024-04-10T09:12:28 < jpa-> octorian: barcode scanners
2024-04-10T09:12:42 < octorian> I have one or two of those.  I'll see how they present.
2024-04-10T09:13:00 < jpa-> specifically netum brand scanners
2024-04-10T09:13:28 < octorian> jpa-, I have a Netum barcode scanner :-)
2024-04-10T09:13:33 < zyp> if you want simple keyboard support, you add support for boot protocol, then all you have to do is check that the keyboard supports it and that it's in boot protocol mode
2024-04-10T09:13:57 < jpa-> but do foot pedals and barcode scanners support boot protocol? :)
2024-04-10T09:14:04 < zyp> if you want to support keyboards that doesn't do boot protocol, you can't really make *any* assumptions on report format
2024-04-10T09:14:35 < zyp> it's entirely legal for keyboards to transmit reports that are entirely bitmaps, not lists of scan codes
2024-04-10T09:14:37 < jpa-> i'm not afraid of making an ass out of you and me, i can assume anything!
2024-04-10T09:14:41 < zyp> I've made such keyboards
2024-04-10T09:14:42 < octorian> Okay, so I'm now reading "Appendix B: Boot Interface descriptors" of the HID spec.
2024-04-10T09:15:33 < zyp> and you can't support a bitmap keyboard without parsing the report descriptor to work out which bit corresponds to which scan code
2024-04-10T09:15:40 < octorian> Seems to say that the first 8 bytes of keyboard reports must conform to the boot protocol spec.
2024-04-10T09:15:51 < zyp> no, it doesn't
2024-04-10T09:16:03 < octorian> So I guess if its >8 bytes, then ignore the report ID and parse the first 8 bytes.
2024-04-10T09:16:55 < zyp> there's a SET_PROTOCOL request to select between boot protocol and regular HID, and in the latter mode the first 8 bytes can be anything
2024-04-10T09:17:08 < zyp> and keyboards aren't required to support the former
2024-04-10T09:17:52 < octorian> But they need to support the boot protocol to work with any PC BIOS, so I suspect most do.
2024-04-10T09:18:01 < octorian> And I don't *have* to switch out of the boot protocol, do I?
2024-04-10T09:18:59 < zyp> foot pedals and barcode scanners don't have to work with BIOS
2024-04-10T09:19:10 < zyp> and I don't think they default to boot protocol
2024-04-10T09:20:06 < octorian> Good point.  I really don't have a reason to support barcode scanners here, and footpedals are really only a "nice to have" and this one at least emulates a keyboard by default.
2024-04-10T09:22:41 < zyp> the way I see it, there's three reasonable ways to do a host with keyboard support
2024-04-10T09:23:11 < zyp> one is to have support for boot protocol keyboards, that's the simplest
2024-04-10T09:23:41 < zyp> the second is to have a full HID report descriptor parser, capable of supporting arbitrary devices
2024-04-10T09:26:02 < zyp> and if you still need some weird in between, you could hardcode support for specific devices, filtered by vid/pid (and potentially hashing the report descriptor to check it matches what you've hardcoded against in case the device vendor pull the rug under you)
2024-04-10T09:27:36 < zyp> I wouldn't bother making any attempt at supporting arbitrary HID devices that don't claim to support boot protocol and that I also haven't already tested before
2024-04-10T09:30:48 < octorian> Yeah, for my current expected use cases, option 1 is probably the way to go.  (with some minor fixes to be aware of what I learned tonight)
2024-04-10T09:31:37 < octorian> No reason to bother with the rest, unless I find myself suddenly wanting to add support for different kinds of peripherals where it may be an issue.
2024-04-10T09:32:00 < octorian> (and now off to bed)
2024-04-10T09:38:18 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-59aa-7547-f755-4ead.fixed6.kpn.net] has joined ##stm32
2024-04-10T09:39:49 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-10T09:40:59 -!- c10ud [~c10ud@95.236.174.95] has joined ##stm32
2024-04-10T09:40:59 -!- c10ud [~c10ud@95.236.174.95] has quit [Changing host]
2024-04-10T09:40:59 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-10T09:58:15 < jpa-> weird benefits of magnetic mat for SMD assembly: it automatically orients SMD passives
2024-04-10T09:58:32 < jpa-> weirdly resistors seem to be oriented along X axis, while capacitors are aligned to Y axis
2024-04-10T09:58:55 < jpa-> also you can easily flip resistors by pushing them sideways until it hits magnetic reversal zone
2024-04-10T10:05:08 < qyx> heh
2024-04-10T10:06:35 -!- HelloShitty [~psysc0rpi@bl15-1-78.dsl.telepac.pt] has quit [Ping timeout: 264 seconds]
2024-04-10T10:08:50 -!- HelloShitty [~psysc0rpi@bl15-1-78.dsl.telepac.pt] has joined ##stm32
2024-04-10T10:16:10 < jpa-> https://www.youtube.com/watch?v=EVe3Ve6DpcE  like this
2024-04-10T10:22:37 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-59aa-7547-f755-4ead.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-10T10:37:44 -!- HelloShitty [~psysc0rpi@bl15-1-78.dsl.telepac.pt] has quit [Ping timeout: 260 seconds]
2024-04-10T10:45:15 -!- HelloShitty [~psysc0rpi@2001:8a0:e143:a100:478:7dff:fe45:a43d] has joined ##stm32
2024-04-10T11:07:41 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-10T12:08:18 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T12:09:12 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T12:21:30 -!- ventYl_ [~ventyl@bband-dyn218.95-103-122.t-com.sk] has joined ##stm32
2024-04-10T12:22:30 -!- ventYl [~ventyl@adsl-dyn58.78-98-154.t-com.sk] has quit [Ping timeout: 255 seconds]
2024-04-10T12:24:27 -!- ventYl_ is now known as ventYl
2024-04-10T12:34:56 < karlp> look at all the shit they pil einto this for the price. it's selling a pre-packaged demo app, I love it.. https://www.aliexpress.com/item/1005006120708226.html
2024-04-10T12:41:45 < sauce> that is pretty impressive
2024-04-10T12:50:35 < zyp> nice
2024-04-10T12:54:45 < zyp> this looks like the sort of shit I would have loved as a kid
2024-04-10T14:55:14 -!- ventYl [~ventyl@bband-dyn218.95-103-122.t-com.sk] has quit [Ping timeout: 252 seconds]
2024-04-10T15:19:14 < qyx> hm trying polyurethane pcb varnish
2024-04-10T15:19:31 < qyx> I need a dust-free container of some sort
2024-04-10T15:19:54 < qyx> also I will probably apply it with a brush instead of spraying, it makes a lot of mess
2024-04-10T15:20:02 < qyx> and I need to mask all connectors and stuff
2024-04-10T15:20:11 < qyx> actually much more work than to paint it manually
2024-04-10T15:37:48 -!- Flea86 [~maomao@user/Flea86] has quit [Ping timeout: 255 seconds]
2024-04-10T15:51:04 -!- ventYl [~ventyl@bband-dyn78.178-40-23.t-com.sk] has joined ##stm32
2024-04-10T16:03:38 -!- scrts [~scrts2@23.28.144.38] has quit [Quit: The Lounge - https://thelounge.chat]
2024-04-10T16:04:02 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-10T16:04:45 < jpa-> yeah, i have also come into the conclusion that brushing is much easier than spraying
2024-04-10T16:20:27 < Ecco> I'm looking at different energy harvesting ICs
2024-04-10T16:21:07 < Ecco> Essentially they try and suck the max amount of power out of a small solar panels to recharge a battery
2024-04-10T16:21:57 < Ecco> Each datasheet mentions several parameters (efficiency, minimum input voltage of solar panel, etc…), among which a "quiescent current". Does that mean that the IC also draws current from the battery?
2024-04-10T16:22:23 < qyx> of course it does
2024-04-10T16:22:26 < qyx> every IC does
2024-04-10T16:23:23 < qyx> which ones did you look at?
2024-04-10T16:23:33 < Ecco> for example spv1050 from ST
2024-04-10T16:23:43 < Ecco> But wait, I agree that every IC does
2024-04-10T16:23:51 < Ecco> but those are supposed to *feed* the battery *from* the panel
2024-04-10T16:24:05 < karlp> using what, superconducting imps?
2024-04-10T16:24:06 < Ecco> so I would assume they'd draw/leak power from the panel, not from the battery
2024-04-10T16:24:24 < Ecco> well, if the panel doesn't provide any current, then the IC could just do nothing?
2024-04-10T16:24:49 < qyx> there still is some leakage current
2024-04-10T16:24:53 < qyx> it is mentioned in the datasheed
2024-04-10T16:24:56 < zyp> at the very least, whatever it'd have to block reverse current flow would have a leakage spec
2024-04-10T16:25:10 < Ecco> Yes it is mentioned in the datasheet
2024-04-10T16:25:12 < qyx> also spv1050 has some startup charge pump iirc
2024-04-10T16:25:15 < Ecco> but I'm wondering where it comes from
2024-04-10T16:25:47 < zyp> have you ever seen the I/V current of a diode?
2024-04-10T16:25:58 < zyp> just a regular simple semiconductor
2024-04-10T16:25:58 < Ecco> yes
2024-04-10T16:26:09 < zyp> noticed how it leaks some current in reverse?
2024-04-10T16:26:15 < Ecco> yeah, gotcha
2024-04-10T16:26:24 < Ecco> ok, so let me rephrase my initial question :)
2024-04-10T16:26:27 < zyp> that's a general semiconductor thing
2024-04-10T16:26:31 < Ecco> I feel like the leakage current is really high
2024-04-10T16:26:43 < Ecco> for a device that's supposed to *very efficiently* gather micro-watts of power
2024-04-10T16:27:01 < zyp> quiescent current could vary depending on mode, it's not necessarily one fixed number
2024-04-10T16:27:19 < zyp> you might very well be looking at the spec for the current lost while the solar panel is active
2024-04-10T16:27:56 < Ecco> ok that makes sense
2024-04-10T16:28:06 < Ecco> I'm also looking at another part from ti
2024-04-10T16:28:11 < Ecco> bq25570
2024-04-10T16:28:34 < zyp> I don't see an Iq spec in SPV1050 datasheet, it lists Isd, Isb and Iop however
2024-04-10T16:28:56 < Ecco> zyp: yeah, I was looking at those numbers too
2024-04-10T16:29:17 < jpa-> SP1050 has them on page 7
2024-04-10T16:29:18 < Ecco> the ti one behaves in a similar way
2024-04-10T16:29:25 < jpa-> *SPV1050
2024-04-10T16:29:43 < Ecco> jpa-: yeah, that's what zyp was mentionning too I think
2024-04-10T16:29:52 < zyp> yep
2024-04-10T16:29:56 < Ecco> both devices seem to have a "shutdown mode" that you drive witha  pin
2024-04-10T16:29:59 < Ecco> and a "standby" mode
2024-04-10T16:30:07 < Ecco> but then I guess my question is: why two modes?
2024-04-10T16:30:19 < zyp> bq25570 has three different Iq specs, correspondingly
2024-04-10T16:30:30 < Ecco> Couldn't the IC just switch itself on and off when there is/isn't power coming from the panel?
2024-04-10T16:30:40 < zyp> sure
2024-04-10T16:31:04 < zyp> and the logic to do that draws power
2024-04-10T16:31:06 < jpa-> it does, but it takes some power to monitor whether power is coming
2024-04-10T16:31:27 < qyx> you are doing it wrong. just connect 8 cells to a single lifepo4 with a diode, done
2024-04-10T16:31:27 < zyp> so ship mode is when you turn off that, for devices that's in a box on a shelf for a long time waiting to be sold
2024-04-10T16:31:29 < jpa-> if you have more efficient way to do that, you can raise the pin :)
2024-04-10T16:31:30 < qyx> #profit
2024-04-10T16:31:54 < Ecco> qyx: I didn't get the joke :)
2024-04-10T16:31:57 < Ecco> jpa-: ok, makes sense
2024-04-10T16:32:12 < qyx> Ecco: the joke is it works and surprisingly efficiently
2024-04-10T16:32:31 < jpa-> hmm.. but BATT_CONN is output..
2024-04-10T16:32:58 < qyx> for bonus points put a zener parallel to the lifepo4 to burn the redundant current to keep Vbat under 4V
2024-04-10T16:33:05 < qyx> but the battery is fine even with more
2024-04-10T16:33:13 < Ecco> jpa-: duh, weird. I also assumed it was an input
2024-04-10T16:33:51 < qyx> of course I am talking about small lifepos4 in the order of 100-300 mAh, not the big ones
2024-04-10T16:35:29 < Ecco> I've never looked at lifepos4 batteries. In a gist, how do they compare to LiPo?
2024-04-10T16:35:38 < zyp> more boring
2024-04-10T16:35:39 < jpa-> less explosions
2024-04-10T16:35:46 < zyp> exactly
2024-04-10T16:35:49 < Ecco> damn, so they remove all the fun!
2024-04-10T16:36:21 < Ecco> Why aren't there in every single consumer product then?
2024-04-10T16:36:35 < jpa-> less energy per mass/volume
2024-04-10T16:36:38 < Ecco> I guess they must have some drawbacks
2024-04-10T16:36:42 < qyx> even more basic circuit is a LTO cell + 5 solar cells in series + a diode
2024-04-10T16:37:06 < qyx> and directly feed a mcu because LTO has 1.8-2.9 V range
2024-04-10T16:37:24 < Ecco> what about MPTT in those scenarios?
2024-04-10T16:37:27 < qyx> so you can omit UVLO required for lifepo4 and use PVD in a STM32 instead
2024-04-10T16:37:29 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-10T16:38:09 < qyx> it is doing "passive mppt", that is, the battery is keeping the solar cell voltage in the optimal range with it's own voltage
2024-04-10T16:38:42 < qyx> 8 cells are about 4.5 V open-circuit, MPP is about 80% of that
2024-04-10T16:38:51 < qyx> that is 3.6
2024-04-10T16:39:00 < qyx> usable range for lifepo4 is 3.3-3.6 V
2024-04-10T16:39:26 < qyx> so you are basically balancing on top of the curve most of the battery SoC
2024-04-10T16:39:35 < qyx> (account for some diode drop)
2024-04-10T16:44:24 < qyx> but don't get me wrong, i am not advocating using such make:r quality tricks in a serious product
2024-04-10T16:45:08 < qyx> spv1040 and spv1050 both work, I have a PCB for some bqxxx but I didn't try it yet
2024-04-10T16:45:24 < qyx> also lt3652
2024-04-10T16:50:12 < Ecco> thanks, that was about to be my next question :)
2024-04-10T16:51:43 < Ecco> spv1040 is a big chip!
2024-04-10T16:52:09 < qyx> I have field proven lt3652 with 19V panels, uhm, for about 7 years now on multiple boards
2024-04-10T16:52:36 < qyx> with a single liion on th eoutput
2024-04-10T16:53:08 < Ecco> sweet
2024-04-10T16:53:34 < Ecco> Also, I was wondering: do LiPo battery age well?
2024-04-10T16:53:48 < Ecco> I mean, I know they lose some capacity as you cycle them
2024-04-10T16:54:07 < Ecco> but in a low-power environment, it may take a really long time to even do 100 cycles
2024-04-10T16:56:56 < Ecco> Ha, this one seems pretty neat: https://www.analog.com/en/products/adp5090.html
2024-04-10T16:57:18 -!- rpifan [~rpifan@p54ac5a96.dip0.t-ipconnect.de] has joined ##stm32
2024-04-10T16:57:18 -!- rpifan [~rpifan@p54ac5a96.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-10T16:57:18 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T17:08:21 < Ecco> (I just looked it up, it costs about $6 to get an STM32's datasheet as a book on lulu.com)
2024-04-10T17:12:44 < Ecco> I'm reading the datasheet of ADP5090, and it seems like this IC has two seperate "outputs"
2024-04-10T17:13:17 < Ecco> 1/ The battery, with very little leakage (just a few nA)
2024-04-10T17:13:36 < Ecco> 2/ An additional external "sys" capacitor, that is used to power the chip itself
2024-04-10T17:14:00 < Ecco> My understanding is that the IC will need to charge the sys capacitor before being functional (that's what they call cold-start)
2024-04-10T17:14:16 < Ecco> does that make sense or am I reading the datasheet completely wrong?
2024-04-10T17:17:42 < zyp> doesn't look like lulu.com does more than 800 pages, the last stm32 manual I read was like 3700 pages
2024-04-10T17:18:18 < Ecco> zyp: yeah, that's why I mentionned the datasheet that's only 133 pages
2024-04-10T17:18:40 < qyx> split by peripherals!
2024-04-10T17:18:42 < Ecco> actually I'm not sure I'd want to handle a 3000 pages book
2024-04-10T17:18:49 < Ecco> Yeah, that's an option for sure
2024-04-10T17:19:12 < qyx> "in this bookshelf I have G0 and G4"
2024-04-10T17:19:28 < zyp> :)
2024-04-10T17:19:29 < Ecco> :-D
2024-04-10T17:19:43 < Ecco> zyp: which ref has a 3.7k pages refman?
2024-04-10T17:40:28 < zyp> h7s7
2024-04-10T17:40:46 < zyp> it's 3791 pages to be exact
2024-04-10T17:55:56 < qyx> that lulu is pretty cheap
2024-04-10T17:57:11 < qyx> I am curious if they allow copying/printing copyrighted material without consent
2024-04-10T17:59:33 < karlp> did we hav ea favourite json encoding library for little C land?
2024-04-10T18:00:20 < zyp> just encoding?
2024-04-10T18:00:28 < karlp> well, both eventually.
2024-04-10T18:00:40 < zyp> I think I've heard good stuff about cJSON
2024-04-10T18:00:41 < karlp> we'r eusing two shitty unmaintained libraries for encoding/decoding at the moment
2024-04-10T18:00:48 < karlp> yeah, mosquitto uses cJSON.
2024-04-10T18:01:04 < karlp> I'd forogtten that one, started with "awesome-json" on github, but it felt slim
2024-04-10T18:01:24 < karlp> oh right, it's decoding 
2024-04-10T18:01:49 < zyp> there's a cJSON_Print()
2024-04-10T18:02:10 < karlp> that's for after youv'e aprsed it.
2024-04-10T18:02:26 < karlp> oh no, they do seem to have creation as wewll.
2024-04-10T18:02:46 < zyp> I would assume you could build a tree yourself in the same way and output it
2024-04-10T18:02:50 < karlp> hrm, uses heap allocated strings though.
2024-04-10T18:03:31 < zyp> there's a cJSON_PrintPreallocated() too, but doesn't seem to support any sort of streaming/chunking
2024-04-10T18:03:43 < zyp> so yeah
2024-04-10T18:04:03 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 255 seconds]
2024-04-10T18:04:21 < zyp> that's a bit weird, JSON is easy to stream out
2024-04-10T18:05:16 < zyp> maybe use your own streaming handler for objects/arrays and just call the lib to format basic types
2024-04-10T18:25:17 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-10T18:44:35 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 252 seconds]
2024-04-10T19:44:04 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T19:44:13 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-10T19:45:09 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T19:59:56 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 252 seconds]
2024-04-10T20:03:20 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 260 seconds]
2024-04-10T21:09:02 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-10T21:12:54 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T21:21:23 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T22:23:41 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-10T22:45:58 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-10T22:56:37 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-10T22:56:56 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T23:00:07 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-10T23:00:19 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-10T23:04:29 -!- HelloShitty [~psysc0rpi@2001:8a0:e143:a100:478:7dff:fe45:a43d] has quit [Ping timeout: 240 seconds]
2024-04-10T23:07:22 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has joined ##stm32
2024-04-10T23:13:17 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-10T23:14:21 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-10T23:15:29 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-4006-8bc4-6a96-bcc4.fixed6.kpn.net] has joined ##stm32
2024-04-10T23:28:40 < qyx> I must say the urethane coating cures too long for my liking
2024-04-10T23:37:22 < qyx> The chemical cross linking reactions take several days under ambient
2024-04-10T23:37:24 < qyx> conditions.
2024-04-10T23:41:14 < BrainDamage> does uv accelerates it?
2024-04-10T23:42:27 < qyx> no, baking at 60°C for 24h
2024-04-10T23:42:44 < qyx> but I didn't try
2024-04-10T23:43:08 < qyx> they say it is fluorescent under uv-a
--- Day changed to huhti 11 2024
2024-04-11T00:06:28 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 260 seconds]
2024-04-11T00:14:12 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T00:36:23 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-11T00:36:35 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T00:40:30 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has joined ##stm32
2024-04-11T00:42:43 -!- c10ud [~c10ud@user/c10ud] has quit [Ping timeout: 246 seconds]
2024-04-11T00:42:45 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-11T00:42:48 < Laurenceb_> reee truly epic fail
2024-04-11T00:43:07  * Laurenceb_ isnt a hyperfail shareholder and company is prob being sold for megabux
2024-04-11T00:43:22 < zyp> how's the motor for your lighter going?
2024-04-11T00:43:54 < Laurenceb_> lighter? lulwut
2024-04-11T00:44:10 < zyp> idk, you said you were gonna build a lighter motor
2024-04-11T00:44:29 < Laurenceb_> oh yeah I didnt follow you lol
2024-04-11T00:44:50 < Laurenceb_> stator backing is going to be 6000 alu now, with silicone gel thermal compound
2024-04-11T00:47:25 < Laurenceb_> this would be for Shinkansen
2024-04-11T00:51:23 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-11T00:51:35 -!- rpifan [~rpifan@p200300d26718e2004ebeb26ccc0a7c2a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-11T00:51:35 -!- rpifan [~rpifan@p200300d26718e2004ebeb26ccc0a7c2a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-11T00:51:35 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T00:52:32 < qyx> qt hjalp, pyqtSignal(type) requires a type of the signal parameter
2024-04-11T00:53:02 < zyp> and?
2024-04-11T00:53:03 < qyx> I want to pass an object instance, but I cannot use pyqtSignal(MyClass)
2024-04-11T00:53:15 < qyx> it works with an int for example
2024-04-11T00:53:22 < zyp> hmm, there's probably a proxy type for that
2024-04-11T00:53:50 < zyp> try object
2024-04-11T00:53:54 < qyx> oh
2024-04-11T00:54:04 < zyp> ref. https://stackoverflow.com/questions/4523006/pyqt-signal-with-arguments-of-arbitrary-type-pyqt-pyobject-equivalent-for-new
2024-04-11T00:54:25 < qyx>     deleted = pyqtSignal(Object)
2024-04-11T00:54:25 < qyx> NameError: name 'Object' is not defined
2024-04-11T00:54:31 < zyp> lowercase
2024-04-11T00:54:38 < zyp> object is a python builtin
2024-04-11T00:54:50 < qyx> sorry dumb me
2024-04-11T00:54:52 < qyx> yeah
2024-04-11T00:56:08 < qyx> wroks thanks
2024-04-11T00:59:57 < qyx> also reading docs helps sometimes
2024-04-11T00:59:59 < qyx> Removes the toolbar from the main window layout and hides it. Note that the toolbar is not deleted.
2024-04-11T01:00:11 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-11T01:01:31 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T01:15:18 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has quit [Read error: Connection reset by peer]
2024-04-11T01:15:37 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-11T01:20:05 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-4006-8bc4-6a96-bcc4.fixed6.kpn.net] has quit [Read error: Connection reset by peer]
2024-04-11T01:29:05 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Remote host closed the connection]
2024-04-11T01:35:03 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has joined ##stm32
2024-04-11T01:37:04 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-11T01:39:14 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-11T01:51:30 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T02:45:24 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-11T02:52:25 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-11T03:13:34 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 246 seconds]
2024-04-11T03:14:13 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-11T03:17:44 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Remote host closed the connection]
2024-04-11T03:19:12 -!- BrainDamage [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-11T03:21:31 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-11T03:24:03 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-11T04:06:13 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-11T04:07:01 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T04:18:17 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-11T06:52:03 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-11T07:49:46 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-11T08:22:39 < qyx> goal for the summer: get realtime sound recording and encoding in opus on U5 under 1 mA
2024-04-11T08:28:55 < jpa-> sounds good, will you make me a walkie-talkie?
2024-04-11T08:51:12 < antto> a walkopus-talkopus
2024-04-11T09:31:05 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-11T10:00:56 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-11T10:04:21 -!- rob_w_ [~bob@2001:a61:60d8:5e01:250a:26de:6146:cadb] has joined ##stm32
2024-04-11T12:49:16 < qyx> jpa-: do you need some? I can ship
2024-04-11T12:56:45 < BrainDamage> what target bitrate?
2024-04-11T12:58:45 < qyx> 32kbps
2024-04-11T12:59:05 < qyx> I tried on overlocked L4 once and it worked, but definitely not sub-1mA
2024-04-11T12:59:41 < qyx> on the other hand, I did PDM filtering in software on that L4 too
2024-04-11T13:00:08 < qyx> but U5 claims 18 uA/MHz in run mode
2024-04-11T13:07:22 < jpa-> qyx: i need something i can strap on bike helmet and talk with kid and wife
2024-04-11T13:10:38 < qyx> when the war started, I wanted to buy a CB radio, just in case
2024-04-11T13:11:03 < qyx> basically the only long range possibility for a non-ham
2024-04-11T13:31:01 < BrainDamage> jpa-: aren't there motorbike walkie talkies exacty for this porpoise?
2024-04-11T14:25:05 -!- quinor [08c0f10716@2a03:6000:1812:100::dad] has quit [Remote host closed the connection]
2024-04-11T14:25:14 -!- quinor [08c0f10716@2a03:6000:1812:100::dad] has joined ##stm32
2024-04-11T14:25:49 -!- quinor [08c0f10716@2a03:6000:1812:100::dad] has quit [Remote host closed the connection]
2024-04-11T14:28:10 -!- quinor [08c0f10716@2a03:6000:1812:100::dad] has joined ##stm32
2024-04-11T14:42:37 -!- akaWolf [~akaWolf@akawolf.org] has quit [Ping timeout: 256 seconds]
2024-04-11T14:44:12 -!- akaWolf [~akaWolf@akawolf.org] has joined ##stm32
2024-04-11T14:44:54 < jpa-> BrainDamage: some yeah, but they tend to be ridicuously expensive
2024-04-11T14:45:02 < jpa-> at least last time i searched
2024-04-11T14:45:22 < jpa-> 100+ EUR per each; so clearly worth spending 50 hours of my time making a custom one ;)
2024-04-11T15:03:46 < Steffanx> Didn't you already make it to HaD with one?
2024-04-11T15:05:05 < qyx> exactly my attitude to home improvement projects
2024-04-11T15:15:18 < jpa-> Steffanx: yes, but i made into fail of the week
2024-04-11T15:18:55 -!- rpifan [~rpifan@p200300d26718e200e1681b1f79f3294f.dip0.t-ipconnect.de] has joined ##stm32
2024-04-11T15:18:55 -!- rpifan [~rpifan@p200300d26718e200e1681b1f79f3294f.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-11T15:18:55 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T16:13:18 < PaulFertser> 100 EUR for a pair https://www.aliexpress.com/item/4000960739831.html
2024-04-11T16:14:10 < PaulFertser> IP65
2024-04-11T16:17:53 < PaulFertser> 75 EUR https://www.aliexpress.com/item/1005003151669040.html
2024-04-11T16:19:45 < PaulFertser> And yes, Sena and Scala are expensive as fuck.
2024-04-11T16:21:16 < qyx> jpa-: what max distance, behind the corner too?
2024-04-11T16:24:41 < PaulFertser> That thing from the last link claims 1 km with clear LOS.
2024-04-11T16:36:45 < jpa-> qyx: 100m would be plenty, but head may be between antennas
2024-04-11T16:38:25 < jpa-> PaulFertser: thanks for the links; i still need to figure out if they work for 3 riders
2024-04-11T16:40:40 < PaulFertser> jpa-: the first one does, the last one no.
2024-04-11T16:41:30 < PaulFertser> jpa-: also for a rider and passenger the first one can connect to a regular Bluetooth HFP device it seems.
2024-04-11T16:42:42 < PaulFertser> There must be some discussions on cheap chinese intercoms on advrider or similar places.
2024-04-11T16:42:56 < PaulFertser> This fodsports brand I've seen mentioned many times.
2024-04-11T16:46:33 < PaulFertser> Probably the last one can connect to a regular headset too, one would need to check the manual to tell.
2024-04-11T16:50:53 < jpa-> yeah.. the separate headphone & mic installation is kind of annoying for a bike helmet
2024-04-11T16:51:04 < jpa-> having it be like a regular bluetooth handsfree would be great
2024-04-11T16:51:22 < jpa-> though i guess i could also just try whether mobile data has improved to the point that it wouldn't drop connection all the time
2024-04-11T16:54:50 < PaulFertser> You want it on a push bike moving at slow speeds?
2024-04-11T16:54:58 < PaulFertser> Because separate mic is there for a reason.
2024-04-11T16:55:24 < jpa-> phone calls seem fine on my regular bluetooth handsfree
2024-04-11T16:58:38 < PaulFertser> On motor bike speeds?
2024-04-11T16:58:49 < jpa-> .. this is for bicycle
2024-04-11T17:01:26 < PaulFertser> That's why I asked about the speed. Probably you're talking about your cool electric bicycle that goes over 50 km/h.
2024-04-11T17:02:13 < jpa-> no, just the normal 25 km/h that is legal here, and it's fine if it works at 20km/h
2024-04-11T17:04:29 < PaulFertser> That's an odd speed limit, I think it doesn't take to be a trained sports man to go above that for like an hour straight if the road is nice and no elevation.
2024-04-11T17:04:53 < PaulFertser> Without electric assist of any sorts.
2024-04-11T17:05:00 < jpa-> yeah, but why would you need electric assist in tailwind on flat road anyway?
2024-04-11T17:05:06 < jpa-> it's for the hills
2024-04-11T17:05:08 < karlp> yeah, the 25kph euro llimit is bullshit low.
2024-04-11T17:05:31 < PaulFertser> I think I was able to do 100 km in less than 4 hours...
2024-04-11T17:05:54 < karlp> I love my scooter braking down hills to keep to 25....
2024-04-11T17:06:22 < PaulFertser> At least it charges the battery instead of wasting all the energy doing it?
2024-04-11T17:06:39 < karlp> no idea, not my problem.
2024-04-11T17:06:57 < jpa-> on ebikes at least you can pedal faster no problem
2024-04-11T17:09:05 < PaulFertser> jpa-: how about a wifi link between two smartphones, I wonder if can be any good
2024-04-11T17:09:19 < jpa-> we've tried that, it drops out constantly
2024-04-11T17:09:46 < jpa-> mobile data worked better but that also drops out time to time and the programs we tried were slow or unable to reconnect
2024-04-11T17:09:52 < jpa-> but that was some years ago
2024-04-11T17:10:44 < jpa-> regular phonecall would be fine but gets a bit costly on longer trips :)
2024-04-11T17:11:16 < jpa-> i guess considering phonecalls are now ip-based in LTE, then maybe mobile data should work too?
2024-04-11T17:11:34 < PaulFertser> These days you have webrtc and jitsi meet servers so can probably hack the backend for faster automatic reconnection without too much effort?
2024-04-11T17:12:07 < PaulFertser> Yes, VoLTE goes over LTE when it doesn't drop to UMTS.
2024-04-11T17:13:26 < PaulFertser> I mean even if the mobile station is normally using LTE all the time it might be downgrading to UMTS or even GSM for phone calls if there's some configuration issue on either side, and that's not actually uncommon.
2024-04-11T17:21:32 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-11T17:24:00 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 255 seconds]
2024-04-11T17:26:03 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Remote host closed the connection]
2024-04-11T17:32:58 < ventYl> there are compatibility issues. LTE is just a standard packet-switched network with no circuit switching capabilities, unlike UMTS and GSM
2024-04-11T17:33:18 < ventYl> so the VoLTE is in fact just a VoIP specific to LTE
2024-04-11T17:33:56 < ventYl> both cellphone, station and network have to supprot VoLTE with compatible parameters, otherwise cellphone will reconnect either to UMTS or GSM (whichever available) in order to make a call
2024-04-11T17:34:16 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-11T17:34:27 < ventYl> LTE as a network is technically not capable of making calls :>
2024-04-11T17:41:13 < PaulFertser> ventYl: yes, that's what I'm saying, and sometimes the operator side just whitelists a set of mobile stations and everything else has to downgrade the connection even if all parameters would be matching.
2024-04-11T17:42:14 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-11T17:45:33 -!- NEYi_ [~NEYi@109.251.216.38] has joined ##stm32
2024-04-11T17:49:47 -!- NEYi [~NEYi@109.251.216.38] has quit [Ping timeout: 264 seconds]
2024-04-11T17:57:26 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-11T17:59:08 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T18:19:02 -!- rob_w_ [~bob@2001:a61:60d8:5e01:250a:26de:6146:cadb] has quit [Remote host closed the connection]
2024-04-11T18:23:12 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-4145-53e9-ccb7-6f27.fixed6.kpn.net] has joined ##stm32
2024-04-11T18:43:25 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 268 seconds]
2024-04-11T18:52:48 < zyp> PaulFertser, assuming there even is GSM or UMTS coverage
2024-04-11T19:05:59 -!- specing [~specing@user/specing] has quit [Remote host closed the connection]
2024-04-11T19:07:05 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-11T19:08:03 < ventYl> in networks which are not fully VoLTE-enabled, there usually is at least one of them available
2024-04-11T19:18:52 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T19:28:48 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has joined ##stm32
2024-04-11T19:56:32 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 252 seconds]
2024-04-11T20:01:56 -!- specing [~specing@user/specing] has quit [Quit: ZNC - https://znc.in]
2024-04-11T20:02:17 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-11T20:27:09 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-11T20:39:41 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-4145-53e9-ccb7-6f27.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-11T21:04:51 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Ping timeout: 260 seconds]
2024-04-11T21:07:58 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has quit [Remote host closed the connection]
2024-04-11T21:08:24 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has joined ##stm32
2024-04-11T21:10:24 < antto> i asked my $distributor about those RGB LEDs, OSS85ST63, MOQ is 2000 and the price is $0.071215
2024-04-11T21:12:16 < antto> so that's roughly twice the price of those XL-3528RGBW-WS2812B, altho i have no clue how much the customs penalty would be
2024-04-11T21:14:25 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has quit [Remote host closed the connection]
2024-04-11T21:14:52 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has joined ##stm32
2024-04-11T21:17:00 -!- NEYi_ [~NEYi@109.251.216.38] has quit [Read error: Connection reset by peer]
2024-04-11T21:17:22 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-11T21:23:41 -!- jtj [~jtj@212.66.207.170] has quit [Quit: Konversation terminated!]
2024-04-11T21:27:15 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-11T21:37:16 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-11T21:42:36 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-101-b1d2-7cc1-f274.fixed6.kpn.net] has joined ##stm32
2024-04-11T21:50:10 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T21:51:09 -!- MGF_Fabio [~MGF_Fabio@2a01:827:1a93:cf03:4cb9:ff15:f49f:5f4e] has quit [Ping timeout: 256 seconds]
2024-04-11T21:52:02 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-11T22:19:50 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-11T22:20:52 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T22:22:38 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-11T22:50:38 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-11T22:52:13 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-11T23:00:54 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-11T23:13:28 < Steffanx> You gave custom issues when you order from AliExpress and co antto? You can pay VAT and stuff in advance right?
2024-04-11T23:13:32 < Steffanx> Have*
2024-04-11T23:26:10 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-11T23:26:27 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T23:26:46 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Ping timeout: 268 seconds]
2024-04-11T23:30:08 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-11T23:30:14 < nomorekaki> hello steff
2024-04-11T23:32:20 < Steffanx> Hello Mr nomorekaki 
2024-04-11T23:32:29 < nomorekaki> musics: https://www.youtube.com/watch?v=6n_de7Ieoh0
2024-04-11T23:33:14 < nomorekaki> do you have any musics for musics exchange?
2024-04-11T23:34:55 < Steffanx> Uhm.
2024-04-11T23:38:45 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-11T23:39:04 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-11T23:40:23 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-11T23:40:24 < qyx> so I aggregated about 20 incidents/customer reports of a device misbehaving and identified several exact locations causing all the mentioned issues
2024-04-11T23:40:31 < qyx> https://bin.jvnv.net/file/ApBnQ/Screenshot_2024-04-11_22-38-38.png
2024-04-11T23:41:27 < zyp> what issues?
2024-04-11T23:41:28 < qyx> all those red marks are in a part of the PCB which is the closest-most one to a non-waterproof connector
2024-04-11T23:42:14 < qyx> LSE stopping not signalling a semaphore to release measured data, HSE intermittently stopping causing muxing artifacts, NTC thermistor offsets, random resets
2024-04-11T23:43:15 < zyp> what's up with the HSE crystal? series resistors?
2024-04-11T23:43:19 < qyx> random comm failures on the can bus
2024-04-11T23:43:36 < qyx> it is not a crystal, it is jpa- approved connection of a clipped sine oscillator
2024-04-11T23:43:45 < antto> Steffanx, aliexpress?!
2024-04-11T23:43:48 < zyp> oh, oscillator
2024-04-11T23:43:50 < zyp> makes sense
2024-04-11T23:43:55 < qyx> cap in series, resistor between osc_in/osc_out
2024-04-11T23:44:20 < zyp> wat
2024-04-11T23:44:51 < qyx> so I am suspecting oxidation
2024-04-11T23:44:52 < qyx> https://bin.jvnv.net/file/e3E5f/Screenshot_2024-04-11_22-44-37.png
2024-04-11T23:45:34 < zyp> I don't get it, but if it's jpa- approved it's probably good
2024-04-11T23:45:55 < qyx> I think I got it from him and verified on interwebs
2024-04-11T23:46:01 < qyx> also used a few times in the past
2024-04-11T23:47:47 < qyx> so I am sad, because I need to redo about 50 devices, conformal coating + replace that damned HRS connector
2024-04-11T23:49:13 < qyx> so I need a 6 pin push-pull connector
2024-04-11T23:49:35 < qyx> probably a (cough) lee-hmo
2024-04-11T23:50:07 < nomorekaki> show us the connector
2024-04-11T23:50:29 < qyx> HRS HR30
2024-04-11T23:50:51 < qyx> it is waterproof when mated
2024-04-11T23:51:52 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-11T23:52:15 < nomorekaki> so you didn't have anything connected+
2024-04-11T23:52:18 < nomorekaki> ?
2024-04-11T23:52:32 < qyx> no, it is a diagnostic console
2024-04-11T23:53:28 < nomorekaki> you didn't have a plug on it?
2024-04-11T23:53:54 < qyx> no :>
2024-04-11T23:54:18 < nomorekaki> does the system have waterproof plug?
2024-04-11T23:54:31 < nomorekaki> connector system*
2024-04-11T23:55:19 < qyx> yes it does but the original cap is just a rubberis one, not the kind attaching to the connector as it should
2024-04-11T23:57:46 < zyp> hrm, the pin allocations on h7s7-dk is a bit annoying
2024-04-11T23:58:03 < qyx> meh lemo ECG.0B is IP50
2024-04-11T23:59:36 < zyp> RMII_MDC is connected to a jumper that can pick between PB2 and PC1
--- Day changed pe huhti 12 2024
2024-04-12T00:00:08 < zyp> PB2 doesn't even look like it has a MDC AF, so idk if the intention is that you should bitbang MDIO
2024-04-12T00:01:03 < zyp> PC1 on the other hand has MDC AF, but is shared with TRACE_D0
2024-04-12T00:01:16 < qyx> is the dk for h7s only?
2024-04-12T00:01:26 < zyp> yes
2024-04-12T00:02:24 < zyp> so okay, maybe I don't need TRACE, I could potentially get by with SWO
2024-04-12T00:02:38 < qyx> wasn't there h7 with two ethernets?
2024-04-12T00:02:41 < zyp> but SWO is on PB3 which is by default shared with LCD_R4
2024-04-12T00:02:59 < qyx> who cares, you just get a bit of red flickering
2024-04-12T00:03:08 < zyp> it has solder jumpers that could remap it to PA7, however
2024-04-12T00:03:23 < zyp> but PA7 is already in use for RMII_CRS_DV
2024-04-12T00:03:45 < qyx> is crsdv even used?
2024-04-12T00:03:58 < zyp> yes, that's the RX data valid signal in RMII
2024-04-12T00:04:08 < qyx> hm what's the other one unused then
2024-04-12T00:04:28 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-101-b1d2-7cc1-f274.fixed6.kpn.net] has quit [Ping timeout: 246 seconds]
2024-04-12T00:04:32 < qyx> rxerr maybe, there were some optional
2024-04-12T00:04:45 < qyx> but rmii is one of the things I always need to look up when drawing
2024-04-12T00:04:47 < zyp> yes
2024-04-12T00:05:09 < zyp> RMII requires six signals, two data signals and one valid signal in each direction
2024-04-12T00:05:31 < zyp> so [TR]XD[01], TXEN and CRS_DV
2024-04-12T00:05:42 < zyp> plus REF_CLK and MDIO/MDC
2024-04-12T00:06:29 < zyp> I don't get why they've used PC1 for TRACE_D0, the natural choice would be PE3
2024-04-12T00:07:03 < zyp> PE3 is hooked up to LCD_INT, presumably just a plain GPIO, maybe EXTI, no special AF
2024-04-12T00:09:15 < aandrew> I'm pretty sure they do it to fuck with the layout people
2024-04-12T00:15:44 < zyp> yeah, the intention is to bitbang mdio
2024-04-12T00:15:50 < zyp> «PB2 GPIO needs to simulate RMII_MDC when JP6 pin [1-2] is ON (default setting).»
2024-04-12T00:16:02 < zyp> «If dedicated RMII_MDC(PC1) is needed, JP6 pin2-3 ON, but the ARDUINO® A5 ADC cannot be used.»
2024-04-12T00:17:40 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has quit [Read error: Connection reset by peer]
2024-04-12T00:24:19 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has joined ##stm32
2024-04-12T00:29:38 < zyp> hmm, looks like they've managed to have an actually usable internal USB HS PHY in this chip
2024-04-12T00:29:55 < zyp> as in not just available on special balls on some undesireable BGA package
2024-04-12T00:36:07 < qyx> ok hr30 doesn't leak along pins
2024-04-12T00:42:26 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-12T01:09:36 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-12T01:14:40 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-101-b1d2-7cc1-f274.fixed6.kpn.net] has joined ##stm32
2024-04-12T01:29:07 -!- Alexer [~alexer@alexer.net] has quit [Ping timeout: 255 seconds]
2024-04-12T01:30:57 < karlp> series to osc-out seems weird. I've never looked at stm32 osc stuff, kinetis it is very explicitly NC for the osc-out pin.
2024-04-12T01:35:35 < zyp> for stm32 it's «In bypass mode the HSE oscillator is switched off and the input pin is a standard I/O.»
2024-04-12T01:36:01 -!- Alexer [~alexer@alexer.net] has joined ##stm32
2024-04-12T01:36:15 < zyp> but presumably «clipped sine» means qyx is not using bypass mode
2024-04-12T01:36:16 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-101-b1d2-7cc1-f274.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-12T01:37:23 < qyx> no I am not using bypass mode, the resistor is biasing the clock wherever stm32 needs it to be
2024-04-12T01:37:41 < qyx> apparently.
2024-04-12T01:38:19 < zyp> hmm, max rise/fall times are 0.3 / freq, at least for h7rs which is the datasheet I've got open
2024-04-12T01:38:31 < zyp> that sounds like it should accomodate a clipped sine
2024-04-12T01:38:44 < zyp> idk if that's 0.3 rise and 0.3 fall, or both combined
2024-04-12T01:39:17 < qyx> https://www.eevblog.com/forum/microcontrollers/stm32-and-external-active-oscillator/
2024-04-12T01:40:09 < zyp> «STM32 software won't toggle the HSE ready flag when HSE is configured in bypass mode» is it even supposed to when bypassed?
2024-04-12T01:41:23 < qyx> no but CSS should work
2024-04-12T01:41:30 < qyx> and I am not using bypass mode
2024-04-12T01:41:38 < qyx> bypass doesn't work with clipped sine
2024-04-12T01:41:58 < zyp> idk, looks like h7rs would be happy with it
2024-04-12T01:42:36 < qyx> no because in bypass mode it is a schmitt trigger gpio input, 0.8 Vpp is not enough for it
2024-04-12T01:42:52 < qyx> you need a cmos output oscillator for bypass mode
2024-04-12T01:42:56 < zyp> h7rs says Vpp min = 0.2V
2024-04-12T01:43:18 < zyp> «200 mV is the minimum peak-to-peak amplitude @25°C (0.1 V< Vdc < VDD IO -0.1 V where Vdc is the DC component of the input signal).»
2024-04-12T01:44:29 < qyx> which DS and section?
2024-04-12T01:44:32 < zyp> there's even a diagram showing a sine: https://www.st.com/resource/en/datasheet/stm32h7s3a8.pdf page 189
2024-04-12T01:46:28 < qyx> no they are saying 0.3/0.7 Vdd for "external digital clock"
2024-04-12T01:46:34 < qyx> but also "external low swing clock"
2024-04-12T01:46:39 < qyx> which is new to me
2024-04-12T01:46:46 < zyp> oh, right
2024-04-12T01:47:30 < qyx> yes table 54 is for bypass mode, digital cmos input
2024-04-12T01:47:35 < qyx> table 55 is for analol clock
2024-04-12T01:48:01 < qyx> that's what I am using
2024-04-12T01:48:14 < qyx> but without the resistor the Vcm can be anything
2024-04-12T01:50:09 < zyp> right, so low swing analog is a separate mode
2024-04-12T01:50:20 < zyp> from RM:
2024-04-12T01:50:20 < zyp> External clock source (HSE bypass)
2024-04-12T01:50:21 < zyp> In this mode the oscillator is not used, and an external clock source must be provided to OSC_IN pin. The external clock can be low swing (analog) or digital. The OSC_OUT pin must be left HI-Z (see Figure 41).
2024-04-12T01:50:41 < zyp> block diagram shows it having an internal «clock squarer»
2024-04-12T01:51:27 < qyx> wat, the DS says explicitly it is called bypass mode when digital output osc is used
2024-04-12T01:51:43 < zyp> there's two bypass modes
2024-04-12T01:51:51 < qyx> In bypass mode the HSE oscillator is switched off and the input pin is a standard I/O.
2024-04-12T01:52:04 < qyx> However,
2024-04-12T01:52:04 < qyx> the recommended clock input waveform is shown in Figure 31
2024-04-12T01:52:10 < qyx> from fig 31 it is crystal clear
2024-04-12T01:52:12 < zyp> RCC_CR has HSEBYP to select bypass mode, and HSEEXT to choose which bypass mode
2024-04-12T01:52:46 < zyp> HSEEXT:externalhighspeedclocktypeinBypassmode
2024-04-12T01:52:46 < zyp> Set and reset by software to select the external clock type (analog or digital).
2024-04-12T01:52:46 < zyp> The external clock must be enabled with the HSEON bit to be used by the device. The HSEEXT bit can be written only if the HSE oscillator is disabled.
2024-04-12T01:52:49 < zyp> 0: HSE in analog mode (default after reset)
2024-04-12T01:52:51 < zyp> 1: HSE in digital mode
2024-04-12T01:53:18 < zyp> that's your low swing input
2024-04-12T01:54:13 < qyx> that's interesting, no such things on G4 for example
2024-04-12T01:54:41 < zyp> yeah, h7rs is one of the newest parts
2024-04-12T01:54:47 < qyx> whereas g4 says "The external clock signal (square, sinus or triangle) with ~40-60 % duty
2024-04-12T01:54:50 < qyx> cycle depending on the frequency (refer to the datasheet) has to drive the OSC_IN"
2024-04-12T01:54:59 < qyx> so it accepts sinus too in bypass
2024-04-12T01:55:04 < zyp> did you see what I mentioned the other day about this thing also having a sane RTC mode?
2024-04-12T01:55:26 < zyp> as in, you can set it to be just a plain counter instead of calendar, or even do both
2024-04-12T01:56:06 < qyx> yeah I noticed
2024-04-12T01:56:12 < qyx> I hope U5 has that too
2024-04-12T01:56:16 < qyx> but lazy to check
2024-04-12T01:57:20 < qyx> U5 has that analol clock mode too
2024-04-12T01:57:54 < zyp> I wonder if I can bring up ethernet on this thing without bothering switching sysclk to something other than default HSI
2024-04-12T01:58:25 < qyx> ar you doing the classical ethernet?
2024-04-12T01:58:35 < zyp> HSI is 64MHz, so I figure data rate wise it should be fine, and the RMII REF_CLK should be provided by the PHY anyway
2024-04-12T01:58:52 < zyp> classical
2024-04-12T01:58:53 < zyp> ?
2024-04-12T01:59:42 < qyx> not 802.3cg
2024-04-12T02:00:01 < zyp> currently I'm protoing on h7s7-dk
2024-04-12T02:00:13 < qyx> I have t1s board already here and haven't had any time to test it yet
2024-04-12T02:00:19 < qyx> also I have only one so not much useful
2024-04-12T02:00:32 < zyp> I'm planning to attempt writing a little ipv6 stack
2024-04-12T02:00:46 < qyx> yeah I am all in
2024-04-12T02:00:58 < qyx> fuk lwip
2024-04-12T02:01:08 < zyp> smolt was kinda a detour from this, because I want to be able to efficiently log what it's doing
2024-04-12T02:02:11 < zyp> so far I have one planned use for it, I grabbed some relay card from aliexpress a while ago that came with an at32
2024-04-12T02:02:23 < zyp> I wanna set that up to be controlled over MQTT or something
2024-04-12T02:03:01 < qyx> I just want ipv6/tcp/mqtt
2024-04-12T02:03:09 < qyx> and optionally ipv6/udp for other protocols
2024-04-12T02:03:23 < qyx> tbh I am questioning the use of tcp in anything
2024-04-12T02:03:40 < qyx> I would happily replace mqtt with anything saner over udp
2024-04-12T02:05:19 < zyp> the idea of having mqtt is so I can just have it announce itself to e.g. home-assistant
2024-04-12T02:06:47 < zyp> i.e. https://www.home-assistant.io/integrations/mqtt/#mqtt-discovery
2024-04-12T02:07:31 < zyp> I think I hacked up some python stuff for that a few years ago, IIRC it worked well, would be convenient to have MCU code that can do the same
2024-04-12T02:08:51 < zyp> also, once I have a network stack, I'm probably gonna do some NCM experiments
2024-04-12T02:09:46 < qyx> ppp!
2024-04-12T02:09:53 < qyx> (for nbiot)
2024-04-12T02:11:00 < zyp> I have the impression NCM is reasonably well supported nowadays, so I figure a USB-device with a webui and mdns and shit over link-local ipv6 over NCM could be fairly feasible
2024-04-12T02:13:09 < qyx> that's fine, even some jabbascripts for the backend
2024-04-12T02:13:20 < qyx> or python
2024-04-12T02:13:44 < zyp> backend?
2024-04-12T02:14:00 < qyx> hm ok you don't need much on a mcu
2024-04-12T02:14:10 < zyp> yeah, idk
2024-04-12T02:15:05 < zyp> for higher level web stuff I kinda like graphql, but I don't really think I want to do that on a mcu
2024-04-12T02:15:22 < zyp> although it'd be kinda cool
2024-04-12T02:17:56 < zyp> could also use my protobuf RPC I guess, it's pretty much gRPC when you put it on HTTP
2024-04-12T02:38:04 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-12T02:38:35 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 252 seconds]
2024-04-12T03:04:46 -!- nuxil_ is now known as nuxil
2024-04-12T03:37:31 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-12T04:00:05 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-12T04:01:36 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-12T04:27:18 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 255 seconds]
2024-04-12T04:57:58 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Quit: Leaving]
2024-04-12T04:58:29 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-12T04:58:59 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Remote host closed the connection]
2024-04-12T04:59:38 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-12T06:09:47 -!- qyx [~qyx@84.245.120.207] has quit [Ping timeout: 252 seconds]
2024-04-12T06:11:28 -!- qyx [~qyx@84.245.120.232] has joined ##stm32
2024-04-12T07:11:43 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-12T07:15:33 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 272 seconds]
2024-04-12T07:16:06 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-12T07:17:30 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-12T07:19:40 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-12T07:20:15 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-12T08:08:07 -!- fenugrec [~f@192.214.232.39] has quit [Ping timeout: 256 seconds]
2024-04-12T08:21:08 -!- fenugrec [~f@192.214.232.39] has joined ##stm32
2024-04-12T08:24:00 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Quit: ZNC 1.7.5 - https://znc.in]
2024-04-12T08:27:31 < jpa-> qyx: that resistor between PF0 and PF1 looks weird to me, but maybe past jpa has been weird or something
2024-04-12T08:28:13 < jpa-> it's point is probably to keep voltage over C46 reasonable, but i think STM32 has biasing resistor internally
2024-04-12T08:28:16 < jpa-> shouldn't hurt though
2024-04-12T08:32:06 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-12T08:34:01 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-12T08:34:02 < qyx> idk
2024-04-12T09:17:31 -!- jtj [~jtj@212.66.207.170] has quit [Quit: Konversation terminated!]
2024-04-12T09:30:44 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-12T09:31:10 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-12T10:05:20 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Ping timeout: 260 seconds]
2024-04-12T10:56:47 < qyx> eteze si kvoli nemu celu noc nespal
2024-04-12T10:59:24 < qyx> sry
2024-04-12T11:00:36 < jpa-> i slept very well, jbo didn't disturb me at all
2024-04-12T11:32:27 -!- martinmoene [~martinmoe@132.229.46.129] has quit [Read error: Connection reset by peer]
2024-04-12T11:41:21 -!- martinmoene [~martinmoe@132.229.46.129] has joined ##stm32
2024-04-12T11:42:49 < Steffanx> He didn't disturb you jpa- ? He stopped snoring?
2024-04-12T11:43:26 < jpa-> i have earplugs
2024-04-12T11:43:52 < Steffanx> I tried that too, but I can't sleep with them.
2024-04-12T12:55:39 < ventYl> does he snore that loud it reaches you too?
2024-04-12T13:04:46 < Steffanx> I sleep on his couch once ;)
2024-04-12T13:04:49 < Steffanx> *slept
2024-04-12T13:12:18 < stgl> Hi, I'd like to record sound from a PDM microphone using the DFSDM peripheral. I used ST's excel file to calculate the values for my target sampling rate of 16 kHz
2024-04-12T13:12:54 < stgl> My problem is that the data jumps around periodically, looks like it's missing samples or something: https://file-share.ch/files/LOHy5dWR2MgfSKvnMCcd3lv5_screenshot-Fr%2012%20Apr%202024%2012:05:31%20CEST.png
2024-04-12T13:13:25 < stgl> I use a manually triggered DMA to record 500 ms and then dump the data out over serial 
2024-04-12T13:13:56 < stgl> Any ideas what the issue could be? I don't know whether it's a problem with the DFSDM filtering or the DMA transfer
2024-04-12T13:13:56 < karlp> (prime use case for SWO....)
2024-04-12T13:16:57 < BrainDamage> the dc offset changing is extremely suspicious too, if the offset is from the 50Hz noise baseline from the angle change you should be able to calculate the amount of samples dropped which might help you debug this
2024-04-12T13:16:58 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-12T13:34:09 < stgl> karlp: true, I never used that before... but it should work with j-link and Ozone
2024-04-12T13:34:28 < stgl> BrainDamage: good idea, thanks
2024-04-12T14:20:48 < jpa-> stgl: you could try with a slower sinewave or triangle wave, then it would be clearer if samples are dropped
2024-04-12T15:03:03 < karlp> fecks sake. dual mode bt/btle in esp32 needs to be turned in in like 4 different places in menuconfig.
2024-04-12T15:16:18 < jpa-> that sounds like every kconfig/menuconfig project ever
2024-04-12T15:30:13 < Steffanx> So how's the day in ##stm32-land?
2024-04-12T15:47:40 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-12T15:47:40 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-12T15:47:40 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-12T16:13:37 -!- benishor [~benishor@188.166.85.117] has quit [Quit: tah tah!]
2024-04-12T16:24:16 < zyp> good
2024-04-12T16:24:50 < zyp> went to a client to help them reproduce a sporadic issue and demonstrate that my fix worked
2024-04-12T16:25:36 < zyp> they had a setup of 27 devices, some with the fix
2024-04-12T16:27:03 < zyp> took a while to get it properly reproduced, but then we managed to get all the devices with the older version to fail at once, and none of the newer
2024-04-12T16:27:27 < zyp> that was pretty satisfying
2024-04-12T16:29:25 < sauce> nice
2024-04-12T16:30:36 -!- benishor [~benishor@scene.ro] has joined ##stm32
2024-04-12T16:37:02 < PaulFertser> Was that easy in retrospect? :)
2024-04-12T16:44:01 < zyp> it's one of those «it's easy when you know how» cases
2024-04-12T16:45:31 < zyp> race condition in some networking code, corner case where it'd run out of free buffers, turn off the rx interrupt and end up in a state where it doesn't turn it on again
2024-04-12T16:46:30 < sauce> reproducible just with traffic over a certain threshold?
2024-04-12T16:46:43 < PaulFertser> Oh still quite tricky then.
2024-04-12T16:47:05 < PaulFertser> I mean the mechanics behind the bug was tricky.
2024-04-12T16:48:07 < zyp> when I originally worked the issue a few weeks ago, I managed to hit it consistently with a nmap ping scan of the local /24
2024-04-12T16:48:40 < zyp> which I discovered by accident before I'd even started, because I ran the ping scan because I was unsure what ip the device I was working on was configured to
2024-04-12T16:49:19 < PaulFertser> Nice :)
2024-04-12T16:49:23 < zyp> but it's pretty timing sensitive, when I tried it with my laptop, I had less success
2024-04-12T16:53:50 < qyx> I like races in network code
2024-04-12T16:57:31 < zyp> my understanding of the corner case is that it requires just the right combination of traffic
2024-04-12T16:59:40 < zyp> first it needs a ton of small broadcast packets, they must be small enough that you run out of firmware buffers before the FIFO in the MAC fills up, and they also need to be ignored because responding to them keeps the network stack going
2024-04-12T17:00:33 < zyp> and then it also needs a few requests for it to actually do something, because otherwise it manages to throw away the incoming packets fast enough that it doesn't run out of buffers
2024-04-12T17:01:27 < zyp> a nmap ping scan achieves both because it produces a couple hundred broadcast ARP requests, along with a few requests directed to this particular device
2024-04-12T17:02:21 < sauce> was it happening in a production environment?
2024-04-12T17:03:55 < zyp> yeah, sporadically, likely related to network topology changes due to failover/failback
2024-04-12T17:04:07 < PaulFertser> Was there a chance to attach with SWD to a device which stopped responding that way to inspect internal state?
2024-04-12T17:04:27 < zyp> the suspicion is packet storms due to temporary loops during failback
2024-04-12T17:04:56 < zyp> and indeed, the way we managed to properly reproduce it today on a full setup was by deliberately introducing a network loop
2024-04-12T17:05:22 < PaulFertser> A rather educational story, thanks for sharing!
2024-04-12T17:05:28 < sauce> yeah
2024-04-12T17:05:49 < sauce> for the unknown address issue I am a big fan of simple discovery protocols - a unicast response on the device to a broadcast request or just periodic broadcast beacons from the device
2024-04-12T17:06:06 < zyp> I couldn't get neither nmap nor ping to reliably reproduce when there were switches in the mix, but creating the network loop did the trick
2024-04-12T17:07:32 < zyp> and yes, when I had the device on my desk a few weeks ago that I could reliably reproduce on with nmap, I worked it out with the debugger and live logging
2024-04-12T17:11:53 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-12T17:13:16 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-12T17:14:45 < karlp> mdns for me.
2024-04-12T17:32:24 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-12T19:02:40 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 256 seconds]
2024-04-12T19:09:36 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-12T19:20:05 -!- Livio [~livio@user/livio] has quit [Ping timeout: 240 seconds]
2024-04-12T19:26:25 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has quit [Ping timeout: 256 seconds]
2024-04-12T19:47:07 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 246 seconds]
2024-04-12T20:07:09 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-12T20:07:09 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-12T20:07:09 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-12T20:41:25 -!- _nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-12T20:41:25 -!- nuxil_ [~nuxil@141.195.51.41] has quit [Read error: Connection reset by peer]
2024-04-12T21:13:48 < antto> bluh, these fancy RGB LEDs have such a weird choice of control signal
2024-04-12T21:14:47 < antto> couldn't they design them so they easily work with some very common MCU peripheral
2024-04-12T21:20:33 < Steffanx> There's so many ways to drive em
2024-04-12T21:20:57 < Steffanx> Or use HD107/8/whatever similar which uses a clock and data
2024-04-12T21:26:53 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-12T21:32:25 < antto> they're less bright
2024-04-12T21:32:36 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-12T21:32:40 < antto> and if that's 5x5mm it won't fit
2024-04-12T21:34:11 -!- Livio [~livio@user/livio] has quit [Ping timeout: 264 seconds]
2024-04-12T21:34:42 < zyp> they have both 5x5 and 2x2mm variants
2024-04-12T21:36:33 < antto> 2x2mm with 6 pins would make my iron raise an eyebrow o_O
2024-04-12T21:37:00 < antto> but high intensity is a priority
2024-04-12T21:37:06 < antto> i mean luminance
2024-04-12T21:37:53 < antto> anyway, i was thinking of wiring an SPI MOSI to these things, but i might add also an alternative route to a PWM periph
2024-04-12T21:38:04 < zyp> 2x2mm qfns with like 12 pins are common enough
2024-04-12T21:38:34 < zyp> you could use a UART, IIRC that's what I did last time
2024-04-12T21:38:47 < antto> zyp, i have to put 104 of these on one board, and i wanted to make 2 or 3 boards for $home
2024-04-12T21:39:04 < antto> hm, UART?
2024-04-12T21:39:18 < antto> lemme re-think it
2024-04-12T21:40:50 < zyp> https://paste.jvnv.net/view/92VJP
2024-04-12T21:42:02 < zyp> each nibble is a pulse, 0x8 is 0b1000, i.e. short pulse and 0xc is 0b1100, i.e. long pulse
2024-04-12T21:45:32 < zyp> also the UART config: https://paste.jvnv.net/view/binF8
2024-04-12T21:45:33 < antto> i might fiddle with this in desmos
2024-04-12T21:46:49 < antto> well, i guess i might also add a usart TXD as a third route
2024-04-12T21:47:00 < antto> having 3 possible periphs to drive it will be better
2024-04-12T21:47:03 < zyp> TXINV and DATAINV makes it so that start bit will be high, M=7b cuts off the upper bit (effectively replacing it with the start bit since it's always 1 anyway)
2024-04-12T21:48:08 < antto> not sure if i can do some of these things on the SERCOM
2024-04-12T21:48:32 < antto> thus i was thinking to toss it into desmos and fiddle with it
2024-04-12T21:49:10 < zyp> and if you wanted to be really efficient and have a highend stm32 with DMA2D, you could use its CLUT feature to do the entire 8 bit to four byte expansion, so you could take raw RGB values, feed them to DMA2D to make a byte buffer and then just feed that to regular DMA to feed it to the UART
2024-04-12T21:49:46 < Steffanx> Atmel doesn't know about all those fancy perihperals.
2024-04-12T21:49:49 < Steffanx> *didnt
2024-04-12T21:50:00 < zyp> aww
2024-04-12T21:50:16 < zyp> speaking of, what's the lowest end stm32 with DMA2D?
2024-04-12T21:50:18 < antto> zyp, no clue what that even means
2024-04-12T21:50:31 < Steffanx> When will you move to the great STM32 mr antto ?
2024-04-12T21:50:38 < antto> not for this clock
2024-04-12T21:50:56 < Steffanx> The time is ticking.
2024-04-12T21:50:57 < antto> Steffanx, i already have began a board with H723
2024-04-12T21:51:07 < antto> but it's not finished
2024-04-12T21:51:31 < antto> i even used the keewwwb
2024-04-12T21:51:54 < antto> before that i went to try NXP IMXRT
2024-04-12T21:52:25 < antto> ...on linux
2024-04-12T21:52:34 < antto> the ST kewb is better
2024-04-12T21:52:48 < antto> NXP in general just seems so unfriendly
2024-04-12T21:53:19 < antto> "oh, so you want to look at a datasheet? WHO ARE YOU AND WHO SENT YOU HERE?! REGISTER IMMEDIATELY!"
2024-04-12T21:53:53 < zyp> yeah, I hate that
2024-04-12T21:54:21 < antto> and then their thing looked like rocket science, i couldn't figure out sh*t, so i gave up
2024-04-12T21:54:30 < zyp> anyway, speaking of adressable leds, what's the brightest one I can get?
2024-04-12T21:54:44 < antto> i was looking for "bright"
2024-04-12T21:55:38 < antto> and small so that they'll fit in 4mm including any solder joints... these are the ones i'm gonna get: XL-3528RGBW-WS2812B
2024-04-12T21:56:06 < zyp> I was discussing an idea with somebody a couple of years ago and today they said somebody might want to actually do it
2024-04-12T21:56:49 < antto> these seem brighter: https://www.optosupply.com/uppic/2023421763886.pdf
2024-04-12T21:57:26 < zyp> project would be something like 400 LEDs, spaced a meter apart or something
2024-04-12T21:57:40 < zyp> individually controllable
2024-04-12T21:57:56 < antto> my $distributor told me they can sell me 2000 of these for $140-ish so they are about twice the price of the XL ones
2024-04-12T21:58:23 < antto> one meter?!
2024-04-12T21:58:32 < antto> what's the goal here?
2024-04-12T21:59:01 < zyp> so I figure I'd make a little PCB with a LED, a stepdown from 48V or something and a RS485 transceiver for the signal
2024-04-12T21:59:04 < antto> did you mean small LEDs or whole RGB light bulbs E27 style
2024-04-12T21:59:20 < Steffanx> Why you use this weird $distributor instead of buying them from aliexpress/lcsc/wahtever chinashop directly antto ?
2024-04-12T21:59:27 < Steffanx> Assuming $distributor isnt from china
2024-04-12T21:59:38 < antto> $distributor is $local
2024-04-12T21:59:38 < zyp> I mean a LED bright enough to be reasonably visible outdoors in daylight
2024-04-12T22:00:01 < antto> they already work with OptoSupply, they just don't happen to sell this RGB LED
2024-04-12T22:00:11 < antto> thus i asked them
2024-04-12T22:00:52 < Steffanx> Yes why use this $distributor.bg antto ?
2024-04-12T22:01:07 < antto> Steffanx, are you saying you know where to buy OSS85ST63 from?
2024-04-12T22:01:08 < zyp> power budget is a potential issue, both since I want 400 of them, and want to run that power over a decent distance
2024-04-12T22:01:41 < antto> ugh, zyp i guess you need big bulbs
2024-04-12T22:01:44 < Steffanx> oh, nevermind antto  :p i thought you were still talking about the ws2812b you mentioned
2024-04-12T22:02:05 < zyp> antto, I don't want bulbs, I want SMT LEDs
2024-04-12T22:02:12 < antto> Steffanx, nah, and these aren't real WSxxxx they are some XINGLIGHT brand
2024-04-12T22:03:13 < zyp> actually
2024-04-12T22:03:35 < antto> * THT enters the chat?
2024-04-12T22:03:39 < zyp> power budget might not be all that much of an issue, when I think about it there probably won't be many LEDs lit at the same time
2024-04-12T22:03:57 -!- qyx [~qyx@84.245.120.232] has quit [Ping timeout: 256 seconds]
2024-04-12T22:05:10 < antto> so, you wanna make like a grid of 400 small RGB LEDs 1 meter apart, and their light has to be visible in daylight?
2024-04-12T22:05:26 < zyp> string, not lights, and yeah
2024-04-12T22:05:32 < zyp> not grid, I mean
2024-04-12T22:05:44 -!- qyx [~qyx@84.245.121.80] has joined ##stm32
2024-04-12T22:05:46 < antto> color me skeptical
2024-04-12T22:06:04 < antto> do not underestimate the power of daylight
2024-04-12T22:06:30 < zyp> people effectively want to do a cheapo https://www.wavelight-technologies.com/
2024-04-12T22:07:08 < zyp> and I told them I can do the circuitry if somebody else works out the mechanics
2024-04-12T22:08:26 < antto> oh, so this is sorta like the ghost replay in some racing games
2024-04-12T22:09:05 < qyx> what the hell is it
2024-04-12T22:09:10 < zyp> AIUI it's for pacing so you can hold a steady speed
2024-04-12T22:09:14 < qyx> I mean, I see, but what's the prupose
2024-04-12T22:09:23 < antto> where when you start the 2nd lap, the game generates a "ghost" of your car from the previous lap so you can compare how you're doing now
2024-04-12T22:09:39 < zyp> I don't run so I don't really care about the why, I'm more interested in the how :)
2024-04-12T22:10:13 < antto> well, bring a single LED outside at noon and observe
2024-04-12T22:10:41 < qyx> 400 transceivers may impose some problems
2024-04-12T22:10:58 < zyp> how so?
2024-04-12T22:11:09 < antto> so many slaves
2024-04-12T22:11:13 < qyx> rs485 uis usually rated for 32
2024-04-12T22:11:19 < qyx> the cool ones 128/256
2024-04-12T22:11:19 < antto> this will be the beginning of the new slavery
2024-04-12T22:12:05 < zyp> qyx, they wouldn't all be a single shared signal though
2024-04-12T22:13:09 < zyp> I figure normal ws2812-style pulse signal, just putting the LEDs between the receive and transmit halves of a duplex transceiver
2024-04-12T22:13:12 < qyx> hm, I would consider radio.. maybe
2024-04-12T22:13:18 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Ping timeout: 268 seconds]
2024-04-12T22:14:23 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-12T22:14:33 < zyp> a microcontroller on each board would cost more
2024-04-12T22:14:49 < qyx> is the cost an issue?
2024-04-12T22:14:52 < qyx> $1 more or less
2024-04-12T22:15:19 < zyp> that adds up to $400 on 400 boards
2024-04-12T22:16:09 < zyp> I figure a stepdown, transceiver and one or more LEDs would be practically free
2024-04-12T22:16:32 < qyx> 48V stepdown is not that cheap
2024-04-12T22:16:53 < qyx> also when using 48 V nominal you basically need at least 60 V stepdown
2024-04-12T22:17:01 -!- BrainDamage [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-12T22:17:35 < qyx> and considering 400 m of cable running across all other sh.t, I would not go lower than 90 V or 100 V, what is practical, + a big transient suppression
2024-04-12T22:18:45 < qyx> also, maybe considering a dedicated 3 channel LED driver with LEDs + your favourite I2C programmable thing
2024-04-12T22:18:49 < qyx> how is it called
2024-04-12T22:18:51 < qyx> greenpak
2024-04-12T22:19:10 < zyp> the HVPAK ones are already suitable for LED driving
2024-04-12T22:19:13 < zyp> but nah
2024-04-12T22:19:30 < qyx> I am reluctant to believe 1. 400 m of addressable LEDs would be problem-free
2024-04-12T22:19:38 < qyx> 2. you are gonna find LEDs bright enough
2024-04-12T22:23:47 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-12T22:25:48 < antto> i wonder if i it would be possible to make a silly analog(ue) circuit that takes two logic signals and just produces one combined logic signal with the "right" timings for those dumb LEDs ;P~
2024-04-12T22:26:38 < Steffanx> imagine already using an MCU that can do magic.
2024-04-12T22:29:14 < zyp> antto, I've made a greenpak that can drive them
2024-04-12T22:29:33 < antto> a wat
2024-04-12T22:29:52 < qyx> a magic
2024-04-12T22:30:30 < qyx> Steffanx: great use case for C0
2024-04-12T22:30:43 < zyp> qyx, this shit looks usable: https://jlcpcb.com/partdetail/Gosemicon-GBI1633SMAR/C5451749
2024-04-12T22:31:40 < zyp> bit annoying that it's non-synchronous, but AIUI most higher voltage bucks are anyway, especially if you want cheap
2024-04-12T22:32:40 < qyx> nah 70%-ish eff
2024-04-12T22:33:29 < qyx> stm32c011 is great and cheap
2024-04-12T22:33:54 < qyx> but you could also use some $0.1 8051
2024-04-12T22:34:16 < antto> so i fiddled in desmos, i guess an inverted UART should work
2024-04-12T22:35:11 < qyx> zyp: I would maybe try 2-wire approach, sending data using polarity reversal
2024-04-12T22:35:20 < qyx> same as DCC does for model railroads
2024-04-12T22:35:21 < Steffanx> What is desmod Mr antto ?
2024-04-12T22:35:32 < qyx> a slovakian pop rock band
2024-04-12T22:35:34 < antto> https://www.desmos.com/calculator
2024-04-12T22:35:59 < antto> for people who like to fiddle with curves
2024-04-12T22:36:16  * antto <----
2024-04-12T22:36:52 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-12T22:36:55 < zyp> qyx, that's a cute idea, but doesn't solve addressing
2024-04-12T22:39:04 < zyp> hmm, C011 is $.45 at jlcpcb in volume
2024-04-12T22:42:48 < zyp> polarity reversal should be fairly robust comms wise and cheap to implement, so I kinda like the idea, but the lack of inherent addressing is an issue
2024-04-12T22:43:24 < antto> i think 5bit USART with inverted output is close
2024-04-12T22:43:56 < antto> that is, 1 "char" per data bit for those LEDs
2024-04-12T22:44:40 < qyx> zyp: you could do address discovery by measuring the consumption on the master
2024-04-12T22:45:07 < zyp> I thought about it, i.e. load modulation
2024-04-12T22:45:13 < qyx> use a broadcast command to light up first half
2024-04-12T22:45:15 < zyp> but still, how do you work out the order?
2024-04-12T22:45:18 < qyx> check if the load increased
2024-04-12T22:45:25 < qyx> hm
2024-04-12T22:46:13 < qyx> a really stupid idea but a tdr?
2024-04-12T22:46:26 < zyp> haha
2024-04-12T22:47:10 < zyp> if they were pluggable, I could require them to be plugged in in order while powered
2024-04-12T22:47:27 < qyx> just a broadcast command/mask to all slaves, the selected ones assert a cap between VIN-/VIN+
2024-04-12T22:47:32 < qyx> and you measure the position
2024-04-12T22:47:46 < zyp> but I don't think I wanna work out cheap enough weatherproof connectors
2024-04-12T22:47:47 < qyx> you could spend $20 more on the master to implement that
2024-04-12T22:48:17 < zyp> I figure easiest way to weatherproof them is to just pot them with cables sticking out
2024-04-12T22:48:51 < qyx> maybe a t1l transceiver could be persuaded to do a tdr on a cable
2024-04-12T22:48:56 < zyp> e.g. put a pair of XHs or something on the board, plug cheap cables between them, then pot it with connector and all
2024-04-12T22:50:16 < qyx> I like this sort of autisming
2024-04-12T22:50:25 < zyp> :)
2024-04-12T22:51:05 < zyp> let's see what the cheapest duplex rs485 transceiver is
2024-04-12T22:51:14 < antto> https://www.desmos.com/calculator/yo9xzrbsyg
2024-04-12T22:51:24 < zyp> simplex are $.1 in volume
2024-04-12T22:52:10 < antto> U3 (the black line) is an inverted 5bit usart
2024-04-12T22:54:36 < zyp> looks like two simplex transceivers is cheaper than a single duplex transceiver
2024-04-12T22:59:29 < Laurenceb_> twisted pervert japs did it again
2024-04-12T22:59:43 < Laurenceb_> > D subs with M2.5 jack screws
2024-04-12T23:00:09 < qyx> whoa U545 is my new G491
2024-04-12T23:05:27 < Laurenceb_> MDF and 14bit ADC looks hawt
2024-04-12T23:05:40  * Laurenceb_ is making Shinkansen cables
2024-04-12T23:06:15 < Laurenceb_> >37 way D subs with M2.5 jack screws and about 12 pins used
2024-04-12T23:06:17 < Laurenceb_> wtf
2024-04-12T23:14:02 < Laurenceb_> https://nitter.poast.org/tomough/status/1778744545619911066#m orbital sides
2024-04-12T23:21:30 < zyp> hmm, interesting, some of the ws2811 variants can actually run strings of 6 leds from a 24V supply
2024-04-12T23:26:58 < Laurenceb_> wtf is this lmao https://nitter.poast.org/pic/orig/media%2FGK592kQXAAEqwXE.jpg
2024-04-12T23:27:08 < Laurenceb_> I dont get this command promp shit
2024-04-12T23:31:13 < Laurenceb_> it doesnt seem to fit format of transformer architecture - like there is some config mode that operates at another level.. supposedly gab edgelords have cloned it
2024-04-12T23:36:28 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-914-45d0-f490-aec2.fixed6.kpn.net] has joined ##stm32
2024-04-12T23:38:17 < antto> meh, for my UART scheme imma need 3.5MHz but the SERCOM can only do 3MHz
2024-04-12T23:45:51 < antto> but in sync mode i can get 3.428571...MHz
2024-04-12T23:53:15 < Laurenceb_> >I deadlift 130kg every day... when I get out of my chair
2024-04-12T23:56:22 < antto> put down the chicken wing... SLOWLY
2024-04-12T23:57:52 < antto> lifting your own body shouldn't count as an achievement normally
--- Day changed la huhti 13 2024
2024-04-13T00:04:38 < Laurenceb_> https://nitter.poast.org/Submetallic/status/1778074516544966726#m
2024-04-13T00:04:40 < Laurenceb_> lolllll
2024-04-13T00:32:32 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-13T00:35:57 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-914-45d0-f490-aec2.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-13T00:36:23 -!- scrts [~scrts2@23.28.144.38] has quit [Ping timeout: 252 seconds]
2024-04-13T01:00:15 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T01:01:43 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T01:10:56 < Steffanx> nomorekaki: the song you requested https://www.youtube.com/watch?v=K5bucpPiPKU
2024-04-13T01:16:50 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-13T01:32:44 < Ecco> I'm looking at this ADP5090 energy harvesting chip
2024-04-13T01:32:57 < Ecco> They use an inductor to boost the solar panel's voltage
2024-04-13T01:32:58 < Ecco> https://i.imgur.com/w6MPPFy.png
2024-04-13T01:33:15 < Ecco> I'm trying to figure out which inductor to use, and more specifically what current rating it should have
2024-04-13T01:33:55 < Ecco> And… I'm kind of clueless. I know what is the max current the panel can provide, but how will it split between the VIN and SW pins?
2024-04-13T01:38:06 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-13T01:39:49 < Ecco> Maybe we can assume that i(Vin)=0? I don't know, it's weird
2024-04-13T01:40:00 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T01:41:39 < qyx> it is not weird, apply the common boostconverter logic
2024-04-13T01:41:57 < qyx> there are many appnotes how to compute i ductors
2024-04-13T01:42:32 < qyx> and I would also bet they have an appnote for that exactchip
2024-04-13T01:46:21 < qyx> all info is in the datasheet
2024-04-13T01:47:27 < qyx> the low side boost mosfet hard current limit is 100 mA which effectively limits your peak inductor current, which must always be lower than the saturation current
2024-04-13T01:47:54 < qyx> 100 mA is so low that you simply don't filter based on this oarameter
2024-04-13T01:48:09 < qyx> inductance is given as 22 uH
2024-04-13T01:48:43 < qyx> the next two important things are size and DCR
2024-04-13T01:49:06 < qyx> the lower the DCR, the higher the efficiency and physical size
2024-04-13T01:49:40 < qyx> when selecting, also keep an eye on the switching freq to not hit inductor limits for someweirdinductors
2024-04-13T01:51:00 < qyx> tl;dr personally I would look at 2010 chip inductors first
2024-04-13T01:52:48 < qyx> the recommended lps4018-223m looks good
2024-04-13T01:55:48 < qyx> 1210sorry
2024-04-13T01:56:25 < Ecco> excellent
2024-04-13T01:56:30 < Ecco> thank you so much for your help
2024-04-13T01:56:56 < Ecco> they do have a paragraph about inductor selection in the datasheet, but it essentially says "pay attention to max current"
2024-04-13T01:57:36 < Ecco> Size and DCR, yeah, this I understand the implications of.
2024-04-13T01:57:38 < qyx> normally you should analyse your min/max duty cycle in corner cases
2024-04-13T01:57:55 < qyx> then your peaj current depe ding on thag
2024-04-13T01:58:10 < Ecco> wat? :-D
2024-04-13T01:58:21 < catphish> i've got into the habit of using STM32G4, it's modern and nice to work with, and cheap, but it seems that it doesn't support ethernet, is there an obvious choice of newer stm32 series that supports ethernet?
2024-04-13T02:00:20 < Ecco> (also, for what it's worth: Analog Devices have a really cool samples program)
2024-04-13T02:00:43 < qyx> idk what I wanted to say even
2024-04-13T02:00:50 < Ecco> (I just asked for 10 samples, and I got them in like … a day? No question asked, they used FedEx-ultra-express-lightspeed-whatever)
2024-04-13T02:00:57 < Ecco> qyx: :-D
2024-04-13T02:01:00 < qyx> catphish: me like g4 too, check h5
2024-04-13T02:01:09 < qyx> new and cheap and should have ethernets
2024-04-13T02:02:15 < qyx> Ecco: looks like you should be able to get a decent and vheap inductor in a 6x6 mm package with dcr under 100 mohm
2024-04-13T02:02:40 < Ecco> yeah, but 6x6mm might be too large for my app
2024-04-13T02:02:45 < Ecco> I'd like to use a chip resistor
2024-04-13T02:02:49 < Ecco> crap, inductor
2024-04-13T02:03:00 < Ecco> I guess I'll need to get up to speed on how boost converters work
2024-04-13T02:03:25 < Ecco> but I'm mainly confused about the fact that there seem to be two input pins, VIN and SW
2024-04-13T02:03:35 < zyp> what
2024-04-13T02:03:43 < qyx> 22u is quite much for chip inductors
2024-04-13T02:03:51 < zyp> SW would be the switch node
2024-04-13T02:04:02 < catphish> qyx: is H5 available now, last i looked it was an awesome dream still in the pipeline
2024-04-13T02:04:13 < Ecco> qyx: I've found a few on lcsc, but they have a rather low current rating
2024-04-13T02:04:14 < qyx> should be
2024-04-13T02:04:15 < zyp> I think it's been available for months
2024-04-13T02:04:23 < catphish> cool, i'll have a look
2024-04-13T02:04:40 < Ecco> They do provide a block diagram tho: https://i.imgur.com/UWSr1ta.png
2024-04-13T02:05:12 < Ecco> But I'm not familiar enough with charge pumps and boost converters to have a good understanding of the whole thing just yet
2024-04-13T02:06:29 < catphish> JLC don't have a single G5 in stock :(
2024-04-13T02:06:36 < qyx> h5
2024-04-13T02:06:40 < catphish> *h5
2024-04-13T02:06:54 < qyx> so check relevant suppliers now
2024-04-13T02:06:59 < catphish> i meant h5, they have 9 listed, none in stock
2024-04-13T02:07:02 < zyp> Ecco, you see the two transistors on SW hooked up to the boost controller block?
2024-04-13T02:07:21 < Ecco> yes
2024-04-13T02:07:32 < catphish> ah yes, normal suppliers have them, cool
2024-04-13T02:08:03 < zyp> Ecco, when the low side transistor turns on, you get VIN -> 0V across the inductor, this charges the inductor
2024-04-13T02:08:48 < zyp> once the inductor is charged, the low side transistor turns off and the high side transistor turns on
2024-04-13T02:09:00 < Ecco> (btw, what's that weird symbol on top of this low-side transistor?)
2024-04-13T02:09:20 < zyp> idk, current sense probably
2024-04-13T02:09:28 < Ecco> oh, makes sense
2024-04-13T02:09:37 < Ecco> ok, and then the inductor wants to keep current flowing
2024-04-13T02:09:52 < zyp> yeah, so when the high side is on, you get SW = VIN + VL
2024-04-13T02:09:52 < Ecco> so it will push current into what's after "BSTO"?
2024-04-13T02:10:14 < zyp> correct
2024-04-13T02:10:44 < Ecco> What's VL in the above equation?
2024-04-13T02:11:00 < Ecco> the voltage across the inductor?
2024-04-13T02:11:10 < zyp> yeah
2024-04-13T02:11:20 < Ecco> In that case, that equation is simply *always* true, right?
2024-04-13T02:11:35 < Ecco> It's just the value of VL that is likely to change over time, right?
2024-04-13T02:12:12 < zyp> correct
2024-04-13T02:12:44 < zyp> when low side is on, VL = -VIN
2024-04-13T02:13:22 < zyp> the point is, you need voltage to push a current
2024-04-13T02:14:01 < qyx> Ecco: did you check bq25504?
2024-04-13T02:14:04 < qyx> it looks similar
2024-04-13T02:14:14 < qyx> it even lists the same u ductor asrecommended
2024-04-13T02:14:23 < qyx> inductor
2024-04-13T02:14:27 < zyp> when low switch is on, VIN is pushing a current into the inductor
2024-04-13T02:15:03 < zyp> when high switch turns on, the inductor wants to keep the same current flowing, and creates the required voltage to achieve that
2024-04-13T02:15:29 < zyp> that's why stuff like solenoids and relay coils give you a big current spike when they turn off
2024-04-13T02:15:39 < zyp> big voltage spike*
2024-04-13T02:16:37 < zyp> they want to keep the current flowing, and if you simply open the circuit so the current can't go anywhere, the coil will create a rather high voltage trying
2024-04-13T02:16:57 < zyp> to the point where it can break down insulation to make current flow
2024-04-13T02:19:25 < zyp> boost and buck regulators are fairly similar in how they work, the main difference is that a boost regulator charges the inductor directly from the supply and discharges it in series with the supply, increasing the voltage, while a buck regulator charges in series with the load (decreasing the voltage) and discharges directly into the load
2024-04-13T02:19:30 < Ecco> qyx: I did, both parts look very similar
2024-04-13T02:19:43 < Ecco> zyp: ok, makes sense
2024-04-13T02:19:47 < Ecco> thank you so much for the help!
2024-04-13T02:20:15 < zyp> which means that a boost regulator is pretty much just a reversed buck regulator, just swap the roles of load and supply
2024-04-13T02:21:10 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T02:21:59 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T02:44:51 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-13T03:04:54 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-13T03:05:58 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T03:06:13 -!- Livio [~livio@user/livio] has quit [Ping timeout: 272 seconds]
2024-04-13T03:09:00 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T03:10:00 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T04:08:39 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-13T05:01:47 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 264 seconds]
2024-04-13T05:02:28 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-13T05:31:11 -!- ferdna [~ferdna@user/ferdna] has quit [Ping timeout: 252 seconds]
2024-04-13T05:37:56 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-13T05:39:31 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-13T05:53:43 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 256 seconds]
2024-04-13T06:44:01 -!- Somebody3456 [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T06:44:42 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-13T06:54:18 -!- Somebody3456 is now known as SystemError
2024-04-13T07:13:09 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-13T07:27:21 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-13T07:57:52 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-13T08:02:10 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-13T08:03:46 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T08:10:19 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-13T08:35:08 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T09:00:09 -!- mrec_ [~mrec@user/mrec] has quit [Remote host closed the connection]
2024-04-13T09:00:47 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T09:01:34 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T09:20:15 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-13T09:28:16 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-5997-8f97-39aa-475.fixed6.kpn.net] has joined ##stm32
2024-04-13T09:31:27 < antto> what's like SPI but with more flexible bitrate and more bits? I2S ;P~
2024-04-13T09:35:13 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has quit [Quit: bye]
2024-04-13T09:37:41 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has joined ##stm32
2024-04-13T09:38:05 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T09:38:59 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T09:39:41 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-13T09:40:54 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-13T09:46:42 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has quit [Quit: bye]
2024-04-13T09:49:08 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has joined ##stm32
2024-04-13T10:07:04 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T10:52:43 -!- Posterdati [~Posterdat@user/Posterdati] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
2024-04-13T10:53:38 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-13T10:54:03 -!- Posterdati [~Posterdat@user/Posterdati] has joined ##stm32
2024-04-13T10:59:31 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-13T11:24:29 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-5997-8f97-39aa-475.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-13T12:08:28 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-13T12:28:47 -!- nerozero [~nerozero@87.253.63.54] has quit [Quit: Leaving]
2024-04-13T12:29:29 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-13T13:30:46 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 246 seconds]
2024-04-13T13:32:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-13T13:46:39 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-13T14:05:44 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T14:25:58 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-1fa-8f2c-36bf-3ddc.fixed6.kpn.net] has joined ##stm32
2024-04-13T14:55:35 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-13T15:01:33 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 268 seconds]
2024-04-13T15:02:32 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-13T15:06:42 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Client Quit]
2024-04-13T15:08:19 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-13T15:17:16 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-13T15:40:50 -!- flom84 [~flom84@user/flom84] has joined ##stm32
2024-04-13T15:47:47 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 260 seconds]
2024-04-13T15:49:20 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-13T15:50:26 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-13T15:59:47 -!- flom84 [~flom84@user/flom84] has quit [Quit: Leaving]
2024-04-13T16:13:52 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 246 seconds]
2024-04-13T16:16:10 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-13T16:22:41 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-13T16:39:25 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-1fa-8f2c-36bf-3ddc.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-13T17:17:07 -!- _nuxil_ is now known as nuxil
2024-04-13T17:25:13 -!- flom84 [~flom84@user/flom84] has joined ##stm32
2024-04-13T17:25:20 -!- flom84 [~flom84@user/flom84] has quit [Client Quit]
2024-04-13T17:27:34 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-13T17:39:49 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-13T18:40:57 -!- ferdna_ [~ferdna@user/ferdna] has joined ##stm32
2024-04-13T18:43:44 -!- ferdna [~ferdna@user/ferdna] has quit [Ping timeout: 252 seconds]
2024-04-13T19:17:51 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-13T19:17:55 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T20:01:01 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 256 seconds]
2024-04-13T20:01:57 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-13T20:06:33 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-13T20:28:19 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-13T20:31:46 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-13T20:35:45 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-60c6-9fe2-2879-1dea.fixed6.kpn.net] has joined ##stm32
2024-04-13T20:56:24 -!- ferdna__ [~ferdna@user/ferdna] has joined ##stm32
2024-04-13T20:59:13 -!- ferdna_ [~ferdna@user/ferdna] has quit [Ping timeout: 268 seconds]
2024-04-13T21:10:47 -!- nerozero [~nerozero@87.253.63.54] has quit [Read error: Connection reset by peer]
2024-04-13T21:10:58 -!- boB_K7IQ [~boB_K7IQ@174-26-242-222.phnx.qwest.net] has joined ##stm32
2024-04-13T21:25:56 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-13T21:27:44 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-13T21:34:31 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 260 seconds]
2024-04-13T21:56:43 -!- rpifan [~rpifan@p200300d26718e200daeba2c1381aadcf.dip0.t-ipconnect.de] has joined ##stm32
2024-04-13T21:56:43 -!- rpifan [~rpifan@p200300d26718e200daeba2c1381aadcf.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-13T21:56:43 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-13T22:33:53 -!- fenugrec [~f@192.214.232.39] has quit [Ping timeout: 268 seconds]
2024-04-13T22:43:46 -!- qyx [~qyx@84.245.121.80] has quit [Ping timeout: 246 seconds]
2024-04-13T22:45:48 -!- qyx [~qyx@84.245.120.48] has joined ##stm32
2024-04-13T22:56:01 -!- NEYi [~NEYi@109.251.216.38] has quit [Ping timeout: 246 seconds]
2024-04-13T23:04:54 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-13T23:04:57 < Laurenceb_> https://nitter.poast.org/pic/media%2FGLAssogXcAAuwcf.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-13T23:09:40 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-13T23:17:09 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 272 seconds]
2024-04-13T23:20:18 -!- alan_o [~alan_o@2600:1700:1902:210f:6905:e17d:2044:5d0e] has quit [Remote host closed the connection]
2024-04-13T23:20:37 -!- alan_o [~alan_o@2600:1700:1902:210f:2845:ed91:d20d:8a28] has joined ##stm32
2024-04-13T23:23:01 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-13T23:23:13 -!- rpifan [~rpifan@p200300d26718e2002511227ff64a6e3e.dip0.t-ipconnect.de] has joined ##stm32
2024-04-13T23:23:13 -!- rpifan [~rpifan@p200300d26718e2002511227ff64a6e3e.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-13T23:23:13 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-13T23:32:52 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-60c6-9fe2-2879-1dea.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-13T23:35:43 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
--- Day changed su huhti 14 2024
2024-04-14T00:14:29 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T00:26:03 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-14T00:26:20 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-14T00:39:37 -!- yukam [~yukam@user/yukam] has quit [Ping timeout: 255 seconds]
2024-04-14T00:47:18 -!- yukam [~yukam@user/yukam] has joined ##stm32
2024-04-14T01:02:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-14T01:03:47 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-14T01:04:21 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 268 seconds]
2024-04-14T01:25:01 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-14T01:27:30 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-14T03:00:04 -!- Livio [~livio@user/livio] has quit [Quit: leaving]
2024-04-14T03:15:12 -!- rpifan [~rpifan@user/rpifan] has quit [Quit: Leaving]
2024-04-14T03:35:21 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-14T04:34:54 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Ping timeout: 256 seconds]
2024-04-14T04:44:26 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-14T04:59:43 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-14T07:04:30 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-14T08:04:44 -!- Xeroine [~Xeroine@user/xeroine] has quit []
2024-04-14T08:05:02 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-14T09:54:04 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-b06f-f7f6-2041-85cb.fixed6.kpn.net] has joined ##stm32
2024-04-14T10:29:02 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-14T10:49:52 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-14T11:57:36 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-14T12:18:59 -!- ferdna__ [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-14T12:32:36 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-14T12:42:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T12:56:58 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 264 seconds]
2024-04-14T12:59:27 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T13:07:00 < Steffanx> Ok I lolled Laurenceb_ 
2024-04-14T13:14:53 -!- Mangy_Dog [~Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-14T13:22:39 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 272 seconds]
2024-04-14T14:06:10 < polprog> what is the way to enable PC14/PC15 as GPIOs on the STM32F4 series?
2024-04-14T14:06:39 < polprog> i have LSE disabled, according to the datasheet this should enable them as GPIOs yet I cannot drive them
2024-04-14T14:06:46 < polprog> there is a 20k external pullup on them
2024-04-14T14:08:10 < zyp> there's nothing special you have to do, they are GPIOs by default
2024-04-14T14:09:03 < zyp> to drive them, the only thing you have to do is ensure GPIOC is enabled in RCC and that they're configured as outputs, just like for every other GPIO
2024-04-14T14:09:22 < polprog> hm, ill check the board for shorts then..
2024-04-14T14:09:32 < polprog> all other GPIOC pins work, these two are stuck high
2024-04-14T14:11:55 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T14:19:09 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-14T14:25:36 < polprog> thats weird. a 24k pullup should be high enough resistance to not over load the driver
2024-04-14T14:45:45 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-14T15:02:43 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 246 seconds]
2024-04-14T15:05:44 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T15:08:36 < qyx> polprog: those gpio are driven by the vbat domain
2024-04-14T15:08:52 < qyx> and require something additional I don't remember now
2024-04-14T15:08:59 < qyx> maybe enabling the backup domain?
2024-04-14T15:20:49 < polprog> The way i have it set up is Vbat connected to Vdd
2024-04-14T15:27:36 < polprog> backup power domain is now on, but still no change
2024-04-14T15:28:45 < polprog> backup rtc register contents are all reset on boot
2024-04-14T15:30:35 < polprog> http://paste.debian.net/1314073/
2024-04-14T15:30:38 < polprog> this is my code now
2024-04-14T16:02:53 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 240 seconds]
2024-04-14T16:05:34 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T16:49:31 < polprog> i gave up and rewired these two chip selects to other gpios
2024-04-14T17:10:00 -!- fenugrec [~f@192.214.232.39] has joined ##stm32
2024-04-14T17:26:22 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-14T17:32:15 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-14T17:34:30 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T18:31:36 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-14T18:48:38 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-14T18:57:06 -!- fenugrec [~f@192.214.232.39] has quit [Quit: fenugrec]
2024-04-14T19:01:20 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-14T19:08:27 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-14T19:10:43 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T19:54:31 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-14T20:09:20 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-14T20:24:12 -!- fenugrec [~f@192.214.232.39] has joined ##stm32
2024-04-14T20:32:19 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-b06f-f7f6-2041-85cb.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-14T20:54:08 -!- drzacek [~quassel@2a01:3d8:4b6:5200:f378:b7f4:178f:eb94] has joined ##stm32
2024-04-14T20:55:40 -!- drzacek [~quassel@2a01:3d8:4b6:5200:f378:b7f4:178f:eb94] has quit [Client Quit]
2024-04-14T21:01:11 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 272 seconds]
2024-04-14T21:13:30 -!- drzacek [~quassel@2a01:3d8:4b6:5200:f378:b7f4:178f:eb94] has joined ##stm32
2024-04-14T21:18:37 -!- drzacek [~quassel@2a01:3d8:4b6:5200:f378:b7f4:178f:eb94] has quit [Remote host closed the connection]
2024-04-14T21:27:47 -!- CygniX [~CygniX@user/CygniX] has joined ##stm32
2024-04-14T21:56:53 < zyp> huh, I'm failing to make a blinky
2024-04-14T21:57:34 < zyp> h7s7-dk has four «user leds», so I've turned on RCC, configured them as outputs and try toggling them and nothing is happening…
2024-04-14T21:59:48 < zyp> the registers respond when I poke them in the debugger, so I know RCC is enabled, GPIOx_MODER is set to output and I see it responding to writes in the debugger
2024-04-14T22:00:01 < zyp> and yet nothing happens when I poke ODR
2024-04-14T22:08:02 < jpa-> i think STM32H7S7 is just some ST joke, the part number is too ridiculous
2024-04-14T22:08:28 < zyp> :)
2024-04-14T22:11:08 < qyx> I would check if the leds are really connected where you expect them to be
2024-04-14T22:11:18 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-14T22:11:22 < jpa-> so the GPIOM2/3 and GPIOO1/5 are what you are writing?
2024-04-14T22:11:29 < zyp> yes
2024-04-14T22:12:28 < jpa-> at what addresses?
2024-04-14T22:12:34 < qyx> gpioo o_O
2024-04-14T22:12:52 < jpa-> funnily RM0468 doesn't seem to give the GPIOM and GPIOO register addresses
2024-04-14T22:12:59 < zyp> (gdb) p/x &$mmio_ref(GPIOM).ODR
2024-04-14T22:12:59 < zyp> $110 = 0x58023014
2024-04-14T22:13:06 < zyp> 468?
2024-04-14T22:13:17 < zyp> I'm reading RM0477
2024-04-14T22:13:31 < jpa-> oh crap, google gave me wrong pdf
2024-04-14T22:13:44 < qyx> 100MB wasted
2024-04-14T22:19:02 < jpa-> zyp: does IDR react to ODR? does ODR read back correctly?
2024-04-14T22:20:20 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 252 seconds]
2024-04-14T22:21:28 < zyp> no, yes, respectively
2024-04-14T22:21:33 < zyp> I think it's a power gating thing
2024-04-14T22:21:34 < zyp>   IDR = 0x0,
2024-04-14T22:21:34 < zyp>   ODR = 0xc,
2024-04-14T22:22:58 < jpa-> funny, considering GPIOM leds are active low, so IDR should float high even if not driven
2024-04-14T22:23:08 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T22:23:27 < zyp> they measure around 1.5V
2024-04-14T22:24:42 < zyp> GPIOO is part of the XSPIM1 power domain, which is off by default, but that doesn't explain why GPIOM doesn't work
2024-04-14T22:26:39 < zyp> ah, PM2/3 are UCPD_DBx, they're probably reserved for that until released
2024-04-14T22:27:58 < jpa-> UCPD seems to go through normal altfunc selection, so if you set mode, it shouldn't matter
2024-04-14T22:28:13 < zyp> not the DB pins
2024-04-14T22:28:25 < zyp> UCPD_DBDIS in PWR_UCPDR
2024-04-14T22:28:50 < jpa-> hmm yeah
2024-04-14T22:29:01 < jpa-> very fun, considering they show up in the alternate functions table
2024-04-14T22:29:38 < jpa-> the UCPD_DBDIS bit is not quite explicit on what it does to the DBx pins
2024-04-14T22:30:15 < zyp> hrm, that was not enough
2024-04-14T22:30:22 < qyx> I failed witn that too in the pasy (ucpd)
2024-04-14T22:30:28 < qyx> past
2024-04-14T22:30:52 < qyx> don't they have an example for the dk?
2024-04-14T22:31:12 < zyp> idk, that's no fun
2024-04-14T22:31:18 < qyx> oh
2024-04-14T22:33:30 < jpa-> do other GPIOM bits work in ODR -> IDR roundtrip? that could tell whether it is related to UCPD
2024-04-14T22:33:57 < zyp> haven't tried
2024-04-14T22:34:26 < zyp> ah, found it
2024-04-14T22:34:47 < zyp> PWR_CSR2.USB33DEN: Note: This bit should be set if USB HS, FS, or a GPIO on PMx is used
2024-04-14T22:35:14 < zyp> yep, that did the trick
2024-04-14T22:35:19 < jpa-> heh, i was searching for "GPIOM".. but of course they have to write "GPIO on PMx" :)
2024-04-14T22:36:08 < zyp> kinda makes sense, GPIOM block is where all the usb stuff is, and that's powered by VDD33USB
2024-04-14T22:36:45 < jpa-> definitely sounds like a fun stm32 to work with, i'll expect daily blogs about it
2024-04-14T22:37:36 < zyp> so regular banks A-H are powered by VDD, O-P are powered by VDDXSPI1, M is powered by VDD33USB and N is powered by VDDXSPI2
2024-04-14T22:37:45 < zyp> yeah, it does seem fun
2024-04-14T22:37:58 < zyp> only thing that'd make it even more fun would be dualcore :)
2024-04-14T22:38:43 < zyp> okay, let's see if we can get the GPIOO leds working too while we're at it
2024-04-14T22:39:47 < zyp> setting EN_XSPIM1 in the same register did the trick
2024-04-14T22:41:36 < zyp> https://bin.jvnv.net/file/POvqq.jpg
2024-04-14T23:31:55 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 272 seconds]
2024-04-14T23:33:55 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-14T23:41:23 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 264 seconds]
2024-04-14T23:43:46 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
--- Day changed ma huhti 15 2024
2024-04-15T00:02:00 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-15T00:03:09 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T00:33:40 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-15T00:36:14 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 256 seconds]
2024-04-15T00:38:47 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-15T00:39:55 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-15T00:46:37 < qyx> have you ever seen a tcp/ip over gatt?
2024-04-15T00:46:47 < qyx> *implementation
2024-04-15T00:50:44 < zyp> hmm, this thing has an ETH_CLK output that can generate a clock for the PHY, so you can share a single crystal without even using MCO
2024-04-15T00:51:25 < zyp> phy got dedicated crystal on the devkit though
2024-04-15T01:26:50 < zyp> uh oh, looks like the dwc ethernet block in h7rs is pretty different to the one in f4
2024-04-15T01:27:23 < zyp> I was hoping I could just start by bringing up the code I once wrote for f4
2024-04-15T01:27:42 < zyp> (and I was also hoping to port this to an at32f4 board afterwards)
2024-04-15T01:32:18 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-15T01:32:31 < jbo> let's say you got three clock sources (STM32 MCO, external (via SMA), 100 MHz on-board oscillator). how how would one go about selecting which one goes into an FPGA? I assume a regular 2.54mm jumper is not really suitable for 100 MHz?
2024-04-15T01:33:11 < zyp> correct
2024-04-15T01:33:35 < zyp> you're doing a board with stm32+fpga?
2024-04-15T01:33:44 < jbo> yeah
2024-04-15T01:33:46 < jbo> F4 + iCE40
2024-04-15T01:33:51 < zyp> oh
2024-04-15T01:33:57 < zyp> why not ecp5?
2024-04-15T01:34:11 < zyp> idk about ice40, but for ecp5 the answer to your question would be obvious
2024-04-15T01:34:15 < jbo> because iCE40 only does simple slow I/O
2024-04-15T01:34:28 < jbo> I mean the iCE40 has multiple clock inputs, so I wonder whether I should just hook up one to each.
2024-04-15T01:34:43 < jbo> and then choose during synthesis
2024-04-15T01:34:49 < jbo> (it's sort of a "dev kit" board)
2024-04-15T01:34:53 < zyp> yeah, for ecp5 the obvious answer is: do all of them, you've got a ton of clock inputs
2024-04-15T01:35:06 < jbo> yeah AFAIK iCE40 has 8
2024-04-15T01:35:09 < zyp> for ice40 I think you're a lot more limited in how clocks need to hook up
2024-04-15T01:35:34 < jbo> from what I can tell there are GBIN0 to GBIN7 which are "clock inputs"
2024-04-15T01:36:11 < zyp> but again, what's the benefit of doing ice40 over ecp5? last I checked you could get ecp5 just as cheap
2024-04-15T01:36:11 < jbo> > The iCE40 device has eight high drive buffers called global buffers (GBUFx). These are connected to eight low-skew
2024-04-15T01:36:11 < jbo> global lines, designed primarily for clock distribution, but also useful for other high-fanout signals such as set/reset and
2024-04-15T01:36:11 < jbo> enable signals.
2024-04-15T01:36:30 < jbo> customer has like 5k iCE40 units in stock and is persistent. I tried.
2024-04-15T01:36:36 < zyp> can either of those feed into PLL?
2024-04-15T01:36:50 < jbo> https://www.latticesemi.com/-/media/LatticeSemi/Documents/ApplicationNotes/IK2/FPGA-TN-02052-1-4-iCE40-sysCLOCK-PLL-Design-User-Guide.ashx?document_id=47778
2024-04-15T01:36:54 < jbo> page 7
2024-04-15T01:37:20 < jbo> iCE40HX to be specific
2024-04-15T01:37:27 < zyp> hx8k?
2024-04-15T01:37:32 < jbo> hx4k
2024-04-15T01:37:35 < jbo> (again, don't ask)
2024-04-15T01:37:44 < jbo> customer insists on not using BGA
2024-04-15T01:37:48 < zyp> hmm, I think that might be relabelled hx8k
2024-04-15T01:38:02 < zyp> not sure
2024-04-15T01:38:04 < jbo> so iCE40HX4K-TQ144 it is
2024-04-15T01:38:52 < zyp> there's a bunch of that in the fpga market apparently, same dies sold as multiple sizes and just artifically limited during synthesis
2024-04-15T01:39:08 < zyp> ecp5-12 is ecp5-25
2024-04-15T01:39:23 < zyp> and I think hx4k might be hx8k
2024-04-15T01:39:54 < jbo> possible. client insists on using QFP so can't change that
2024-04-15T01:39:57 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-15T01:40:01 < zyp> yup
2024-04-15T01:40:02 < zyp> «However this is actually an 8K die that has been knobbled by Lattice's proprietary programming software.»
2024-04-15T01:40:23 < zyp> so if you build with yosys/nextpnr, it'll be an 8k ;)
2024-04-15T01:40:24 < jbo> anyway, I guess the answer is indeed: have each clock source feed into one GBINx pin on the FPGA and not doing fancy clock switcher on PCB
2024-04-15T01:40:31 < jbo> that's nice
2024-04-15T01:40:37 < jbo> where did you get that intel from?
2024-04-15T01:40:48 < zyp> https://www.fpgarelated.com/thread/799/mystorm-a-30-ice40-arm-m3-dev-board
2024-04-15T01:42:07 < jbo> neat
2024-04-15T01:42:35 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-15T01:42:42 < zyp> hmm, let's see, «To determine which GBIN pin can be used as a PLL Reference Clock Input, refer to the Hardware Design Considerations section.»
2024-04-15T01:42:45 < zyp> this is important
2024-04-15T01:43:49 < zyp> oh, doesn't really matter: «If the PLL requires an external reference clock source driven through the Global Buffer Input (GBIN), use any GBIN for the PLL reference clock input.»
2024-04-15T01:43:57 < jbo> yeah but that is listed in the Radiant software section
2024-04-15T01:44:07 < jbo> why on earth would that be different if using iCEcube2 tho
2024-04-15T01:44:30 < zyp> software might be more limiting than hardware
2024-04-15T01:45:22 < jbo> hmpf
2024-04-15T01:45:39 < jbo> I am also still very confused on how to hook up the external SPI configuration flash if there's an on-board FTDI chip for programming
2024-04-15T01:45:44 < zyp> or I guess the limitations described might be the same, just presented in different ways
2024-04-15T01:45:54 < zyp> that's a shitshow
2024-04-15T01:45:58 < jbo> I looked at the icebreaker schematics and it has a bunch of solder bridges, "cut trace here" and jumpers
2024-04-15T01:46:17 < zyp> exactly, I remember I was talking with esden about that some time
2024-04-15T01:46:21 < jbo> Ideally I'd like to be able to programm from both the on-board FTDI and the on-board STM32
2024-04-15T01:46:25 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T01:46:42 < jbo> as it's a dev-kit it's not really cost sensitive so not sure whether it's worth adding some sort of bus switch/mux
2024-04-15T01:46:51 < jbo> but it indeed seems to be just a shitshow
2024-04-15T01:47:05 < zyp> why do you have a FTDI?
2024-04-15T01:47:28 < jbo> to have a way to program the FPGA config flash without using the MCU (dev-kit -> options)
2024-04-15T01:47:53 < jbo> also, want an UART bridge to the FPGA too. the FT2232HL is dual channel, so I can use one for SPI and one for UART
2024-04-15T01:48:09 < zyp> since you've got an onboard STM32, I suggest just setting the ice40 up as a SPI slave and letting the STM32 do everyting else (i.e. STM32 can get FPGA bitstream either from a flash, from USB or whatever)
2024-04-15T01:49:37 < jbo> I get that, jup. would still like to have the option to mess with the FPGA without the MCU being invovled tbh.
2024-04-15T01:50:14 < jbo> I'd rather be able to ONLY program the FPGA via the FTDI and never via the MCU than the other way around.
2024-04-15T01:50:29 < zyp> if you were doing ecp5, sure, then you'd have separate signals for JTAG and config flash/mcu
2024-04-15T01:50:59 < zyp> but ice40 only have the single SPI interface and setting it up to both be able to be a master and a slave is a shitshow
2024-04-15T01:51:58 < zyp> the thing about the solder bridges on the icebreaker, AIUI, is that the FTDI can either talk to the FPGA or the flash
2024-04-15T01:52:32 < zyp> AIUI if it's set up as FTDI -> FPGA, the flash isn't used at all, and the FPGA must always be loaded over USB
2024-04-15T01:53:03 < zyp> if it's set up as FTDI -> flash, you also get FPGA -> flash, and to reload it you must always rewrite flash and have FPGA reload it from flash
2024-04-15T01:53:55 < zyp> it's really inflexible, so the easiest solution to be able to do anything is to just let the mcu manage it
2024-04-15T01:54:11 < jbo> hmm
2024-04-15T01:54:32 < jbo> so if I ALWAYS program FTDI -> flash and the FPGA is always loading from flash, can I have a static hardware setup?
2024-04-15T01:54:51 < zyp> I think that might be the icebreaker default
2024-04-15T01:55:17 < zyp> not sure how fun it'll be to try adding the mcu into that mix
2024-04-15T02:01:43 < jbo> yeah I guess I'll just drop the requirement of also being able to program form the MCU
2024-04-15T02:13:28 -!- Mangy_Dog [~Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 260 seconds]
2024-04-15T02:53:16 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Read error: Connection reset by peer]
2024-04-15T02:55:06 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-15T03:05:47 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-15T03:37:17 < Ecco> I'm considering doing a N-bits adder using discrete transistors, just for fun. Should I use BJTs or MOSFETs?
2024-04-15T03:39:05 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-15T07:56:06 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-15T07:58:44 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a016-b53d-a492-3f09.fixed6.kpn.net] has joined ##stm32
2024-04-15T08:08:43 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a016-b53d-a492-3f09.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-15T08:21:33 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-15T08:21:44 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-15T08:26:36 < jpa-> Ecco: mosfets are probably easiest
2024-04-15T09:13:14 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-15T09:21:35 -!- Xeroine [~Xeroine@user/xeroine] has quit [Ping timeout: 264 seconds]
2024-04-15T09:22:57 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-15T09:25:17 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-15T09:37:25 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-15T10:39:01 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-15T10:48:58 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-15T10:56:17 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-15T11:52:41 -!- specing [~specing@user/specing] has quit [Ping timeout: 268 seconds]
2024-04-15T12:23:43 < karlp> qyx: why tcp over gatt? not over l2cap? I'm pretty sure there's a PAN v6 shit bluetooth class, I've not looked real h ard though
2024-04-15T12:26:46 < karlp> but related, todays shortsighted standard is weight scales in btle: https://bin.jvnv.net/file/sgAcd.png
2024-04-15T12:27:01 < karlp> 5g, best I can offer, but we'll build imperial and SI switchign into the fucking protocol :)
2024-04-15T12:35:59 -!- catphish [~quassel@user/catphish] has quit [Ping timeout: 272 seconds]
2024-04-15T12:40:07 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has joined ##stm32
2024-04-15T12:40:13 < machinehum> SUP
2024-04-15T12:41:22 -!- catphish [~quassel@user/catphish] has joined ##stm32
2024-04-15T12:41:35 < karlp> da sky brah
2024-04-15T12:44:42 < machinehum> Sure is
2024-04-15T13:06:17 < PaulFertser> karlp: huh, I'm not sure it feels natural that resolution can be affected by changing the units.
2024-04-15T13:06:37 < karlp> I believe the thinking is that you don't change the units.
2024-04-15T13:06:41 < karlp> but that's not actually at all uncommon
2024-04-15T13:07:06 < PaulFertser> All non-standard units are naturally just a way to display, nothing more...
2024-04-15T13:07:10 < karlp> it's a problem we have here a lot too, the listed ratings for divisions in pounds/metric are not at all interchangeable
2024-04-15T13:07:20 < karlp> PaulFertser: that would be sensible....
2024-04-15T13:08:04 < PaulFertser> Probably your company can enroll to become part of standards group there? At least ISO usually seeks expert opinions.
2024-04-15T13:08:15 < karlp> I've tried to insist that it's all numbers internally, why shuld it be re-calibrated to be in pounds/oz instead of metric, but.... "no, must be calibrated in correct units"
2024-04-15T13:08:27 < karlp> lol.
2024-04-15T13:08:49 < karlp> I'd have to get everybody _here_ on the same page first :)
2024-04-15T13:09:23 < karlp> and apparently I thought UL/ETL testing for safety stuff was inane pixel peeping.
2024-04-15T13:10:03 < PaulFertser> Damn
2024-04-15T13:10:34 < karlp> there's still a lot less harmonization of standards than I was expecting too.
2024-04-15T13:10:45 < PaulFertser> I wonder if you have no metrologist there at all.
2024-04-15T13:11:08 < PaulFertser> It's like a separate set of knowledge and skills.
2024-04-15T13:11:10 < karlp> "this is how it has always been, you're wrong"
2024-04-15T13:11:19 < PaulFertser> Dickheads
2024-04-15T13:11:23 < karlp> correct 
2024-04-15T13:13:00 < karlp> still, bluetooth class doc that made a scale profile, and then tied it _hard_ to measuring weights of humans is just fucking silly.
2024-04-15T13:13:19 < karlp> there's class docs for alllll soooorts of things, but some of them are just fucked in the head.
2024-04-15T13:16:05 < machinehum> I don't have time to work on a DDR/MPU routing thing so I'm looking for someone on freecancer. This was the "sample work" someone sent me: https://imgur.com/a/aOVopws
2024-04-15T13:16:09 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-15T13:17:01 < machinehum> What do you guys think? Looks pretty good to me SMD components are a bit to mainstream for my hipster product so I'll probably go with this guy.
2024-04-15T13:18:53 < PaulFertser> :D
2024-04-15T13:19:16 < PaulFertser> The micro USB there is SMD though ;)
2024-04-15T13:21:10 < machinehum> PaulFertser: lol
2024-04-15T13:27:38 < BrainDamage> karlp: there's a tiny difference in uncertainty if you use one unit or another due to truncation error, but probablly nothing you care about
2024-04-15T13:38:43 < karlp> yeah, the dynamic range of a scale isnt' enough for that to matter.
2024-04-15T13:38:51 < karlp> at least, not IMO.
2024-04-15T13:39:08 < karlp> but, I'm jus there to fix the most egregarious shit.
2024-04-15T13:39:20 < karlp> fundamentals I'm not even looking at right now.
2024-04-15T13:47:18 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-15T14:19:06 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T14:19:35 < rpifan> hi
2024-04-15T14:24:34 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-15T14:24:45 -!- rpifan [~rpifan@p200300d26718e20034fd039342e5b7ea.dip0.t-ipconnect.de] has joined ##stm32
2024-04-15T14:24:45 -!- rpifan [~rpifan@p200300d26718e20034fd039342e5b7ea.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-15T14:24:45 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T14:26:46 -!- rpifan_ [~rpifan@p200300d26718e200cccc9f0ced2f6c7a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-15T14:29:48 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-15T14:42:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T14:45:43 -!- rpifan_ [~rpifan@p200300d26718e200cccc9f0ced2f6c7a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-15T14:45:43 -!- rpifan_ [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T14:45:47 -!- rpifan_ is now known as rpifan
2024-04-15T14:57:50 < polprog> machinehum: neon-, neon+, is that supposed to be a watch or something?
2024-04-15T14:58:05 < polprog> i dont think i saw a THT microusb port
2024-04-15T14:58:19 < polprog> proooobabyl exists so that these ultra cheap single sided boards can work
2024-04-15T14:58:41 < machinehum> polprog: I have no clue
2024-04-15T14:59:08 < polprog> lol the usb is just used for power with no negotiation resistors
2024-04-15T14:59:29 < polprog> i guess i should not laugh at a newbie, i was on that level a few years ago :<
2024-04-15T15:34:14 < qyx> a newbie doesn't ask for money
2024-04-15T15:34:40 < qyx> you can't be a newbie if you go on freelancer to earn money as a self-employed
2024-04-15T15:34:51 < qyx> that's at least a half-pro level
2024-04-15T15:36:56 < qyx> karlp: don't get me started with imperial
2024-04-15T15:40:54 < machinehum> qyx: https://imgur.com/a/rodLO4F
2024-04-15T15:40:57 < machinehum> ?
2024-04-15T15:41:18 < machinehum> I don't think that's a thing in Europe
2024-04-15T15:54:00 < ventYl> no four-letter curse words please! no inch, mile, yard, pint, acre
2024-04-15T15:54:06 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-65a5-c665-55e-ce13.fixed6.kpn.net] has joined ##stm32
2024-04-15T15:57:36 < karlp> ounce is ok though ;)
2024-04-15T15:57:44 < qyx> machinehum: never seen that
2024-04-15T15:57:59 < ventYl> that's a 5-letter curse word
2024-04-15T15:58:59 < machinehum> qyx: Potentially outlawed in Europe
2024-04-15T15:59:14 < karlp> what is it?
2024-04-15T15:59:22 < machinehum> You know ritz bits?
2024-04-15T15:59:41 < machinehum> They're two little crackers with cheese in between
2024-04-15T16:00:18 < machinehum> That is the type of cheese that is in between
2024-04-15T16:00:20 < machinehum> It's not good
2024-04-15T16:00:57 < qyx> so basically an imperial shit
2024-04-15T16:12:53 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-15T16:24:06 < jbo> moinser
2024-04-15T16:28:11 < ColdKeyboard> Is there a library for stepper motors that works with bare metal? Something like FlexyStepper or SpeedyStepper for Arduino
2024-04-15T16:28:32 < jbo> huh?
2024-04-15T16:28:40 < jbo> what driver are you using?
2024-04-15T16:30:45 < ColdKeyboard> I have the generic DRV8825 
2024-04-15T16:31:07 < jbo> that has a simple step/dir interface. what kind of library are you looking for? it's just two GPIOs
2024-04-15T16:32:12 < ColdKeyboard> Instead of (or before I endavour in) making my own library for speed up/down, goto position, absolute/relative move etc.. I was hoping there is a library for bare metal that someone could recommend
2024-04-15T16:34:01 < jbo> so you have an encoder for feedback?
2024-04-15T16:35:41 < ColdKeyboard> No
2024-04-15T16:35:57 < jbo> then it's still just two GPIOs.
2024-04-15T16:36:03 < jbo> direction controls CW/CCW
2024-04-15T16:36:10 < jbo> and then step for each step
2024-04-15T16:36:22 < jbo> so goto position etc. is really just an internal step counter, nothing else.
2024-04-15T16:36:29 < ColdKeyboard> Thanks
2024-04-15T16:37:12 < ColdKeyboard> But the question is; Anyone aware of a library like FlexyStepper (or similar Arduino ones) that work on bare metal?
2024-04-15T16:38:46 < jbo> seriously
2024-04-15T16:38:53 < jbo> just use the arduino library then
2024-04-15T16:39:18 < jbo> :)
2024-04-15T16:41:08 < jbo> maybe we can tackle this from the other way around: can you explain what kind of problem you're hitting or what questions you have regarding driving this interface without a library?
2024-04-15T16:42:05 < qyx> I would opt for a library too
2024-04-15T16:42:26 < qyx> because move-to is not just a bunch of pulses
2024-04-15T16:42:54 < qyx> observing kinetics too, proper timing too,etc.
2024-04-15T16:43:16 < jbo> that is motion & trajectory control on top of the stepper interface
2024-04-15T16:43:32 < jbo> which has technically nothing to do with the stepper motor
2024-04-15T16:43:42 < qyx> maybe that's what ColdKeyboard is asking for
2024-04-15T16:43:57 < qyx> idk what that flexiplexi is doi g
2024-04-15T16:44:28 < jbo> aye, ColdKeyboard are you looking for a complete motion control library?
2024-04-15T16:44:37 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has quit [Ping timeout: 256 seconds]
2024-04-15T16:45:06 < qyx> apparently that's what flexystepperis doing
2024-04-15T16:45:12 < qyx> on top if a step-dir iface
2024-04-15T16:45:15 < ColdKeyboard> It's doing exactly that. You can set the motor speed, acceleration etc. And then you can issue move by specifying steps, rotation or "distance"
2024-04-15T16:46:17 < ColdKeyboard> I agree it's not difficuilt to make but it is time consuming and I would rather use a library if one already exists rather than reinvent the wheel
2024-04-15T16:46:43 < jbo> sure that's fine. I "assumed" you're only talking about the stepper interface and ingesting a library for that would be more time consuming that just toggling two GPIOs IMHO
2024-04-15T16:48:07 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-15T16:49:16 < ColdKeyboard> No that part is luckily dirt simple, just direction pin and step toggle...
2024-04-15T16:49:35 < ColdKeyboard> Any suggestions for the library before I start porting this Arduino one?
2024-04-15T17:00:00 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has joined ##stm32
2024-04-15T17:04:32 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has quit [Ping timeout: 252 seconds]
2024-04-15T17:08:46 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-15T17:15:32 -!- specing [~specing@user/specing] has quit [Ping timeout: 252 seconds]
2024-04-15T17:16:41 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-15T17:16:57 -!- Mangy_Dogg [Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has joined ##stm32
2024-04-15T17:17:11 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 252 seconds]
2024-04-15T17:20:32 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-15T17:21:33 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T17:21:59 -!- Mangy_Dogg [Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-15T17:32:28 -!- specing [~specing@user/specing] has quit [Ping timeout: 268 seconds]
2024-04-15T17:35:10 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-15T17:36:07 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has joined ##stm32
2024-04-15T17:44:28 -!- c10ud [~c10ud@user/c10ud] has quit [Read error: Connection reset by peer]
2024-04-15T17:45:23 -!- c10ud [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has joined ##stm32
2024-04-15T17:45:23 -!- c10ud [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has quit [Changing host]
2024-04-15T17:45:23 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-15T17:51:13 < machinehum> https://www.youtube.com/watch?v=8AbtCE-sePc
2024-04-15T17:51:16 < machinehum> HOly ahahaha
2024-04-15T17:51:18 < ColdKeyboard> I know Arduino is not famous for it but I love how some Arduino libraries are YOLOing resources :)
2024-04-15T17:51:46 < ColdKeyboard> FlexyStepper... almost every function uses either long or float or even unsigned long :)
2024-04-15T17:52:16 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:cbfd:1bd4:8b0:2d5e] has joined ##stm32
2024-04-15T17:52:50 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:cbfd:1bd4:8b0:2d5e] has quit [Client Quit]
2024-04-15T17:53:04 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:cbfd:1bd4:8b0:2d5e] has joined ##stm32
2024-04-15T18:01:07 < karlp> all arduinos are 32bit right? :)
2024-04-15T18:01:22 < karlp> would you rather they used int_least8_t?
2024-04-15T18:01:29 -!- rpifan [~rpifan@p200300d26718e200fd010ba5fe57d534.dip0.t-ipconnect.de] has joined ##stm32
2024-04-15T18:01:29 -!- rpifan [~rpifan@p200300d26718e200fd010ba5fe57d534.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-15T18:01:29 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T18:01:56 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-15T18:02:53 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T18:13:52 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-15T18:14:03 -!- rpifan [~rpifan@p200300d26718e200737e5bb43a74f423.dip0.t-ipconnect.de] has joined ##stm32
2024-04-15T18:14:03 -!- rpifan [~rpifan@p200300d26718e200737e5bb43a74f423.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-15T18:14:03 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T18:17:43 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-15T18:22:43 < ColdKeyboard> Nope. You can run Arduino on 8bit micros
2024-04-15T18:24:30 < karlp> I know :)
2024-04-15T18:38:58 -!- machinehum [~machinehu@ip-185-104-138-52.ptr.icomera.net] has quit [Ping timeout: 264 seconds]
2024-04-15T18:40:59 -!- machinehum [~machinehu@213.55.225.161] has joined ##stm32
2024-04-15T18:49:11 < jpa-> little point using anything 8-bit nowadays though
2024-04-15T18:51:48 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-15T18:52:40 -!- machinehum [~machinehu@213.55.225.161] has quit [Ping timeout: 255 seconds]
2024-04-15T18:53:39 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-15T18:53:59 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T18:56:19 < jbo> zyp, ping
2024-04-15T18:56:43 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-15T19:08:14 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-15T19:36:51 < qyx> jpa-: yesterday I was thinking about doing some proj with a 8051 or 68hc11
2024-04-15T19:37:52 < jpa-> putting them upside down on the floor as a joke?
2024-04-15T19:38:09 < jpa-> or maybe crushing them to recover the gold from bond wires?
2024-04-15T19:38:18 < qyx> No.
2024-04-15T19:57:46 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-15T20:02:52 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-15T20:16:47 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-15T20:18:15 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has joined ##stm32
2024-04-15T20:18:15 -!- rpifan [~rpifan@p200300d26718e200172be97bf075286a.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-15T20:18:15 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-15T20:23:00 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-15T20:26:11 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T20:26:58 -!- System_Error [~SystemErr@user/systemerror] has quit [Read error: Connection reset by peer]
2024-04-15T20:41:16 < zyp> jbo, pong
2024-04-15T21:02:24 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-15T21:03:48 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T21:06:59 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-15T21:08:37 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-15T21:19:39 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T21:22:55 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-65a5-c665-55e-ce13.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-15T21:40:51 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 260 seconds]
2024-04-15T21:47:28 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-15T21:52:05 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-15T21:56:16 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 255 seconds]
2024-04-15T22:04:39 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-15T22:26:35 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-65a5-c665-55e-ce13.fixed6.kpn.net] has joined ##stm32
2024-04-15T22:40:22 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 255 seconds]
2024-04-15T22:44:24 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-15T22:44:35 -!- qyx [~qyx@84.245.120.48] has quit [Ping timeout: 256 seconds]
2024-04-15T22:46:28 -!- qyx [~qyx@84.245.120.167] has joined ##stm32
2024-04-15T22:50:08 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-15T22:50:30 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-15T22:51:17 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-15T23:11:51 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Ping timeout: 260 seconds]
2024-04-15T23:37:36 < Ecco> Back on that "discrete transistor" PCB, what would be some fun stuff to build that'd still be manageable? I intend to use a custom PCB with SMT transistors (as opposed to breadbords + thru-hole)
2024-04-15T23:37:49 < Ecco> I feel like an 8-bit adder might be fun, but I'm still not quite sure about the I/O part
2024-04-15T23:38:29 < Ecco> I guess LEDs + some buttons for a start, but the problem is you're still explicitely manipulating base 2 which is cool but nowhere near what a "normal" person would call a calculator
2024-04-15T23:46:17 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-15T23:46:45 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-15T23:49:55 < qyx> so do a BCD one?
2024-04-15T23:50:16 < qyx> some selector 0-9 rotary switches and for the output 7seg LED displays?
2024-04-15T23:50:50 < qyx> you could BCD-decode to 7seg with transistors too
2024-04-15T23:51:31 < qyx> I have seen some circuit with 4 transistors and a couple of diodes for one digit somewhere
2024-04-15T23:52:23 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
--- Day changed ti huhti 16 2024
2024-04-16T00:09:29 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 272 seconds]
2024-04-16T00:11:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-16T00:20:21 < Ecco> hmm, wait, is it complicated to do a BCD adder?
2024-04-16T00:20:50 < Ecco> bcd-decode to 7seg -> Yeah, that I guess I could figure out how to do, doesn't seem too hard. I was really worried about the base conversion
2024-04-16T00:24:19 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-65a5-c665-55e-ce13.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-16T00:25:51 < qyx> idk, isn't it just a bunch of 4 bit adders with a slightly modified carry logic?
2024-04-16T00:51:32 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-16T00:54:10 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-16T00:58:53 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 256 seconds]
2024-04-16T01:20:35 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T01:20:52 -!- rpifan [~rpifan@p200300d26718e200939ccb8f560f29ff.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T01:20:52 -!- rpifan [~rpifan@p200300d26718e200939ccb8f560f29ff.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T01:20:52 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T01:27:46 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:cbfd:1bd4:8b0:2d5e] has quit [Ping timeout: 255 seconds]
2024-04-16T01:30:10 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T01:30:22 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T01:42:40 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T01:42:52 -!- rpifan [~rpifan@p200300d26718e2002696122877ea56fc.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T01:42:52 -!- rpifan [~rpifan@p200300d26718e2002696122877ea56fc.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T01:42:52 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T01:58:46 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-16T01:59:32 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-16T02:54:08 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 252 seconds]
2024-04-16T03:32:54 -!- rpifan [~rpifan@user/rpifan] has quit [Quit: Leaving]
2024-04-16T03:51:21 -!- nuxil [~nuxil@141.195.51.41] has joined ##stm32
2024-04-16T03:52:21 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-16T03:56:07 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 256 seconds]
2024-04-16T04:00:13 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-16T04:04:52 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-16T06:51:00 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-16T06:55:46 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 264 seconds]
2024-04-16T07:19:58 < ColdKeyboard> Huh... weird thing that I'm first time seeing
2024-04-16T07:20:17 < ColdKeyboard> I'm using an 8-bit micro and when I use switch, only the first case is executed properly
2024-04-16T07:20:31 < ColdKeyboard> the second, third, default... all of them don't execute
2024-04-16T07:20:44 < ColdKeyboard> And if I switch to if/else-if, everything works fine
2024-04-16T07:36:10 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-16T07:36:47 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Ping timeout: 260 seconds]
2024-04-16T07:37:40 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-16T07:41:15 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has quit [Read error: Connection reset by peer]
2024-04-16T08:04:57 -!- jmcgnh [~jmcgnh@wikipedia/jmcgnh] has joined ##stm32
2024-04-16T08:07:02 < antto> ColdKeyboard, is your switch written properly?
2024-04-16T08:13:25 < qyx> code?
2024-04-16T08:16:54 < antto> long ago i found a bug in gcc involving switch statements in compile-time-executed functions, but i forgot to report it..
2024-04-16T08:46:54 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-16T08:53:00 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-16T08:57:35 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 256 seconds]
2024-04-16T09:10:39 < qyx> so starting to jpa- a bq25756 today
2024-04-16T09:10:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-16T09:15:45 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-16T09:22:16 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-16T09:39:56 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-16T09:44:33 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Ping timeout: 272 seconds]
2024-04-16T10:03:57 < qyx> but now idk, maybe I should make the converter as a doughterboard
2024-04-16T10:04:01 -!- machinehum [~machinehu@xcpe-62-167-160-167.cgn.res.adslplus.ch] has joined ##stm32
2024-04-16T10:04:10 < qyx> because all the components are flat
2024-04-16T10:12:11 < Steffanx> You should make it work first. After all the success stories I heard ....
2024-04-16T10:36:40 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-16T10:37:04 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-16T12:44:55 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-16T12:48:23 -!- machinehum [~machinehu@xcpe-62-167-160-167.cgn.res.adslplus.ch] has quit [Quit: WeeChat 4.2.1]
2024-04-16T13:06:09 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 255 seconds]
2024-04-16T13:06:57 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-16T14:11:20 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-16T14:55:19 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-16T14:59:36 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-16T15:21:56 < jbo> new bq25756 pcbs arrived yesterday, qyz
2024-04-16T15:21:57 < jbo> qyx
2024-04-16T15:22:13 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-16T15:48:54 < jpa-> thanks NXP.. https://jpa.kapsi.fi/stuff/pix/imx_nvcc.png   of course one of the pins that needs only capacitor is named NVCC while all other NVCC pins are power inputs; also thanks snapeda for grouping all NVCC together
2024-04-16T15:50:48 < jbo> yeah screw snapeda
2024-04-16T15:51:10 < jpa-> though i would have made exact same mistake if i had made the symbol myself..
2024-04-16T15:52:01 < qyx> wha timx
2024-04-16T15:53:17 < qyx> also, anyone familiar with http://gerrithub.io/?
2024-04-16T15:53:26 < qyx> or any other gerrit-like review for github
2024-04-16T15:54:21 < jbo> what do you mean? you're looking for a review tool?
2024-04-16T15:54:29 < jbo> phabricator works really, really well (nowdays called phorge)
2024-04-16T15:55:24 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-16T15:55:46 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Read error: Connection reset by peer]
2024-04-16T16:02:34 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-16T16:25:31 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-16T16:36:49 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-16T16:41:05 -!- NEYi [~NEYi@109.251.216.38] has quit [Read error: Connection reset by peer]
2024-04-16T16:42:11 -!- josuah [~josuah@46.23.94.12] has joined ##stm32
2024-04-16T16:54:08 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-16T17:00:20 -!- Luggi09498284764 [~lux@host-82-48-197-127.retail.telecomitalia.it] has joined ##stm32
2024-04-16T17:11:08 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-16T17:19:50 -!- rpifan [~rpifan@p200300d26731bd000f42ed7403c554d9.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T17:19:50 -!- rpifan [~rpifan@p200300d26731bd000f42ed7403c554d9.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T17:19:50 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T17:34:20 < karlp> well, filing phy driver support upstream had benefits.  they pointed to a test suite, and discovered that we'd hard strapped autoneg off.
2024-04-16T17:34:33 < karlp> doesn't _normallly_ matter, but... pretty unnecessary, 
2024-04-16T17:34:43 < karlp> also, loopback differences, who uses loopback anyway?!
2024-04-16T17:41:17 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T17:41:29 -!- rpifan [~rpifan@p200300d26731bd000373067f8e5afe18.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T17:41:29 -!- rpifan [~rpifan@p200300d26731bd000373067f8e5afe18.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T17:41:29 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T18:23:53 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 272 seconds]
2024-04-16T18:36:09 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-16T18:39:57 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T18:43:59 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 252 seconds]
2024-04-16T18:53:03 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-16T18:54:45 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-16T18:55:45 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T18:55:57 -!- rpifan [~rpifan@p200300d26731bd0092a92b1816867043.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T18:55:57 -!- rpifan [~rpifan@p200300d26731bd0092a92b1816867043.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T18:55:57 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T19:49:57 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-16T19:56:06 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 256 seconds]
2024-04-16T19:57:18 -!- scrts [~scrts2@23.28.144.38] has quit [Quit: The Lounge - https://thelounge.chat]
2024-04-16T19:57:46 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-16T19:58:40 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T20:17:06 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-16T20:18:28 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-16T20:48:03 -!- Ecco [~user@user/Ecco] has quit [Read error: Connection reset by peer]
2024-04-16T20:52:14 -!- Ecco [~user@user/Ecco] has joined ##stm32
2024-04-16T21:05:57 -!- rpifan [~rpifan@user/rpifan] has quit [Remote host closed the connection]
2024-04-16T21:06:10 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T21:44:59 -!- Posterdati [~Posterdat@user/Posterdati] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
2024-04-16T21:45:23 -!- Posterdati [~Posterdat@user/Posterdati] has joined ##stm32
2024-04-16T22:00:08 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f8b9-bd8c-4f2-ff79.fixed6.kpn.net] has joined ##stm32
2024-04-16T23:06:53 -!- rpifan [~rpifan@user/rpifan] has quit [Ping timeout: 240 seconds]
2024-04-16T23:11:48 -!- rpifan [~rpifan@p200300d26731bd0057c6bfcff46482d5.dip0.t-ipconnect.de] has joined ##stm32
2024-04-16T23:11:48 -!- rpifan [~rpifan@p200300d26731bd0057c6bfcff46482d5.dip0.t-ipconnect.de] has quit [Changing host]
2024-04-16T23:11:48 -!- rpifan [~rpifan@user/rpifan] has joined ##stm32
2024-04-16T23:44:39 -!- Livio [~livio@user/livio] has joined ##stm32
--- Day changed ke huhti 17 2024
2024-04-17T00:16:34 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 246 seconds]
2024-04-17T00:50:49 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-17T01:00:15 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T01:00:56 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T01:06:46 -!- rpifan [~rpifan@user/rpifan] has quit [Quit: Leaving]
2024-04-17T01:08:59 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f8b9-bd8c-4f2-ff79.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-17T01:09:15 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has quit [Quit: bye]
2024-04-17T01:09:31 -!- LFSveteran [~LFSvetera@keymaker.msrv.nl] has joined ##stm32
2024-04-17T01:09:51 -!- Livio [~livio@user/livio] has quit [Ping timeout: 272 seconds]
2024-04-17T01:31:28 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 246 seconds]
2024-04-17T01:45:44 -!- ventYl [~ventyl@bband-dyn78.178-40-23.t-com.sk] has quit [Ping timeout: 256 seconds]
2024-04-17T01:58:48 -!- yukam [~yukam@user/yukam] has quit [Ping timeout: 260 seconds]
2024-04-17T02:08:23 -!- yukam [~yukam@user/yukam] has joined ##stm32
2024-04-17T02:26:59 -!- ventYl [~ventyl@adsl-dyn149.78-99-59.t-com.sk] has joined ##stm32
2024-04-17T02:27:11 -!- ventYl [~ventyl@adsl-dyn149.78-99-59.t-com.sk] has quit [Client Quit]
2024-04-17T02:27:40 -!- ventYl [~ventyl@adsl-dyn149.78-99-59.t-com.sk] has joined ##stm32
2024-04-17T02:31:46 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Read error: Connection reset by peer]
2024-04-17T02:48:00 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-17T03:52:47 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-17T04:02:10 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T04:03:29 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T04:40:21 < nomorekaki> https://www.youtube.com/watch?v=27x446o8RMg you know more mecha anime like this?
2024-04-17T04:47:01 -!- Luggi09498284764 [~lux@host-82-48-197-127.retail.telecomitalia.it] has quit [Ping timeout: 256 seconds]
2024-04-17T05:30:05 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-17T06:29:07 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-17T07:05:37 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T07:06:23 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T07:32:26 -!- rkta [~rkta@user/rkta] has quit [Quit: sysupgrade]
2024-04-17T07:52:39 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-17T07:53:28 -!- rkta [~rkta@user/rkta] has joined ##stm32
2024-04-17T08:00:23 -!- scrts [~scrts2@23.28.144.38] has quit [Ping timeout: 252 seconds]
2024-04-17T08:30:00 < antto> nomorekaki, does an electric horse count?
2024-04-17T09:39:33 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-17T09:41:06 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-17T10:02:39 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-17T10:09:46 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-17T10:55:03 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-17T10:55:27 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-17T10:55:30 < Laurenceb_> kek https://nitter.poast.org/Cobratate/status/1780180832369402005#m
2024-04-17T10:57:18 < Laurenceb_> >how do I engrish
2024-04-17T10:57:55 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Client Quit]
2024-04-17T11:01:18 < specing> lol
2024-04-17T11:49:51 -!- specing [~specing@user/specing] has quit [Remote host closed the connection]
2024-04-17T11:51:19 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-17T11:59:38 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Read error: Connection reset by peer]
2024-04-17T12:00:15 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-17T12:38:52 < karlp> well, I'm not getting that time back.
2024-04-17T12:40:51 < jpa-> would you want it back?
2024-04-17T12:42:57 < karlp> if I could use it fresh? sure, why not get time refunds :)
2024-04-17T12:43:53 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T12:44:38 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T13:16:18 -!- Miyu [~hackkitte@94.31.73.186] has joined ##stm32
2024-04-17T13:16:57 -!- hackkitten [~hackkitte@94.31.99.136] has quit [Ping timeout: 255 seconds]
2024-04-17T13:27:17 < qyx> css wizards, any method to select the first h2 after a h1? (not direct sibling)
2024-04-17T13:28:44 < qyx> maybe I shouldn't ask this on a channel full of embedded spaghetti programmers
2024-04-17T13:34:20 < karlp>  ifrst and only?
2024-04-17T13:34:27 < karlp> as in, not any other h2s?
2024-04-17T13:34:41 < jpa-> h1 + h2 would work if it comes right after h1
2024-04-17T13:35:12 < benishor> h1 > h2
2024-04-17T13:35:26 < qyx> I want break-before: page; on the first h2 after a h1
2024-04-17T13:35:33 < qyx> but there is a couple of <p> between them
2024-04-17T13:35:41 < qyx> I want them to stay together with the h1
2024-04-17T13:35:50 < benishor> use :first-child
2024-04-17T13:36:31 < qyx> so parent > h2?
2024-04-17T13:36:38 < benishor> but yeah, jpa has the better solution
2024-04-17T13:36:46 < karlp> parent > h2: first?
2024-04-17T13:36:56 < benishor> parent > h2 does not specify that the h2 must be the first child
2024-04-17T13:37:11 < karlp> it's just all h2 after parent, but can't yo uadd :first to that?
2024-04-17T13:37:16 < benishor> jpa-'s solution works there
2024-04-17T13:37:26 < jpa-> h1 + h2, h1 + :not(h2) + h2, h1 + :not(h2) + :not(h2) + h2 {     is the best i can come up with if there is other elements in between
2024-04-17T13:37:28 < qyx> benishor: jpa's doesn'ŧ work, it is not the first descendant
2024-04-17T13:37:37 < benishor> then try mine
2024-04-17T13:37:44 < benishor> h1 > h2 
2024-04-17T13:37:52 < karlp> that will take all h2 after h1 though?
2024-04-17T13:38:10 < benishor> yes, also needs the first
2024-04-17T13:38:14 < qyx> no
2024-04-17T13:38:21 < qyx> I want "first h2 after h1"
2024-04-17T13:38:29 < qyx> not "first element after h1 to be h2"
2024-04-17T13:38:34 < benishor> or 
2024-04-17T13:38:44 < benishor> h1 > h2:first-of-type
2024-04-17T13:38:48 < benishor> ah
2024-04-17T13:38:55 < benishor> try that one
2024-04-17T13:39:39 < jpa-> like   http://jpa.kapsi.fi/stuff/other/h1h2.html
2024-04-17T13:39:45 < benishor> it should target first h2 as a direct descendent of h1
2024-04-17T13:41:16 < qyx> jpa wins
2024-04-17T13:42:37 < benishor> well done jpa. I brainfarted. my take relied on h2 being a descendent of h1, which is silly
2024-04-17T13:43:20 < qyx> except it doesn't work in sphinx
2024-04-17T13:43:38 < jpa-> make sure you have enough :not(h2) copies :)
2024-04-17T13:44:16 < qyx> yeah I am expecting to be surprised in the future
2024-04-17T13:47:31 < jpa-> it seems that sphinx puts things inside spans, which would make it much easier
2024-04-17T13:47:42 < qyx> and <section>s
2024-04-17T13:47:46 < jpa-> yeah
2024-04-17T13:47:59 < jpa-> so just section h2:first-child  or something
2024-04-17T13:48:41 < qyx> yeah that works
2024-04-17T13:51:52 < qyx> ok not that easy, that matches ever h2 section, not just the first
2024-04-17T13:52:18 < jpa-> ah, i meant :first-of-type
2024-04-17T13:52:22 < jpa-> hmm
2024-04-17T13:52:38 < jpa-> but if there is intermediate elements between h1 and h2, you must use the :first-of-type on that
2024-04-17T13:53:29 < karlp> yeah, I was just fucking around and also went to "descendent of h1" and realized that was all wrong :)
2024-04-17T13:53:46 < jpa-> maybe   section section:first-of-type h2   or something
2024-04-17T13:54:58 < karlp> this works too? https://jsfiddle.net/gj7th1xk/
2024-04-17T13:57:12 < jpa-> karlp: only if you have only one h1 per container
2024-04-17T13:57:46 < jpa-> but sphinx seems to put   <section> <h1>foo</h1> .... <section> <h2>bar</h2> </section> </section>  which breaks h2:first-of-type
2024-04-17T13:58:42 < karlp> qyx never said it should only be first h2 after first h1 :) just first h2 after h1 :)
2024-04-17T13:58:54 < qyx> first h2 after h1
2024-04-17T13:59:00 < qyx> the latter is correect
2024-04-17T13:59:12 < jpa-> karlp: but yours does the first one
2024-04-17T13:59:31 < qyx> computers are hard
2024-04-17T13:59:52 < jpa-> karlp: https://jsfiddle.net/p8uLn6e7/
2024-04-17T13:59:53 < karlp> oh, I thought mine would do all, and you thought it should onyl do first.
2024-04-17T13:59:55 < karlp> lame...
2024-04-17T14:00:06 < karlp> yeah, i just did the same thing to try it out
2024-04-17T14:00:28 < jpa-> but it's all moot, with sphinx doing different nesting of elements
2024-04-17T14:00:28 < karlp> fuck front end.
2024-04-17T14:00:35 < karlp> back to cmake....
2024-04-17T14:01:27 < jpa-> it's also kinda fun that in a channel full of embedded devs, pretty much everyone still needs to know html and css and bit of javascript
2024-04-17T14:02:03 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-17T14:02:13 < benishor> https://xkcd.com/349/
2024-04-17T14:02:32 < karlp> now, why have I always implemented _write() and why does this other project implement _write_r? those end up in the same place right? yuou just have the "struct _reent *ptr," extra param?
2024-04-17T14:05:03 < karlp> enwlib docs only mention write and _write_r, no mentiong of _write...
2024-04-17T14:05:27 < jpa-> avoids one wrapper if you do _write_r
2024-04-17T14:13:48 < qyx> h1 + section seems to work
2024-04-17T14:14:07 < qyx> because sphinx places a <section> enclosing h2's directly after h1
2024-04-17T14:15:01 < qyx> ok no
2024-04-17T14:15:02 < qyx> fukit
2024-04-17T14:15:36 < jpa-> what about the   section section:first-of-type h2
2024-04-17T14:19:09 < qyx> matches all h2 for some reason
2024-04-17T14:20:34 < jpa-> funny, does it put multiple h2 in same sub-section?
2024-04-17T14:20:53 < jpa-> then maybe    section section:first-of-type h2:first-of-type   would work :)
2024-04-17T14:21:04 < karlp> fucking newlib still uses this fucking caddr_t type in their docs. thanks arseholes
2024-04-17T14:21:38 < zyp> karlp, the _r stands for reentrant, IIRC there's an option when newlib is built to decide whether you want a threadsafe newlib or not
2024-04-17T14:23:07 < zyp> https://sourceware.org/newlib/libc.html#Reentrant-Syscalls
2024-04-17T14:23:18 < zyp> better reference: https://sourceware.org/newlib/libc.html#Reentrancy
2024-04-17T14:24:25 < karlp> yeah, I was more curious what the order was in, you can define "either" of them in your own code.
2024-04-17T14:25:09 < zyp> I think when you've got both, foo is typically a shim over _foo_r(_impure_ptr)
2024-04-17T14:25:39 < karlp> ok. I've got cmake, m0+, using floating point, with syscalls.  now to work out why work doesn't work unless I link with this hardcoded mcuxpresso libc.a....
2024-04-17T14:25:56 < zyp> work doesn't work?
2024-04-17T14:26:02 < karlp> work does not work.
2024-04-17T14:26:17 < zyp> what's work?
2024-04-17T14:26:48 < karlp> work does this, which I have severe reservations about, but does "work" https://paste.jvnv.net/view/ASm93
2024-04-17T14:27:37 < karlp> when I remove those, and try and use -lc -lm sorty of normal shit, it fails to link all the floating point. 
2024-04-17T14:28:15 < karlp> I'll figure it out, just lots of cmake layers and macros and functions and the various different options available, with nostartfiles vs freestanding and nocommon and blah blah lots of ways to end up in much the same place....
2024-04-17T14:28:23 < zyp> no thanks
2024-04-17T14:28:28 < zyp> enjoy
2024-04-17T14:29:05 < karlp> I am endeavouring to look on the bright side :)
2024-04-17T14:29:07 < zyp> speaking of build systems, I encountered a custom ./configure with a bunch of weird makefile oddities yesterday
2024-04-17T14:29:51 < karlp> this weird mess is only for this particular cm0+ board, the multiple other cm4f boards are all a more "normal" https://paste.jvnv.net/view/tTuZO
2024-04-17T14:30:09 < karlp> so just trying to unravel it.
2024-04-17T14:30:50 < karlp> been bothering me for months, decided it was time.
2024-04-17T14:31:23 < zyp> I'm working on stuffing some stuff into a set of docker containers and putting a nice little pyqt gui in front of it to spin it all up, interact with it and clean it up afterwards, and getting one of the tool packages to build was fairly annoying
2024-04-17T14:32:11 < zyp> it was calling /sbin/libtool which somehow failed, so I ended up just patching the whole /sbin/libtool in the build container to be a no-op before running the build
2024-04-17T14:32:20 < karlp> libtool is always a sign of a bad day.
2024-04-17T14:32:36 < karlp> autohell 
2024-04-17T14:32:42 < karlp> worse that autohell.
2024-04-17T14:33:16 < zyp> no autohell involved, just homebrewed stuff that tried to pretend to be autohell and doing an even worse job
2024-04-17T14:33:38 < karlp> well, pretending ot be autohell sounds like autohell still :)
2024-04-17T14:35:11 < zyp> the makefiles had no sane control of permissions either, so to build something that would dynamically link against a library in /usr/lib/, it had to have write access to /usr/lib/
2024-04-17T14:35:36 < zyp> luckily I'm building it in a throwaway container, so it can shit all over the place for all I care
2024-04-17T14:35:52 < karlp> heh.
2024-04-17T14:36:24 < karlp> I'm watching a train wreck building up steam on another project, they're using alllll sorts of mixed case stuff, and they're currently all on windows....
2024-04-17T14:36:47 < zyp> it also had a makefile rule to collect all the build outputs in a folder for zipping up a binary release, but that didn't copy the library symlinks, so I didn't get the name it actually linked against
2024-04-17T14:37:03 < karlp> classic....
2024-04-17T14:37:31 < karlp> oh, I found out my ethernet phy isn't 802.3 compliant for loopback yesterday, that was cool.
2024-04-17T14:37:47 < karlp> it needs the 802.3 loopback bit set, but -also_ needs another bit set in another register...
2024-04-17T14:37:56 < karlp> I mean, who needs loopback anyway amirite
2024-04-17T14:38:21 < ventYl> isn't libtool like... 25 years obsolete?
2024-04-17T14:38:30 < zyp> but yeah, I managed to beat that into submission, now I just need to ensure I get all the container environments correct, add stuff to get files into and out of containers, and some stuff to select/load specific container images from files
2024-04-17T14:38:54 < ventYl> as I understand it, the whole purpose of libtool was to hack something into a.out format before elf became widespread
2024-04-17T14:41:40 < karlp> libtool is as obsolete as autohell, yes.
2024-04-17T14:41:47 < karlp> well, worse.
2024-04-17T14:41:52 < zyp> this is some silly enterprise level setup where my python gui lets the user pick files which gets fed to a typescript thing running in a container, the typescript uploads the file as a job in some sort of azure queue (azurite running in a different container)
2024-04-17T14:42:48 < ventYl> autohell is bad, yet at least it had a purpose. libtool is just a dirty hack
2024-04-17T14:43:08 < zyp> a C# process in another container pulls the job from the queue, and starts a data analysis routine in it (which is in C, I wrote the C/C# bindings some time last year) and pushes the result back to another queue
2024-04-17T14:43:42 < zyp> finally another typescript thing gets the result and creates a report from it, which my gui tool needs to retrieve and present to the user
2024-04-17T14:44:38 < ventYl> please stop
2024-04-17T14:44:53 < karlp> enterprising baby.
2024-04-17T14:46:12 < zyp> AIUI this thing is designed to handle a ton of data running on the azure cloud with a bunch of worker nodes, but people wanted a way to run it standalone on a laptop, so somebody decided the trick is to stick it all into docker and put a GUI over it
2024-04-17T14:46:33 < benishor> microservices++
2024-04-17T14:46:52 < benishor> next you wanna go to a docker swarm
2024-04-17T14:47:00 < benishor> and scale your services accordingly
2024-04-17T14:47:28 < ventYl> so it was designed ass-forward. first make it run in cloud and then uh-oh, we might probably want to run it locally. sheeeeeit
2024-04-17T14:48:33 < zyp> I kinda feel like hooking up a GUI directly to the actual C analysis code and cutting out all the middle layers would be less work than getting this whole stack to run reliably, but I'm not gonna argue that point
2024-04-17T14:49:14 < zyp> that'd require me to understand what the middle layers actually do, and I don't wanna get into that
2024-04-17T14:50:08 < zyp> (and the report generation stuff might still be doing stuff that's annoying to run without sticking it into a container)
2024-04-17T14:52:46 < benishor> it depends on what needs to scale and how
2024-04-17T14:53:17 < benishor> and what the change sources are
2024-04-17T14:54:24 < karlp> hrm, using my own error syscalls is both a) smaller, and b) doens't genertate "warning: _write is not implemented and will always fail" type warnings than using --specs=nosys.specs.
2024-04-17T14:54:35 < karlp> but, had to write stub syscalls.
2024-04-17T14:54:41 < karlp> anywya, that's a distraction...
2024-04-17T14:57:23 < \dev\ice> what does RTC_TS actually does? saves current rtc time to some registers on rising edge?
2024-04-17T14:57:39 < karlp> yeah timestamp
2024-04-17T14:57:45 < karlp> whne something happened...
2024-04-17T15:20:56 -!- Livio [~livio@user/livio] has quit [Ping timeout: 252 seconds]
2024-04-17T15:21:44 < \dev\ice> how it can be used for PPS?
2024-04-17T15:23:15 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-17T15:31:30 -!- Miyu is now known as hackkitten
2024-04-17T15:35:07 -!- Livio [~livio@user/livio] has quit [Ping timeout: 255 seconds]
2024-04-17T15:53:38 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 268 seconds]
2024-04-17T16:27:25 < karlp> if you mean pulse per ssecond output, it isn't? iiuc
2024-04-17T16:28:07 < karlp> "Timestamp feature which can be used to save the calendar content. This function can
2024-04-17T16:28:09 < karlp> be triggered by an event on the timestamp pin, or by a tamper event, or by a switch to
2024-04-17T16:28:11 < karlp> VBAT mode.
2024-04-17T16:28:31 < karlp> if you want to count the time that you got a PPS event from externally, to comapre clocks, tha tcould work?
2024-04-17T16:28:47 < karlp> if you want to use an external PPS as the _clock_ foryour rtc, that's not TS stuff.
2024-04-17T16:30:58 < karlp> I don't actually even think you can, refin is for 50&60hz stuff, 
2024-04-17T17:01:57 < qyx> for 1/32768 PPS resolution you can use any EXTI and capture the subsecond counter manually
2024-04-17T17:02:56 < qyx> if you want anything serious, use a multi-MHz free running counter and input capture
2024-04-17T17:03:24 < \dev\ice> karlp: yup. gps pulse per secondf
2024-04-17T17:03:59 < qyx> and if you want to use the PPS signal as an input clock for RTC, don't
2024-04-17T17:04:31 < \dev\ice> nope. main idea is to have it for synchronizing
2024-04-17T17:04:59 < \dev\ice> and I got board with PPS conected to RTC_TS pin :-)
2024-04-17T17:05:25 < \dev\ice> guy designed the board trying me to convince - this is the way to go :-)
2024-04-17T17:05:29 < qyx> so timestamp it and adjust until you get consistent zero
2024-04-17T17:06:09 < qyx> rtc can be adjusted either by trimming cap on LSE inputs
2024-04-17T17:06:25 < qyx> or by software, it has functionality for that
2024-04-17T17:06:26 < \dev\ice> but can I get interrupt during PPS, not just snapshot of RTC registers? 
2024-04-17T17:06:35 < \dev\ice> It's LSE on board
2024-04-17T17:06:51 < qyx> I guess you can get interrupt after snapshot?
2024-04-17T17:06:59 < qyx> haven't used so idk
2024-04-17T17:10:13 < karlp> if you tie pps to rtc ts, you can get an irq on it, and it will have saved the timestamp for you. 
2024-04-17T17:10:38 < qyx> yes justchecked on g4, you can enable timestamp interrupt
2024-04-17T17:10:41 < karlp> what you do with it is then your problem.
2024-04-17T17:10:56 < qyx> and there is also a subsecond timestamp register
2024-04-17T17:11:35 < karlp> on parts newer than, like, f0 or something, yeah
2024-04-17T17:11:38 < qyx> in your interrupt just check if your timestamped subsecond is 0 and adjust the reload calib value to get 0
2024-04-17T17:11:47 < karlp> and you can have as much fun as yu like calibrating the rtc 
2024-04-17T17:12:44 < qyx> adjusting normal timers is much more fun
2024-04-17T17:12:55 < qyx> because you need to manipulate ARR
2024-04-17T17:13:04 < qyx> or get vctcxo
2024-04-17T17:13:16 < \dev\ice> ok, thanks guys
2024-04-17T17:13:31 < \dev\ice> will read RM
2024-04-17T17:14:06 < qyx> gxti was a timing pro here
2024-04-17T17:14:14 < qyx> idk where is he now
2024-04-17T17:21:38 < Ecco> Dumb question: I have this PCB that I'd like to power and debug from my computer. My debug probe uses USB but doesn't provide power. My PCB could use USB vbus to power itself. What would be the easiest way to connect this?
2024-04-17T17:23:04 < zyp> is there a USB connector on the PCB?
2024-04-17T17:23:38 < Ecco> I could add one
2024-04-17T17:24:01 < Ecco> Yeah, sorry: my question should have started with "I'm *designing* this PCB that..."
2024-04-17T17:24:24 < Ecco> The easy way is to have two usb cables from the PC: one to the debug probe and the other to the PCB
2024-04-17T17:24:26 < zyp> if there are, the easy answer is «plug a second USB cable between your computer and the PCB»
2024-04-17T17:24:27 < Ecco> but that's two cables
2024-04-17T17:24:38 < zyp> yes
2024-04-17T17:24:43 < Ecco> Yes, that works, but it's kind of annoying
2024-04-17T17:25:04 < zyp> idk, I often work like that
2024-04-17T17:25:17 < Ecco> There's a VCC pin on my debug probe
2024-04-17T17:25:23 < Ecco> but I really think it only does sensing
2024-04-17T17:25:47 < Ecco> so I don't think the debug probe can power the PCB
2024-04-17T17:25:56 < zyp> you mean the Vtref on pin 1 on the 10-pin connector?
2024-04-17T17:25:57 < Ecco> that's kind of lame
2024-04-17T17:26:28 < Ecco> I guess. It's labeled "VCC" on the probe
2024-04-17T17:27:36 < zyp> yeah, the reason it's also called vtref is because it stands for «target reference», and its main purpose is for the target PCB to give a reference of its IO voltage to the probe so the probe can also run its IO at the same voltage
2024-04-17T17:28:15 < zyp> it's slightly different and more important than just sensing, but in any case it's intended as an input
2024-04-17T17:28:39 < Ecco> Oh, makes sense
2024-04-17T17:29:26 < zyp> now, some probes can *also* use it as an output, which can be convenient if the thing you're working on can handle being backfed that way
2024-04-17T17:29:57 < zyp> I sometimes use it
2024-04-17T17:30:30 < qyx> I usually put 1k in series on pin 10
2024-04-17T17:30:36 < qyx> to make people angry
2024-04-17T17:30:47 < zyp> the reset pin?
2024-04-17T17:30:58 < qyx> *pin 1
2024-04-17T17:31:14 < zyp> you're only hurting yourself :)
2024-04-17T17:31:33 < qyx> why
2024-04-17T17:31:53 < zyp> because you're the one having to develop on the boards you design
2024-04-17T17:32:44 < qyx> yeah and then a random developer uses a probe back-feeding 3v3 on my precious 1v8 device
2024-04-17T17:33:35 < zyp> it's a 3.3v device if you feed 3.3v into it
2024-04-17T17:34:04 < Ecco> :-)
2024-04-17T17:36:57 < zyp> when I'm handing a project over to a customer and have to teach them how to flash their boards, my favorite solution is to just give them a standalone flasher with the instructions «plug in cable here, wait for LED to turn green, unplug cable»
2024-04-17T17:37:41 < zyp> we've got a board that'll feed power through vtref, detect a connected target and run a preloaded flashing sequence
2024-04-17T17:37:49 < zyp> very idiot proof
2024-04-17T17:38:08 < qyx> did you mean in ""
2024-04-17T17:38:46 < qyx> also, segger flasher is such device with a big white button on it, nothing more
2024-04-17T17:38:55 < qyx> push to flash
2024-04-17T17:39:36 < zyp> I very much prefer that over some «so go download this windows software for your debugger probe, connect the board like so, run this .bat script, pray all goes well»
2024-04-17T17:39:39 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-17T17:40:40 < zyp> and yeah, the thing we've got is similar in function to that, except what we've got is much cheaper
2024-04-17T17:41:05 < qyx> I would put a rpi + stlink in a box
2024-04-17T17:41:48 < zyp> it's just a custom firmware on a common prototyping/production test/debug probe board that my company designed before I got started
2024-04-17T17:42:20 < qyx> reminds me of my oob-debug board I once did and never finished of course
2024-04-17T17:42:27 < qyx> (out-of-band)
2024-04-17T17:42:27 -!- rkta_ [~rkta@user/rkta] has joined ##stm32
2024-04-17T17:42:46 < qyx> it had a 2G modem, F7 running uclinux, openocd
2024-04-17T17:42:50 < zyp> some of my coworkers use them as debug probes and they've been used for a few production test setups before
2024-04-17T17:43:07 -!- rkta_ [~rkta@user/rkta] has quit [Client Quit]
2024-04-17T17:43:11 < Ecco> F7 running uclinux -> how well does that run?
2024-04-17T17:43:19 < Ecco> Also, didn't uclinux make it to mainline?
2024-04-17T17:43:20 < zyp> the more recent production test setup I've been working on lately is based on a derivate of that that's more streamlined for production testing
2024-04-17T17:43:28 < Ecco> (as in mainline kernel has a no-mmu option)
2024-04-17T17:43:36 < qyx> not very well because uclinux userspace without mmu is shitty
2024-04-17T17:43:41 < zyp> but yeah
2024-04-17T17:43:46 < Ecco> how would the userspace know?
2024-04-17T17:43:58 < qyx> easily, eg. fork doesn't work as intended
2024-04-17T17:44:09 < Ecco> Actually that's a good question: I knew it exists, but I never questionned the implications
2024-04-17T17:44:17 < Ecco> I just assumed the kernel would do "some magic"
2024-04-17T17:44:36 < Ecco> how would fork behave differently?
2024-04-17T17:44:37 < zyp> I'm intending to make a lcd+navpad+sdcard addon for orbtrace at some point and run micropython on it
2024-04-17T17:44:43 < qyx> my main motivation was to be able to put F7 into standby to get near zero consumption
2024-04-17T17:45:01 < zyp> can put python scripts on the sd card to do stuff like automatic flashing sequences
2024-04-17T17:45:05 < zyp> or unattended logging
2024-04-17T17:45:34 < qyx> oh yeah I want to do the same with my platform, except with imx6
2024-04-17T17:46:08 < qyx> no more mmu-less linuces
2024-04-17T17:46:17 < zyp> there's been so many times where something is going wrong in the field and it'd be really convenient if I could just send a prescripted orbtrace to be deployed and later get it back full of logs
2024-04-17T17:46:34 < qyx> same intention here
2024-04-17T17:46:54 < qyx> except nowadays I would add a nbiot or cat1.bis modem on it to interact with it remotely
2024-04-17T17:47:11 < zyp> true
2024-04-17T17:47:24 < qyx> but I am investing quite a lot of time doing in-band debug properly
2024-04-17T17:47:56 < qyx> so the device itself can catch most fails, log them and send the outputs over the commonly used comms iface
2024-04-17T17:51:09 < zyp> my docker stuff is starting to work pretty well
2024-04-17T17:52:00 < zyp> it was failing before because one of the typescript things was calling a bunch of shell commands and the busybox sed in the container didn't behave exactly like regular sed
2024-04-17T17:54:01 -!- rkta [~rkta@user/rkta] has quit [Quit: brb]
2024-04-17T17:54:15 -!- rkta [~rkta@user/rkta] has joined ##stm32
2024-04-17T18:21:48 < karlp> great, 40k saving switching to nanospecs and the toolchain libc, (or a 30k increase, using non-nano) but.... it doesn't work anymore :)
2024-04-17T18:21:57 < karlp> I wonder what UB I've hit...
2024-04-17T18:28:18 < qyx> I want such payjob too
2024-04-17T18:28:22 < qyx> breaking things daily
2024-04-17T18:33:43 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-17T18:40:33 < karlp> well, ti's all latent bugs I'm just uncovering...
2024-04-17T18:40:36 < karlp> or, so I believe.
2024-04-17T18:44:05 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 240 seconds]
2024-04-17T19:01:24 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T19:02:50 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T19:46:53 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-8082-3d02-ff39-3225.fixed6.kpn.net] has joined ##stm32
2024-04-17T20:08:15 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Ping timeout (120 seconds)]
2024-04-17T20:44:46 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-17T20:52:36 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T21:05:48 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-17T21:06:47 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-17T22:29:13 < karlp> (not UB, just plain old heap/stack collision, sbrk implementation changed...)
2024-04-17T22:46:10 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-17T22:49:56 -!- qyx [~qyx@84.245.120.167] has quit [Ping timeout: 260 seconds]
2024-04-17T22:51:33 -!- qyx [~qyx@84.245.120.249] has joined ##stm32
2024-04-17T23:03:27 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-17T23:30:45 -!- jbo [~jbo@user/tct] has quit [Read error: Connection reset by peer]
2024-04-17T23:31:08 -!- jbo [~jbo@user/tct] has joined ##stm32
2024-04-17T23:32:12 < qyx> regarding heap/stack collision, is there any reason why they are not allocated the other way around? (that is, stack on the bottom, heap the rest)
2024-04-17T23:32:38 < qyx> at least for a single stack it would be possible to catch stack overflows
2024-04-17T23:33:41 < zyp> the main reason is that it'd require you to have an active relationship to how much stack vs how much heap you need
2024-04-17T23:34:02 < qyx> you should have that either way
2024-04-17T23:34:05 < zyp> whereas if you let them grow towards each other you don't need to care before it blows up
2024-04-17T23:34:33 < zyp> idk, I tend to not care
2024-04-17T23:35:20 < zyp> running out of memory happens so rarely for me that it's generally not worth worrying about
2024-04-17T23:38:26 < zyp> hmm, I got the h7 ethernet configured, but DMA refuses to start, I suspect I've pointed it at a memory area it can't reach
2024-04-17T23:39:47 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 252 seconds]
2024-04-17T23:40:24 < zyp> yeah, I'm currently putting everything in DTCM and while HPDMA can reach that, the other DMA engines can't
2024-04-17T23:43:34 < zyp> I should probably just dedicate the AHBSRAM blocks to DMA buffers
2024-04-17T23:48:52 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-17T23:50:57 < qyx> have you ever seen these boxes https://bin.jvnv.net/file/hzPjo
2024-04-17T23:54:26 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
--- Day changed to huhti 18 2024
2024-04-18T00:11:58 < zyp> heh, I managed to crash smolt
2024-04-18T00:12:15 < zyp> somehow I've managed to make a tag that the parser doesn't handle
2024-04-18T00:12:54 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-18T00:27:01 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-18T00:36:28 < zyp> it's fucking up a substitution, substituting the wrong thing
2024-04-18T00:51:46 < jbo> zyp
2024-04-18T00:52:05 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-18T01:00:15 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T01:00:52 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-8082-3d02-ff39-3225.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-18T01:01:13 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T01:22:27 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-18T01:31:18 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-18T01:51:35 < zyp> ok, took me a while to work out the substitution issue, turned out that for untemplated types it was learning it twice, causing an offset in the substitution table
2024-04-18T01:52:57 < zyp> templated types are learned twice, once without template args and once with
2024-04-18T01:54:27 < zyp> untemplated types don't have template args, but ended up at the same place in the substitution transformer
2024-04-18T01:56:15 < zyp> ethernet is also receiving stuff: https://paste.jvnv.net/view/GqCza
2024-04-18T02:04:26 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-18T02:05:23 -!- Livio [~livio@user/livio] has quit [Ping timeout: 264 seconds]
2024-04-18T02:08:35 < zyp> with packet contents too: https://paste.jvnv.net/view/ptWLL
2024-04-18T02:09:40 < zyp> https://paste.jvnv.net/view/hc6MF <- I love this logging code
2024-04-18T02:26:22 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-18T02:27:15 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T02:54:23 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-18T03:05:09 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T03:21:49 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T03:34:23 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-18T04:01:43 < Ecco> Found this cool resource: https://www.quinapalus.com/cm7cycles.html
2024-04-18T04:02:25 < Ecco> (it's a summary of the cycles taken per instruction on M7)
2024-04-18T04:02:36 < Ecco> I wonder if it's very different on M3 and co
2024-04-18T05:01:46 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T05:02:41 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T06:01:16 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T06:49:02 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-18T07:15:21 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-18T07:54:56 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-18T08:24:12 < jpa-> Ecco: https://developer.arm.com/documentation/ddi0337/h/programmers-model/instruction-set-summary/cortex-m3-instructions   has for cortex-m3
2024-04-18T08:25:18 < jpa-> and similarly ARM has for cortex-m0 etc; for M7 they don't publish timings, presumably because it is hard to specify them for an out-of-order CPU; but the page you linked seems to be a reasonable estimate
2024-04-18T09:08:56 < ds2> publishing it seems to be the exception... found a post before where they have said they do not publish it for the A cores at all
2024-04-18T09:11:25 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f562-efef-7719-5bd8.fixed6.kpn.net] has joined ##stm32
2024-04-18T09:23:41 < jpa-> ds2: i think most OOO CPU manufacturers don't attempt to publish timings, because the actual behavior is too complex
2024-04-18T09:25:19 < jpa-> intel used to do so back in 2010 at least, but the latest optimization reference manual no longer has any timings
2024-04-18T09:25:48 < jpa-> cortex-m0-4 are simple because the pipeline is quite straightforward
2024-04-18T09:26:04 < zyp> my impression is that the timings themselves aren't all that interesting because they're unlikely to accurately add up, and if you're optimizing a particular piece of code it's more useful to benchmark it specificly
2024-04-18T09:27:13 < zyp> I mean, you're not gonna be writing fixed delay loops taking an exact time, what you care about is how little time you can get the code to run at as a whole
2024-04-18T09:27:26 < qyx> cough-cough
2024-04-18T09:28:11 < jpa-> i have mostly found use in the floating point timing
2024-04-18T09:28:23 < jpa-> e.g. at which point it makes sense to do reciprocals instead of division
2024-04-18T09:28:29 < qyx> once I found generating an ultrasonic excitation wave with nops in one codepase and when I pointed that out I was called being arrogant and ellitist
2024-04-18T09:28:46 < qyx> *codebase
2024-04-18T09:29:06 < jpa-> i've written nop delays every now and then, but just by experimentation
2024-04-18T09:29:55 < jpa-> https://github.com/ZuluSCSI/ZuluSCSI-firmware/blob/main/lib/ZuluSCSI_platform_GD32F205/scsi_accel_sync.cpp#L176   though i've found ldr is more predictable than nop
2024-04-18T09:33:29 < qyx> rabbit hole computing sounds like a good software development workflow
2024-04-18T09:34:49 < qyx> have we discussed peripheral/core wakelocks yet?
2024-04-18T09:35:32 < qyx> I am slowly coming to a conclusion that in a more complex codebase it is hard to determine which parts of the system want to sleep and when
2024-04-18T09:35:44 < qyx> it is much easier to determine when they want to stay awake for various reasons
2024-04-18T09:35:59 < qyx> eg. ADC is still converting, comms is still running until TC flag is set, etc.
2024-04-18T09:37:06 < qyx> and when no part is requesting that the core/periph stays awake, it may safely go to some of the low power modes until there is a request to come back
2024-04-18T09:38:21 < jpa-> yeah, that's how i usually do it
2024-04-18T09:38:51 < qyx> I would abstract that to FULL, SLEEP+CLK, SLEEP, STANDBY, OFF
2024-04-18T09:39:07 < qyx> SLEEP+CLK being the core sleep, peripherals are still clocked
2024-04-18T09:39:20 < qyx> for bonus points I could turn off busses too but who cares
2024-04-18T09:41:07 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-18T09:42:55 < jpa-> hmm, whether to recommend gdb-multiarch or deadsnakes python3.8 + arm-none-eabi-gdb
2024-04-18T09:43:36 < jpa-> i have had gdb-multiarch crash a couple of times when disassembling / instruction stepping on specific instruction sequences
2024-04-18T09:43:51 < qyx> even in gcc13?
2024-04-18T09:44:47 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 252 seconds]
2024-04-18T09:45:45 < jpa-> gcc12, but not sure if it matters; is there some known problem with gcc12 and gdb-multiarch?
2024-04-18T09:46:02 < qyx> no but with python 3.8 is
2024-04-18T09:46:23 < jpa-> ah, yeah, even newest arm-none-eabi-gdb links against python3.8
2024-04-18T09:46:44 < qyx> arm-gnu-toolchain-13.2.Rel1-aarch64-arm-none-eabi
2024-04-18T09:46:49 < qyx> libpython3.8.so.1.0: cannot open shared object file: No such file or directory
2024-04-18T09:46:51 < qyx> meh
2024-04-18T09:54:00 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-18T10:08:39 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T10:13:33 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-18T10:50:14 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-18T10:55:14 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T11:00:57 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T11:57:02 -!- drkow [~k\o\w@2607:fea8:1d00:89f0:4662:11c8:f36e:f0ce] has quit [Read error: Connection reset by peer]
2024-04-18T12:06:10 -!- boB_K7IQ [~boB_K7IQ@174-26-242-222.phnx.qwest.net] has quit [Read error: Connection reset by peer]
2024-04-18T12:06:39 < karlp> cute, now as well as having cameras staring at us all day, we're going to have other people beaming at our faces all day too: https://www.eenewseurope.com/en/ultrasound-eye-gaze-and-expression-detection-in-smart-glasses/
2024-04-18T12:08:36 -!- boB_K7IQ [~boB_K7IQ@174-26-251-232.phnx.qwest.net] has joined ##stm32
2024-04-18T12:08:45 < karlp> zyp: you have to do the loc() call in every call? I thought you had that magicked?
2024-04-18T12:11:02 < karlp> meh, quinaplus sit eis blocked on corpo.
2024-04-18T12:11:07  * karlp reports an "incorrect block"
2024-04-18T12:13:11 -!- boB_K7IQ [~boB_K7IQ@174-26-251-232.phnx.qwest.net] has quit [Ping timeout: 264 seconds]
2024-04-18T12:13:25 < karlp> jpa-: down on line 213 of that zulu file though, you use ASM_DELAY1, which is nops not ldr?
2024-04-18T12:13:30 -!- boB_K7IQ [~boB_K7IQ@174-26-246-35.phnx.qwest.net] has joined ##stm32
2024-04-18T12:14:18 < karlp> oh god no, you keep redefining what asm_delay1 and 2 are...
2024-04-18T12:14:44 < karlp> I use gdb multiarch from fedora for everything except esp32.
2024-04-18T12:14:49 < karlp> only the xtensa one works there.
2024-04-18T12:15:37 < karlp> qyx: that wakelock rathe than sleep thinking is how all of the efm32 stuff works.
2024-04-18T12:19:25 < qyx> good, so it isn't that stupid idea
2024-04-18T12:25:03 < qyx> https://github.com/hrshygoodness/EFM32-Library/blob/master/v2/an/an0007_efm32_energymodes/em4_wakeup_stk.c#L109
2024-04-18T12:25:14 < qyx> wut, do they use the same convention as stdperiph
2024-04-18T12:30:32 < jpa-> karlp: isn't it beautiful? and yeah, i used nop when the ldr seemed too slow :)
2024-04-18T12:31:49 < jpa-> (ldr is 2 cycles and nop is 1 cycle, at least most of the time; it seems sometimes nop is 0 cycles)
2024-04-18T12:49:10 < karlp> jpa-: speaking of reciprocals vs divs, does gcc not handle that mostly for you? how often were you running into that?
2024-04-18T12:49:56 < karlp> I've sort of generally assumed that I can write blah / 42.69f and it will do blah * 1 / 42.69f for me, but perhaps that was naiive.
2024-04-18T12:50:00 < karlp> I guess godbolt....
2024-04-18T12:50:55 < qyx> TIL reciprocals
2024-04-18T12:50:57 < qyx> https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html#index-funsafe-math-optimizations
2024-04-18T12:51:00 < qyx> lol
2024-04-18T12:52:01 < karlp> (at least quickly, no, it doesn't automatically do recipricals)
2024-04-18T12:52:26 < karlp> ah right, -freciprical-math
2024-04-18T12:52:32 < karlp> hadn't actually seen that one, 
2024-04-18T12:52:47 < karlp> i guess of course recipricals aren't always safe for the full edges of +-inf and nan shit.
2024-04-18T12:54:19 < jpa-> yeah, and sometimes you might even need to adjust the expression to be able to use reciprocals
2024-04-18T12:55:18 < karlp> yeah, I'm not sure turning on reciprocal-math project wide is going to be popular anyway.
2024-04-18T12:55:25 < karlp> ok, good to remember, do them by hand...
2024-04-18T12:55:32 < karlp> https://godbolt.org/z/K45hj1zbG works though.... 
2024-04-18T12:55:45 < jpa-> yeah, constants are easy
2024-04-18T12:56:52 < karlp> sure, but I had assumed constnats already worked, but they don't, unless you turn on -freciprocal-math.
2024-04-18T12:57:46 < jpa-> ah, yeah
2024-04-18T12:57:53 < jpa-> i would have assumed they might have worked
2024-04-18T12:57:57 < karlp> anyway, back to my stack/heap.
2024-04-18T12:58:04 < jpa-> but rounding errors
2024-04-18T12:58:06 < karlp> one day I'll get back to the work work...
2024-04-18T12:58:13 < karlp> rounding schmounding.
2024-04-18T12:58:22 < karlp> 32bit adcs to make a 13bit result...
2024-04-18T13:02:15 < karlp> that cortx-m7 page is interesting though, (wayback is not blocked...)  vmul is 1 cycle, and vadd is 1 cycle, but vmla is 11?  so you can do mul+acc interleaved manually faster than you can do a  single using the instructions?
2024-04-18T13:02:32 < karlp> oh no fuck
2024-04-18T13:02:39 < karlp> I was looking at single and double wrong.
2024-04-18T13:03:03 < karlp> 1,1,3, not 1,1,11.
2024-04-18T13:04:59 < jpa-> vadd with dependency is 3
2024-04-18T13:06:03 < jpa-> but yeah, if you can interleave muls and adds so that there is independent instructions in between, it is a whopping 1 cycle faster than vmla :)
2024-04-18T13:08:54 < karlp> I've generally still been "lucky" that there's been enough low(er) hanging fruit that I've not needed to resort to that sort of thing...
2024-04-18T13:09:35 < jpa-> yeah, the reciprocal thing has been the only real cycle-counting thing that has had real benefit
2024-04-18T13:10:00 < jpa-> i sometimes do use the DSP instructions manually but they are mostly single cycle like most other instructions
2024-04-18T13:11:28 < karlp> right, stack/heap collision is codered being fucked in the head.
2024-04-18T13:14:22 < karlp> fucking acquired in 2013. 
2024-04-18T13:19:39 < jpa-> the "let's collide!" stack + heap arrangement is so silly
2024-04-18T13:20:00 < jpa-> for single thread stuff, stack at bottom and heap at top and predefined stack size is so much more predictable
2024-04-18T13:21:05 < karlp> yeah, well, what actually happened here is that codered's "nohost" that was being binary linked in here in the past has a sbrk impl that uses _pvHeapStart.
2024-04-18T13:21:11 < karlp> and newlib looks for "end"
2024-04-18T13:21:29 < karlp> sooooo malloc just started mallocing from "not good places"
2024-04-18T13:22:38 < karlp> nxp's mcuxpresso github even has linker scripts for both "mcuxpresso" which uses this codered shit, and "gcc" with basically that diference. but we've got some in house bastardized version of the codered dependnet one.
2024-04-18T13:24:32 < qyx> somehow I think the global malloc()/free() is a flawed concept
2024-04-18T13:24:58 < qyx> individual heaps + heap referenced malloc/free metods look more reasonable to me
2024-04-18T13:25:57 < qyx> and a heap defined like uint8_t heap[size] or what
2024-04-18T13:25:59 < jpa-> and hope that the free() asserts that it is actually freed to correct heap :)
2024-04-18T13:26:44 < qyx> that's what DI and MPU is for
2024-04-18T13:28:07 < qyx> actually it is a requirement to use dynamic memory and MPU together, you can't MPU a bunch of randomly spread regions of malloc'd ram
2024-04-18T13:29:29 < qyx> *for using
2024-04-18T13:45:23 < qyx> LAN8671 finally available
2024-04-18T13:45:33 < qyx> now I can make a t1s switch without doing a line of code
2024-04-18T13:56:04 < zyp> karlp, it's optional, you can leave it out :)
2024-04-18T13:56:18 < zyp> but yeah, if you want location in the tag, you have to include it
2024-04-18T13:58:19 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-18T14:01:22 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Quit: Leaving]
2024-04-18T14:14:37 < karlp> hrm, trying to comapre and merge linker scripts, so I can put this to bed really, not just leave my patcha nd have to come back later.
2024-04-18T14:15:13 < karlp> nxp's sample linker script places the mtb buffer in .data,
2024-04-18T14:15:39 < karlp> that's pretty wasteful.
2024-04-18T14:19:54 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-18T14:23:50 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T14:27:10 < zyp> so I finished the «put a whole pile of stuff in docker» thing, and now all that remains is to test it on windows
2024-04-18T14:27:49 < zyp> so I asked my coworker if he had a windows computer to test it on, and he dragged up this obsolete laptop
2024-04-18T14:35:22 < zyp> which doesn't seem to want to run docker…
2024-04-18T14:36:50 < karlp> that sounds pretty old...
2024-04-18T14:37:04 < karlp> probably runs IE too :)
2024-04-18T14:37:45 < karlp> ok, codered libc -> gcc newlib => .data goes _down_ 2k, .bss goes _up_ 300 bytes.  (.text goes down 30k!)
2024-04-18T14:38:37 < zyp> it's a 5th gen i3, so yeah
2024-04-18T14:42:26 < karlp> ok, newlib has _wayyyy_ smaller "impure_data" blob. 
2024-04-18T14:46:57 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-18T14:52:07 < qyx> 5th gen i3? my current notebook is something like that
2024-04-18T14:53:47 < karlp> and "__malloc_av_" is just gone, another 400 bytes of .data 
2024-04-18T14:53:53 < karlp> lol, not the first: https://community.nxp.com/t5/LPC-Microcontrollers/Strange-memory-consumption/m-p/646596
2024-04-18T14:56:25 < qyx> I am fairly certain that objects instantiated within the scope of a function are not really placed on the stack.  Malloc is called to create the object and then destroyed when the function goes out of scope. 
2024-04-18T14:56:29 < qyx> really?
2024-04-18T14:56:36 < qyx> I am dumb and can't c++,so idk
2024-04-18T15:06:54 < ventYl> that's not true
2024-04-18T15:07:16 < ventYl> automatic variables are automatic, regardless if they are structs or objects. they end up on stack
2024-04-18T15:08:19 < zyp> I gave up
2024-04-18T15:13:15 < jpa-> qyx: most objects go on stack, but some classes do internal allocations (like std::vector or std::unique_ptr)
2024-04-18T15:13:51 < zyp> unique_ptr doesn't allocate anything itself, just free
2024-04-18T15:14:22 < jpa-> heh, true, i guess i meant make_unique() :)
2024-04-18T15:14:36 < ventYl> unique_ptr in itself is just a raw pointer
2024-04-18T15:14:50 < ventYl> all the shiny stuff is just semantics created on top of it
2024-04-18T15:14:55 < zyp> yeah
2024-04-18T15:15:03 < ventYl> and that's really amazing
2024-04-18T15:15:11 < jpa-> i wouldn't call it "just a raw pointer" because the surprise call to free() when it leaves scope
2024-04-18T15:15:22 < jpa-> better make sure you are not leaving scope with unique_ptr when holding a lock
2024-04-18T15:15:25 < qyx> so never believe random pros on nxp's forums
2024-04-18T15:16:02 < ventYl> jpa-: yeah, that's the added semantics part. but unique_ptr doesn't have any additional storage other than the address of the object pointed to
2024-04-18T15:19:26 < zyp> I'm a big fan of std::span, which is another «just a raw pointer» and optionally a length
2024-04-18T15:20:07 < ventYl> we will see more of this and it will probably make a revolution in C++ similar to C++11
2024-04-18T15:20:27 < zyp> more of what?
2024-04-18T15:20:40 < ventYl> zero-copy APIs
2024-04-18T15:20:49 < ventYl> like string_views
2024-04-18T15:21:30 < zyp> well, a span is effectively a generalized string view, it's C++17 so not directly new
2024-04-18T15:21:49 < jpa-> i wish C++ had covariant templates like C#
2024-04-18T15:22:02 < jpa-> ETL has poly_span which is halfway there
2024-04-18T15:22:09 < qyx> I wish c++ was just python
2024-04-18T15:22:12 < zyp> before spans, people were doing begin/end iterator pairs
2024-04-18T15:22:45 < zyp> and then C++20 added std::ranges which is another go at improving iterator stuff
2024-04-18T15:23:17 < zyp> jpa-, what's that?
2024-04-18T15:24:40 < jpa-> zyp: it lets you make a span<BaseClass> out of vector<SubClass>
2024-04-18T15:27:49 < zyp> hmm, looking at the interfaces, I figure it'll have to store stride internally then, to handle sizeof(SubClass) > sizeof(BaseClass)
2024-04-18T15:28:12 < jpa-> yeah
2024-04-18T15:28:24 < zyp> std::ranges probably have something that could do the same
2024-04-18T15:31:15 < ventYl> I guess it wouldn't be safe at all
2024-04-18T15:32:46 < ventYl> why would you need that anyway? You can use pointer or reference to SubClass wherever pointer or reference to BaseClass is needed
2024-04-18T15:33:15 < ventYl> in all other scenarios, SubClass may have so different semantics of its virtual functions, replacement actually doesn't make sense
2024-04-18T15:33:35 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T15:33:51 < zyp> presumably if you have a function taking an iterable of parents, but they're stored as an array of children
2024-04-18T15:34:09 < zyp> I'm also not sure when you'll have that in practice though
2024-04-18T15:34:25 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T15:34:48 < ventYl> that's exactly case where the iterator / container has to adjust for different size as simple pointer++ won't do it
2024-04-18T15:35:17 < zyp> exactly
2024-04-18T15:36:42 < jpa-> and that's the point of poly_span
2024-04-18T15:39:39 < jpa-> one place where i used it was  poly_span<ConfigItem>   and then the base class had virtual methods to format/parse from strings, and subclasses were per datatype; though i'd have to agree that std::variant would usually work as well
2024-04-18T15:39:58 < jpa-> i don't like the idea of constructing a new vector of pointers just for polymorphism
2024-04-18T15:41:57 < jpa-> ah, and another unfinished project where each module had a bunch of IO interfaces; so i would have an interface like GPIOPin and then implementations like MCP23017_I2C_GPIOPin, and the module would return a span of all pins it has
2024-04-18T15:42:17 < zyp> hmm, ranges seems to mostl be about concepts: https://godbolt.org/z/x5oTca8vE
2024-04-18T15:42:24 < jpa-> with some funniness i was able to put those arrays in ROM
2024-04-18T15:42:29 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-18T15:42:42 < jpa-> (funniness = put constexpr everywhere randomly)
2024-04-18T15:42:56 < ventYl> jpa-: that sounds like vastly wrong architecture where too many things know way too much about physical layout
2024-04-18T15:43:29 < jpa-> ventYl: maybe, but on the other hand the whole point of the project was to allow access to a bunch of pins etc.
2024-04-18T15:43:44 < jpa-> so i couldn't abstract the pins away
2024-04-18T15:49:12 < karlp> (I only shared that as an example of someone finding ~800 bytes of data that seemed unnecessary, not commenting on their other claims)
2024-04-18T15:51:58 < karlp> but, now I've found that my new style libc has eliminated the display buffer I guess.  menus work, but weight display is now just blank all the time. boo.
2024-04-18T15:57:31 < qyx> try returning those missing bytes
2024-04-18T16:16:05 < karlp> indeed :)
2024-04-18T16:20:58 < qyx> but the guy was pretty determined running c++ + freertos with 8k/32k
2024-04-18T16:21:18 < ventYl> there's nothing exceptionally wrong on that
2024-04-18T16:21:40 < karlp> when stupid libc settings take up ~1k of flash+ram it's nasty.
2024-04-18T16:21:47 < karlp> but those things he's referring to are not c++ related,
2024-04-18T16:21:56 < karlp> just "mcuxpresso/codered/lol"
2024-04-18T16:22:01 < qyx> freertos alone takes 10k, printf&friends about 22k
2024-04-18T16:31:29 < qyx> I wonder why this kind of issues wasn't addressed at all
2024-04-18T16:31:38 < qyx> date functions, 14k
2024-04-18T16:32:03 < qyx> function-sections no worky for those?
2024-04-18T16:32:31 < ventYl> everything references everything, functions-sections are powerless
2024-04-18T17:12:47 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 256 seconds]
2024-04-18T17:15:59 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-18T17:21:22 < karlp> except the things you want kept, those are discarded.
2024-04-18T17:26:35 -!- splud is now known as chatgpt
2024-04-18T17:26:48 -!- chatgpt is now known as splud
2024-04-18T17:37:09 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f562-efef-7719-5bd8.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-18T17:53:36 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-18T18:12:33 -!- nuxil_ is now known as nuxil
2024-04-18T18:30:11 -!- specing [~specing@user/specing] has quit [Ping timeout: 268 seconds]
2024-04-18T18:37:51 < karlp> god, two fucking printf implementations.
2024-04-18T18:38:05 < karlp> some internal "light weight" one, but not being used in all palces, so libc version as well.
2024-04-18T18:44:51 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 260 seconds]
2024-04-18T19:00:15 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-18T19:19:18 -!- Kerr [~quassel@174.31.47.68] has quit [Remote host closed the connection]
2024-04-18T19:24:43 < qyx> well done, so 22k + nk
2024-04-18T19:25:02 < qyx> I guess 9k for float too, since you are metrologists
2024-04-18T19:25:14 < karlp> it's ok, 512kB flash on this...
2024-04-18T19:25:23 < karlp> no-ones proposing using the 256k version
2024-04-18T19:25:34 < karlp> and we're at around 256 right now...
2024-04-18T19:32:55 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 260 seconds]
2024-04-18T20:08:00 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-18T20:08:37 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-18T20:36:07 -!- scrts [~scrts2@23.28.144.38] has joined ##stm32
2024-04-18T20:44:33 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-18T20:55:12 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-18T21:02:35 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T21:03:44 -!- qyx [~qyx@84.245.120.249] has quit [Ping timeout: 268 seconds]
2024-04-18T21:08:18 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T21:10:25 -!- qyx [~qyx@84.245.121.219] has joined ##stm32
2024-04-18T21:13:09 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-389a-4771-c12b-d7a4.fixed6.kpn.net] has joined ##stm32
2024-04-18T21:49:10 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-18T23:05:13 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-18T23:40:53 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-18T23:41:42 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-18T23:47:52 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 255 seconds]
--- Day changed pe huhti 19 2024
2024-04-19T00:08:54 -!- drfff [~k\o\w@2607:fea8:1d00:89f0:1051:315e:8d62:4fc] has joined ##stm32
2024-04-19T00:18:27 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-19T00:18:57 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-19T00:29:44 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-19T00:42:09 -!- sauce [~sauce@free.and.open.sauce.icu] has quit [Remote host closed the connection]
2024-04-19T00:43:20 -!- sauce [~sauce@free.and.open.sauce.icu] has joined ##stm32
2024-04-19T00:50:03 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-389a-4771-c12b-d7a4.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-19T01:22:21 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 255 seconds]
2024-04-19T01:59:58 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 264 seconds]
2024-04-19T02:02:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-19T02:28:00 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-19T03:09:27 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-19T03:10:05 < nomorekaki> https://www.youtube.com/watch?v=5eDQ2SSsmEk short film for blade runner fans
2024-04-19T03:33:58 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-19T03:34:49 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-19T03:39:47 < Ecco> Wow, the aspect ratio
2024-04-19T03:39:49 < Ecco> 3283:1
2024-04-19T07:03:43 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-19T08:00:32 < ds2> What I wish CPU/SoC makers would publish is joules per instruction at some particular clock
2024-04-19T08:01:47 < ds2> ideally at max clock so it becomes an upper envelope
2024-04-19T08:02:01 < qyx> they are publishing uA/MHz and dmips/MHz
2024-04-19T08:03:13 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-19T08:04:17 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-85e7-df2d-cd07-9a87.fixed6.kpn.net] has joined ##stm32
2024-04-19T08:05:23 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-19T08:15:02 < ds2> but a load is likely to be less energy then say a divide
2024-04-19T08:17:03 < jpa-> and any GPIO use is likely to be more than the difference
2024-04-19T08:17:09 < jpa-> as is memory bus use
2024-04-19T08:17:30 < jpa-> and it will depend also on the data, so seems like lot of work for only slightly better estimate
2024-04-19T08:26:16 -!- boB_K7IQ [~boB_K7IQ@174-26-246-35.phnx.qwest.net] has quit [Ping timeout: 255 seconds]
2024-04-19T08:26:58 -!- boB_K7IQ [~boB_K7IQ@174-26-246-35.phnx.qwest.net] has joined ##stm32
2024-04-19T08:30:26 < qyx> how would you use the data anyway
2024-04-19T08:30:48 < qyx> tracing instructions and calculating?
2024-04-19T08:47:13 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-85e7-df2d-cd07-9a87.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-19T09:07:35 < jpa-> don't even need tracing, PC sampling would give reasonable estimate of the instruction distribution
2024-04-19T09:07:55 < jpa-> but in any case, at that point you already have the code running on the CPU, so could just as well measure the power consumption
2024-04-19T09:09:23 < zyp> yeah, power consumption is even more of a «just benchmark it» thing than execution time
2024-04-19T09:37:45 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-19T09:42:22 -!- drkow [~k\o\w@2607:fea8:1d00:89f0:5018:7245:91f6:7898] has joined ##stm32
2024-04-19T09:42:55 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-19T09:45:47 -!- drfff [~k\o\w@2607:fea8:1d00:89f0:1051:315e:8d62:4fc] has quit [Ping timeout: 272 seconds]
2024-04-19T10:27:05 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-19T10:36:37 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-19T10:40:07 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Ping timeout: 260 seconds]
2024-04-19T11:00:38 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-19T11:43:22 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-19T11:44:50 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-19T12:13:58 -!- boB_K7IQ [~boB_K7IQ@174-26-246-35.phnx.qwest.net] has quit [Ping timeout: 256 seconds]
2024-04-19T12:14:21 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-19T13:10:38 -!- PaulFertser [paul@paulfertser.info] has quit [Read error: Connection reset by peer]
2024-04-19T13:10:47 -!- PaulFertser [paul@paulfertser.info] has joined ##stm32
2024-04-19T13:22:02 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-19T13:51:57 < karlp> heh, cute. updating freertos, got a weird warning. spot the difference: https://github.com/FreeRTOS/FreeRTOS-Kernel/blob/main/include/task.h#L1943 vs https://www.freertos.org/Stacks-and-stack-overflow-checking.html
2024-04-19T13:55:35 < qyx> I don't see it?
2024-04-19T13:56:10 < qyx> on signed char for a string?
2024-04-19T13:56:34 < karlp> yeah, signed vs unspecified.
2024-04-19T13:56:52 < karlp> "error: conflicting types for 'vApplicationStackOverflowHook'; have 'void(struct tskTaskControlBlock *, signed char *)" vs .... blah 
2024-04-19T13:57:01 < qyx> anyway, stack overflow detection in freertos never worked for me
2024-04-19T13:58:18 < karlp> it's caught stuff for me 
2024-04-19T13:58:28 < karlp> not as much as I'd like it to, but any catch is better than no catches
2024-04-19T14:06:32 < karlp> it has approved my report of incorrect filtering, whee...
2024-04-19T14:15:13 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-19T14:49:24 < jpa-> https://jpa.kapsi.fi/stuff/pix/bga_fix.jpg   8 out of 9 imx boards successfully fixed (first one burned because powering it on with wrong NVCC_PLL connection)
2024-04-19T14:51:35 < qyx> which imx?
2024-04-19T14:54:00 < Steffanx> Why didn't you cut/remove the pad and the nearby traces jpa- ?
2024-04-19T14:54:32 < zyp> because it's routed through the pad to other places that need the power
2024-04-19T14:54:43 < Steffanx> oh ofcourse -_-
2024-04-19T14:54:53 < jpa-> yeah; could have jumperwired it from a via though, but this was easier
2024-04-19T14:55:50 < jpa-> qyx: IMX1062
2024-04-19T15:02:19 < qyx> nxp's web is even worse than st's
2024-04-19T15:02:39 < qyx> 1M ram but I wasn't able to find the flash size
2024-04-19T15:03:02 < jpa-> external
2024-04-19T15:03:14 < qyx> oh
2024-04-19T15:03:32 < jpa-> on this board 64 MB external flash + 64 MB hyperbus ram
2024-04-19T15:03:38 < qyx> MB?
2024-04-19T15:04:07 < jpa-> megabyte
2024-04-19T15:06:01 < qyx> thats pretty unusual, 64 MB hyperbus ram
2024-04-19T15:06:15 < jpa-> well 2x32 MB actually
2024-04-19T15:06:28 < jpa-> IS66WVO32M8DBLL
2024-04-19T15:06:35 < qyx> on that dual-quad iface?
2024-04-19T15:06:49 < qyx> o don't see any dedicated hyperbus i terface
2024-04-19T15:06:58 < jpa-> err hmm, octospi actually
2024-04-19T15:07:08 < qyx> but otherwise looks like a capable mcu
2024-04-19T15:07:09 < jpa-> but IIRC it supports both modes
2024-04-19T15:07:29 < jpa-> two chips on same bus with different chipselects
2024-04-19T15:08:46 < jpa-> i think i looked at hyperbus chips but it was difficult to find big chips with 3 V supply, and the IO power domains weren't working in my favor
2024-04-19T15:09:33 < jpa-> fortunately i don't need to figure out the software, just the PCB :)
2024-04-19T15:47:48 < Ecco> Heya :)
2024-04-19T15:47:57 < Steffanx> Haye!
2024-04-19T15:47:57 < Ecco> I've been trying to use orbuculum, but failed
2024-04-19T15:48:33 < Ecco> I would like to profile code execution. I'm using a WBA chip with a ST-link v3. Is it possible? Is orbuculum the right tool for the job?
2024-04-19T15:52:30 < karlp> yeah, orbtop should be fine.
2024-04-19T15:53:35 < Ecco> ok cool
2024-04-19T15:54:02 < Ecco> ok so if in *theory* it's the right tool, I couldn't get it to work in practice
2024-04-19T15:54:18 < Ecco> I just realized there's a lot of doc at orbcode.org
2024-04-19T15:54:33 < Ecco> So let me read this first, and if I still can't get it to work I'll ask you guys more questions:)
2024-04-19T15:54:58 < karlp> what have you got so far?
2024-04-19T15:55:20 < karlp> I've never used stlinkv3, but aiui, it collects swo justa s well as stlinkv2,
2024-04-19T15:56:22 < karlp> I source this file in my openocd.cfg: https://paste.centos.org/view/aea28620
2024-04-19T15:57:03 < karlp> then I h ave a two lines in my openocd.cfg like this:
2024-04-19T15:57:06 < karlp> $_TARGETNAME.tpiu configure -protocol uart -traceclk 120000000 -output :3443 -formatter 0
2024-04-19T15:57:08 < karlp> $_TARGETNAME.tpiu enable
2024-04-19T15:57:19 < karlp> set traceclk to your sys clock.
2024-04-19T15:57:34 < karlp> then from gdb: "mon dwt_pc_sampling on 1024"
2024-04-19T15:57:46 < karlp> then orbtop -e myelf
2024-04-19T15:58:49 < karlp> I personally find the https://github.com/orbcode/orbuculum/blob/main/Support/gdbtrace.init a little too overwhelming, as it combines all the trace setup for capture as well.
2024-04-19T15:59:15 < karlp> and etm stuff as well,
2024-04-19T15:59:25 < karlp> I normally am more than happy with just pc sampling over swo...
2024-04-19T16:00:05 < Ecco> oh wow, that's a long openocd config file!
2024-04-19T16:00:18 < karlp> it's just a lot of functions for trace related stuff
2024-04-19T16:00:20 < karlp> and tcl is verbose.
2024-04-19T16:00:27 < Ecco> yeah, ok, so
2024-04-19T16:00:32 < karlp> you  can pick out the bits you like if you want, I've just had that file around for a while now.
2024-04-19T16:00:35 < Ecco> I kind of tried doing something similar
2024-04-19T16:00:49 < Ecco> First problem: openocd errors-out when I try to configure tpiu
2024-04-19T16:01:07 < karlp> well, no point going any further then...
2024-04-19T16:01:29 < Ecco> indeed
2024-04-19T16:01:39 < Ecco> But I'm not even quite sure about the theory
2024-04-19T16:01:41 < karlp> (if you want help, you need to say more than "errors out")
2024-04-19T16:01:50 < Ecco> yes yes I'm pulling out the error right now :)
2024-04-19T16:02:09 < Ecco> I understand openocd acts as a bridge between GDB and the debug probe
2024-04-19T16:02:21 < Ecco> Now I'm not quite sure how orbtop enters the game here
2024-04-19T16:02:38 < Ecco> and I'm not quite sure if I need any software provision for profiling on the device
2024-04-19T16:03:54 < Ecco> The error openocd gives is "stm32wbax.tpiu does not support protocol uart"
2024-04-19T16:04:58 < karlp> hah, that's ... cute.
2024-04-19T16:05:10 < karlp> try setting it to manchester
2024-04-19T16:05:14 < Ecco> I tried too
2024-04-19T16:05:17 < Ecco> same error
2024-04-19T16:05:24 < Ecco> apparently the default option is "sync", and this one is accepted
2024-04-19T16:05:27 < Ecco> (no idea what it means)
2024-04-19T16:05:33 < Ecco> but then I get another error about bit_width
2024-04-19T16:05:44 < karlp> you need to supply a bit width for sync
2024-04-19T16:05:46 < karlp> try 1.
2024-04-19T16:06:00 < Ecco> "stm32wbax.tpiu does not support protocol manchester"
2024-04-19T16:06:03 < karlp> I've  no idea about using sync, that's nominally etm over 1 wire, not itm, iirc.
2024-04-19T16:06:26 < Ecco> (I tried to look at the source of openocd, and what's weird is that they don't seem to be doing anything too specific for wba)
2024-04-19T16:06:45 < karlp> no, it' doesn't care, it's just debug coresight stuff.
2024-04-19T16:06:51 < Ecco> (which, apart from the radio stuff, seems to be a pretty standard, U5-like chip with a M33 core)
2024-04-19T16:07:05 < karlp> the "stm32wbax" name comes from the rest of the config, it's not really relevant.
2024-04-19T16:07:17 < Ecco> > TPIU does not support port-width of 1 bits
2024-04-19T16:07:38 < karlp> yeah, I've only ever used unart/machester, if it wants sync, you're into stuff I've simply never tried.
2024-04-19T16:07:42 < Ecco> (I tried 1,2,4,8 -> same error)
2024-04-19T16:07:51 < Ecco> But every doc I read says to use uart
2024-04-19T16:07:56 < karlp> no, it's 1,2,3,4, it' show many wires are used for the trace.
2024-04-19T16:08:03 < Ecco> oh, ok
2024-04-19T16:08:13 < Ecco> well, I really only wired SWD/SWO anyway
2024-04-19T16:08:40 < karlp> yeah, you've gotten too far from what I've done for me help much wihtout me opening up a lot of docs.
2024-04-19T16:08:43 < Ecco> I actually have no idea what this TPIU thing is configuration
2024-04-19T16:08:49 < karlp> maybe try #orbuculum discord?
2024-04-19T16:08:54 < karlp> trace port interface unit.
2024-04-19T16:08:54 < Ecco> ok :)
2024-04-19T16:09:03 < karlp> it's one of the blocks in the big complicated chain of debug.
2024-04-19T16:09:12 < Ecco> But the thing is, I'm really trying to do something simple
2024-04-19T16:09:21 < Ecco> I don't understand why uart doesn't work, for example
2024-04-19T16:12:05 < karlp> https://community.silabs.com/s/article/etm-and-itm-swo-trace-in-cortex-m3-and-cortex-m4-efm32-and-efr32-x?language=en_US has  bit of an example of the block diagram
2024-04-19T16:12:21 < karlp> uart is just one of the protocols coresight defines for the wire level handling of the pin
2024-04-19T16:12:30 < karlp> it's "non-returntozero" 
2024-04-19T16:12:38 < karlp> there's also manchester for the pin,
2024-04-19T16:12:54 < karlp> uart is "eaiser" for some trace _capture_ side tools to work with,
2024-04-19T16:13:18 < karlp> manchester for others, manchester lets you determine the clock auomatically, but at the expense of having faster data to handle.
2024-04-19T16:13:26 < karlp> sync is the other style altogether.
2024-04-19T16:13:43 < karlp> if smt32wbax doesn't support uart or manchester, something's a little odd...
2024-04-19T16:13:57 < Ecco> yeah, that's weird
2024-04-19T16:16:23 < karlp> rm0493 section 43.10.5 says it should work...
2024-04-19T16:16:34 < karlp> somethign else is going on?
2024-04-19T16:16:40 < karlp> what version of openocd is this?
2024-04-19T16:17:28 < karlp> it looks form the manual like it should behave like anything else....
2024-04-19T16:17:41 < karlp> maybe a coresight version check or somethign is goign wrong in openocd?
2024-04-19T16:19:41 < Ecco> yeah I guess
2024-04-19T16:19:45 < Ecco> it's a recent openocd version
2024-04-19T16:19:55 < Ecco> Open On-Chip Debugger 0.12.0+dev-01563-g04154af5d (2024-04-09-15:32)
2024-04-19T16:20:03 < Ecco> *very* recent actually
2024-04-19T16:39:04 < karlp> can you paste  your config for me?
2024-04-19T16:41:06 < karlp> or, can you use gdb or whatever to read the register at TPIU_DEVIDR  0xfc8 offset, 
2024-04-19T16:41:15 < karlp> um, maybe 0xE004fc8?
2024-04-19T16:41:30 < karlp> I don't really understand table 23 rom table.
2024-04-19T16:42:21 < karlp> yeah, try reading 0xe04fc8 for me?
2024-04-19T16:43:10 < karlp> if it has bits 10 and 11 set, you need to chase it up with oepnocd why it's failing to accept the uart/manchester protocols
2024-04-19T16:45:53 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-19T16:54:15 < Ecco> (gdb) x 0xE0040fc8
2024-04-19T16:54:16 < Ecco> 0xe0040fc8:	0x00000ca0
2024-04-19T16:54:39 < Ecco> so… yeah, bit 10 and 11 are set
2024-04-19T16:58:09 < Ecco> One thing I've still to understand: what does the ITM do? How is it different from ETM?
2024-04-19T16:58:38 < Ecco> I've used the SWO pin before, kind of like an UART/console, essentially as a way to output printf-like debug statements
2024-04-19T16:58:59 < Ecco> but that's pretty much it. My understanding is that the ITM can take control of the SWO pin and output… stuff?
2024-04-19T16:59:48 < karlp> have you dont that on this board?
2024-04-19T16:59:52 < karlp> have you done that on this board?
2024-04-19T17:00:27 < Ecco> nope
2024-04-19T17:00:42 < karlp> right.
2024-04-19T17:01:07 < karlp> on that silabs page I linked, it has a bit of a diagram, there's other ones, but you can see that itm goes to tpiu, which goes out the pin.
2024-04-19T17:01:13 < Ecco> The code that runs on this board is very vanilla: it doesn't do anything *at all* regarding debug/swd/swo/tracing
2024-04-19T17:01:55 < karlp> here, orbcode has a digram too https://orbcode.org/orbuculum/single-wire-output/
2024-04-19T17:02:19 < karlp> so, when you were doing "printf" you were using one of those "ITM software channels"
2024-04-19T17:02:22 < karlp> and writing character data to it.
2024-04-19T17:02:41 < karlp> you can also just tell the DWT to do PC sampling, and it stuffs it out the same pin, with different header bits.
2024-04-19T17:03:21 < karlp> https://orbcode.org/wp-content/uploads/2022/07/Orbtrace_KPLabs.pdf has a digram too
2024-04-19T17:04:00 < karlp> (none of this will help with why openocd isn't recognizing it, can  you share your config for your board please?)
2024-04-19T17:04:48 < Ecco> oh ok cool that's great doc
2024-04-19T17:05:53 < Ecco> ok, so I now have a better understanding
2024-04-19T17:06:04 < Ecco> one thing I still don't understand is what does the ETM do?
2024-04-19T17:07:38 < Ecco> Apparently it's better than ITM? Also WBA doesn't seem to have one?
2024-04-19T17:18:37 < zyp> < karlp> I've  no idea about using sync, that's nominally etm over 1 wire, not itm, iirc.
2024-04-19T17:19:01 < zyp> you should know better by now, ETM != sync/parallel trace :)
2024-04-19T17:21:33 < zyp> Ecco, ETM is instruction trace, it essentially streams a full log of exactly what the CPU is executing, whereas ITM is more like a sort of event trace, where events can be e.g. PC samples, or stuff that the firmware itself decides to output
2024-04-19T17:22:09 < zyp> they're both trace sources, which is orthogonal to trace transports
2024-04-19T17:22:35 < zyp> SWO and parallel trace are trace transports
2024-04-19T17:23:03 < zyp> SWO is async since it's uart or manchester encoded without a dedicated clock signal
2024-04-19T17:23:16 < zyp> parallel trace has a dedicated clock pin plus one or more data pins
2024-04-19T17:23:43 < zyp> most probe hardware can only collect SWO
2024-04-19T17:25:10 < zyp> it varies how the trace sources and the trace transports are interconnected inside the MCU, sometimes SWO is connected directly to ITM
2024-04-19T17:27:03 < zyp> usually low/mid-range MCUs have a simple dual-mode TPIU that can do both SWO and parallel, so they can do either, whereas highend MCUs often have more trace blocks in between and separate SWO/TPIU blocks
2024-04-19T17:27:31 < zyp> stm32f4 is an example of the former, stm32h7 is an example of the latter
2024-04-19T17:28:35 < Ecco> ok, crystal clear explanation
2024-04-19T17:29:22 < zyp> in any case, ETM is a firehose of data, so even if you have a MCU that can output it over SWO, you probably don't want to
2024-04-19T17:30:11 < Ecco> how is the ETM data more comprehensive than, say, PC samples?
2024-04-19T17:30:17 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-19T17:30:20 < aandrew> SWO is considerably 
2024-04-19T17:30:30 < aandrew> SWO is considerably more than that... take a look at the Orbuculum project
2024-04-19T17:30:50 < Ecco> aandrew: not quite sure what you mean :)
2024-04-19T17:31:08 < zyp> like I said, ETM makes a full log of what the CPU is doing
2024-04-19T17:31:31 < Ecco> but wouldn't I get the same info with PC sampling?
2024-04-19T17:31:38 < zyp> no
2024-04-19T17:31:47 < Ecco> I can then look at the code to see what is happening when PC=something
2024-04-19T17:31:48 < zyp> PC sampling is just that, sampling
2024-04-19T17:32:05 < Ecco> oh, you mean with PC sampling I can miss some samples?
2024-04-19T17:32:24 < jpa-> Ecco: ITM vs. ETM: https://essentialscrap.com/tips/arm_trace/arm_itm.png  https://essentialscrap.com/tips/arm_trace/arm_etm.png
2024-04-19T17:32:28 < Ecco> or is the problem deeper than this
2024-04-19T17:32:43 < zyp> PC sampling means that every so many cycles, DWT goes and grabs the current PC value and feeds it to an ITM channel
2024-04-19T17:33:00 < zyp> you have no idea of what the CPU is doing between two samples
2024-04-19T17:33:18 < Ecco> ok, so it's a speed issue, right?
2024-04-19T17:33:24 < jpa-> https://essentialscrap.com/tips/arm_trace/overflows.png  often also this if you don't stall.. 
2024-04-19T17:33:44 < zyp> PC sampling is useful for statistical profiling of where the CPU is most likely to spend its time, see e.g. orbtop for a practical use of this
2024-04-19T17:33:45 < jpa-> ETM uses a fancy encoding so that it doesn't need to retransmit PC values all the time
2024-04-19T17:33:54 < Ecco> jpa-: in your ETM trace, is the protocol streaming all the executed code, or is this infered from the value of PC?
2024-04-19T17:34:10 < zyp> ETM is useful for figuring out what the CPU did right before it crashed, see e.g. orbmortem for a practical use
2024-04-19T17:34:15 < jpa-> Ecco: ETM protocol just says "i executed 5 instructions and then took a conditional branch"
2024-04-19T17:34:51 < Ecco> ok, that was in fact my question indeed: what does the ETM protocol contain, really?
2024-04-19T17:35:06 < Ecco> so I undestand that with ITM one option is to get regular PC samples
2024-04-19T17:35:16 < Ecco> whereas ETM gives you *every single value* of PC
2024-04-19T17:35:22 < Ecco> (possibly with exact timing)
2024-04-19T17:35:27 < Ecco> is this corerct?
2024-04-19T17:35:33 < zyp> pretty much
2024-04-19T17:35:39 < jpa-> neither has exact timing, because there is output buffering
2024-04-19T17:35:42 < Ecco> ok, that's super clear
2024-04-19T17:35:56 < jpa-> https://essentialscrap.com/tips/arm_trace/theory.html   is my intro to the subject
2024-04-19T17:36:04 < Ecco> oh awesome
2024-04-19T17:39:02 < Ecco> ok, so it sucks that WBA doesn't have an ETM then
2024-04-19T17:39:20 < jpa-> in practice i rarely bother with ETM
2024-04-19T17:39:43 < Ecco> why?
2024-04-19T17:40:07 < zyp> same, due to availability, mainly
2024-04-19T17:40:08 < Ecco> I guess for performance work, "Program counter value every N instructions, N >= 32" is good enough indeed
2024-04-19T17:40:08 < jpa-> it gives too much information.. trying to dig what went wrong gets difficult if you don't exactly when the problem occurs
2024-04-19T17:40:26 < jpa-> and if you do know when the problem occurs, usually you can just breakpoint it
2024-04-19T17:40:34 < Ecco> hmm, makes sense
2024-04-19T17:40:44 < jpa-> ITM is perfect for performance profiling
2024-04-19T17:40:51 < Ecco> cool
2024-04-19T17:40:56 < jpa-> and ITM exception trace is nice too, seeing when interrupts occur
2024-04-19T17:41:11 < Ecco> What are real uses cases for ETM traces?
2024-04-19T17:41:36 < zyp> when a problem makes you go «how the fuck did I get here?»
2024-04-19T17:41:57 < jpa-> and especially when it refuses to occur when you put breakpoints into the code
2024-04-19T17:42:42 < zyp> I run into those sort of issues at work now and then, the issue is just that it haven't yet happened on a board where ETM is actually available
2024-04-19T17:42:45 < jpa-> i use ETM when i'm desperate enough to spend a day looking at endless traces..
2024-04-19T17:43:41 < zyp> nrf52 doesn't do ETM over SWO, and I haven't talked people into breaking out parallel trace yet (and given how few GPIO nrf52 has, I kinda doubt I can)
2024-04-19T17:44:08 < Ecco> hmm, ok :)
2024-04-19T17:46:06 < zyp> this stm32h7s7-dk that I recently picked up for prototyping stuff has parallel trace, so I figure I'll eventually run into some issue that makes me break out orbmortem and learn how to use it
2024-04-19T17:47:23 < jpa-> if it is anything like STM32H743, you definitely will.. i always ended up having some bus clock or kernel clock disabled and the whole CPU would just freeze and debugger would also freeze if you try to access that part of address space
2024-04-19T17:47:24 < karlp> (pc sampling is normally _more_ than sufficient for finding unexpected cpu loads and things though
2024-04-19T17:47:29 < karlp> ETM will _drown_ you 
2024-04-19T17:47:38 < karlp> and you can get so much befor eyou go there
2024-04-19T17:47:56 < karlp> none of this helps explain why openocd is rejecting configuring the WBA for async trace, which is the "real" problem.
2024-04-19T17:47:59 < jpa-> even PC sampling over SWD is often enough for basic performance monitoring
2024-04-19T17:48:24 < karlp> jpa-: yeah, it's not bad these days, but if you ahve swo, it's "free" to do it better....
2024-04-19T17:49:44 < Ecco> ok, now wait, regarding this openocd problem. My understanding is that openocd just use the debug probe to write to some registers to config the ITM trace / TPIU block
2024-04-19T17:50:06 < Ecco> I guess I could do that 1/ Manually with custom mww commands in openocd or 2/ Directly in my binary
2024-04-19T17:50:16 < jpa-> yes, but karlp will call you stupid
2024-04-19T17:50:29 < Ecco> Now I assume openocd also somewhat "grabs" SWO output and delivers it to orbuculum
2024-04-19T17:50:31 < jpa-> .. if you want to, here's the codes  https://github.com/PetteriAimonen/STM32_Trace_Example/
2024-04-19T17:50:46 < Ecco> Yeah I was precisely reading https://github.com/PetteriAimonen/STM32_Trace_Example/blob/master/configure-trace.openocd
2024-04-19T17:50:52 < jpa-> depends on your debug probes, orbuculum talks directly to some probes
2024-04-19T17:51:00 < Ecco> ok
2024-04-19T17:51:03 < zyp> I suspect that going forward, a fairly big amount of what I trace would just be smolt over ITM
2024-04-19T17:51:34 < jpa-> you can also capture low speed ITM with usb-usart adapter
2024-04-19T17:51:35 < Ecco> I have an stlinkv3, so I guess for this one orbuculum talks through openocd
2024-04-19T17:51:48 < jpa-> yes, i think that is true
2024-04-19T17:51:50 < Ecco> or even use a logic analyzer I guess, like you mentionned
2024-04-19T17:52:03 < Ecco> ok, good, now I understand how it works in theory
2024-04-19T17:52:12 < Ecco> I can try and figure out what the practical problem is
2024-04-19T17:52:14 < Ecco> thanks guys!
2024-04-19T17:53:35 < zyp> jpa-, by the way, did you try smolt yet? I wrote the readme just for you
2024-04-19T17:54:59 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-19T17:59:14 < Ecco> Is there any way that RTT logging would interfere with TPIU config in openocd?
2024-04-19T17:59:27 < zyp> no
2024-04-19T17:59:38 < Ecco> My binary logs data using RTT. My understanding is that it's just writing to a buffer in memory
2024-04-19T17:59:45 < zyp> yes
2024-04-19T17:59:51 < Ecco> then it's the probe's job to peek at this memory (if it wants to)
2024-04-19T17:59:52 < Ecco> ok good
2024-04-19T18:12:13 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 268 seconds]
2024-04-19T18:22:55 < mawk> if VBAT is floating what is the corresponding ADC channel going to give
2024-04-19T18:23:02 < mawk> random crap?
2024-04-19T18:23:25 < mawk> I want to detect when the assembly people forget to put a RTC battery 
2024-04-19T18:24:25 < Ecco> well, that seems like a reasonable assumption
2024-04-19T18:24:38 < mawk> if I take several readings and toggle some stuff in-between to make the floating voltage hopefully float around maybe I can detect 
2024-04-19T18:24:41 < Ecco> I guess measuring 0 would mean that there would be a pull-down on VBAT
2024-04-19T18:24:55 < Ecco> which would yield some leakage current, which I think is highly undesireable
2024-04-19T18:25:04 < Ecco> so if I were to design the MCU I would not put one
2024-04-19T18:25:28 < Ecco> well, I guess you could add an external pull-down
2024-04-19T18:25:30 < mawk> cpuld be a gigaohm pulldown 
2024-04-19T18:25:30 < zyp> doesn't it measure VBAT/2? I'd suspect it had a switched divider, which would also act as a pulldown when switched on
2024-04-19T18:25:33 < Ecco> (or, maybe, pull-up?)
2024-04-19T18:25:38 < mawk> VBAT/4 zyp
2024-04-19T18:25:46 < zyp> which particular chip is this?
2024-04-19T18:25:53 < mawk> f767ni
2024-04-19T18:26:00 < qyx> mawk: with vbat floating your mcu will not boot
2024-04-19T18:26:06 < qyx> or?
2024-04-19T18:26:14 < mawk> if the divider is not behind a buffer yes zyp maybe 
2024-04-19T18:26:16 < qyx> oh vbat
2024-04-19T18:26:16 < Ecco> qyx: why?
2024-04-19T18:26:17 < qyx> sorry
2024-04-19T18:26:19 < mawk> yeah it will
2024-04-19T18:26:22 < Ecco> oh ok :)
2024-04-19T18:26:26 < mawk> the clock will just be wrong
2024-04-19T18:26:31 < qyx> I meant vdda, which is common problem
2024-04-19T18:26:46 < Ecco> oh yeah, I got bitten by this one recently
2024-04-19T18:26:48 < mawk> so I could ask the user operator to physically cycle power
2024-04-19T18:26:51 < mawk> but it's annoying
2024-04-19T18:26:55 < zyp> the divider is 50k, according to the datasheet
2024-04-19T18:27:02 < Ecco> (and I think you helped with this very piece of advice, qyx)
2024-04-19T18:28:19 < mawk> right
2024-04-19T18:30:58 < Ecco> ok so I'm reading openocd's source code
2024-04-19T18:31:00 < Ecco> https://github.com/openocd-org/openocd/blob/04154af5d6cd5fe76a2583778379bdacb5aa6fb0/src/target/arm_tpiu_swo.c#L669
2024-04-19T18:31:19 < Ecco> and indeed it makes no sense why it would not see a UART-compatible TPIU
2024-04-19T18:32:45 < zyp> mawk, it's controlled by ADC_CCR.VBATE, I would expect that when you set that bit, you'll see a 50k load on VBAT
2024-04-19T18:33:58 < zyp> i.e. 60uA if you're powering it from a 3V battery
2024-04-19T18:35:08 < zyp> Ecco, on wba?
2024-04-19T18:35:10 < zyp> which wba?
2024-04-19T18:37:49 < Ecco> wba52
2024-04-19T18:37:55 < zyp> huh, WBA *only* has SWO, not parallel trace
2024-04-19T18:37:58 < Ecco> yes
2024-04-19T18:38:05 < Ecco> and no ETM
2024-04-19T18:38:11 < Ecco> why?
2024-04-19T18:38:12 < zyp> not sure I've seen that before
2024-04-19T18:38:22 < zyp> well, no ETM makes sense
2024-04-19T18:38:26 < Ecco> indeed
2024-04-19T18:38:40 < zyp> like I said, ETM is a firehose of data, without parallel trace it's not very useful
2024-04-19T18:39:16 < zyp> no parallel trace is probably a result of it not having a lot of pins
2024-04-19T18:39:58 < zyp> other stm32s often only have parallel trace available in packages with 100 pins or more
2024-04-19T18:40:03 < Ecco> make sense
2024-04-19T18:40:37 < zyp> smaller packages contain the same dies, so they still got everything internally, but they don't break out GPIOE where the trace signals usually are
2024-04-19T18:40:50 < zyp> but wba has no 100 pin packages…
2024-04-19T18:41:46 < Ecco> I guess I'll have to rebuild and debug openocd
2024-04-19T18:41:50 < Ecco> annoying
2024-04-19T18:42:28 < jpa-> zyp: i know and i appreciate the readme, but no, did not; i haven't been writing much firmware lately
2024-04-19T18:42:39 < zyp> aww
2024-04-19T18:58:55 -!- Xeroine [~Xeroine@user/xeroine] has quit []
2024-04-19T19:06:29 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-19T19:19:28 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-19T19:32:44 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-19T19:46:32 -!- Xeroine [~Xeroine@user/xeroine] has quit [Remote host closed the connection]
2024-04-19T19:51:29 < Ecco> damn
2024-04-19T19:51:36 < Ecco> ok, so openocd is reading address 0x00000000e0040fc7
2024-04-19T19:51:42 < Ecco> and reads a zero there
2024-04-19T19:51:54 < Ecco> that's why it doesn't let me configure a UART tpiu
2024-04-19T19:52:04 < Ecco> so, I don't know why it's reading at the wrong address
2024-04-19T19:52:22 < Ecco> (I hope it's not one of those tagged-thumb-function-pointers BS)
2024-04-19T19:53:13 < Ecco> "obj->spot.base" seem to be off by one already
2024-04-19T19:57:10 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has joined ##stm32
2024-04-19T21:11:14 < Ecco> Apparently this changes it by one? https://github.com/openocd-org/openocd/blob/04154af5d6cd5fe76a2583778379bdacb5aa6fb0/src/target/arm_adi_v5.c#L2405
2024-04-19T21:17:47 < Ecco> It looks like a bug in the parsing of options to the tpiu command :-/
2024-04-19T21:35:44 < Ecco> oh, no my bad
2024-04-19T21:35:54 < Ecco> but there is indeed a problem probing memory
2024-04-19T21:36:39 < Ecco> https://github.com/openocd-org/openocd/blob/04154af5d6cd5fe76a2583778379bdacb5aa6fb0/src/target/arm_tpiu_swo.c#L669
2024-04-19T21:37:00 < Ecco> This line returns 0, when it should in fact have bits 10 and 11 set
2024-04-19T21:37:11 < Ecco> (and mdw show it as such indeed)
2024-04-19T21:48:25 < aandrew> jpa-: very excellent writeup, thank you for that
2024-04-19T22:12:46 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-19T22:17:33 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 272 seconds]
2024-04-19T22:38:19 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-19T23:06:41 -!- polprog [~ath0@user/polprog] has quit [Ping timeout: 268 seconds]
2024-04-19T23:07:43 -!- rajkosto [~rajkosto@178.220.164.104] has joined ##stm32
2024-04-19T23:08:15 -!- polprog [~ath0@user/polprog] has joined ##stm32
2024-04-19T23:18:21 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 272 seconds]
2024-04-19T23:18:45 -!- Mangy_Dogg [~Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has joined ##stm32
2024-04-19T23:19:29 -!- Mangy_Dogg [~Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has quit [Remote host closed the connection]
2024-04-19T23:19:37 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-19T23:19:41 -!- Mangy_Dogg [Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has joined ##stm32
2024-04-19T23:19:56 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-19T23:20:01 -!- Mangy_Dogg [Mangy_Dog@82-69-39-176.dsl.in-addr.zen.co.uk] has quit [Remote host closed the connection]
2024-04-19T23:20:11 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-19T23:21:10 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-19T23:32:32 -!- boB_K7IQ [~boB_K7IQ@184-98-176-228.phnx.qwest.net] has joined ##stm32
2024-04-19T23:35:29 -!- nuxil [~nuxil@141.195.51.41] has joined ##stm32
2024-04-19T23:37:30 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-19T23:37:30 -!- nuxil [~nuxil@141.195.51.41] has quit [Read error: Connection reset by peer]
--- Day changed la huhti 20 2024
2024-04-20T00:05:03 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-20T00:09:57 -!- Xeroine [~Xeroine@user/xeroine] has quit [Remote host closed the connection]
2024-04-20T00:13:04 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-20T00:14:30 -!- boB_K7IQ [~boB_K7IQ@184-98-176-228.phnx.qwest.net] has quit [Ping timeout: 245 seconds]
2024-04-20T01:38:08 -!- rajkosto [~rajkosto@178.220.164.104] has quit [Quit: Leaving]
2024-04-20T02:01:07 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-20T02:10:02 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-20T02:10:05 < Laurenceb_> lmao https://nitter.poast.org/tomough/status/1781272338475962534#m
2024-04-20T02:10:11 < Laurenceb_> >making anglofuturism real
2024-04-20T02:11:20 < Laurenceb_> kekkkkk https://nitter.poast.org/pic/card_img%2F1781345044693164032%2FM2HFDgZz%3Fformat%3Djpg%26name%3D800x419
2024-04-20T02:12:34 < Laurenceb_> https://nitter.poast.org/pic/media%2FGLdZiKlXEAAvU9d.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-20T02:21:35 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-20T02:21:51 < nomorekaki> hello night crew
2024-04-20T02:24:38 < Laurenceb_> suppp
2024-04-20T02:24:39 < Laurenceb_> https://nitter.poast.org/pic/media%2FGLd8tPFXYAE15Dj.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-20T02:27:56 < Laurenceb_> orbitals sides https://nitter.poast.org/libsoftiktok/status/1780720047515701528#m
2024-04-20T02:31:34 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-20T02:55:08 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Killed (NickServ (GHOST command used by Spirit5320))]
2024-04-20T02:55:13 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-20T02:57:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-20T02:59:10 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-20T03:01:31 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 260 seconds]
2024-04-20T03:32:20 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f529-8dae-d4a-c0ad.fixed6.kpn.net] has joined ##stm32
2024-04-20T03:45:38 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-d97e-3748-e340-e5bd.fixed6.kpn.net] has joined ##stm32
2024-04-20T03:48:15 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-f529-8dae-d4a-c0ad.fixed6.kpn.net] has quit [Ping timeout: 245 seconds]
2024-04-20T03:53:40 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-d97e-3748-e340-e5bd.fixed6.kpn.net] has quit [Ping timeout: 245 seconds]
2024-04-20T04:13:40 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-20T07:06:23 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-20T08:20:16 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-20T08:47:06 -!- nuxil_ is now known as nuxil
2024-04-20T09:48:10 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a430-438a-6517-78d3.fixed6.kpn.net] has joined ##stm32
2024-04-20T09:59:19 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-20T10:23:14 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-20T10:42:00 -!- Livio [~livio@user/livio] has quit [Ping timeout: 245 seconds]
2024-04-20T11:00:58 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 246 seconds]
2024-04-20T11:19:56 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has joined ##stm32
2024-04-20T12:22:56 -!- Xeroine [~Xeroine@user/xeroine] has quit []
2024-04-20T13:52:17 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-20T14:12:52 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a430-438a-6517-78d3.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-20T14:16:29 -!- machinehum [~machinehu@5.226.148.123] has joined ##stm32
2024-04-20T14:16:35 < machinehum> Good news everyone
2024-04-20T14:17:05 < machinehum> Seems like this UPS actually work, except some shitty esp32 modules I bought off some sketchy Chinese lady
2024-04-20T14:18:06 < machinehum> I was worries the PWM stuff wouldn't be accurate enought to control the current during the end of the charging range
2024-04-20T14:18:40 < machinehum> https://imgur.com/a/s2wjOg9
2024-04-20T14:18:42 < machinehum> Seems fine
2024-04-20T14:21:30 -!- ventYl [~ventyl@adsl-dyn149.78-99-59.t-com.sk] has quit [Ping timeout: 268 seconds]
2024-04-20T14:33:40 -!- machinehum [~machinehu@5.226.148.123] has quit [Ping timeout: 245 seconds]
2024-04-20T14:43:17 -!- machinehum [~machinehu@5.226.148.123] has joined ##stm32
2024-04-20T14:43:37 < qyx> suh!
2024-04-20T14:44:04 < qyx> sorry cat
2024-04-20T14:59:33 -!- dima [~dima@2a02:8010:679f:0:ca85:6068:42e3:2044] has quit [Ping timeout: 255 seconds]
2024-04-20T15:19:22 < Steffanx> Hi cat!
2024-04-20T15:47:59 -!- machinehum [~machinehu@5.226.148.123] has quit [Ping timeout: 264 seconds]
2024-04-20T16:01:52 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-20T16:02:01 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-20T16:11:50 < qyx> uguh stuffiig
2024-04-20T16:45:20 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-20T17:35:29 < catphish> morning
2024-04-20T17:41:22 < Steffanx> Gooday sir catphish . long time no see
2024-04-20T17:41:53 < catphish> i've been rather busy with non-electronicsy things lately
2024-04-20T17:42:07 < catphish> and probaby will continue to do so since i just this week bought a house!
2024-04-20T17:42:59 < jpa-> did it come with square meters or square feet?
2024-04-20T17:43:42 < jpa-> i've always felt that having some weird feet in your house would be disgusting, while you can never have too many meters in your house
2024-04-20T17:44:08  * catphish checks
2024-04-20T17:44:59 < catphish> looks like mine comes with both, so i'll have to keep an eye out for weirdly shaped feet when i get the keys
2024-04-20T17:45:27 < jpa-> i guess you can pile them up in some corner
2024-04-20T17:45:47 < jpa-> apparently some home buyers really like them, because i have seen comparisons of "dollars per square feet"
2024-04-20T17:58:43 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-6d33-a749-9341-82cd.fixed6.kpn.net] has joined ##stm32
2024-04-20T18:02:33 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-20T18:08:55 < Steffanx> What's your favourite meter, jpa- ?
2024-04-20T18:19:24 < jpa-> my favourite meter was a probability meter i saw at university; unfortunately i haven't been able to find one for myself
2024-04-20T18:20:35 < jpa-> it had two inputs and a rotary switch to measure either A, B, A ∪ B or A ∩ B, and analog meter 0 to 100%
2024-04-20T18:48:06 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-20T18:50:45 < qyx> catphish: haven't you had one already?
2024-04-20T18:52:33 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 272 seconds]
2024-04-20T18:56:14 < nomorekaki> how is muscle building going catphish? how come you bought a house you had a house already?
2024-04-20T18:59:18 -!- ventYl [~ventyl@adsl-dyn-114.95-102-65.t-com.sk] has joined ##stm32
2024-04-20T18:59:20 < jbo> moin moin
2024-04-20T19:01:56 < nomorekaki> that reminds me of some guy years ago here responding to me mockingly "yeah muscles" at random times
2024-04-20T19:04:35 < nomorekaki> jbo: moite
2024-04-20T19:06:52 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-20T19:08:27 < nomorekaki> machinehum: did you get the power regulator thing working and into production?
2024-04-20T19:23:06 < Steffanx> That was stvn/crt nomorekaki 
2024-04-20T19:23:25 < nomorekaki> ah yes
2024-04-20T19:24:16 < nomorekaki> do you miss him?
2024-04-20T19:24:19 < Steffanx> The guy found out he has MS and pretty much disappeared after that.
2024-04-20T19:24:49 < Steffanx> Nomorekaki: He was a fun and smart guy, so ofcourse
2024-04-20T19:25:46 < nomorekaki> I dont remember him but I remember you saying you miss stvn
2024-04-20T19:28:43 < nomorekaki> I remember the nick but not the person
2024-04-20T19:29:13 < BrainDamage> ms?
2024-04-20T19:29:33 < nomorekaki> you don't know MS disease?
2024-04-20T19:30:49 < jbo> stvn/crt - wasn't that the guy constantly talking about penis pumps?
2024-04-20T19:32:09 < jpa-> jbo: no, that was you
2024-04-20T19:32:40 < Steffanx> oh, I'm not sure if that was public knowledge BrainDamage ....
2024-04-20T19:33:06 < BrainDamage> I knew what multiple sclerosis is, not the MS shortening
2024-04-20T19:33:36 < Steffanx> Ah ok, so perhaps it was public knowledge
2024-04-20T19:34:00 < jpa-> i don't think it was, but it certainly is now :)
2024-04-20T19:34:19 < Steffanx> Not penis pumps nomorekaki . Just pumpers.
2024-04-20T19:34:51 < jpa-> also, do you miss dongs?
2024-04-20T19:35:08 < Steffanx> jbo talking about penis pumps wasn't public knowledge either, jpa- :D
2024-04-20T19:35:14 < jpa-> oops
2024-04-20T19:35:15 -!- nomorekaki96 [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-20T19:35:46 < jpa-> zyp told jbo to give me "the talk", but i think something went wrong because i had a dirty dream about BGAs last night..
2024-04-20T19:36:10 < jbo> pump proudly
2024-04-20T19:39:07 < nomorekaki96> I have been talking about pumps and pumping
2024-04-20T19:39:11 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-20T19:39:18 -!- nomorekaki96 is now known as nomorekaki
2024-04-20T19:39:46 < nomorekaki> and being pumped
2024-04-20T19:41:33 < Steffanx> I hope you all are alright
2024-04-20T19:41:43 < jbo> u2 <3
2024-04-20T19:42:16 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 260 seconds]
2024-04-20T19:44:07 < zyp> jpa-, sounds about right
2024-04-20T19:44:22 < jbo> yeah, zyp didn't want to share his intimite knowledge and told me to ask you
2024-04-20T19:49:55 -!- haritzondo [~hrtz@2a02:8010:65b5:0:6009:6384:e3cb:2220] has joined ##stm32
2024-04-20T19:51:15 -!- haritz [~hrtz@user/haritz] has quit [Ping timeout: 256 seconds]
2024-04-20T20:02:00 -!- haritzondo is now known as haritz
2024-04-20T20:02:01 -!- haritz [~hrtz@2a02:8010:65b5:0:6009:6384:e3cb:2220] has quit [Changing host]
2024-04-20T20:02:01 -!- haritz [~hrtz@user/haritz] has joined ##stm32
2024-04-20T20:04:04 -!- qyx [~qyx@84.245.121.219] has quit [Ping timeout: 268 seconds]
2024-04-20T20:05:27 -!- qyx [~qyx@84.245.121.194] has joined ##stm32
2024-04-20T20:13:27 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-20T20:18:41 < machinehum> nomorekaki: Yeah works great
2024-04-20T20:19:14 < machinehum> My biggest issues is wrestling a webserver onto this esp32
2024-04-20T20:19:16 < machinehum> Zephyr
2024-04-20T20:19:33 < machinehum> The networking stack is a little thick
2024-04-20T20:19:55 < machinehum> Getting heap overflows and shit, running out of ram
2024-04-20T20:22:27 < nomorekaki> the regulator has webserver running while regulating?
2024-04-20T20:22:37 < machinehum> no lol
2024-04-20T20:23:02 < machinehum> This was the whole "holy shit you're a dumbass why did you use three MUC
2024-04-20T20:23:17 < machinehum> MCU debacle"
2024-04-20T20:23:27 < machinehum> Because the regulator shouldn't also be a webserver
2024-04-20T20:24:13 < fenugrec> I'm sure one could find enough node.js garbage to piece together a regulator
2024-04-20T20:24:32 < machinehum> fenugrec: Made me laugh
2024-04-20T20:24:36 < machinehum> node.os
2024-04-20T20:25:26 < machinehum> I don't know what the general consensus is here about Zephyr but I've been pretty impressed with it's networking stack
2024-04-20T20:26:04 < machinehum> You need dhcp-client, turn that on in defconfig, server is there to, full wifi shell
2024-04-20T20:26:20 < machinehum> It responds to a ping out of the box
2024-04-20T20:26:25 < machinehum> That's some serious high living
2024-04-20T20:26:41 < fenugrec> (I had similar argument with someone about using an esp32 to webserve some shit as well as run a motor control loop. "but it can do it all no problem !")
2024-04-20T20:27:27 < machinehum> fenugrec: Yes it can also handle the motor overcurrent relays (safety rated) no problem as well
2024-04-20T20:28:36 < fenugrec> to be clear, I was arguing that packing web stuff and critical code on the same esp32 is a Bad Idea
2024-04-20T20:29:10 < machinehum> I know I'm agreeing with you 
2024-04-20T20:29:30 < fenugrec> right
2024-04-20T20:29:30 < machinehum> Just more of my sarcastic bullshit
2024-04-20T20:34:25 < qyx> aren't you both canadians
2024-04-20T20:34:38 < machinehum> I am
2024-04-20T20:35:41 < qyx> I am happy that it works for you,mainly the bidir part which I hadn't tested thoroughly
2024-04-20T20:36:21 < machinehum> qyx: It's a cool chip, and my CM didn't complain about soldering it at all
2024-04-20T20:36:29 < qyx> cm?
2024-04-20T20:36:55 < machinehum> Contract manufacturer, shop in Canada that did PCBa
2024-04-20T20:37:38 < qyx> how many did you do?
2024-04-20T20:37:56 < qyx> and was it enough to make some money at least?
2024-04-20T20:38:25 < machinehum> We did a small run of ten, but then it will be 100 and eventually 1000.
2024-04-20T20:40:23 < Ecco> What do you guys think of https://github.com/probe-rs/rusty-probe ?
2024-04-20T20:41:28 < Ecco> (I'm curious: how can it do 62 MHz JTAG with 12MBit/s USB?)
2024-04-20T20:48:49 < machinehum> seems cool
2024-04-20T21:22:03 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-20T21:23:36 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-20T21:27:25 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 245 seconds]
2024-04-20T21:37:48 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-6d33-a749-9341-82cd.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-20T21:44:33 < Ecco> PaulFertser: I'm running into a bug in openocd. I'm trying to get ITM data over SWO out of a STM32WBA. I've narrowed the problem down to openocd trying to read TPIU_DEVID and getting an incorrect value from the chip. The value should be 0xCA1 but I get 0. Which is super weird is that if I issue a "mdw" command manually in openocd/telnet, I *do* get the right value. Here's the line where the bug happens: 
2024-04-20T21:44:39 < Ecco> https://github.com/openocd-org/openocd/blob/04154af5d6cd5fe76a2583778379bdacb5aa6fb0/src/target/arm_tpiu_swo.c#L669
2024-04-20T21:49:12 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-20T22:03:28 -!- nomorekaki81 [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-20T22:05:13 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-20T22:09:38 < qyx> does jpa- have an aurora now?
2024-04-20T22:09:46 < qyx> oh maybe tomorrow
2024-04-20T22:12:16 < BrainDamage> is that the name of the child with jbo?
2024-04-20T22:12:51 < jbo> no
2024-04-20T22:15:08 < qyx> she is already maybe 30 and clear blonde
2024-04-20T22:18:08 < jpa-> no, we have clouds
2024-04-20T22:19:15 < zyp> hmm, I wonder how I should handle ethernet buffering, I can think of a bunch of different ways, and there doesn't seem to be an obviously best way
2024-04-20T22:21:42 < zyp> consider e.g. rx buffers (ownership/lifetime)
2024-04-20T22:22:46 < zyp> I could let the rx task own them, call handlers and recycle them immediately afterwards, no dynamic handlers required
2024-04-20T22:22:55 < zyp> dynamic allocation*
2024-04-20T22:24:32 < zyp> OTOH that'd require any data that needs to be kept to be processed immediately, e.g. another task must have called recv() on a socket providing a buffer to copy the data into
2024-04-20T22:24:54 < zyp> which wouldn't be very performant
2024-04-20T22:26:23 < zyp> handing over ownership to a handler would let it defar handling for later but still be zerocopy
2024-04-20T22:28:38 < zyp> I wonder if I should just hack up some shit using heap allocated buffers for maximum flexibility, and then work out clever allocation strategies later when I know what exactly should be allocated
2024-04-20T22:29:21 < qyx> slab no good?
2024-04-20T22:30:05 < zyp> that's effectively a pool allocator, right?
2024-04-20T22:30:27 < qyx> hm yeah, there are multiple different variations
2024-04-20T22:30:44 < qyx> but basically pools of different sizes
2024-04-20T22:30:58 < zyp> I figure some sort of pool allocation strategy is what I really want, but I'm not sure I know enough to work out a good one yet
2024-04-20T22:31:33 < zyp> so I guess I can prototype with everything on the heap, and refactor to other allocators later
2024-04-20T22:32:59 < catphish> qyx: i do already have a house, i bought a bigger one
2024-04-20T22:33:04 < catphish> i like houses :)
2024-04-20T22:33:44 < qyx> aren't they expensive?
2024-04-20T22:34:24 < qyx> zyp: laks allocators when?
2024-04-20T22:34:56 < qyx> LaksAlloc foo(16384);
2024-04-20T22:35:15 < qyx> foo->alloc(size)
2024-04-20T22:35:26 < zyp> 2012 or so: https://cgit.jvnv.net/laks/tree/os/pool.h
2024-04-20T22:40:10 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-20T22:44:42 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-20T22:45:49 < qyx> law pros, how do license combine when using git submodules or dynamically fetched code? (during the build process)
2024-04-20T22:47:09 < zyp> I don't see how that would matter, most opensource licenses only cares about distribution
2024-04-20T22:47:34 < qyx> yes and I don't want to distribute that third party library code
2024-04-20T22:47:55 < zyp> unless you've got an AGPL licensed dependency I'm not sure it matters at all
2024-04-20T22:48:35 < qyx> so lets say I have bsd licensed main project and use a gpl submoduke
2024-04-20T22:48:47 < qyx> does that mean I have to relicense my codebase?
2024-04-20T22:48:48 < Ecco> you're fucked
2024-04-20T22:48:55 < Ecco> static linking is considered redistribution
2024-04-20T22:48:59 < Ecco> yes, your whole code becomes GPL
2024-04-20T22:49:06 < Ecco> GPL is a terrible license
2024-04-20T22:49:26 < qyx> my main proj is gpl
2024-04-20T22:49:32 < qyx> and I want it that way
2024-04-20T22:49:37 < Ecco> ok, then you don't care
2024-04-20T22:49:53 < qyx> I am dealing with gplv3 main + lgpl libraries
2024-04-20T22:49:58 < Ecco> GPL is pretty much the worst case scenario anyway
2024-04-20T22:50:03 < Ecco> so it can't get any worst
2024-04-20T22:50:22 < zyp> I still don't think it matters at all whether it's a submodule or anything else, what matters is what you're using it for
2024-04-20T22:50:34 < Ecco> submodule makes no difference whatsoever indeed
2024-04-20T22:50:36 < qyx> and I don't really understand if I had to redistribute the original lgpl library code in my repo as gpl3 too or what
2024-04-20T22:50:41 < Ecco> what matters is *distribution*
2024-04-20T22:50:53 < Ecco> no, you don't get to relicense 3rd-party code anyway
2024-04-20T22:51:03 < zyp> you could.
2024-04-20T22:51:22 < qyx> yes I am distributing my code, isn't it what matters?
2024-04-20T22:51:23 < Ecco> you can redistribute under the original license
2024-04-20T22:51:31 < qyx> but it doesn't contain 3rd party code
2024-04-20T22:51:35 < qyx> only binaries do
2024-04-20T22:51:47 < Ecco> you can redistribute your own code however you see fit
2024-04-20T22:51:48 -!- alan_o [~alan_o@2600:1700:1902:210f:2845:ed91:d20d:8a28] has quit [Remote host closed the connection]
2024-04-20T22:52:06 < Ecco> exactly, that's the problem; if you distribute static binaries, then you're de-facto redistributing other code
2024-04-20T22:52:07 -!- alan_o [~alan_o@2600:1700:1902:210f:22a9:73f7:2f7a:950a] has joined ##stm32
2024-04-20T22:52:13 < qyx> but the license requires me ti distribute everything to create the resulting binary
2024-04-20T22:52:15 < Ecco> so you need to comply with *their* license too 
2024-04-20T22:52:16 < zyp> qyx, it's the binaries that becomes a derivate work of the dependencies
2024-04-20T22:52:32 < qyx> so
2024-04-20T22:52:35 < Ecco> ok, what's the exact scenario?
2024-04-20T22:52:48 < Ecco> Your code is GPLv3, you have some LGPL dependencies and some BSD dependencies?
2024-04-20T22:53:01 < qyx> yes and I distribute compiled binaries
2024-04-20T22:53:05 < Ecco> yes
2024-04-20T22:53:07 < Ecco> no problem at all
2024-04-20T22:53:12 < Ecco> the licenses are compatible
2024-04-20T22:53:24 < Ecco> your binary will be distributed under GPLv3
2024-04-20T22:53:26 < zyp> right, so the binaries are GPL and thereby all the code to build them must be available
2024-04-20T22:53:34 < qyx> yes zyp
2024-04-20T22:53:39 < Ecco> and you will need to make the source of the libs available, per BSD license
2024-04-20T22:53:40 < qyx> tahts what I am asking
2024-04-20T22:53:55 < Ecco> you need to probably mention the lib's license in the binary too (another requirement of the BSD license)
2024-04-20T22:54:02 < zyp> so where does these non-GPL dependencies come from?
2024-04-20T22:54:02 < qyx> so distributing it as a git repo with submodules is not enough?
2024-04-20T22:54:12 < zyp> sure it is
2024-04-20T22:54:22 < zyp> as long as it's *available*, it is
2024-04-20T22:54:26 < Ecco> yes
2024-04-20T22:54:30 < Ecco> agreed with zyp
2024-04-20T22:54:51 < qyx> ok so the source tarball doesn't need to contain all tbe sources required to buikd the binary?
2024-04-20T22:54:57 < Ecco> honestly, if you're just using a "standard" library nobody could reasonably argue that *you* had to provide an in-tree copy
2024-04-20T22:55:02 < zyp> technically I think the GPL only requires you to make it available on request, so if anybody asks you can point them at the existing third party repos
2024-04-20T22:55:08 < zyp> correct
2024-04-20T22:55:18 < Ecco> also, it's really a moot point if you haven't patched the library
2024-04-20T22:55:28 < qyx> I do
2024-04-20T22:55:28 < zyp> nobody cares about *how* you make it available
2024-04-20T22:55:37 < Ecco> ok so if you do patch the library
2024-04-20T22:55:46 < Ecco> then you need to make the patch available somewhere
2024-04-20T22:55:47 < qyx> patches are in my repo
2024-04-20T22:55:53 < Ecco> then it's all good
2024-04-20T22:55:54 < zyp> what matters is that you *do* make it available *somehow*
2024-04-20T22:56:16 < qyx> they are being applied before the build starts
2024-04-20T22:56:18 < qyx> ok
2024-04-20T22:56:29 < Ecco> Honestly, like I said earlier, if your code is GPLv3, you're very very unlikely to have any dependency problem
2024-04-20T22:56:30 < zyp> submodules or a big tarball or whatnot are unimportant details
2024-04-20T22:56:37 < Ecco> the #1 issue is GPL being contagious
2024-04-20T22:56:49 < Ecco> and people using GPL code in proprietary product
2024-04-20T22:57:06 < qyx> Ecco: and I consider gpl actually good for specific use cases and I have one very suitable business use case when gpl protects my IP in various ways
2024-04-20T22:57:11 < Ecco> if your code is OSI compliant to begin with, you're extremely unlikely to have any problem. Even moreso if it's GPLv3
2024-04-20T22:57:24 < Ecco> qyx: Yeah, agreed, there are use cases where GPL works
2024-04-20T22:57:33 < Ecco> (for instance, to me it makes sense for the Linux kernel)
2024-04-20T22:57:40 < Ecco> but *generally* speaking I think it's a terrible license
2024-04-20T22:58:06 < qyx> I am using it in proprietary products
2024-04-20T22:58:09 < qyx> :p
2024-04-20T22:58:23 < Ecco> well, if it's *your* code then you don't give a shit
2024-04-20T22:58:42 < Ecco> (because technically you can add a new custom license just for yourself)
2024-04-20T22:59:18 < Ecco> My experience with GPL code
2024-04-20T22:59:31 < qyx> the key point here is that I don't want anyone to make a derived work and sell it in their devices, at least without the ability to get those changes back to the mainline
2024-04-20T22:59:52 < Ecco> is that when it prevents reuse in proprietary products (which is most of the time) it results in the long-term death of the project taken over by a more permissive (MIT/BSD) one
2024-04-20T23:00:02 < qyx> for all the others, I don't care, anyone can use it
2024-04-20T23:00:15 < Ecco> "I don't want anyone to make a derived work and sell it in their devices" -> fundamentally not open-source
2024-04-20T23:00:31 < Ecco> at its *core*, OSI license don't discriminate on commercial reuse
2024-04-20T23:00:40 < Ecco> (if you do, you're not OSI-compliant)
2024-04-20T23:00:54 < Ecco> (which is totally fine by me BTW. It's just not what OSI defines as open source)
2024-04-20T23:00:59 < qyx> it definitely is, they definitely can, wih releasig their improvements
2024-04-20T23:01:12 < qyx> under my terms
2024-04-20T23:01:15 < Ecco> well, yes
2024-04-20T23:01:16 < qyx> gpl terms
2024-04-20T23:01:27 < Ecco> because you added "without the ability to get those changes back to the mainline
2024-04-20T23:01:39 < Ecco> but GPL doesn't prevent someone from just cloning your product
2024-04-20T23:01:39 < zyp> I think GPL can make sense for end products like that, but for stuff that ends up as a minor dependency in something larger it's mostly just a hassle
2024-04-20T23:01:47 < Ecco> zyp: agreed
2024-04-20T23:01:51 < Ecco> it works well for the linux kernel
2024-04-20T23:02:04 < Ecco> essentially because it's actually not really contagious
2024-04-20T23:02:14 < Ecco> you build a proprietary product using linux
2024-04-20T23:02:19 < qyx> the project I am considering it for is the "large" (cough) oart
2024-04-20T23:02:22 < Ecco> you really only have to distribute your kernel patches
2024-04-20T23:02:27 < qyx> not any lib
2024-04-20T23:02:44 < Ecco> You can keep all your userland proprietary (so 99% of the project really)
2024-04-20T23:03:12 < Ecco> so it's kind of weird, because you're "escaping" the idea of the GPL on a technicality ("static" vs "dynamic" linking)
2024-04-20T23:03:47 < Ecco> I still believe the GPL license is shit, precisely because it has such a hard time defining what is a "derivative work"
2024-04-20T23:04:25 < qyx> on embedded there is usually no dynamic linking possible
2024-04-20T23:04:53 < Ecco> indeed
2024-04-20T23:05:00 < qyx> and whenever I add that possibility, I am probably making an addendum saying it applies to that too
2024-04-20T23:05:41 < zyp> you can still mix proprietary/lgpl by providing linkable objects
2024-04-20T23:05:56 < Ecco> honestly, apart from the Linux kernel, I wonder if there are some really popular *building blocks*
2024-04-20T23:06:07 < Ecco> like sure, there are GPL apps, but what about *dependencies*
2024-04-20T23:06:22 < zyp> tons of gpl libraries
2024-04-20T23:06:27 < Ecco> are they popular?
2024-04-20T23:06:42 < Ecco> honestly, I feel like most are being replaced by non-copyleft variants
2024-04-20T23:08:08 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-20T23:08:36 < qyx> so,all good then
2024-04-20T23:09:06 < Ecco> yeah, you're good :)
2024-04-20T23:09:12 < catphish> qyx: somewhat expensive, yes
2024-04-20T23:10:16 < catphish> this one certainly looks expensive https://i.imgur.com/kpdinuJ.jpg
2024-04-20T23:19:03 < qyx> jbo's daily car https://m.youtube.com/watch?v=POWGNndjuQk
2024-04-20T23:19:42 < jbo> I used to drive those around, jup :)
2024-04-20T23:20:00 < jbo> we customized one with some train rails
2024-04-20T23:20:12 < zyp> hmm, is that in norway?
2024-04-20T23:20:24 < qyx> I guess alps
2024-04-20T23:21:09 < zyp> yup, it's norwegian: https://www.vegvesen.no/en/vehicles/buy-and-sell/vehicle-information/sjekk-kjoretoyopplysninger/?registreringsnummer=SV7089
2024-04-20T23:21:16 < zyp> thought that looked like a norwegian license plate
2024-04-20T23:21:19 < qyx> I am looking for a chassis to convert to electro-hydrostatic
2024-04-20T23:21:22 < qyx> oh
2024-04-20T23:21:59 < jbo> AEBI is swiss to :)
2024-04-20T23:24:30 < zyp> hmm
2024-04-20T23:26:11 < zyp> I was thinking I should make some dynamically allocatable buffer class that also has some metadata like current size and later maybe next-pointers for linking multiple buffers
2024-04-20T23:26:46 < zyp> but putting metadata+data in a single contiguous allocation gets annoying quickly
2024-04-20T23:52:28 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-20T23:57:05 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 252 seconds]
--- Day changed su huhti 21 2024
2024-04-21T00:03:00 < fenugrec> zyp re buffers : "bip buffer" ? IIRC a fifo that guarantees a contiguous block
2024-04-21T00:06:37 < fenugrec> specing would be happy to point out https://blog.adacore.com/from-rust-to-spark-formally-proven-bip-buffers
2024-04-21T00:12:43 < zyp> yeah, I was thinking that if I decide I can guarantee that the rx buffers get freed in order, I could take a ringbuffer and segment it into small buffers so that all packets end up contiguous and nothing gets split on the wraparound
2024-04-21T00:13:29 < zyp> but then I'm not sure I even need the buffers to be contiguous :)
2024-04-21T00:14:15 < zyp> the ethernet DMA does scatter/gather, so it could make sense to actually utilize that rather than trying to work around it
2024-04-21T00:15:36 < zyp> especially for sending, it could make a lot of sense to use gather dma to assemble the various minor buffers
2024-04-21T00:39:09 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has quit [Quit: Bye]
2024-04-21T00:40:32 < qyx> what minor buffers?
2024-04-21T00:40:41 < qyx> tcp/ip headers?
2024-04-21T00:40:46 < zyp> yeah
2024-04-21T00:41:05 < qyx> is there any reason to use dma for 40 bytes?
2024-04-21T00:41:26 < qyx> configuring it takes probably longer than actually copying the data manually
2024-04-21T00:41:50 < zyp> have you ever done ethernet on stm32? the only way to get data into it is to tell it where to go grab it
2024-04-21T00:42:18 < qyx> yes but didn't code it
2024-04-21T01:00:54 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-21T01:02:08 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-21T01:16:24 < aandrew> qyx: pretty much. I had a dude use the CRC peripheral to compute CRC8 on a 4 byte packet. I replaced it with a lookup table and it was faster but dude was worried it'd not only be slower but indeterminate
2024-04-21T01:16:55 < aandrew> I told him put the damn thing in a protected section if he's worried about that but it's really no different in risk from the peripheral setup/readout
2024-04-21T01:38:49 < karlp> hey zyp , have you seen stonerberg's approach to peripherap addressing in c++? https://github.com/azonenberg/stm32-cpp/blob/master/devices/link/stm32h735.ld#L113  (I'm not sure at all what I think abotu this right now.... it's ... novel, I'll give it that up front...)
2024-04-21T01:42:14 < karlp> hrm, headers that don't include what the header needs, always makes me grumpy.
2024-04-21T01:52:10 < qyx> the hell is that
2024-04-21T01:52:43 < qyx> one single little bug and the whole thing just breaks to your surprise
2024-04-21T02:00:51 < nomorekaki81> https://www.youtube.com/watch?v=NSS6yAMZF78
2024-04-21T02:01:01 < aandrew> admittedly if there is a tiny mistake it makes itself obvious very fast
2024-04-21T02:01:09 -!- nomorekaki81 is now known as nomorekaki
2024-04-21T02:53:21 < zyp> karlp, heh, that's cute
2024-04-21T02:53:34 < zyp> doesn't seem worthwhile though
2024-04-21T02:57:26 < karlp> no, I'm not really convinced it's beneficial. seems..... cute, but ... not any sort of improvement.
2024-04-21T02:57:29 < karlp> definitely novel though
2024-04-21T02:57:57 < karlp> kinda feels like, "lets spread complexityt into all the places that _can_ rather than keeping it one place"
2024-04-21T02:59:03 < zyp> well, the benefit is that it (presumably) doesn't involve macros
2024-04-21T03:00:36 < zyp> and last I checked, you couldn't make a constinit pointer/reference cast from an integer literal
2024-04-21T03:01:00 < zyp> hence why laks is storing it as an integer instead and doing the cast on access
2024-04-21T03:06:31 < zyp> hrm, constinit works, constexpr doesn't
2024-04-21T03:41:27 < aandrew> I mean I don't mind that, and I like how moving to another device in the family is more or less automatic without macros, but generally porting to a different device in the family has other subtleties which you have to handle in software anyway
2024-04-21T05:16:42 < karlp> I'm generallyon the side of "linker + startup is complex enough already, why add more in that space" 
2024-04-21T07:43:09 < aandrew> sure, but the idea is still sound. Instead of ensuring the next label is at the correct address with .align(), he could create a memory region for each peripheral. Then the one mistake killing everything just kills the peripheral
2024-04-21T07:43:32 < aandrew> that's now not much different than how it's done today, except it's in the linker file instead of a header file full of macros
2024-04-21T09:13:25 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-21T09:32:18 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-21T09:33:06 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-bddb-2179-935d-34c1.fixed6.kpn.net] has joined ##stm32
2024-04-21T09:34:06 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-21T10:07:59 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has joined ##stm32
2024-04-21T10:13:59 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-21T10:26:28 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 268 seconds]
2024-04-21T12:17:17 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-bddb-2179-935d-34c1.fixed6.kpn.net] has quit [Ping timeout: 240 seconds]
2024-04-21T12:35:27 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-21T13:08:21 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-21T13:08:24 < Laurenceb_> cool story bro https://science.slashdot.org/story/24/04/19/223206/nasa-veteran-behind-propellantless-propulsion-drive-announces-major-discovery
2024-04-21T13:56:21 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has quit [Ping timeout: 272 seconds]
2024-04-21T14:13:17 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-21T14:48:53 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-21T17:41:30 -!- IanW_ [~IceChat9@jindivik.force9.co.uk] has quit [Quit: Bye]
2024-04-21T17:44:41 < fenugrec> at least a header can be used on different compilers... really don't see the interest in doing that in the .ldscript. It's also way more trouble to verify correctness since the locations depend on section size + .align directive. Just sounds like more ways to subtly mess up
2024-04-21T17:45:58 < fenugrec> also, if you're doing .ldscript trickery for other purposes (bootloader area, non-BSS RAM area, whatever), then it's going to be even more fun to maintain
2024-04-21T18:14:19 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-21T18:18:40 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 245 seconds]
2024-04-21T18:26:50 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-21T18:59:05 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 245 seconds]
2024-04-21T19:14:53 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-21T19:19:14 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 252 seconds]
2024-04-21T20:52:38 < ColdKeyboard> I'm sure I'm doing something wrong but just to double check; If I have an array of four uint8_t, I should be able to declare a uint32_t *pointer, point it to first byte of the array and read the uint32_t value via that pointer without having to do bit shifting and or-ing. Right?
2024-04-21T20:54:08 < qyx> the exact line would help
2024-04-21T20:54:24 < qyx> and the expected result
2024-04-21T20:55:35 < ColdKeyboard> Something like this -> u32Value = (uint32_t*)&data[0];
2024-04-21T20:55:49 < ColdKeyboard> pointer is defined as uint32_t *u32Value;
2024-04-21T20:56:59 < qyx> that's ok
2024-04-21T20:57:08 < qyx> I guess you are considering endianness
2024-04-21T20:57:41 < ColdKeyboard> Good point, I don't know if this 8bit micro is little or big endian :\
2024-04-21T20:59:19 < qyx> that's the wrong part, your code should not depend on it and there are various differing opinions on how to achieve it
2024-04-21T20:59:52 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 260 seconds]
2024-04-21T21:00:00 < ColdKeyboard> I was being lazy but that's just an excuse...
2024-04-21T21:00:00 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-21T21:00:40 < ColdKeyboard> Weird part is that for some reason I can't convert to uint32_t with simple one line
2024-04-21T21:01:20 < qyx> uint32_t val = *(uint32_t *)&data[0]
2024-04-21T21:01:24 < ColdKeyboard> it. uint32_t value = (uint_32t) ( (uint_32t)(data[0]<<24) | (uint_32t)(data[0]<<16) ...)
2024-04-21T21:01:33 -!- duude__- is now known as duude__
2024-04-21T21:02:37 < qyx> I would say you have to convert data[0] to uint32_t first, then shift
2024-04-21T21:03:12 < qyx> but I failed at it recently as was publicly observed here
2024-04-21T21:05:03 < ColdKeyboard> Ah right, I have to move the (uint32_t) *inside*...
2024-04-21T21:26:51 < sauce> https://i.imgflip.com/8njffh.jpg
2024-04-21T21:34:42 -!- HelloShitty [~psysc0rpi@bl6-131-96.dsl.telepac.pt] has joined ##stm32
2024-04-21T21:38:34 < fenugrec> ColdKeyboard you're exposing yourself to endianness problems and unaligned-access problems
2024-04-21T21:46:03 -!- PaulFertser [paul@paulfertser.info] has quit [Ping timeout: 256 seconds]
2024-04-21T21:47:40 -!- PaulFertser [paul@paulfertser.info] has joined ##stm32
2024-04-21T21:53:06 < zyp> doing some more thinking about network stack; having a rx task that blocks while it handles a request and sends a response seems like a pretty bad idea
2024-04-21T21:53:35 < qyx> yes
2024-04-21T21:54:16 < qyx> I am doing it with my message broker and it is a broker idea
2024-04-21T21:55:30 < zyp> because sending the response might involve a NDP lookup, and if you end up in a situation where the receive task is blocked doing a NDP lookup, blocking again while waiting for the response, you're stuck
2024-04-21T21:55:54 < qyx> exactly happens with my impl
2024-04-21T21:56:05 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-21T21:56:34 < qyx> for other things of course, but become stuck is a very easy and not much obvious dead end
2024-04-21T21:56:40 < qyx> *becoming
2024-04-21T21:57:20 < qyx> but I didn't find any better wayof handling request without complicated buffer management and/or copying data
2024-04-21T21:58:38 < qyx> so I am mandating that clients defer sending data back, that is not sending immediately in the rx handler
2024-04-21T21:59:35 < zyp> I'm not entirely sure how to fix it, I guess the ethernet rx task needs a nonblocking dispatcher
2024-04-21T22:01:08 < zyp> which means that once it reaches the ipv6 layer, frames needs to be sorted into different queues or something
2024-04-21T22:01:58 < qyx> sorted by what? individual handlers?
2024-04-21T22:02:32 < zyp> NDP NDs can be handled nonblocking (learn new mac addr, wake up any tasks waiting for a NDP resolution), while other packets gets queued for later processing
2024-04-21T22:03:03 < zyp> not sure NDP is the only thing that'll have an issue like that though
2024-04-21T22:03:19 < qyx> dns?
2024-04-21T22:04:21 < zyp> no, DNS requests wouldn't be triggered by incoming traffic
2024-04-21T22:04:46 < zyp> but maybe some of the TCP handshakes would have similar issues
2024-04-21T22:05:22 < qyx> then all incoming traffic
2024-04-21T22:05:35 < zyp> idk
2024-04-21T22:05:42 < qyx> whenever the handling server responds back immediately
2024-04-21T22:05:48 < qyx> tcp, tls
2024-04-21T22:06:27 < zyp> I figure anything involving a socket would have a task managing the socket
2024-04-21T22:06:52 < qyx> but if those protocols are handled by FSMs, it should not happen
2024-04-21T22:07:00 < qyx> if the FSM is run in a dedicated task
2024-04-21T22:07:40 < zyp> question is, what tasks do I need and how do I pass data between them?
2024-04-21T22:08:01 < qyx> or any other mean actually, that is deferring the processing to a dedicated task
2024-04-21T22:08:10 < qyx> idk, I didn't solve that yet
2024-04-21T22:08:47 < qyx> I am calling the client handler with a shortenough timeout to not bother me much
2024-04-21T22:09:06 < qyx> if it blocks atleastthe dispatcher continues handling rx
2024-04-21T22:10:00 < zyp> < qyx> so I am mandating that clients defer sending data back, that is not sending immediately in the rx handler
2024-04-21T22:10:05 < zyp> so like having a tx task?
2024-04-21T22:10:24 < qyx> either that or having a "process" task
2024-04-21T22:10:45 < qyx> client rx handler just sets a state, signals some semaphore, copies some data
2024-04-21T22:10:56 < qyx> and tx happens in the processing task
2024-04-21T22:11:09 < zyp> I guess I could have an ipv6 tx task that I just push a destination+buffer into a queue for, and then the tx task is responsible for doing NDP lookup before passing on to ethernet layer
2024-04-21T22:11:17 < qyx> not in the client rx handler (run in the global rx handler task)
2024-04-21T22:11:42 < zyp> except that comes with other issues, it'd block the queue while waiting for that one NDP response
2024-04-21T22:12:48 < zyp> being able to have multiple tasks with outgoing packets that are each blocked on different NDP lookups sounds more reasonable
2024-04-21T22:20:44 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 252 seconds]
2024-04-21T22:21:43 < zyp> I guess this also means that if the ethernet rx handler can't find any free queue space, it'll have to drop the frame
2024-04-21T22:23:01 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:387:95ce:cacb:2d01] has joined ##stm32
2024-04-21T22:26:55 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Quit: Leaving]
2024-04-21T22:27:44 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-21T22:28:02 -!- cybernaut [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has quit [Read error: Connection reset by peer]
2024-04-21T22:37:31 < qyx> easy dos
2024-04-21T22:37:53 < qyx> but idk how to handke that
2024-04-21T23:18:32 -!- GenTooMan [~cyberman@2601:547:437f:e5c6:21f:5bff:fefe:a883] has joined ##stm32
2024-04-21T23:19:48 < qyx> any recommended free/foss/no-ads pdf viewer for android?
2024-04-21T23:21:31 < PaulFertser> Librera ?
2024-04-21T23:22:33 < qyx> i'll check
2024-04-21T23:22:41 < qyx> can it search and view TOC?
2024-04-21T23:23:26 < qyx> librera says contains ads
2024-04-21T23:25:34 < qyx> https://www.reddit.com/r/androidapps/comments/qpjxik/is_there_a_simple_pdf_viewer_app_with_no_ads_no/
2024-04-21T23:25:42 < qyx> I am using pdf viewer plus now
2024-04-21T23:25:45 < PaulFertser> qyx: https://f-droid.org/packages/com.foobnix.pro.pdf.reader/ where are ads?
2024-04-21T23:26:25 < qyx> o what, yeah it is on fdroid
2024-04-21T23:26:31 < qyx> aurora says ads
2024-04-21T23:27:26 < qyx> installing now
2024-04-21T23:28:11 < PaulFertser> I didn't promise TOC though
2024-04-21T23:29:32 < qyx> np
2024-04-21T23:32:32 < qyx> ok it is definitely a very unusual UI
2024-04-21T23:32:43 < qyx> why is it telling me the battery status on top?
2024-04-21T23:33:13 < PaulFertser> It's highly configurable.
2024-04-21T23:33:33 < PaulFertser> I guess the battery status is there because in full screen mode when reading an e-book you might want to know how much energy left.
2024-04-21T23:34:07 < PaulFertser> It has different scrolling modes and seems to be as suitable as possible for reading PDFs on a smartphone.
2024-04-21T23:37:29 < qyx> nah I just segfaulted, still thanks for the suggestion
2024-04-21T23:40:54 < qyx> mupdf is another one completely ignoring MDI
2024-04-21T23:41:35 < qyx> *MD
2024-04-21T23:43:50 < PaulFertser> Is it a good thing or a bad thing?
2024-04-21T23:47:40 < qyx> a bad thing
2024-04-21T23:47:57 < qyx> I hate apps not adhering to the systwm look and feel
2024-04-21T23:48:59 < qyx> actually Iam not able to use them because I refuse to explore and learn all the unobvious ways to control them
--- Day changed ma huhti 22 2024
2024-04-22T00:01:08 < Steffanx> Lol you qyx 
2024-04-22T00:01:48 < qyx> whats lol, they aren't even able to use a standard toolbar
2024-04-22T00:02:07 < qyx> instead they are reinventing things
2024-04-22T00:03:25 < PaulFertser> qyx: can you recommend a sane proper calculator? So far the best I found is "bc -l" in termux.
2024-04-22T00:04:05 < qyx> I can't, that's another topic making me sad
2024-04-22T00:04:21 < qyx> I tend to usegnome-calculator a lot but it is not sane
2024-04-22T00:05:14 < qyx> on a console I just python
2024-04-22T00:06:59 < PaulFertser> bc -l shows long history, allows editing etc.
2024-04-22T00:07:10 < PaulFertser> I guess it's not MD or MO though.
2024-04-22T00:07:49 < qyx> Steffanx: I'll gibe you an example, we have a gov app for our eID. Someone doing that piece of abomination though it is a good idea to make the main window frameless with round corners and three dots on top left. Green, yellow and red
2024-04-22T00:08:15 < qyx> you could beat me to death but I simply don't know how to use them
2024-04-22T00:08:32 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-22T00:09:12 < qyx> the app completely ignores any UI design conbentions and guidelines and is very hard to use because it is violating the principle of the least surprise
2024-04-22T00:09:36 < qyx> if the convention is to use a X button to close the app on the given OS, then just fu.king use it
2024-04-22T00:09:52 < qyx> or better, don't touch that part and let the DE do it for you
2024-04-22T00:12:35 < qyx> not mentioning they are using java awt in their other apps
2024-04-22T00:18:34 < qyx> installed acrobat, no ads, free, havps toc, search, can print, standard ui, working
2024-04-22T00:19:09 < qyx> negatives: full of subscribe junk and cloud shit, 130 MB, slow to UI
2024-04-22T00:30:47 < zyp> okay, now I've got a bunch of bullshit that in theory should be capable of handling NDP messages and learning mac addrs
2024-04-22T00:30:50 < zyp> https://paste.jvnv.net/view/L4GE8
2024-04-22T00:31:21 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-22T00:33:17 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 252 seconds]
2024-04-22T00:33:46 < nomorekaki> https://www.youtube.com/watch?v=muZnCSTrq0g short film of the night
2024-04-22T00:39:04 < qyx> looks complex
2024-04-22T00:56:30 < zyp> a couple of wrong offsets, but with them fixed, NDP learning now works: https://paste.jvnv.net/view/y76NZ
2024-04-22T00:56:50 < fenugrec> android software is always moneygrabbing schemes
2024-04-22T00:56:59 < fenugrec> f mobile apps
2024-04-22T00:57:43 < fenugrec> PaulFertser , I use galculator but it (like many calcs) annoyingly lacks a "1/x" button
2024-04-22T00:57:50 < fenugrec> 2nd choice is Free42
2024-04-22T00:58:08 < fenugrec> but RPN is... an acquired taste which I'm not sure I've fully acquired
2024-04-22T00:58:49 < zyp> qyx, I'm starting to come to the conclusion that implementing a network stack is an annoying layer spaghetti
2024-04-22T00:59:45 < zyp> but I figure if I manage to put a decent async socket API on top of this, I can end up with something that's actually fairly comfortable to use
2024-04-22T01:00:23 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-22T01:00:43 < zyp> assuming I can identify and eliminate any deadlock hazards while keeping it reasonably performant and not too resource intensive
2024-04-22T01:02:24 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-22T01:09:55 -!- emeb [~emeb@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-22T02:17:37 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-22T02:17:40 < Laurenceb_> noooooo
2024-04-22T02:17:41 < Laurenceb_> https://www.ti.com/product/LM3914
2024-04-22T02:17:45 < Laurenceb_> not muh lm3914
2024-04-22T02:21:55 -!- emeb [~emeb@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-22T02:22:33 < nomorekaki> https://www.youtube.com/watch?v=UShD03eG9IU angry chunk of inconel
2024-04-22T02:23:36 < nomorekaki> I guess inconel but materials must get interesting when you try to contain explosion inside it
2024-04-22T02:27:40 < nomorekaki> Copper Chromium Niobium
2024-04-22T02:28:04 < nomorekaki> GRCop-42
2024-04-22T03:00:48 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-22T03:02:02 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-22T03:03:39 -!- rooferdave [~rooferdav@24.97.76.202] has joined ##stm32
2024-04-22T03:09:27 -!- rooferdave [~rooferdav@24.97.76.202] has quit [Quit: Konversation terminated!]
2024-04-22T03:17:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-22T03:18:49 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:387:95ce:cacb:2d01] has quit [Ping timeout: 256 seconds]
2024-04-22T06:24:49 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-22T07:43:52 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-22T08:51:09 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-22T09:21:41 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 272 seconds]
2024-04-22T09:33:46 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-22T09:38:27 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 260 seconds]
2024-04-22T09:40:51 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has joined ##stm32
2024-04-22T10:03:58 -!- c10ud [~c10ud@user/c10ud] has quit [Quit: Leaving]
2024-04-22T10:04:14 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-22T10:14:38 -!- nuxil_ is now known as nuxil
2024-04-22T10:22:19 -!- machinehum [~machinehu@187.197.2.85.dynamic.wline.res.cust.swisscom.ch] has quit [Ping timeout: 260 seconds]
2024-04-22T10:55:07 -!- machinehum [~machinehu@5.226.148.123] has joined ##stm32
2024-04-22T10:55:25 -!- machinehum [~machinehu@5.226.148.123] has left ##stm32 []
2024-04-22T12:31:32 < karlp> I'm pretty happy with speedcrunch which I think someone here recommended.
2024-04-22T12:32:13 < karlp> heh, nice, got a personal email reply from richard barry after pointing out an error on the freertos website.
2024-04-22T12:37:11 < qyx> signed issue?
2024-04-22T13:13:53 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-54a8-6088-ae99-f381.fixed6.kpn.net] has joined ##stm32
2024-04-22T13:15:57 < karlp> yeah,
2024-04-22T13:18:33 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-54a8-6088-ae99-f381.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-22T13:28:11 < qyx> Note that the following instructions are only for installation of the GPL licensed version of PyQt. If you need to use PyQt in a non-GPL project you will need to purchase an alternative license from Riverbank Computing to release your software.
2024-04-22T13:28:15 < qyx> the hell
2024-04-22T14:00:34 < karlp> hasn't pyqt has always been weirdly licensed?
2024-04-22T14:07:20 < qyx> I forgot about that
2024-04-22T14:11:18 < karlp> ins't pyside the alternatigve, from qt themselves/nokia?
2024-04-22T14:11:42  * karlp closes thos windows, stop distracting me!
2024-04-22T14:33:05 < qyx> I am sorry
2024-04-22T14:33:37 < qyx> your rabbithole index is pretty high
2024-04-22T14:50:30 < karlp> no, it's more that ltos of things are more appealing than some of the work tasks.
2024-04-22T14:50:53 < karlp> retrofitting low power to a design is very non-trivial.
2024-04-22T14:52:38 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-22T14:53:57 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-22T15:16:39 < ventYl> qyx: it is like that since pyqt5, use pyside instead
2024-04-22T15:44:44 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:7638:a112:ca6c:854] has joined ##stm32
2024-04-22T15:52:09 < Ecco> "If you need to use PyQt in a non-GPL project
2024-04-22T15:52:15 < Ecco> > That doesn't seem very accurate
2024-04-22T15:52:32 < Ecco> It's more like "If using PyQt under the terms of the GPL doesn't work for you, then..."
2024-04-22T15:52:43 < Ecco> Anyway, I got ITM-based PC sampling to work
2024-04-22T15:52:55 < Ecco> That's indeed *much* faster than break-backtrace-continue in GDB
2024-04-22T15:53:03 < Ecco> Now… it really only gives a PC
2024-04-22T15:53:05 < Ecco> not a backtrace
2024-04-22T15:53:15 < Ecco> so how am I supposed to figure out what's going on?
2024-04-22T15:54:22 < Ecco> for example, if function A calls function B way too often, then I will get PC samples in function B, but in reality it's function A that I'd need to optimize
2024-04-22T15:54:45 < Ecco> It's hard to get the big picture
2024-04-22T15:57:10 < zyp> that's the limitation of statistical sampling
2024-04-22T15:57:39 < zyp> it tells you where your application is spending time, but not why
2024-04-22T15:58:43 < karlp> Ecco: so what was busted in your setup?
2024-04-22T15:59:19 < zyp> but still, in your example, if statistics shows that you're spending a lot of time in function B, that's still what you'd want to fix, either by making B faster or by ensuring B gets called less often
2024-04-22T15:59:54 < zyp> optimizing A to call B less often is an example of the latter
2024-04-22T16:00:17 < zyp> inlining B might also help
2024-04-22T16:01:25 < zyp> last I profiled some code, I found that significant time was spent doing ringbuffer operations, which weren't inlineable
2024-04-22T16:01:47 < zyp> making them inlineable gave a massive speedup
2024-04-22T16:02:17 < zyp> not only because it got rid of the function call over head, but because inlining also lets the compiler fold redundant operations
2024-04-22T16:06:44 < Ecco> karlp: openocd is bugged on my chip. For some reason it fails to properly read TPIU_DEVID on startup. I ended up using probe-rs from the Rust ecosystem and it worked right away.
2024-04-22T16:07:17 < Ecco> zyp: Yes agreed, that is still valuable information
2024-04-22T16:07:38 < Ecco> and it does help. It's just less info that what I'm used to working with
2024-04-22T16:08:15 < Ecco> karlp: (Patching the TPIU_DEVID read command in openocd to hard-code the right return value also "works", but well…)
2024-04-22T16:08:36 < Ecco> I have no idea why the normal read doesn't work though
2024-04-22T16:09:44 < Ecco> The problem in my case is that function B can be called from a million different places (essentially it's "uint32_t volatile_register_read(void * address)". In that case I agree that it *should* be inlined. But generally speaking if function B can be called from multiple places, it's hard to know what's going on.
2024-04-22T16:12:06 < zyp> yeah, profiles with caller statistics is easier to read, but requires more data to be collected
2024-04-22T16:13:10 < Ecco> is it possible at all on an stm32?
2024-04-22T16:14:47 < Ecco> Unrelated question: is it expected to not be able to find a clock configuration where every peripheral is running at max speed?
2024-04-22T16:14:48 < zyp> anything is possible, the question is if there's any tools doing it
2024-04-22T16:15:29 < zyp> but you'd need to collect more data than just PC samples
2024-04-22T16:15:35 < Ecco> yes indeed
2024-04-22T16:16:06 < Ecco> On the PLL, is it better to do "/2 then x10" or "/4 then x20", or does it make zero difference?
2024-04-22T16:16:55 < qyx> VCO consumption differs
2024-04-22T16:17:02 < zyp> if you're okay with intrusive profiling, you could use -finstrument-functions or something that logs calls/returns to ITM or something
2024-04-22T16:17:49 < Ecco> zyp: Interesting!
2024-04-22T16:18:34 < Ecco> qyx: Where can I read about this?
2024-04-22T16:18:43 < Ecco> (and is it significant)
2024-04-22T16:18:50 < qyx> oh vco is the same, you are referring to M and N?
2024-04-22T16:19:06 < zyp> Ecco, https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html
2024-04-22T16:21:00 < zyp> also, orbuculum has an «orbprofile», which I'm not sure exactly how works, but might be non-intrusive profiling over ETM
2024-04-22T16:22:07 < Ecco> zyp: Yeah I googled that page but for some reason it doesn't work ("gcc.gnu.org refused to connect.")
2024-04-22T16:22:36 < Ecco> qyx: Yes
2024-04-22T16:22:47 < Ecco> But it's a general question
2024-04-22T16:22:50 < zyp> hmm, yeah, looks like orbprofile is ETM, not sure how mature it is
2024-04-22T16:22:58 < Ecco> I feel like there are too many options
2024-04-22T16:23:01 < karlp> what tools were you using that have been giving you rmore than this in the past?
2024-04-22T16:23:03 < zyp> and you need an ETM capture solution to use that
2024-04-22T16:23:20 < Ecco> karlp: non-embedded tools. Standard profilers on computers.
2024-04-22T16:23:23 < zyp> karlp, gprof? anything host side profiling
2024-04-22T16:23:27 < Ecco> yeah
2024-04-22T16:23:32 < karlp> right, yeah, those are totally different though
2024-04-22T16:23:36 < Ecco> yes
2024-04-22T16:24:05 < Ecco> for the PLL thing, I feel like there are just too many parameters
2024-04-22T16:24:25 < Ecco> I can get seemingly identical outcomes with different inputs (e.g. /2 x10 or /4 x20)
2024-04-22T16:24:31 < Ecco> so I *assume* I'm missing something
2024-04-22T16:24:41 < Ecco> otherwise, why would they bother providing that many settings
2024-04-22T16:24:47 < karlp> doh, I have this, but not the call-graphing.cpp file: https://github.com/karlp/l2-krv-miniblink/blob/main/SConstruct#L53-L54
2024-04-22T16:24:53 < karlp> probably sitting at home somewhere
2024-04-22T16:25:13 < zyp> Ecco, a PLL typical have min/max limits on VCO freq
2024-04-22T16:25:52 < Ecco> Also, why this error in CubeMX: https://i.imgur.com/c9EOLYQ.png
2024-04-22T16:26:03 < zyp> and inputs/outputs typically also have min/max limits
2024-04-22T16:26:10 < Ecco> Apparnetly SYSCLK is… wrong? Yet if I change the AHB1 prescaler to /1 then it's ok?
2024-04-22T16:26:15 < karlp> ah, here it is: https://github.com/karlp/l2-st-ble/blob/fc29716ca5f62408eca7a4b04fa4f372e3de5dd0/src/call-graphing.cpp#L4
2024-04-22T16:26:39 < karlp> that does some ITM tracing of entry7exit nicely 
2024-04-22T16:27:48 < zyp> Ecco, it's usual for there to be multiple possible PLL configs, the thing is that you need to find a possible solution that satisfies all the output freqs you want without violating any limits
2024-04-22T16:28:15 < zyp> sometimes it's easy and you've got multiple valid solutions, sometimes there's no valid solution
2024-04-22T16:29:19 < karlp> hrm, blogs got moved, it's https://orbcode.org/orbuculum/swo-code-instrumentation/ now.
2024-04-22T16:29:50 < zyp> e.g. for stm32f4, if you want USB, the VCO freq needs to be a multiple of 48 MHz, and with SYSCLK typically being VCO/2, it means that to use USB, SYSCLK must in practice be a multiple of 24 MHz
2024-04-22T16:30:37 < zyp> for f407, this means that you set VCO to 336 MHz, sysclk is 336/2 = 168 MHz and usb is 336/7 = 48 MHz
2024-04-22T16:31:44 < zyp> and then f427 bumped up the max sysclk to 180 MHz, which is *not* a multiple of 24, so IIRC it can't possibly run at 180 and generate 48 MHz for USB at the same time without violating any limits
2024-04-22T16:31:55 < zyp> which makes that speed increase pretty pointless
2024-04-22T16:32:01 < Ecco> ok, good to know
2024-04-22T16:32:25 < zyp> which chip are you on now?
2024-04-22T16:32:52 < Ecco> WBA
2024-04-22T16:33:01 < zyp> right, haven't used that
2024-04-22T16:33:13 < zyp> anyway, as for having multiple solutions
2024-04-22T16:33:18 < Ecco> But it's very similar to what you just said - and my previous experience too
2024-04-22T16:33:31 < zyp> IIRC on f4, VCO input must be 1-2MHz
2024-04-22T16:34:20 < zyp> so if we assume you've got an 8 MHz crystal, the two obvious solutions for VCO freq is 8/8*336 = 336 and 8/4*168 = 336
2024-04-22T16:34:56 < Ecco> yeah, and would any of these solutions be "better"?
2024-04-22T16:35:03 < zyp> I like the former, means you can use any integer MHz crystal and just divide it to 1 MHz and run a fixed config on the rest
2024-04-22T16:35:15 < zyp> but IIRC the datasheet recommends the latter, due to slightly better jitter specs
2024-04-22T16:35:48 < zyp> so given multiple valid solutions, they might differ in jitter and as qyx said power consumption
2024-04-22T16:36:47 < zyp> probably not significantly so, unless you're pushing some kind of edge somewhere, so I wouldn't worry too much about it
2024-04-22T16:37:54 < zyp> given multiple valid solutions, unless I had a specific reason to do otherwise, I'd simply just pick the one with the most readable math
2024-04-22T16:38:03 < zyp> most readable/most obvious/etc
2024-04-22T16:39:37 < Ecco> excellent, thanks!
2024-04-22T16:39:58 < Ecco> Finally, any idea about this CubeMX error? https://i.imgur.com/c9EOLYQ.png
2024-04-22T16:40:16 < zyp> what error? the magenta color?
2024-04-22T16:40:17 < Ecco> Not that I care about CubeMX, but I'm concerned that this clock configuration is indeed problematic
2024-04-22T16:40:20 < Ecco> yeah
2024-04-22T16:40:36 < Ecco> If I change the AHB1 Prescaler to /1 then the error goes away
2024-04-22T16:40:50 < Ecco> But… technically the AHB1 prescaler is *after* sysclk
2024-04-22T16:41:08 < zyp> I don't use cubemx, so I'm not sure what's going on
2024-04-22T16:41:15 < zyp> but, uh, you want 8 MHz on AHB1?
2024-04-22T16:41:42 < zyp> 6, I mean
2024-04-22T16:41:51 < Ecco> Well, I'm messing around
2024-04-22T16:42:01 < Ecco> Looking at what a slowed-down variant could look like
2024-04-22T16:42:11 < zyp> mouseover doesn't tell you what the error is?
2024-04-22T16:42:28 < Ecco> So, it does indeed! It says "Freq should be <= 16MHz"
2024-04-22T16:42:33 < Ecco> But… why?
2024-04-22T16:42:53 < Ecco> Also, are all those frequency limits listed somewhere in the datasheet/refman?
2024-04-22T16:43:00 < Ecco> I assume they are but I couldn't find them
2024-04-22T16:43:08 < zyp> yes
2024-04-22T16:43:32 < zyp> RCC chapter in refman should have at least some of them and the ones you can't find there would be in the datasheet
2024-04-22T16:43:36 < Ecco> ok
2024-04-22T16:43:54 < zyp> which refman number is this?
2024-04-22T16:43:57 < Ecco> for example, SYSCLK has two ranges, Range 1 where it should be <= 100MHz and Range 2 <= 16MHz
2024-04-22T16:44:00 < Ecco> WBA
2024-04-22T16:44:02 < Ecco> oh sorry
2024-04-22T16:44:09 < Ecco> 0493
2024-04-22T16:44:54 < zyp> right, so 12.4 has the main clock limits
2024-04-22T16:44:55 < Ecco> but anyway, I don't understand why suddenly CubeMX would insist that SYSCLK should be lower than 16MHz. I assume it enforces Voltage RAnge 2 for some reason?
2024-04-22T16:45:11 < Ecco> 11.4 I assume?
2024-04-22T16:45:26 < zyp> it's 12.4 in RM0493 rev 3
2024-04-22T16:45:34 < zyp> i.e. https://www.st.com/resource/en/reference_manual/rm0493-multiprotocol-wireless-bluetooth-lowenergy-armbased-32bit-mcu-stmicroelectronics.pdf
2024-04-22T16:46:50 < Ecco> waaat
2024-04-22T16:46:57 < Ecco> Wait, they changed the refman?
2024-04-22T16:47:28 < zyp> they revise them all the time, that's why you can't really refer to specific pages/chapters without also specifying document revision
2024-04-22T16:47:41 < Ecco> hmm, ok had no idea!
2024-04-22T16:47:48 < Ecco> They added a new chapter for newer chips
2024-04-22T16:47:57 < zyp> yep
2024-04-22T16:48:13 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-22T16:48:26 < Ecco> fun
2024-04-22T16:51:20 < zyp> I don't see any reason you couldn't run 96/16 if you wanted to, but I also don't see why you'd want to
2024-04-22T16:52:20 < zyp> probably cubemx are making some weird assumptions, maybe it sees AHB1 <= 16 and assumes you want voltage range 2, then sees the 96 MHz SYSCLK violating range 2
2024-04-22T16:54:58 < zyp> and yeah, while SYSCLK max is 100 MHz, I see how it could make more sense to run AHB5 at 96/3 than 100/4
2024-04-22T16:55:28 < Ecco> So, you're making a good point
2024-04-22T16:55:37 < Ecco> AHB5 is only used for the RADIO 
2024-04-22T16:55:49 < Ecco> So I guess it's a compromise between CPU speed vs Bluetooth speed?
2024-04-22T16:56:23 < Ecco> Also, what's the conceptual problem with 96/16?
2024-04-22T16:56:28 < zyp> hmm, how is this hooked up
2024-04-22T16:56:43 < zyp> I don't see any problem with 96/16, probably just cubemx being stupid
2024-04-22T16:57:01 < zyp> (and it seems pointless, I don't see why you'd want to turn it down)
2024-04-22T16:57:33 < zyp> leave it at /1 and if you want to save power either reduce the entire SYSCLK or just finish up quickly and go to sleep
2024-04-22T16:58:31 < zyp> reducing AHB1 seems like it'd just bottleneck the CPU, making it waste more time waiting for the peripherals (and wasting time wastes power)
2024-04-22T16:58:54 < qyx> I don't use pll for low power stuff, I usuallu run everythung at 16 MHz in such scenario
2024-04-22T16:59:12 < qyx> and switching msi/hse or hsi/hse
2024-04-22T16:59:23 < qyx> because it always wakes up with hsi or msi
2024-04-22T16:59:39 < zyp> disclaimer: I'm pulling this out of my ass, actual power consumption numbers must be evaluated/measured in the actual application to be useful
2024-04-22T16:59:43 < qyx> I would say u5 is gonna be similar to l4
2024-04-22T17:01:09 < Ecco> makes total sense though
2024-04-22T17:01:12 < Ecco> thanks!!
2024-04-22T17:02:05 < zyp> my ethernet project on stm32h7s7 is still running on HSI
2024-04-22T17:02:58 < zyp> I haven't bothered setting up a clock config yet, just configured the ethernet block to be RMII, and the PHY part is then clocked from RMII_REF_CLK
2024-04-22T17:03:05 < zyp> the rest happily works on HSI
2024-04-22T17:05:03 < qyx> the point of avoiding hse+pll on low power is that you either run on hsi/msi or just forget wakeup time figures
2024-04-22T17:05:29 < zyp> yep
2024-04-22T17:05:33 < qyx> yes they say a couple of us to wake from stop0
2024-04-22T17:05:39 < qyx> but pll lock is 40 us or so
2024-04-22T17:05:47 < qyx> and hse ready is around a ms
2024-04-22T17:05:56 < Ecco> hmm
2024-04-22T17:06:04 < Ecco> ok I never thought of that
2024-04-22T17:06:17 < zyp> that's part of why I'm saying power optimization is application specific and that you have to benchmark it
2024-04-22T17:06:50 < Ecco> So what you're saying is that there are scenarios where you'd rather avoid HSE/PLL altogether in order to be able to wake-up super quickly
2024-04-22T17:06:58 < Ecco> And that, even though the chip might run slower
2024-04-22T17:07:13 < qyx> yes, it all depends on the specific scenario
2024-04-22T17:07:14 < Ecco> it's worth it because the total time, including wake-up is still going to be lower
2024-04-22T17:07:22 < Ecco> yes indeed
2024-04-22T17:07:33 < Ecco> I want to buy a picoprobe
2024-04-22T17:07:34 < zyp> I'd rather state it as: there's scenarios where it doesn't make sense to start up the PLL because you're just gonna check a flag or increment a variable or whatever and go right back to sleep
2024-04-22T17:07:38 < Ecco> but it's super expensive
2024-04-22T17:07:45 < Ecco> I think I'll start with a Power Profiler Kit
2024-04-22T17:08:14 < Ecco> yeah, got it. It depends on the payload when waking up
2024-04-22T17:08:32 < Ecco> But I mean if the wakeup time is up to a milisecond, well, that's quite a few cycles at HSI
2024-04-22T18:03:34 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-22T18:03:54 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-22T19:06:29 -!- nuxil_ [~nuxil@141.195.51.41] has joined ##stm32
2024-04-22T19:09:52 -!- nuxil [~nuxil@141.195.51.41] has quit [Ping timeout: 260 seconds]
2024-04-22T20:03:47 -!- qyx [~qyx@84.245.121.194] has quit [Ping timeout: 260 seconds]
2024-04-22T20:05:31 -!- qyx [~qyx@84.245.121.111] has joined ##stm32
2024-04-22T20:36:34 < jpa-> hmph, wife said i need to take 3 months of vacation
2024-04-22T20:36:42 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-22T20:37:16 < Ecco> well, you'd be back just for the olympics :)
2024-04-22T20:37:42 < jpa-> what's that?
2024-04-22T20:37:49 < qyx> why jpa- 
2024-04-22T20:38:03 < jpa-> qyx: apparently i'm "stressed out", whatever that means
2024-04-22T20:38:32 < qyx> mine too, although she also said I should finish other things during the vacation
2024-04-22T20:38:51 < qyx> what will we do
2024-04-22T20:38:52 < benishor> jpa-: got kids too?
2024-04-22T20:38:58 < jpa-> 1
2024-04-22T20:39:03 < benishor> same
2024-04-22T20:39:12 < benishor> maybe you yell at the kid and ignore the wife? :)
2024-04-22T20:39:14 < jpa-> qyx: yeah, already some renovation on the list..
2024-04-22T20:39:17 < benishor> hence being stressed
2024-04-22T20:39:28 < qyx> jpa-: I have a list ready too
2024-04-22T20:39:31 < jpa-> nah, i'm more of the cry in the corner type
2024-04-22T20:39:52 < benishor> I hate holidays. wife wants to go places and I have flight anxiety
2024-04-22T20:40:09 < Ecco> well, depending on where you live you could drive/ride a train
2024-04-22T20:42:05 < jpa-> it's pretty rare to get to drive the train..
2024-04-22T20:42:23 < jpa-> though i agree it would be cool
2024-04-22T20:49:56 < ventYl> I hope to finish renovation TODO list before I die
2024-04-22T20:50:50 < jpa-> that's why you have life insurance, it will cover paying someone to rip it all off and do it properly
2024-04-22T20:55:08 < Ecco> :-D
2024-04-22T21:16:29 < zyp> jpa-, I'm going on vacation to japan in a couple of weeks, wanna come with?
2024-04-22T21:28:23 < Steffanx> There's a museum here in dutchland that lets you drive a train jpa- . Diesel, steam whatever you prefer.
2024-04-22T21:34:40 < qyx> vectron?
2024-04-22T21:45:57 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-22T22:02:03 < Steffanx> Some old dutchy
2024-04-22T22:02:12 < Steffanx> https://usercontent.irccloud-cdn.com/file/50IQhNpC/1000018830.jpg
2024-04-22T22:02:26 < Steffanx> Made in 1956 google told me
2024-04-22T22:06:25 < qyx> oh we have those too, I could ride it but not drive it
2024-04-22T22:07:12 < qyx> kids were in the railway museum this weekend
2024-04-22T22:07:25 < qyx> steam too
2024-04-22T22:20:36 -!- nerozero [~nerozero@87.253.63.54] has quit [Remote host closed the connection]
2024-04-22T22:31:24 -!- nuxil_ is now known as nuxil
2024-04-22T22:38:50 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-7875-4ee5-b26d-7a13.fixed6.kpn.net] has joined ##stm32
2024-04-22T22:45:06 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-22T22:46:13 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
--- Day changed ti huhti 23 2024
2024-04-23T00:24:49 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8dd0:7638:a112:ca6c:854] has quit [Ping timeout: 272 seconds]
2024-04-23T01:15:24 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-23T01:49:10 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has joined ##stm32
2024-04-23T02:08:59 -!- Livio [~livio@user/livio] has quit [Ping timeout: 264 seconds]
2024-04-23T02:16:52 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-7875-4ee5-b26d-7a13.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-23T02:45:39 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-23T03:14:07 -!- nomorekaki [~nomorekak@37-219-222-173.nat.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-23T03:35:32 -!- NEYi [~NEYi@109.251.216.38] has quit [Read error: Connection reset by peer]
2024-04-23T04:49:51 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-23T05:00:22 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-23T05:18:15 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-23T05:52:10 -!- Kerr [~quassel@174.31.47.68] has joined ##stm32
2024-04-23T07:45:49 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-23T07:54:38 -!- nerozero [~nerozero@87.253.63.54] has quit [Remote host closed the connection]
2024-04-23T07:56:29 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-23T08:15:02 < jpa-> zyp: nah, i'm not much into traveling abroad
2024-04-23T08:49:30 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-23T09:52:41 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-23T09:59:07 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has joined ##stm32
2024-04-23T10:02:52 -!- c10ud [~c10ud@user/c10ud] has quit [Ping timeout: 245 seconds]
2024-04-23T10:34:05 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 240 seconds]
2024-04-23T10:39:18 < zyp> aww
2024-04-23T10:54:27 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-23T10:56:39 -!- Kerr [~quassel@174.31.47.68] has quit [Remote host closed the connection]
2024-04-23T11:05:37 < Steffanx> And why is jpa- stressed out? Too many deadlines or.. just jpa- being jpa-?
2024-04-23T11:05:50 < Steffanx> Needs some more +
2024-04-23T11:37:40 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-23T11:37:50 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 268 seconds]
2024-04-23T11:41:03 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-23T11:42:00 -!- duude__- [~duude__@user/duude/x-4676560] has quit [Ping timeout: 255 seconds]
2024-04-23T11:58:54 < qyx> ¿
2024-04-23T12:23:11 < karlp> ah, a rebranding, that is always a classically good move.
2024-04-23T12:23:24 < karlp> get to it jpa, fix all your problems, call yourself jpa+ instead :)
2024-04-23T12:24:22 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-23T12:24:48 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-23T12:28:08 < jpa-> i think + is not possible on IRC
2024-04-23T12:29:05 < jpa-> but yeah, just me being me, i've been ditching customers since last year already, still getting recurring multi-day migraines
2024-04-23T12:38:15 -!- ventYl [~ventyl@adsl-dyn-114.95-102-65.t-com.sk] has quit [Ping timeout: 245 seconds]
2024-04-23T12:46:27 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 272 seconds]
2024-04-23T12:59:02 < qyx> no worky
2024-04-23T13:04:01 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-23T13:04:39 < qyx> so after a full week of illness I need to fix my nerd cave
2024-04-23T13:04:45 < qyx> I haven't been there for a while
2024-04-23T13:08:26 < jpa-> which fun illness?
2024-04-23T13:09:46 < qyx> sore throat, cough, running nose, no covid
2024-04-23T13:09:50 < qyx> the kindergarten one
2024-04-23T13:11:10 < jpa-> we are awaiting a delivery of some vomit sickness from school
2024-04-23T13:13:44 < qyx> fun too
2024-04-23T14:12:38 < Steffanx> Joy o joy.
2024-04-23T14:49:19 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-23T15:25:58 < qyx> so, there was that "routing interface" in older stm32s before TSC became common
2024-04-23T15:26:16 < qyx> then that functionality was moved to TSC apparently
2024-04-23T15:26:34 < qyx> (analol muxing/switching at the gpio level)
2024-04-23T15:26:45 < qyx> where is it now? g4 specifically
2024-04-23T15:26:54 < karlp> syscfg?
2024-04-23T15:59:36 < qyx> checking syscfg and I see only "analol switch voltage control setting"
2024-04-23T16:00:03 < qyx> probably for ADC only
2024-04-23T16:10:05 -!- ventYl [~ventyl@adsl-dyn138.78-98-137.t-com.sk] has joined ##stm32
2024-04-23T17:03:30 < englishman> japan is pretty cool jpa- you should go
2024-04-23T17:11:50 < jpa-> i'm sure many countries are pretty cool, but i'm not in a mood for long travel
2024-04-23T17:22:14 < englishman> i agree, long travel sucks
2024-04-23T17:22:20 < englishman> but i never regret it
2024-04-23T17:27:48 < karlp> we're postponing japan again til the youngest is older.
2024-04-23T17:28:01 < karlp> but looking at nz and "some soth pacific island" over christmas new years.
2024-04-23T18:03:45 < qyx> that one with 6mo snailmail delivery time?
2024-04-23T18:09:50 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-23T18:13:11 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-23T18:18:19 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-23T18:27:07 < karlp> fucking... gcc has decided that "ustime" and "mstime" are the same...
2024-04-23T18:27:25 < karlp> I was tearing my hair out why my mstimes was giving me off by a factor of 1000.
2024-04-23T18:27:34 < karlp> "break mstime" "too many breakpoitns..."
2024-04-23T18:27:43 < karlp> "list mstime" > shows me both ustime and mstime.
2024-04-23T18:28:19 < jpa-> tail call optimization or some inlining?
2024-04-23T18:28:33 < qyx> we need a ##stm32 approved date&time lib
2024-04-23T18:29:29 < karlp> fucked if I know, it "works" on the "full" project
2024-04-23T18:29:39 < karlp> but I've tried to cut a lot of it down to work out what _else_ is busted, 
2024-04-23T18:29:48 < karlp> maybe it doesn't work on the full project, I'd  have to go back and look.
2024-04-23T18:34:12 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 255 seconds]
2024-04-23T18:36:03 < karlp> nm and map file both have them at different locations.
2024-04-23T18:37:37 < karlp> in gdb p /x &mstime and p /x &ustime both give different addresses, and same addresses from map/nm.
2024-04-23T18:37:58 < karlp> but list mstime shows both funcitons, and the code is defintiely callign the wrong one.
2024-04-23T18:38:01 < karlp> wat.
2024-04-23T18:47:22 < zyp> does mstime call ustime?
2024-04-23T18:54:12 < qyx> what are those functions btw?
2024-04-23T18:54:24 < qyx> somethih custom?
2024-04-23T19:11:23 < karlp> ok, so "list" is still weird, not sure how it is working..
2024-04-23T19:11:32 < karlp> my actual problem was this: https://paste.jvnv.net/view/wQkZg
2024-04-23T19:14:20 < zyp> I figure if mstime just invokes ustime and scales it and they both get inlined and folded, you can probably get some assembly where it's not very clear which of the two functions each instruction actually originated from
2024-04-23T19:22:09 < karlp> (pressing alt-tab, and sometimes it does a tab instead)
2024-04-23T19:22:27 < karlp> so that ended up with having the mstime _impl_ not being correct, and being "another" ustime impl.
2024-04-23T19:24:55 < karlp> ustime/mstime are implemented like this: https://godbolt.org/z/sYhfvcszz
2024-04-23T19:25:41 < karlp> this is a kinetis thing, it's chained channels of a timer. 48Mhz in, ch0 counts to 48 and ticks channel 1, channel 2 coutns to 48000 and ticks channel 3.
2024-04-23T19:25:57 < karlp> so you can read usecs in channel 1 and msecs in channel 3.
2024-04-23T19:26:20 < karlp> I have _no_ idea why gdb doe shtis though: https://paste.jvnv.net/view/wg21X
2024-04-23T19:26:38 < karlp> p mstime only knows of one, the right one, but "break" and "list" lose their fucking minds.
2024-04-23T19:35:59 < karlp> I only ran into the latter problem while trying to diagnose the first one.  I think I need to get a new keyboard, I regularly have problems with alt and left shift on this one.
2024-04-23T19:49:34 < PaulFertser> But keyboard:13 references neither mstime nor ustime.
2024-04-23T19:49:51 < karlp> indeed. hence my wat'ing.
2024-04-23T19:51:22 < PaulFertser> objdump -S might give additional clues
2024-04-23T19:52:51 < karlp> nah, looks just like you'd expect, two functions nearby.
2024-04-23T19:52:57 < karlp> looks like the godbolt output
2024-04-23T19:53:04 < PaulFertser> Or the dwarf info is indeed not matching the current source code
2024-04-23T19:53:24 < karlp> well, it would happily keep telling me, "' has changed; re-reading symbols."
2024-04-23T19:53:27  * karlp shrugs
2024-04-23T19:53:30 < PaulFertser> There must be a reason it thinks address 0x00000004 has mstime
2024-04-23T19:54:05 < karlp> out of time for that sort of thing, I've fixed what led to that, now I can actually work on what I _want_ to fix, which is "why doesn't the damn thing actually go to fuckign sleep"
2024-04-23T19:54:36 < PaulFertser> Indeed
2024-04-23T19:55:13 < PaulFertser> I think if you just change the source then GDB doesn't report, it'll only print when ELF is new.
2024-04-23T19:56:32 < PaulFertser> And it can't be inlined anyway without LTO.
2024-04-23T20:44:17 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 252 seconds]
2024-04-23T20:56:32 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-23T21:17:02 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-23T21:17:09 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-23T21:18:00 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-23T21:44:29 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 240 seconds]
2024-04-23T21:58:14 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-23T21:58:15 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Excess Flood]
2024-04-23T21:58:37 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-23T22:27:07 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-23T22:54:28 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-23T23:12:09 < karlp> yeah, I renamed the function, tried a few things, but it always gave me weird extra results for "list" and adding breakpoints.  not going to chase it further.
2024-04-23T23:12:12 < qyx> zyp: don't you plan to offer your mcu testing rig as a service?
2024-04-23T23:13:29 < qyx> basically allocate a specific worker using some rest api, power it up/down and expose a gdb's tcp interface
2024-04-23T23:13:51 < qyx> + grab some serial output during the run
2024-04-23T23:14:09 < karlp> "Voltage: 3x400V - 50 c/s or as required" I don't think I've ever seen a requirement listed in "cycles per second" 
2024-04-23T23:21:23 < qyx> hm I may do a nucleo octopus hanging off my github runner
2024-04-23T23:25:37 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-23T23:25:40 < Laurenceb_> behold the anglofuture
2024-04-23T23:25:41 < Laurenceb_> https://nitter.poast.org/pic/media%2FGLrW4PMXoAAou3D.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-23T23:29:42 < specing> those don't look like the right kind of tyres for the environment
--- Day changed ke huhti 24 2024
2024-04-24T00:27:02 < qyx> STM32U545CEU6Q looks like the right MCU for kid's john deere toy electronics retrofit
2024-04-24T00:27:28 < qyx> at least I can finally play with U5
2024-04-24T00:27:53 < qyx> let's add pam8302 and some diesel sound simulator
2024-04-24T00:28:40 < qyx> or maybe I could use stm32wba
2024-04-24T00:30:13 < qyx> ok no DAC in WBA
2024-04-24T00:33:22 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-24T00:38:23 < qyx> 3MB of SRAM, a wholly different universe
2024-04-24T00:45:27 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-24T03:12:50 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-7869-92b5-9607-c8a1.fixed6.kpn.net] has joined ##stm32
2024-04-24T03:17:29 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-7869-92b5-9607-c8a1.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-24T03:25:02 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-24T03:37:19 < Ecco> What chip has 3MB of SRAM?
2024-04-24T03:37:44 < Ecco> Oh found it https://newsroom.st.com/media-center/press-item.html/n4600.html
2024-04-24T05:11:18 -!- Xeroine [~Xeroine@user/xeroine] has quit []
2024-04-24T05:11:37 -!- Xeroine [~Xeroine@user/xeroine] has joined ##stm32
2024-04-24T05:12:00 -!- alan_o [~alan_o@2600:1700:1902:210f:22a9:73f7:2f7a:950a] has quit [Ping timeout: 245 seconds]
2024-04-24T05:12:30 -!- alan_o [~alan_o@2600:1700:1902:210f:5db2:c05c:cdee:ca5] has joined ##stm32
2024-04-24T05:14:13 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
2024-04-24T05:14:30 -!- qyx [~qyx@84.245.121.111] has quit [Ping timeout: 245 seconds]
2024-04-24T05:15:29 -!- qyx [~qyx@84.245.121.111] has joined ##stm32
2024-04-24T08:21:03 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-24T08:23:16 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-24T08:24:19 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-24T08:38:07 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-24T08:49:48 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-24T08:50:13 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-24T09:09:15 < zyp> qyx, well, I could, but allowing untrusted people to access it comes with a risk of them bricking chips
2024-04-24T09:09:43 < zyp> e.g. enabling protection or writing OTP bits for chips that have that
2024-04-24T09:10:31 < qyx> true, I didn't consider that
2024-04-24T09:12:11 < zyp> if doing it as a paid service, I could charge people for anything they break and the time spent replacing broken stuff though
2024-04-24T09:15:25 < qyx> anyway, i just wanted a test rig and I am lazy to maintain it myself :P
2024-04-24T09:16:26 < zyp> I still only have a single target in mine, so I'm afraid it's not very useful yet, but I don't mind you running stuff on it when it's more useful
2024-04-24T09:17:28 < qyx> your only requirement is to have trace connected if the MCU supports it OR you only want to have trace-enabled MCUs in there?
2024-04-24T09:17:41 < qyx> or whatever yolo
2024-04-24T09:17:55 < zyp> the former
2024-04-24T09:18:19 < zyp> I'm intending to even have non-cortex-m chips in there
2024-04-24T09:19:22 < zyp> it's really more about testing the debug side of things than the trace side
2024-04-24T09:19:26 < qyx> I'll think about it, I could draw and/or donate modules for chips I am interested in
2024-04-24T09:19:39 < qyx> if that helps
2024-04-24T09:19:43 < zyp> absolutely
2024-04-24T09:31:09 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-24T09:47:17 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-24T09:50:15 < qyx> k writing at the end of my to-do list
2024-04-24T10:06:37 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Ping timeout: 255 seconds]
2024-04-24T10:17:19 -!- NEYi [~NEYi@109.251.216.38] has quit [Quit: Leaving]
2024-04-24T11:44:50 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-24T11:47:15 -!- drfff [~k\o\w@pool-99-250-234-193.cpe.net.cable.rogers.com] has joined ##stm32
2024-04-24T11:48:51 -!- drkow [~k\o\w@2607:fea8:1d00:89f0:5018:7245:91f6:7898] has quit [Ping timeout: 268 seconds]
2024-04-24T12:06:58 < karlp> don't forget that you don't get lvgl on those st parts anymore, until/if st pays the ransomware....
2024-04-24T12:09:12 < qyx> I am not interested in anything graphical
2024-04-24T12:09:42 < qyx> I just need large bufforz for a wide fft
2024-04-24T12:10:14 < qyx> also want to play with dynamic relocations and loadable modules
2024-04-24T12:25:27  * karlp is goin to see if he can finally get kinetis to go to sleep...
2024-04-24T12:44:34 < karlp> kientis is so weird:         .div1 = kSCG_AsyncClkDivBy1,              /* Fast IRC Clock Divider 1: divided by 1 */
2024-04-24T12:44:36 < karlp>         .div3 = kSCG_AsyncClkDivBy1,              /* Fast IRC Clock Divider 3: divided by 1 */
2024-04-24T12:44:55 < karlp> if those are configurable dividers, how about not giving them names div1 and div3?!
2024-04-24T12:45:42 < karlp> it's kinda like P,Q, shit in st I guess, just I had assumed for a while it really was "div 3"
2024-04-24T12:46:21 < karlp> it's more obvious on osme other parts in the series that have div1, div2, div3.  obviously I'm just kinetis enoughed...
2024-04-24T12:53:56 -!- drkow [~k\o\w@2607:fea8:1d00:89f0:112a:a703:3b25:2d8a] has joined ##stm32
2024-04-24T12:58:03 -!- drfff [~k\o\w@pool-99-250-234-193.cpe.net.cable.rogers.com] has quit [Ping timeout: 255 seconds]
2024-04-24T13:01:34 < c10ud> karlp, re: lvgl+st, what?
2024-04-24T13:03:00 < karlp> did you miss the fun? :) https://github.com/lvgl/lvgl/issues/5555
2024-04-24T13:03:14 < karlp> (they removed support for it in lvgl v9, pending st paying....)
2024-04-24T13:09:59 < c10ud> holy f-, when I tried using in a project of mine I wrote the dma part from scratch, maybe it wasn't already there..
2024-04-24T13:10:06 < c10ud> so much for open source duh
2024-04-24T13:10:33 < karlp> I saw they were "hoping to finalise an agreement with ST during embedded world" but. *meh*
2024-04-24T13:10:37 < c10ud> i think I experimented with it before 8.x, shame it is a nice lib
2024-04-24T13:10:45 < karlp> not evening mentioning it in the release notes was pretty rude.
2024-04-24T13:11:02 < karlp> they seem to acknowledge it wasn't handled well,but... 
2024-04-24T13:11:05 < qyx> is there any problem with forking v9 and bacporting ST's code back?
2024-04-24T13:11:17  * karlp is not a display person
2024-04-24T13:11:25 < karlp> they claim it was a lot of work for the re-arch in v9?
2024-04-24T13:11:40 < qyx> whatever, it is MIT
2024-04-24T13:11:56 < qyx> oh you mean it is not straightforward to do that?
2024-04-24T13:12:35 < qyx> idk we have TOUCHGFX and CUBE
2024-04-24T13:12:50 < c10ud> i believe using it in a real world™ project would require rewriting a bit of it anyway..i.e. rtos semaphores and stuff if you really want some speed
2024-04-24T13:16:54 < jpa-> qyx: afaik it is easy enough to add custom drivers, it's just that they used to include it
2024-04-24T13:17:21 < jpa-> but yeah, probably needs updates
2024-04-24T14:41:52 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-24T14:45:49 < karlp> well, it was the fucking debugger being attached.  I thought I'd already ruled that out, but ... no, apparentrly..
2024-04-24T14:49:51 < karlp> now, how to make it... not need that.
2024-04-24T14:50:01 < karlp> fuck trying to do this with plugging and unplugging eveyr fucking time.
2024-04-24T15:08:09 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-24T15:23:40 < Ecco> For this mcu-test-as-a-service thing, wouldn't an emulated core be better?
2024-04-24T15:23:54 < Ecco> (I'm not sure I understood what you guys were really talking about)
2024-04-24T15:24:05 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 245 seconds]
2024-04-24T15:25:31 < qyx> I don't want to test if the software is good, that's the purpose of unit tests (or, some higher level integration tests) which may be core-dependent at most, ideally not
2024-04-24T15:25:55 < qyx> I want to test sw-hw integration for a particular hardware platform
2024-04-24T15:26:55 < qyx> if the code manages to init peripherals properly, boot properly, utilize low power modes, etc.
2024-04-24T15:27:29 < Ecco> makes sense. but then it's most likely very device-specific, right?
2024-04-24T15:29:45 < karlp> indeed.
2024-04-24T15:30:41 < Ecco> I'm looking at the datasheet of a MEMS accelerometer by ST
2024-04-24T15:30:49 < Ecco> namely, lis2dw12
2024-04-24T15:30:56 < Ecco> It has some detection functions
2024-04-24T15:31:11 < Ecco> Paragraph 3.2.4 is titled "Activity/Inactivity, Android stationary/motion-detection functions"
2024-04-24T15:31:20 < Ecco> What does "Android" mean here? The mobile OS?
2024-04-24T15:31:24 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-24T15:32:58 < jpa-> Ecco: yeah, android has some specs on what the platform IMU should provide to support low-power motion wake
2024-04-24T15:33:07 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-24T15:40:31 < qyx> now tell me why spinx numbers my TOCs despite I removed all numbering params
2024-04-24T15:40:52 < qyx> after I did that, it stopped numbering *the first* subdoc, the rest stays numbered
2024-04-24T15:41:25 < qyx> where was some sort of cache, wasn't there
2024-04-24T15:43:46 < qyx> of course..
2024-04-24T15:56:56 < Ecco> jpa-: oh, ok!
2024-04-24T16:42:26 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-24T16:57:06 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-24T16:57:16 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-24T16:57:22 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-24T17:07:36 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-24T17:18:38 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-24T17:23:41 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-24T19:05:09 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-24T20:05:33 -!- qyx [~qyx@84.245.121.111] has quit [Ping timeout: 272 seconds]
2024-04-24T20:06:45 -!- qyx [~qyx@84.245.121.81] has joined ##stm32
2024-04-24T20:25:10 -!- Sadale [~Sadale@user/sadale] has quit [Quit: Bye!]
2024-04-24T20:25:30 -!- Sadale [~Sadale@user/sadale] has joined ##stm32
2024-04-24T20:48:15 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has joined ##stm32
2024-04-24T20:50:41 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-4463-2857-bc85-6fc6.fixed6.kpn.net] has joined ##stm32
2024-04-24T20:54:53 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-24T21:00:39 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-24T21:52:42 < qyx> zyp: do you have specs of your 2240 m.2 module handy?
2024-04-24T22:26:07 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-24T22:32:57 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-24T22:44:32 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-24T22:45:00 < Laurenceb_> do audiophools have a way to clean up mains waveforms?
2024-04-24T22:45:08  * Laurenceb_ needs clean 50Hz mains
2024-04-24T22:46:56 < Laurenceb_> I've got a microwave oven transformer hooked up to a motor to look for partial discharge, but the mains is too noisey
2024-04-24T22:48:36 < BrainDamage> there's standard AC filter modules
2024-04-24T22:49:01 < BrainDamage> not perfect, but should get you ~30dB noise suppression to 10MHz or so
2024-04-24T23:03:19 < lemmi> Laurenceb_: oh yes of course. search for power conditioners
2024-04-24T23:03:34 < Laurenceb_> ok
2024-04-24T23:03:55 < Laurenceb_> yeah I'm looking for current spikes from partial discharge in the motor, but there is too much noise
2024-04-24T23:04:03 < lemmi> ranges from simple filters to completely active re-generated 50hz sine with ultra low thd for best clarity and >5 figures prices
2024-04-24T23:04:30 < Laurenceb_> I made a supply with reverse driven toroidal transformer and 30V opamp with bjt gain stage, but it cant drive a whole motor with 2.5kV AC
2024-04-24T23:04:56 < Laurenceb_> so I swapped to microwave oven transformer and 50Hz mains
2024-04-24T23:06:26 < Laurenceb_> looks like triac switching events on the 50Hz waveform
2024-04-24T23:08:33 < Laurenceb_> my issue is that partial discharge looks like spikes at fixed phases relative to the AC waveform, but the triac switching is also fixed, so its hard to tell apart
2024-04-24T23:12:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-24T23:31:42 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-24T23:59:35 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
--- Day changed to huhti 25 2024
2024-04-25T00:00:45 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T00:20:23 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-25T00:39:36 < karlp> tossing up replacing the small odroid that had a usb3 external harddrive (which has failed) with a small x86 box.
2024-04-25T00:39:59 < karlp> it's definitely more power consumption, but the price isn't hugely different, and you get a lot more for it.
2024-04-25T00:40:01 < karlp> more space though.
2024-04-25T00:40:23 < karlp> just not sure that usb attached harddrives for the storage is really a great idea...
2024-04-25T00:40:42 < karlp> I was probably using this thing a bit more than it was intended, it was probably meant to be archving and moving shit, not... main storage.
2024-04-25T00:40:55 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-25T00:42:43 < karlp> hrm, I definitely don't want fans though
2024-04-25T00:45:33 < qyx> quadcore intel low power shit is same power consumption as odroid m1
2024-04-25T00:45:59 < qyx> odroid h2+ (amd64) is < 4 W, odroid m1 is ~3.2 W
2024-04-25T00:46:38 < qyx> both fanless
2024-04-25T00:47:05 < qyx> m1 is more stable
2024-04-25T00:47:14 < qyx> although I am using h2+ with a gui
2024-04-25T00:47:32 < qyx> it sometimes turns off when gpu load is high, probably some power supply something
2024-04-25T00:47:57 < qyx> (ie. if I keep streetview open for > 10 minutes)
2024-04-25T00:54:53 < karlp> hrm, I think I was finding the wrong x86 then.
2024-04-25T00:57:29 < karlp> yeah, m1-s looks ok, 64GB emmc included, but if want like 2TB storage for movies/vidoes, it's either back to usb external, or a shitty pcie2 nvme.
2024-04-25T00:58:48 < karlp> hrm, m1 ok, with a bog standard 3.5" drive
2024-04-25T00:58:59 < qyx> yes it has sata
2024-04-25T00:59:08 < qyx> what's shitty about nvme?
2024-04-25T00:59:34 < qyx> it can safely saturate 1gbit ethernet
2024-04-25T01:00:45 < qyx> check firmware images for m1 before you buy..
2024-04-25T01:00:57 < qyx> the original one is too old for anything serious
2024-04-25T01:01:16 < qyx> I found some newer debianz and upgraded to bootworm
2024-04-25T01:01:19 < qyx> *k
2024-04-25T01:03:06 < karlp> no, I mean, the m1s, it cuts down the pcie socket.
2024-04-25T01:03:36 < karlp> an m1, with an older spare smaller nvme as boot might be fine, then a sata 3.5" rust, and use usb drives for actual backup, 
2024-04-25T01:03:51 < karlp> feck, so many damn options.
2024-04-25T01:04:54 < qyx> idk if it boots from the nvme
2024-04-25T01:04:59 < qyx> I am booting using microsd
2024-04-25T01:05:09 < qyx> but it has that petitboot thing on spi nor
2024-04-25T01:05:16 < qyx> so maybe it can boot nvme using it
2024-04-25T01:05:53 < qyx> I would buy a m1-like thing with 4x m.2 nvme, single lane would be enough
2024-04-25T01:09:36 < karlp> hrm, h4+ looks ok.
2024-04-25T01:10:08 < karlp> I mean, yeah, slow nvme is still super fast compared to rust I guess.
2024-04-25T01:12:22 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-4463-2857-bc85-6fc6.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-25T01:17:02 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-25T01:38:49 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-25T01:49:27 < nomorekaki> there is new good movie from funland steff
2024-04-25T01:51:45 < nomorekaki> relentlessly violent and deliriously gory Sisu https://www.imdb.com/title/tt14846026/
2024-04-25T02:00:05 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-25T02:00:42 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T02:08:14 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-25T02:55:07 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Killed (NickServ (GHOST command used by Spirit5324))]
2024-04-25T02:55:12 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-25T03:33:37 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-25T05:11:32 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Remote host closed the connection]
2024-04-25T05:11:54 -!- duude__ [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-25T07:17:34 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-25T07:23:35 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T07:33:35 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-25T08:25:45 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-9a1-2341-388f-382e.fixed6.kpn.net] has joined ##stm32
2024-04-25T09:13:38 < Steffanx> Let me check it out one day, nomorekaki :)
2024-04-25T09:46:13 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-25T10:03:17 < ventYl> PaulFertser: do you have any tip on where could GDB message "Reply contains invalid hex digit 79" originate from if gdb is used with openocd 0.12 and cmsis-dap compatible probe?
2024-04-25T10:04:03 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-25T10:04:46 < zyp> it's gdb complaining about a reply from openocd
2024-04-25T10:04:58 < zyp> you can try «set debug remote 1» to see what that's about
2024-04-25T10:05:35 < qyx> whoa openocd 0.12, are you from the future
2024-04-25T10:05:54 < qyx> wait what, I am on 0.12 too
2024-04-25T10:06:07 < qyx> when did that happen
2024-04-25T10:06:18 < qyx> I must have been doing something unusual, updates maybe
2024-04-25T10:06:42 < ventYl> zyp: thx, I tried set verbose on but that yielded nothign
2024-04-25T10:12:01 < ventYl> i am no smarter
2024-04-25T10:16:50 < ventYl> nah, I'll throw this raspberry crap into trashcan
2024-04-25T10:51:36 < jpa-> ventYl: maybe your gdb is too old
2024-04-25T10:59:59 < ventYl> jpa-: yeah, plausible but unfortunate, as the newer GDBs from ARM GNU toolchain are built with dependency on python, which they don't distribute
2024-04-25T11:00:07 < ventYl> so the GDB there is generally unusable
2024-04-25T11:04:19 < zyp> stop using toolchain bundled GDBs, use a multiarch gdb from your distro or something
2024-04-25T11:04:24 < zyp> e.g. gdb-multiarch in debian
2024-04-25T11:11:49 < ventYl> unfortunately mine's not multiarch
2024-04-25T11:15:20 < PaulFertser> ventYl: that option is supposed to give you essential info about what command and response exactly makes it fail
2024-04-25T11:15:34 < PaulFertser> ventYl: and yes, just use distro-packaged gdb-multiarch in general.
2024-04-25T11:15:55 < PaulFertser> Usually odd errors like that were from using incompatible GDB from a different arch.
2024-04-25T11:28:09 < ventYl> GDB configure is trying to convince me, that it doesn't know --enable-targets switch
2024-04-25T11:28:12 < ventYl> fuck this shiw
2024-04-25T11:33:25 < PaulFertser> Why do you want to self build it and not use what distro has?
2024-04-25T11:35:57 < zyp> maybe he's using LFS
2024-04-25T11:38:25 < qyx> :>
2024-04-25T11:38:49 < jpa-> i just used deadsnakes repo to install python3.8
2024-04-25T11:39:05 < jpa-> gdb-multiarch keeps crashing for me when disassembling specific instructions
2024-04-25T11:39:50 < jpa-> http://paste.dy.fi/15v/plain  such quality
2024-04-25T11:40:21 < ventYl> PaulFertser: because distro doesn't have gdb-multiarch
2024-04-25T11:40:28 < jpa-> which distro?
2024-04-25T11:40:29 < zyp> jpa-, what arch is that?
2024-04-25T11:40:35 < ventYl> that's why used toolchain-provided GDB
2024-04-25T11:40:39 < ventYl> jpa-: slackware
2024-04-25T11:41:03 < jpa-> zyp: gdb-multiarch  12.1-0ubuntu1~22.04, target is STM32F4
2024-04-25T11:41:30 < PaulFertser> radare2 crashes all the time, I even saw a talk by primary author where it crashed like 3 times during the demo. But reversers still love it.
2024-04-25T11:43:25 < PaulFertser> Building GDB from source manually is usually easy enough.
2024-04-25T11:43:26 < zyp> jpa-, hmm, that looks really weird
2024-04-25T11:43:29 < ventYl> funny enough, the configure.ac knows that switch, so it will be something between the heaven and autohell
2024-04-25T11:43:43 < zyp> really curious why you're even getting this: https://sourceware.org/gdb/current/onlinedocs/gdb.html/Breakpoint_002drelated-Warnings.html
2024-04-25T11:43:59 < zyp> also, shouldn't the addrs have LSB set?
2024-04-25T11:44:18 < PaulFertser> https://forums.raspberrypi.com/viewtopic.php?t=364475 <- ventYl ready made for slackware :)
2024-04-25T11:45:05 < zyp> ventYl, isn't this bullshit kinda what you sign up for when you choose slackware?
2024-04-25T11:45:37 < qyx> I tend to make mistakes too :P
2024-04-25T11:45:46 < qyx> he should have used debian
2024-04-25T11:47:06 < PaulFertser> Probably those who liked Slackware back then can switch to OpenBSD now.
2024-04-25T11:47:38 < ventYl> zyp: in exc hange for not having to deal with other kind of bullshit
2024-04-25T11:47:50 < ventYl> it turns out that configure for GDB is kind-of buggy
2024-04-25T11:47:56 < ventYl> or whatever
2024-04-25T11:48:04 < zyp> maybe
2024-04-25T11:48:15 < ventYl> GDB is Dante's kind of autohell
2024-04-25T11:48:22 < ventYl> so configure runs other configures
2024-04-25T11:48:34 < ventYl> and only some of them recognize --enable-targets
2024-04-25T11:48:42 < zyp> I've settled on just picking debian for anything, seems like the least bullshit option
2024-04-25T11:48:45 < ventYl> hopefully the main configure passess all its arguments downstream
2024-04-25T11:49:08 < ventYl> so if you run ./configure --enable-targets=all --enable-option-checking you get warning that --enable-targets is not recognized
2024-04-25T11:51:53 < PaulFertser> zyp: what's your opinion about Nix(OS)? It looks like a sane approach to problems that people tend to solve with "docker images" and such.
2024-04-25T11:52:20 < zyp> never used it
2024-04-25T11:53:02 < ventYl> ok, so even gdb 14.2 is falling apart
2024-04-25T11:53:14 < PaulFertser> It's a declarative way to specify final reproducible state for a software configuration, and it allows arbitrary number of such configurations to coexist on same system.
2024-04-25T11:53:36 < zyp> PaulFertser, yeah, I like the idea of it, haven't had a reason to try it out yet
2024-04-25T11:53:45 < PaulFertser> ventYl: falling as in not configuring?
2024-04-25T11:54:24 < ventYl> PaulFertser: no, that warning was bogus. I've built GDB 14.2 multiarch, but it has problems talking to openocd / yapicoprobe stack
2024-04-25T11:54:46 < PaulFertser> ventYl: please show the full log connecting after set debug remote 1
2024-04-25T11:58:06 < jpa-> zyp: i think gdb always shows breakpoint addresses without bottom bit
2024-04-25T12:01:25 < ventYl> PaulFertser: https://paste.debian.net/1315134/
2024-04-25T12:04:14 < PaulFertser> ventYl: there's no unexpected hex or similar error in there
2024-04-25T12:05:38 < ventYl> it starts later after I try to load and run the software
2024-04-25T12:06:11 < PaulFertser> And also this log tells you have "mon exit" in the script.
2024-04-25T12:06:28 < ventYl> one unusual thing happening now is that last three lines repeat
2024-04-25T12:07:15 < ventYl> there's only monitor reset halt, no mon exit
2024-04-25T12:10:45 < PaulFertser> OK, probably I was confused by "start_remote_1: exit"
2024-04-25T12:11:17 < PaulFertser> Do you take into account that if you want right data then you have to do "mon gdb_sync" "stepi" after "mon reset halt"?
2024-04-25T12:12:49 < ventYl> PaulFertser: this very script was working for years with older versions of gdb, openocd and picoprobe
2024-04-25T12:13:07 < ventYl> PaulFertser: https://paste.debian.net/1315136/ this happens after load and run
2024-04-25T12:13:38 < ventYl> enter / exit repeat indefinitely. long row of dots (.........) replaces actual data to shorten the log
2024-04-25T12:14:58 < ventYl> and yeah, usually, the first command after launching gdb using this script was load followed by run, so no need to do gdb_sync
2024-04-25T12:15:02 -!- srk [~sorki@user/srk] has quit [Quit: ZNC 1.8.1 - https://znc.in]
2024-04-25T12:15:09 < zyp> jpa-, yeah, you're right
2024-04-25T12:15:19 < zyp> still curious why gdb tries changing the address
2024-04-25T12:15:41 < zyp> is it something like an ite set?
2024-04-25T12:15:51 < ventYl> LSB?
2024-04-25T12:16:02 < zyp> or something corrupt that's not even valid instructions?
2024-04-25T12:18:49 < PaulFertser> ventYl: hm, OpenOCD shouldn't be saying that in response to qOffsets
2024-04-25T12:19:31 < PaulFertser> It has hardcoded reply Text=0;Data=0;Bss=0
2024-04-25T12:19:53 < ventYl> isn't this response probe-dependent?
2024-04-25T12:20:41 < PaulFertser> Not to qOffsets.
2024-04-25T12:21:34 < PaulFertser> 1 was an answer to qAttached
2024-04-25T12:22:27 < PaulFertser> QC0 is the answer to qC
2024-04-25T12:23:24 < PaulFertser> wth GDB sends "Sending packet: $R0#82" and then without waiting for reply "Sending packet: $qC#b4", that surely gets it out of sync.
2024-04-25T12:25:11 < PaulFertser> And BTW, vRun is supported since 0.11.0
2024-04-25T12:25:28  * ventYl shrugs
2024-04-25T12:25:35 < ventYl> both are upstream vanilla release sources
2024-04-25T12:25:57 < PaulFertser> S00 is reply to vRun
2024-04-25T12:26:40 < ventYl> there's one empty reply packet
2024-04-25T12:26:51 < ventYl> which probable made gdb confused
2024-04-25T12:27:39 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-25T12:28:20 < PaulFertser> Probably seeing debug log from OpenOCD too would help to understand how it happens.
2024-04-25T12:29:22 < PaulFertser> Looks like vRun gets a reply that target was restarted but the reply is empty as you use no RTOS awareness.
2024-04-25T12:33:10 < ventYl> there is no RTOS in that image
2024-04-25T12:33:16 < ventYl> at least no recognized RTOS
2024-04-25T12:33:38 < ventYl> http://digital-nomad.sk/openocd.log OpenOCD log from similar but different run of GDB
2024-04-25T12:42:09 < PaulFertser> ventYl: odd thing. Can you probably workaround it by using "remote" rather than "extended-remote"?
2024-04-25T12:44:39 < ventYl> The remote target does not support run
2024-04-25T12:44:40 < ventYl> eh
2024-04-25T12:44:56 < ventYl> OK, what is some reasonably priced probe which works out of the box with openocd?
2024-04-25T12:46:46 < qyx> stlink
2024-04-25T12:47:10 < qyx> zyps orbtrace apparently
2024-04-25T12:47:34 < ventYl> stlink sucks because it is limited to STM cpus
2024-04-25T12:47:40 < PaulFertser> That's strange OpenOCD log shows that it received vRun, answere S00 and only after that R0 was received.
2024-04-25T12:47:51 < qyx> no, only stlink v3
2024-04-25T12:48:02 < PaulFertser> But GDB thought there was a dummy response
2024-04-25T12:48:06 < qyx> I flashed TI and microchip stuff with stlink v2.1
2024-04-25T12:48:15 < PaulFertser> ventYl: yes, you do not really need "run" command for anything do you?
2024-04-25T12:48:27 < PaulFertser> ventYl: since you still have "mon reset halt" there.
2024-04-25T12:48:35 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-25T12:48:47 < PaulFertser> ventYl: this part of code is adapter agnostic, I'd expect same behaviour with any probe.
2024-04-25T12:49:16 < zyp> I don't use openocd much myself, so I can't speak for that, and I'm obviously biased so I'd suggest orbtrace
2024-04-25T12:50:19 < ventYl> PaulFertser: don't you have to "reboot" the CPU after you load new program?
2024-04-25T12:50:46 < ventYl> if not initially, then after subsequent realoads
2024-04-25T12:50:58 < PaulFertser> ventYl: when you do "load" the default scripts already perform "reset init" before and "reset halt" after so you can just "c(ontinue)" right after load.
2024-04-25T12:51:28 < PaulFertser> ventYl: and even if you do need to reboot you can use it with "mon reset halt" and then "c".
2024-04-25T12:51:46 < PaulFertser> Doing "run" after "mon reset halt" makes little sense if at all.
2024-04-25T12:52:05 < PaulFertser> Doing "start" might be handy since it automatically adds a temporary breakpoint on main().
2024-04-25T12:52:12 < ventYl> both switching to "remote" and removing "mon reset halt" ends up doing basically the same.
2024-04-25T12:52:23 < PaulFertser> But if somehow in this GDB version run is problematic then just do not use it for now.
2024-04-25T12:52:41 < PaulFertser> ventYl: just do not use "run" or "start" manually at all.
2024-04-25T12:52:58 < ventYl> I can load and run/continue, but then I get `The target is not responding to interrupt request.`
2024-04-25T12:54:35 < PaulFertser> wtf, looks like this GDB is broken wrt remote serial protocol.
2024-04-25T12:55:52 < ventYl> I can try to build 14.1
2024-04-25T12:55:53 < PaulFertser> ventYl: hm, are you using the pipe mode rather than tcp connection?
2024-04-25T12:56:26 < PaulFertser> ventYl: please retry with regular TCP.
2024-04-25T12:56:41 < PaulFertser> Might have been some regression with pipe, not many test it I guess.
2024-04-25T12:57:00 < ventYl> yes, I am using pipe mode
2024-04-25T12:58:37 < ventYl> PaulFertser: ACK, using TCP it works normally
2024-04-25T13:00:35 < PaulFertser> ventYl: so you have two non-SMP cores there and both connected to a single "pipe", guess that's what confuses everything.
2024-04-25T13:01:51 < ventYl> are they really non-SMP?
2024-04-25T13:02:29 < ventYl> also, I've seen some armv7m_ prefixed functions in the OpenOCD log
2024-04-25T13:02:45 < ventYl> dunno if that's a bug or v6m and v7m both use same code path
2024-04-25T13:03:50 < PaulFertser> ventYl: SMP or not also depends on your target OpenOCD config. If you want to treat them as SMP then only one GDB connection is used. If you want to run different code on them then two GDB sessions are used.
2024-04-25T13:04:07 < PaulFertser> ventYl: v6m was developed after v7m so borrows stuff.
2024-04-25T13:05:11 < PaulFertser> ventYl: you can see different options to handle cores in the current config selected by USE_CORE
2024-04-25T13:05:46 < PaulFertser> But that was added post 0.12.0
2024-04-25T13:05:59 < PaulFertser> You can just grab this newer target config though, I think it's compatible.
2024-04-25T13:07:44 < ventYl> as there are two connections I guess that 0.12.0 config treats them not as SMP
2024-04-25T13:07:48 < ventYl> while they in fact are SMP
2024-04-25T13:08:39 < ventYl> as of now I basically don't care about the second core as long as it doesn't cause any troubles
2024-04-25T13:08:42 < PaulFertser> ventYl: it's a user preference, not a fact
2024-04-25T13:09:20 < PaulFertser> ventYl: because even when you have cores that are really SMP you might be using them separately.
2024-04-25T13:09:56 < ventYl> interesting if it really works, e.g. resets won't affect the other core
2024-04-25T13:10:28 < ventYl> so, the pipe mode is confused because they are not configured as SMP?
2024-04-25T13:11:03 < PaulFertser> ventYl: if you do "set USE_CORE 1" prior to sourcing 0.12.0 target config you'll have just the first core enabled and used and that should work with the pipe.
2024-04-25T13:11:41 < PaulFertser> ventYl: yes, it would seem so, the pipe mode is broken when more than one GDB server is created.
2024-04-25T13:12:28 < ventYl> I wonder how raspberry hacked the support for both cores in
2024-04-25T13:13:11 < PaulFertser> Probably reset will actually reset both cores. OpenOCD kinda expects that system reset leads to reset of all the CPUs on the board so that's not a problem.
2024-04-25T13:14:06 < ventYl> as the very same gdb and openocd scripts were working with raspberry-hacked version of openocd
2024-04-25T13:14:27 < PaulFertser> Doesn't matter.
2024-04-25T13:15:13 < PaulFertser> USE_CORE 0 actually for your case
2024-04-25T13:15:16 < ventYl> kind-of does as the SMP was working with pipe mode there. so the implementation had to circumvent it somehow
2024-04-25T13:15:48 < PaulFertser> ventYl: you can have SMP working properly with pipe and you will. If you enable it!
2024-04-25T13:16:26 < PaulFertser> And current (not 0.12.0 old but current) config actually enables SMP by default!
2024-04-25T13:19:47 < ventYl> yeah I am reading it right now
2024-04-25T13:26:44 < ventYl> hm, the script in git is older than the 0.12.0 release
2024-04-25T13:26:49 < ventYl> they should both be identical
2024-04-25T13:29:37 < jpa-> zyp: arm-none-eabi-gdb decoded it fine, too bad i apparently didn't take note of what it was
2024-04-25T13:30:29 < ventYl> ah, the 0.12 tree has been frozen back in january
2024-04-25T13:37:16 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-25T13:38:25 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-25T13:38:55 < ventYl> PaulFertser: thanks for cooperation. I would never been able to dig in that deep into gdb/openocd/cmsis-dap stack to deal with it
2024-04-25T13:41:31 < PaulFertser> ventYl: welcome
2024-04-25T13:43:30 < ventYl> next boss will be the Pico-SDK
2024-04-25T14:16:10 -!- BrainDamage [~m-t6k752@user/BrainDamage] has quit [Read error: Connection reset by peer]
2024-04-25T14:20:18 -!- BrainDamage [~m-t6k752@user/BrainDamage] has joined ##stm32
2024-04-25T14:52:06 < PaulFertser> ventYl: https://termbin.com/n4ly but I have no way to test
2024-04-25T15:22:01 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-9a1-2341-388f-382e.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-25T16:15:04 < PaulFertser> ventYl: pushed to https://review.openocd.org/c/openocd/+/8222 . Do you want Reported-By tag there? Would be good if somebody tested.
2024-04-25T16:16:31 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-25T16:23:34 < PaulFertser> Probably nice fellows here can find someone to review my LPC55xx flash driver support?
2024-04-25T16:36:00 < Ecco> Hi PaulFertser :) I ran into an openocd bug the other day: on STM32WBA using an ST-Link V3, openocd incorrectly probes the TPIU
2024-04-25T16:36:46 < Ecco> Namely, this line here https://github.com/openocd-org/openocd/blob/04154af5d6cd5fe76a2583778379bdacb5aa6fb0/src/target/arm_tpiu_swo.c#L669 reports a value of 0 even though doing a "mdw" manually yields the correct value (0xCA0 or something similar)
2024-04-25T16:39:48 < fenugrec> you guys know there's an #openocd channel right
2024-04-25T16:43:30 < qyx> this is just a cross channel cooperation
2024-04-25T16:45:23 < PaulFertser> Ecco: that's odd, probably it probes it at wrong time
2024-04-25T16:45:41 < PaulFertser> Ecco: I'm not in a position to debug it, you have that WBA board there and I do not.
2024-04-25T16:51:00 < Ecco> Indeed
2024-04-25T16:51:08 < Ecco> Unfortunately I don't know jack about openocd :)
2024-04-25T16:51:16 < Ecco> I tried
2024-04-25T16:51:28 < Ecco> But couldn't come up with a fix :-(
2024-04-25T16:52:10 < Ecco> fenugrec: I didn't know that. Sorry!
2024-04-25T16:57:06 < PaulFertser> Ecco: is it a multi-core device? Probably the probe in tpiu code is done on the wrong target and that makes the difference?
2024-04-25T16:58:20 < Ecco> It's a single-core M33 device
2024-04-25T16:58:51 < PaulFertser> Hm, I'd be looking at -d3 log to see what might be the difference between reading it from the code and with mdw.
2024-04-25T16:59:09 < PaulFertser> Probably the target state is somehow different.
2024-04-25T16:59:15 < Ecco> ok, I need to prepare a better bug report I guess
2024-04-25T17:07:55 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-25T17:14:48 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-25T19:02:13 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-25T19:28:06 -!- Netsplit *.net <-> *.split quits: skz81_, josuah, boB_K7IQ, Spirit532, jmcgnh, mid-kid, dobson, emeb_mac, aandrew, haritz,  (+15 more, use /NETSPLIT to show all of them)
2024-04-25T19:32:23 -!- Netsplit over, joins: Kamilion, russell--, mid-kid, nohit, jmcgnh, jbo, Livio, emeb_mac, Spirit532, Sadale (+9 more)
2024-04-25T19:33:35 -!- josuah [~josuah@46.23.94.12] has joined ##stm32
2024-04-25T19:33:35 -!- akaWolf [~akaWolf@akawolf.org] has joined ##stm32
2024-04-25T19:33:35 -!- sugarbeet [~barbas@81.4.123.134] has joined ##stm32
2024-04-25T19:33:35 -!- dobson [~dobson@static.38.6.217.95.clients.your-server.de] has joined ##stm32
2024-04-25T19:33:35 -!- noarb [~noarb@user/noarb] has joined ##stm32
2024-04-25T19:33:35 -!- skz81_ [~skz81@vps-68d3ea17.vps.ovh.net] has joined ##stm32
2024-04-25T20:01:55 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-25T20:03:18 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T20:49:33 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-14ab-47a9-f097-8f0c.fixed6.kpn.net] has joined ##stm32
2024-04-25T20:57:50 -!- Livio [~livio@user/livio] has quit [Ping timeout: 245 seconds]
2024-04-25T21:00:15 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-25T21:13:37 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-25T21:14:33 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T21:29:41 < ventYl> PaulFertser: yeah, you can link me
2024-04-25T21:39:14 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-25T22:14:35 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-25T22:16:03 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-25T22:52:50 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 245 seconds]
2024-04-25T23:03:30 -!- boB_K7IQ [~boB_K7IQ@184-98-176-228.phnx.qwest.net] has joined ##stm32
--- Day changed pe huhti 26 2024
2024-04-26T00:01:59 -!- boB_K7IQ [~boB_K7IQ@184-98-176-228.phnx.qwest.net] has quit [Ping timeout: 272 seconds]
2024-04-26T00:34:50 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-26T01:00:05 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T01:01:18 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T01:13:33 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-26T01:14:45 -!- qyx [~qyx@84.245.121.81] has quit [Ping timeout: 268 seconds]
2024-04-26T01:16:09 -!- qyx [~qyx@84.245.120.234] has joined ##stm32
2024-04-26T01:39:31 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-14ab-47a9-f097-8f0c.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-26T02:02:14 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-26T02:28:00 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-26T03:48:07 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T03:50:07 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T04:19:26 -!- System_Error [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-26T04:21:49 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T06:47:31 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T06:49:06 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-26T06:49:12 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T09:28:04 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-26T09:46:28 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-26T10:00:44 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-26T10:31:34 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-26T11:27:00 -!- Livio [~livio@user/livio] has quit [Quit: leaving]
2024-04-26T11:35:54 -!- martinmoene [~martinmoe@132.229.46.129] has quit [Read error: Connection reset by peer]
2024-04-26T11:42:08 -!- martinmoene [~martinmoe@132.229.46.129] has joined ##stm32
2024-04-26T13:29:14 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T13:29:52 -!- rkta [~rkta@user/rkta] has quit [Remote host closed the connection]
2024-04-26T13:30:01 -!- rkta [~rkta@user/rkta] has joined ##stm32
2024-04-26T13:30:31 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T13:44:36 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-26T13:55:45 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T14:05:03 < karlp> ffs, do uart_init, then uart_write_blocking() just like the sample code.  no output.  do it.... "later" and it works. so something somewhere _else_ is required, and I have to go and find out what in the spaghettilayercake.
2024-04-26T14:11:11 < Steffanx> Sometimes I get the idea karlp is reading my code :P
2024-04-26T14:38:05 < jpa-> don't try to take the credit for my spaghetti!
2024-04-26T14:58:20 -!- ferdna [~ferdna@user/ferdna] has joined ##stm32
2024-04-26T15:03:52 < karlp> problem is spaghetti.... but in the schematic: 
2024-04-26T15:03:54 < karlp> https://bin.jvnv.net/file/fWqDq.png
2024-04-26T15:04:34 < karlp> I'd been assuming (for months) that the rs232 port was on it's own supply.  So, on this test setup, I was explicitly _not_ powering up the other 3.3v section (it  consumes a lot of other power and is out of scope)
2024-04-26T15:04:55 < karlp> but... someone tied pin 20 there to the 3.3v supply, not the rs232 3.3v supply...
2024-04-26T15:05:08 < karlp> I've spent far far too much time on that, chasing code down...
2024-04-26T15:17:54 < PaulFertser> But you get professional TVS protection on the exposed lines
2024-04-26T15:22:16 < jpa-> karlp: it's the f***-off pin
2024-04-26T16:11:10 < Steffanx> Lol
2024-04-26T16:47:57 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-26T16:58:00 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-26T17:03:05 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T17:21:55 -!- josuah [~josuah@46.23.94.12] has quit [Remote host closed the connection]
2024-04-26T17:26:25 -!- ferdna [~ferdna@user/ferdna] has quit [Quit: Leaving]
2024-04-26T18:10:10 < karlp> aight, got the demo code goign to stop modes on my hardware, and it happily goes to sleep with the debugger connected.
2024-04-26T18:10:26 < karlp> just can't seem to port it to my code. but, I'll call it progress.
2024-04-26T18:10:40 < karlp> my code doing "the same thing" just wakes up immediately if the debugger is attached.
2024-04-26T18:11:43 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-26T18:28:45 -!- josuah [~josuah@46.23.94.12] has joined ##stm32
2024-04-26T19:36:22 < mawk> zgeg
2024-04-26T20:00:35 < Steffanx> Language mawk
2024-04-26T20:17:30 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T20:19:36 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T20:38:39 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a5e5-150b-393b-1853.fixed6.kpn.net] has joined ##stm32
2024-04-26T20:44:58 -!- yukam [~yukam@user/yukam] has quit [Ping timeout: 264 seconds]
2024-04-26T20:53:06 -!- yukam [~yukam@user/yukam] has joined ##stm32
2024-04-26T21:36:34 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-26T21:59:01 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-26T22:21:40 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T22:22:44 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T22:36:49 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-26T22:37:29 -!- System_Error [~SystemErr@user/systemerror] has joined ##stm32
2024-04-26T23:04:33 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 268 seconds]
2024-04-26T23:04:49 < mawk> you don't know what it means Steffanx 
2024-04-26T23:05:00 < Steffanx> Yes i do mawk 
2024-04-26T23:05:13 < mawk> what does it mean
2024-04-26T23:05:18 < Steffanx> zhe françoise slang
2024-04-26T23:05:38 < mawk> for what
2024-04-26T23:06:03 < mawk> it's slang from arabic I think 
2024-04-26T23:07:09 < Steffanx> piemel in je oor, mawk
2024-04-26T23:07:16 < mawk> :(
2024-04-26T23:07:23 < mawk> just piemel
2024-04-26T23:07:25 < mawk> no oor business 
2024-04-26T23:07:30 < Steffanx> i know..
2024-04-26T23:09:29 < Steffanx> So how's life in the nieuwe drachten mr mawk ?
2024-04-26T23:09:45 < Steffanx> ready for koningsdag?
2024-04-26T23:15:38 < mawk> yes
2024-04-26T23:15:53 < mawk> there is some kind of festival on the city hall plaza
2024-04-26T23:16:04 < mawk> koningsnacht or whatever
2024-04-26T23:30:46 < Steffanx> Fun times. Maybe you can sell some 
2024-04-26T23:33:51 < Steffanx> Stuff on the street 
--- Day changed la huhti 27 2024
2024-04-27T00:36:03 -!- dreamcat4 [uid157427@id-157427.hampstead.irccloud.com] has joined ##stm32
2024-04-27T01:58:52 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-27T01:59:50 -!- System_Error [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-27T02:05:14 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-27T02:37:25 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 245 seconds]
2024-04-27T02:40:23 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a5e5-150b-393b-1853.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-27T02:52:03 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-27T07:42:37 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-27T10:05:43 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Ping timeout: 260 seconds]
2024-04-27T10:24:44 < jpa-> sometimes i hear the claim that chinese PCB fabs will steal your design.. then when i look at my big box of failed PCB prototypes, I pity the fool that would do so
2024-04-27T10:38:54 < antto> they'd only steal mine
2024-04-27T10:48:16 < antto> https://i.imgur.com/BZKmHm2.png
2024-04-27T10:49:19 < antto> but when they do, they'll run out of money
2024-04-27T10:50:01 < antto> if .cn bankrupts, it'd be partially my fault
2024-04-27T11:11:45 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-27T11:41:12 -!- Livio [~livio@user/livio] has quit [Ping timeout: 268 seconds]
2024-04-27T12:21:38 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-27T12:24:18 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-3449-8a28-e0f3-aa4e.fixed6.kpn.net] has joined ##stm32
2024-04-27T12:33:31 < jpa-> qyx: i think i have successfully burned out the last BQ25756 that i will ever buy
2024-04-27T12:34:08 < jpa-> qyx: some thoughts to aid you in your quest: 1) the REGN regulator is supposedly over-current protected, but appears to be very picky and burn at the slightest excuse you give it
2024-04-27T12:34:53 < jpa-> 2) there is something weird going on with the CE pin, REGN and the I2C access; text says CE affects REGN regulator but suggests that I2C will work without it; block diagram seems to indicate that REGN would be on regardless of CE
2024-04-27T12:36:04 < jpa-> 3) the chip easily gets into weird oscillation modes when operated with a current-limited power supply; even if it takes only a few mA when idle, if you try to power it up with 20 mA current limit the REGN gets into some weird power-up loop that seems to burn it out after a short time
2024-04-27T12:36:58 < jpa-> 4) unconnected pins from rework soldering mishaps seem to easily break something
2024-04-27T12:37:55 < jpa-> it could be that there is something stupid that i have done and just never found out, but i'm giving up :)
2024-04-27T12:38:36 < zyp> aww
2024-04-27T12:44:17 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-3449-8a28-e0f3-aa4e.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-27T12:44:53 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-3449-8a28-e0f3-aa4e.fixed6.kpn.net] has joined ##stm32
2024-04-27T12:57:20 < Steffanx> Didn't jbo fancy this ic as well?
2024-04-27T13:02:34 < jpa-> yeah, but he probably succeeds
2024-04-27T13:03:04 < zyp> I'm kinda tempted to have a go at it, but have no use for it
2024-04-27T13:06:12 < qyx> jpa-: now I call it a challenge!
2024-04-27T13:09:25 < jpa-> i seem to always have bad luck with these smart power management chips; MAX77975 took me years to get properly soldered, and even after that the boost converter doesn't get up to rated current; client project has BQ25628 and its boost is also not meeting requirements; and then this BQ25756 which i no longer even expected anything but misery from
2024-04-27T13:16:50 < qyx> same experience with battery management bqxxx here too
2024-04-27T13:26:21 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-27T15:14:24 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 260 seconds]
2024-04-27T15:16:49 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-27T15:27:52 -!- flom84 [~flom84@user/flom84] has joined ##stm32
2024-04-27T15:36:48 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 260 seconds]
2024-04-27T15:37:25 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has joined ##stm32
2024-04-27T15:40:34 -!- flom84 [~flom84@user/flom84] has quit [Ping timeout: 256 seconds]
2024-04-27T17:02:34 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-27T17:06:54 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-27T17:50:37 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a828-aad0-7386-dadd.fixed6.kpn.net] has joined ##stm32
2024-04-27T17:53:03 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-3449-8a28-e0f3-aa4e.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-27T18:19:26 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:9d1e:9da6:5239:1385] has joined ##stm32
2024-04-27T19:03:11 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:9d1e:9da6:5239:1385] has quit [Quit: Konversation terminated!]
2024-04-27T19:07:05 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:3cb2:f8f8:9ff0:f9cd] has joined ##stm32
2024-04-27T19:29:35 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:3cb2:f8f8:9ff0:f9cd] has quit [Quit: Konversation terminated!]
2024-04-27T19:33:52 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:4a67:f238:4a1c:880d] has joined ##stm32
2024-04-27T21:36:26 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-27T21:37:33 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-27T22:31:21 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 255 seconds]
2024-04-27T23:33:52 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-89e7-dabe-7222-f66c.fixed6.kpn.net] has joined ##stm32
2024-04-27T23:36:05 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a828-aad0-7386-dadd.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
--- Day changed su huhti 28 2024
2024-04-28T00:32:02 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 268 seconds]
2024-04-28T00:34:16 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-28T00:34:59 < qyx> so, for ADC-ing to ethernet, H5 is the goto chip now?
2024-04-28T01:14:16 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-89e7-dabe-7222-f66c.fixed6.kpn.net] has quit [Ping timeout: 246 seconds]
2024-04-28T01:14:29 -!- qyx [~qyx@84.245.120.234] has quit [Ping timeout: 252 seconds]
2024-04-28T01:16:23 -!- qyx [~qyx@84.245.121.90] has joined ##stm32
2024-04-28T01:54:03 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 268 seconds]
2024-04-28T02:13:04 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-28T02:14:06 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-28T02:20:07 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:4a67:f238:4a1c:880d] has quit [Ping timeout: 260 seconds]
2024-04-28T03:44:25 -!- Livio [~livio@user/livio] has quit [Quit: leaving]
2024-04-28T05:52:03 -!- emeb_mac [~emeb_mac@ip174-72-120-238.ph.ph.cox.net] has quit [Quit: Leaving.]
2024-04-28T07:22:04 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-28T07:23:21 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-28T07:32:26 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-28T07:37:20 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-28T08:22:37 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-28T08:23:33 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-28T09:12:34 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-28T09:13:14 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-28T09:16:36 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-28T09:26:03 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has joined ##stm32
2024-04-28T09:35:17 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Ping timeout: 250 seconds]
2024-04-28T11:40:15 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-28T12:07:14 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has joined ##stm32
2024-04-28T12:10:20 -!- c10ud [~c10ud@user/c10ud] has quit [Ping timeout: 245 seconds]
2024-04-28T13:33:45 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-28T13:47:01 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:f92:38cc:25d8:72b6] has joined ##stm32
2024-04-28T14:15:35 < Steffanx> Enough ram and flash for sure. For not too much money, qyx. 
2024-04-28T14:15:47 < Steffanx> I would say it's a fine chip for ethernet
2024-04-28T14:17:50 < Steffanx> Will you use HAL code though?
2024-04-28T14:33:55 < qyx> no?
2024-04-28T14:34:47 < qyx> I wouldn't touch ST's stuff even with a 2 meter stick?
2024-04-28T14:36:25 < PaulFertser> Would you hate me for removing a failed common mode choke on SMPS input? Will I regret?
2024-04-28T14:37:08 < BrainDamage> your neighbour ham operators will hate you
2024-04-28T14:37:33 < BrainDamage> and maybe yourself if the noise creates random glitches in proximity appliances
2024-04-28T14:38:27 < PaulFertser> BrainDamage: thank you. Guess I'm safe then because I have no ham operators around and no proximity sensors AFAICT :)
2024-04-28T14:39:49 < BrainDamage> not proximity sensors, appliances in the proximity of it that sense analog voltages
2024-04-28T14:39:52 < PaulFertser> I was using this adjustable power supply for probably 20 years and just yesterday right after I thought how much I liked it it failed. No kidding.
2024-04-28T14:41:15 < PaulFertser> Apparently the manufacturer couldn't have been arsed to add some proper glue to secure heavy/high elements so the choke eventually cracked one of the leads.
2024-04-28T16:02:35 < qyx> PaulFertser: yolo
2024-04-28T16:03:11 < qyx> nobody is gonna hate you because 99.5% of people don't know what a choke is
2024-04-28T16:05:02 < qyx> BrainDamage: have you ever had a close encounter with a fluxgate magnetometer?
2024-04-28T16:05:42 < qyx> is it hard to make with a sensitivity good enough to measure earth magnetic field fluctuations?
2024-04-28T16:05:57 < qyx> (I know you do a lot of phys stuff)
2024-04-28T16:07:03 < BrainDamage> nope sorry
2024-04-28T16:10:53 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-28T16:10:59 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 252 seconds]
2024-04-28T17:15:03 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-28T17:51:19 < zyp> qyx, I recently played with fluxgate current sensors, if that counts
2024-04-28T17:52:35 < zyp> (it doesn't, because the one I got is a fully integrated thing that mainly just provides digital outputs that tells if the AC/DC leakage currents are above the thresholds)
2024-04-28T18:05:25 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has joined ##stm32
2024-04-28T18:13:06 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-28T18:13:15 < Laurenceb_> >high accuracy simulation by X NaN
2024-04-28T18:13:17 < Laurenceb_> orbital sides
2024-04-28T18:18:42 < BrainDamage> Laurenceb_ protesting https://usercontent.irccloud-cdn.com/file/pHtmxrei/1000003107.jpg
2024-04-28T18:22:32 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 256 seconds]
2024-04-28T18:24:34 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-28T18:27:43 < Laurenceb_> literally who
2024-04-28T18:28:09 < Laurenceb_> who could be responsible for this
2024-04-28T18:38:27 < antto> the famous illusionist Whodeeny
2024-04-28T18:45:41 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 240 seconds]
2024-04-28T18:48:28 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-28T19:32:51 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Ping timeout: 250 seconds]
2024-04-28T19:40:56 < qyx> I don't get those protests
2024-04-28T19:41:06 < qyx> people are protesting for the case of protesting
2024-04-28T19:43:35 < qyx> zyp: not much usable for geomagnetic storm detection it seems
2024-04-28T20:15:08 -!- joel135 [uid136450@id-136450.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
2024-04-28T20:22:03 -!- MGF_Fabio [~MGF_Fabio@2a01:e11:200d:8e90:f92:38cc:25d8:72b6] has quit [Ping timeout: 256 seconds]
2024-04-28T20:44:40 < qyx> so apparently the event onset is about 50 nT increase, that's.. a little
2024-04-28T21:06:14 < jpa-> MMC5633 gets close, it has 150 nT RMS noise at 75 Hz samplerate
2024-04-28T21:07:02 < jpa-> so with 5 second averaging 10 nT noise
2024-04-28T21:09:37 < qyx> just reading some eastern low power fluxgate magnetometer desig 
2024-04-28T21:10:10 < qyx> it draws 8 W and is powered by 9 V batteries using 7805 regulators
2024-04-28T21:12:19 < qyx> oh not eastern.. https://www.colorado.edu/center/spacegrant/content/high-precision-single-axis-fluxgate-magnetometer
2024-04-28T21:15:57 < jpa-> weird document, they don't give any lab results on how the magnetometer worked and just finish with "SD card failed; bad luck."
2024-04-28T22:09:47 < qyx> very weird and apparently size is not the key to high sensitivity
2024-04-28T22:10:13 < qyx> I have found small sensors with tiny coils having more than enough sensitivity
2024-04-28T22:10:41 < qyx> FLC-100 for example
2024-04-28T22:15:56 < BrainDamage> since you don't have to mass produce it, what I'd do is to make a differential system
2024-04-28T22:16:03 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 255 seconds]
2024-04-28T22:16:31 < BrainDamage> make 2 identical LC oscillators with variable capacitors, tune them to be in antiphase and then get the sum signal
2024-04-28T22:18:14 < qyx> and place the two coils along tne same axis, in the opposite direction?
2024-04-28T22:18:24 < BrainDamage> yep
2024-04-28T22:19:01 < qyx> will that measurethe magnetic field?
2024-04-28T22:20:48 < BrainDamage> the magnetic field will slightly shift the saturation point in the ferrites
2024-04-28T22:21:22 < BrainDamage> positive for one and negative for the other
2024-04-28T22:21:46 < BrainDamage> increasing the diversity signal
2024-04-28T22:21:55 < qyx> so the freqneeds to be low enough to go into saturation?
2024-04-28T22:22:33 < BrainDamage> they should be at the edge of saturation
2024-04-28T22:22:53 < BrainDamage> if you push them strongly in saturation, your own signal will swamp the external one
2024-04-28T22:26:23 < BrainDamage> the advantage of the differential system is that to a first order you'll be insensitive to EM noise
2024-04-28T22:49:18 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has joined ##stm32
2024-04-28T22:51:29 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-28T22:51:32 < Laurenceb_> https://nitter.poast.org/pic/media%2FGMQr4GEWAAAQD7S.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-28T22:52:16 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has quit [Ping timeout: 256 seconds]
2024-04-28T22:52:48 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has joined ##stm32
2024-04-28T22:54:04 < Laurenceb_> https://nitter.poast.org/pic/media%2FGMPvGM9XIAAZtyu.jpg%3Fname%3Dsmall%26format%3Dwebp
2024-04-28T22:54:46 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-28T23:04:40 < qyx> BrainDamage: I can now officially work on "my stuff" for a bit as an exchange for removing some items from our household to-do list, so I may try multiple approaches
2024-04-28T23:05:56 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-28T23:49:42 < Laurenceb_> wait wut
2024-04-28T23:49:44 < Laurenceb_> https://www.infineon.com/dgdl/Infineon-IMW120R007M1H-DataSheet-v01_20-EN.pdf?fileId=8ac78c8c7f2a768a017f8783822631b2&_gl=1*gyd9ao*_up*MQ..&gclid=76555b20829b1d36c60c3d0e2cbe4b4a&gclsrc=3p.ds
2024-04-28T23:49:47 < Laurenceb_> madness
--- Day changed ma huhti 29 2024
2024-04-29T00:14:35 < Laurenceb_> traction inverter in size of a phone
2024-04-29T00:14:52 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-29T00:15:38 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has joined ##stm32
2024-04-29T00:16:32 < specing> Laurenceb_: my e-bicycle's traction inverter is about the size of a phone
2024-04-29T00:35:56 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-29T00:37:09 < zyp> that's SiC
2024-04-29T00:44:43 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-29T00:54:40 < specing> how repairable are BEV car inverters, btw?
2024-04-29T00:55:00 < specing> can you send it for repair or do they just say "sorry,this will be $5k"
2024-04-29T00:55:20 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Ping timeout: 245 seconds]
2024-04-29T01:00:30 < lemmi> they'll say "sorry, this will be $9k"
2024-04-29T01:01:09 < specing> lol
2024-04-29T01:01:54 < lemmi> but seriously, i don't think they are repariable. they are sealed, filled with gel and what not
2024-04-29T01:11:35 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-a422-67ce-4a9-34a4.fixed6.kpn.net] has quit [Ping timeout: 245 seconds]
2024-04-29T01:12:51 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Quit: Bye]
2024-04-29T01:44:25 -!- phryk [~totallyno@user/phryk] has quit [Quit: ZNC 1.9.0 - https://znc.in]
2024-04-29T01:45:15 -!- phryk [~totallyno@user/phryk] has joined ##stm32
2024-04-29T01:53:25 < qyx> specing: catphish knows
2024-04-29T02:52:55 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-29T02:55:15 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-29T03:29:46 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-29T03:40:01 < nomorekaki> https://www.cameronsworld.net/
2024-04-29T07:50:26 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has joined ##stm32
2024-04-29T07:58:35 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-29T08:37:02 -!- Mangy_Dog [Mangy_Dog@user/mangy-dog/x-7397214] has quit [Remote host closed the connection]
2024-04-29T09:17:45 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-29T09:20:48 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-29T09:21:04 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-29T09:51:16 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has joined ##stm32
2024-04-29T09:58:15 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-29T10:16:19 -!- Livio [~livio@user/livio] has quit [Ping timeout: 256 seconds]
2024-04-29T10:29:21 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-29T10:56:42 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-29T12:35:56 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-29T12:37:52 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-29T13:18:43 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 260 seconds]
2024-04-29T14:37:40 -!- specing [~specing@user/specing] has quit [Quit: ZNC - https://znc.in]
2024-04-29T14:40:01 -!- specing [~specing@user/specing] has joined ##stm32
2024-04-29T15:01:46 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-29T15:57:17 -!- MGF_Fabio [~MGF_Fabio@host-217-58-46-226.business.telecomitalia.it] has quit [Ping timeout: 240 seconds]
2024-04-29T16:24:06 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-29T16:27:59 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-29T16:37:28 -!- c10ud_ [~c10ud@host-79-17-52-47.retail.telecomitalia.it] has quit [Quit: Leaving]
2024-04-29T16:37:43 -!- c10ud [~c10ud@user/c10ud] has joined ##stm32
2024-04-29T16:57:22 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-29T17:04:10 -!- jtj [~jtj@212.66.207.170] has quit [Ping timeout: 268 seconds]
2024-04-29T17:08:47 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-29T17:20:48 -!- jtj [~jtj@212.66.207.170] has quit [Read error: Connection reset by peer]
2024-04-29T17:21:08 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-29T17:23:41 -!- duude__- [~duude__@user/duude/x-4676560] has joined ##stm32
2024-04-29T17:24:35 -!- duude__ [~duude__@user/duude/x-4676560] has quit [Ping timeout: 252 seconds]
2024-04-29T17:25:46 -!- duude__- is now known as duude__
2024-04-29T17:34:29 -!- jtj [~jtj@212.66.207.170] has quit [Ping timeout: 252 seconds]
2024-04-29T18:13:02 < karlp> so... i have an app that goes to a low power state, with free rtos.  (no tickless, plain ordinary freertos)
2024-04-29T18:13:13 < karlp> if the debugger is connected, it wakes immediately.
2024-04-29T18:13:25 < karlp> if thed ebugger is not connected (and the board is power cycled) it goes to sleep properly.
2024-04-29T18:13:43 < karlp> without freertos, it works fine with or without the debugger.
2024-04-29T18:14:34 < karlp> I guess this is systick somehow being kept running because the core is still running for debug?
2024-04-29T18:16:43 < karlp> "If the debug signal is active and the system attempts to
2024-04-29T18:16:45 < karlp> enter Stop or VLPS, FCLK continues to run to support core register access.
2024-04-29T18:16:56 < karlp> that's my problem right? FLCK stays on, so systick keeps ticking,
2024-04-29T18:20:59 < jpa-> i assume this is still kinetis and not stm32?
2024-04-29T18:21:05 < karlp> I feel like this is just kinetis weirdness 
2024-04-29T18:21:17 < jpa-> on STM32 you would set the DBGMCU bits
2024-04-29T18:21:29 < karlp> I'm not concerned about extra cpu consumption of debug or anything, 
2024-04-29T18:21:36 < karlp> just... i wanted it to ... behave rationally.
2024-04-29T18:21:55 < karlp> I was comparing a freertos sample and a non-freertos sample inintially, never considered that it would be relevant.
2024-04-29T18:23:24 < jpa-> all these years with microcontrollers and you still expect them to behave rationally
2024-04-29T18:23:26 < karlp> DBGMCU looks to make it _optional_ to do what kinetis seems to do by default.
2024-04-29T18:23:43 < karlp> dbgmcu-dbg_stop even says: "FCLK and HCLK running, derived from the internal RC oscillator remaining active. If
2024-04-29T18:23:45 < karlp> Systick is enabled, it may generate periodic interrupt and wake up events.
2024-04-29T18:23:49 < karlp> so yeah, harhar, you're getting systick
2024-04-29T18:26:21 < karlp> so I guess I'm just at this hell of unplug and power cycle for every change I want to evaluate.
2024-04-29T18:26:37 < karlp> I've spent "too much" time trying to avoid that..
2024-04-29T18:26:49 < karlp> thought I was missing something, but it's feeling like no.
2024-04-29T18:28:10 < karlp> tickless=2 with some nxp sample code behaves identically, ands tickless1, that I might be able to use, doesn't even run the tasks at all, excelltn.
2024-04-29T18:32:09 < karlp> hrm, they have a note on being able to go from run to vlps via vlpr even when the debugger is connected, perhaps that will workaround sufficiently.
2024-04-29T18:32:12  * karlp tries this.
2024-04-29T18:53:36 < karlp> got it, works. wheee...
2024-04-29T18:53:43 < karlp> time to get  back to towkr then.
2024-04-29T18:54:08 < karlp> hrmm, now it automatically restarts 3 seconds later, not _quite_  right.
2024-04-29T19:20:36 < karlp> lol, presendential elections coming up, total clown show, we had at one stage like 60 people registered as potential candidates collecting signatures.
2024-04-29T19:20:44 < karlp> final hand in date, 13 submitted.
2024-04-29T19:20:51 < karlp> they only needed 1500 each.
2024-04-29T19:20:56 < karlp> one guy submitted..... _9_.
2024-04-29T19:21:15 < karlp> I mean, whoo, he got his minute on tv turning up to turn in his paperwork, but... really?! :)
2024-04-29T19:22:58 < jpa-> here they needed 20000; väyrynen claimed that postal service had delays in delivering his support cards, they give him one week extra, still got only about 10k or so
2024-04-29T19:36:58 < karlp> ok, this shit doesn't work at all. I've just wasted... all up, probably 3 days.
2024-04-29T19:37:19 < karlp> time to just unplug and plug debug.
2024-04-29T19:37:55 < karlp> yet another "bug" that I've spent a bunch of time on, with zero resolution. super thrilled.
2024-04-29T20:22:21 < zyp> the kinetis experience™
2024-04-29T20:44:25 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-29T20:46:09 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-d00c-47e5-4989-4faa.fixed6.kpn.net] has joined ##stm32
2024-04-29T20:48:48 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-29T21:04:55 -!- Livio [~livio@user/livio] has quit [Ping timeout: 260 seconds]
2024-04-29T21:06:57 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-29T21:27:52 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has joined ##stm32
2024-04-29T21:40:17 -!- nerozero [~nerozero@87.253.63.54] has quit [Ping timeout: 256 seconds]
2024-04-29T21:52:25 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 245 seconds]
2024-04-29T22:05:23 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-29T23:47:28 < karlp> well, lets see how far the "cube experience" will get me before I scream at it and say it's not work time...
2024-04-29T23:54:44 < catphish> specing: in theory repairable, in practice, nobody does it
--- Day changed ti huhti 30 2024
2024-04-30T00:06:10 -!- Livio [~livio@user/livio] has quit [Ping timeout: 245 seconds]
2024-04-30T00:11:16 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-d00c-47e5-4989-4faa.fixed6.kpn.net] has quit [Ping timeout: 255 seconds]
2024-04-30T00:11:29 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has joined ##stm32
2024-04-30T00:14:16 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-f46a-2f14-cbbe-5ba7.fixed6.kpn.net] has joined ##stm32
2024-04-30T00:18:37 -!- martinmoene_ [~Martin@77-173-84-114.fixed.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-30T01:08:39 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-f46a-2f14-cbbe-5ba7.fixed6.kpn.net] has quit [Ping timeout: 272 seconds]
2024-04-30T01:18:40 -!- qyx [~qyx@84.245.121.90] has quit [Ping timeout: 245 seconds]
2024-04-30T01:39:55 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Ping timeout: 255 seconds]
2024-04-30T01:40:32 -!- qyx [~qyx@84.245.120.154] has joined ##stm32
2024-04-30T01:45:12 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-30T01:46:24 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-30T02:00:26 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-30T02:08:54 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-30T02:55:05 -!- Spirit532 [~Spirit532@user/Spirit532] has quit [Killed (NickServ (GHOST command used by Spirit5322))]
2024-04-30T02:55:10 -!- Spirit532 [~Spirit532@user/Spirit532] has joined ##stm32
2024-04-30T03:13:17 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-30T03:14:06 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-30T04:08:13 < nomorekaki> https://www.youtube.com/watch?v=EHjgJWYj4Pk scifi short
2024-04-30T04:59:20 -!- digimer [~digimer@198.96.117.124] has joined ##stm32
2024-04-30T05:21:51 -!- SystemError [~SystemErr@user/systemerror] has quit [Remote host closed the connection]
2024-04-30T05:23:36 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-30T06:04:09 < digimer> hello, would anyone be around who might be able to help me debug an i2c issue? 
2024-04-30T06:33:23 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has quit [Ping timeout: 256 seconds]
2024-04-30T06:35:42 -!- boB_K7IQ [~boB_K7IQ@184-98-35-182.phnx.qwest.net] has joined ##stm32
2024-04-30T07:47:31 -!- nerozero [~nerozero@87.253.63.54] has joined ##stm32
2024-04-30T07:55:43 < qyx> digimer: see the last part of the topic
2024-04-30T07:56:14 < digimer> fair enough, lol. Didn't read the URL, clearly
2024-04-30T07:56:38 < digimer> I since got it working, thanks. Was bitten by the address needing to be '<< 1' shifted
2024-04-30T08:49:41 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has joined ##stm32
2024-04-30T09:16:48 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-30T09:17:00 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-30T09:47:35 -!- jtj [~jtj@212.66.207.170] has joined ##stm32
2024-04-30T09:48:20 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-30T10:21:17 -!- Livio [~livio@user/livio] has quit [Ping timeout: 240 seconds]
2024-04-30T10:34:39 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-dd09-9e15-e906-7399.fixed6.kpn.net] has joined ##stm32
2024-04-30T10:42:10 -!- SystemError [~SystemErr@user/systemerror] has quit [Ping timeout: 260 seconds]
2024-04-30T10:43:21 -!- SystemError [~SystemErr@user/systemerror] has joined ##stm32
2024-04-30T12:04:11 < karlp> sometimes it does, sometimes it doesn't....
2024-04-30T12:05:36 < nomorekaki> Steffanx: https://www.youtube.com/watch?v=dA-Y0VgJ5vM musics
2024-04-30T12:06:01 < karlp> https://direct.nuvoton.com/en/numaker-iot-ma35d0-a1 fullsize SD + microsd, quadspi nand plus parallel nand.
2024-04-30T12:38:56 -!- martinmoene__ [~Martin@2a02-a45a-96ba-1-dd09-9e15-e906-7399.fixed6.kpn.net] has quit [Ping timeout: 260 seconds]
2024-04-30T12:46:21 < karlp> nice, today openocd has decided to forget how to do freertos threads.
2024-04-30T12:54:50 -!- nomorekaki [~nomorekak@176-93-39-29.bb.dnainternet.fi] has quit [Quit: Client closed]
2024-04-30T13:33:31 < jpa-> i hate how the RTOS support is so integrated into openocd source and so dependent on RTOS version
2024-04-30T13:34:21 < jpa-> especially when GDB would be in much better position to do it and it could be a python plugin, but apparently there isn't a good way in the python API to do it
2024-04-30T13:36:19 < jpa-> https://github.com/apache/nuttx/blob/master/tools/gdb/thread.py   nuttx fakes it by just defining custom commands
2024-04-30T13:52:26 < polprog> doesnt rtos debugging need some hacks and a resident program that only halts one thread?
2024-04-30T13:53:40 < qyx> why would you halt only a single thread?
2024-04-30T13:56:35 < polprog> sometimes you want to debug just one thread and not freeze the entire device
2024-04-30T13:56:52 < polprog> i mean you can freeze all the rest of it but OS-aware debug works like that
2024-04-30T13:56:55 < jpa-> userspace debugging would do that, but i have rarely seen that done on RTOS
2024-04-30T13:57:13 < jpa-> most RTOS debug systems stop everything, but support switching threads and checking state of other threads
2024-04-30T13:57:19 < polprog> hmm
2024-04-30T13:57:57 < jpa-> you could probably make MRI into userspace debugger
2024-04-30T13:58:05 < polprog> i guess you are right, this kind of debug can be done over serial or whatever else
2024-04-30T13:58:19 < polprog> it could be in the VxWorks manual that i was reading one day
2024-04-30T14:01:39 < jpa-> i used vxworks debug system some decade ago; it was quite advanced IIRC
2024-04-30T14:01:46 < karlp> (it was my own gunfoot, I had badly merged something and removed the openocd support, yet more motivation for me to move us forward to >10.4 here the openocd support is built in.
2024-04-30T14:02:18 < karlp> it's also tmepting with my infinite motivation to rewrite oepnocd's freertos support to just... not work like that, so it doesn't need that weird shim.
2024-04-30T14:05:04 < jpa-> i wish there was some standard for rtos info, because it is mostly the same anyway; registers stored at end of stack, so you get quite far with just list of latest SP per thread
2024-04-30T14:26:44 -!- dkc [~dan@user/dkc] has quit [Remote host closed the connection]
2024-04-30T14:29:49 -!- dkc [~dan@user/dkc] has joined ##stm32
2024-04-30T14:37:38 < ventYl> polprog: that's usually not feasible as many RTOSes are simply not built reentrant
2024-04-30T14:41:29 < zyp> we had an internal RTOS at a place I used to work that had an internal gdb-stub that let you debug individual threads
2024-04-30T14:43:33 < zyp> and yeah, I'm with jpa-, it'd be a lot more convenient to have the RTOS awareness in a python plugin than in the gdbserver
2024-04-30T14:44:00 < ventYl> even for me. i had to hack my creepy RTOS stuff as custom commands
2024-04-30T14:44:28 < zyp> I'd take that approach too
2024-04-30T14:45:38 < polprog> ventYl: i didnt do much with RTOSes so i believe you
2024-04-30T14:45:51 < ventYl> I think that I wasn't able to shadow original commands entirely, or there was some similar undocumented limitation.
2024-04-30T14:51:52 < zyp> I should probably make some python functions to collect information about the state of my coroutine tasks
2024-04-30T15:05:05 < qyx> state of my mind
2024-04-30T15:05:12 < qyx> corrupted
2024-04-30T15:05:56 < qyx> anyway, I always though it is a gdb script and not some openocd internak code
2024-04-30T15:06:02 < qyx> I have never used that
2024-04-30T15:06:28 < qyx> I don't need to debug my stuff, I write working code kn firsttry
2024-04-30T15:06:39 < qyx> *on first try
2024-04-30T15:07:34 < ventYl> for most part it is actually openocd faking some stuff to gdb based on the contents of RAM/firmware
2024-04-30T15:10:27 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-30T15:39:56 -!- jtj [~jtj@212.66.207.170] has quit [Remote host closed the connection]
2024-04-30T15:42:25 -!- Livio [~livio@user/livio] has quit [Ping timeout: 245 seconds]
2024-04-30T16:00:09 -!- Luggi09498284764 [~lux@host-87-4-32-147.retail.telecomitalia.it] has joined ##stm32
2024-04-30T16:04:04 -!- rob_w [~bob@host-82-135-31-73.customer.m-online.net] has quit [Remote host closed the connection]
2024-04-30T17:26:48 < karlp> I saw some slides from a linux conf about gdb vs openocd vs in kernel, and it has somewhat been apparently just a "time to get it rolled out in gdb wasn't feasible, even though we know whta tmight be nicer, and now that it's common in debugger middleware, there's not a lot of push to do it the other way"
2024-04-30T18:04:01 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-8060-3630-e5b6-4e01.fixed6.kpn.net] has joined ##stm32
2024-04-30T18:16:28 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-30T18:17:46 -!- NEYi [~NEYi@109.251.216.38] has joined ##stm32
2024-04-30T18:22:52 < karlp> meh, dudes put a sht30 on a baord, 0.2° accuracy.  didn't plan it at all, board self heats the ground plane it's been bolted to by like 6°C.
2024-04-30T18:23:21 < karlp> got away with on another board because it was just big enough to heat sink enough, so they just did it again :)
2024-04-30T18:31:16 < zyp> ah, that's the embedded-sht I always misread in this project I'm working on
2024-04-30T19:01:33 < qyx> I am also using shtc3 because of the momentum
2024-04-30T19:02:38 < qyx> before I used si7006/21 in all designs
2024-04-30T20:24:55 -!- Livio [~livio@user/livio] has quit [Ping timeout: 245 seconds]
2024-04-30T20:55:20 -!- martinmoene_ [~Martin@2a02-a45a-96ba-1-8060-3630-e5b6-4e01.fixed6.kpn.net] has quit [Ping timeout: 268 seconds]
2024-04-30T21:08:43 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has joined ##stm32
2024-04-30T21:09:05 < Laurenceb_> alsa is failing on rpi lunix :(
2024-04-30T21:09:16 < Laurenceb_> high cpu use causes skipped samples
2024-04-30T21:09:33 -!- con3 [~con3@164.90.228.156] has joined ##stm32
2024-04-30T21:10:15 -!- con3 [~con3@164.90.228.156] has quit [Remote host closed the connection]
2024-04-30T21:12:33 -!- con3 [~con3@164.90.228.156] has joined ##stm32
2024-04-30T21:13:25 < Laurenceb_> aplay using 27% cpu wtf
2024-04-30T21:31:01 -!- con3 [~con3@164.90.228.156] has quit [Quit: ZNC 1.8.2 - https://znc.in]
2024-04-30T21:33:07 -!- con3 [~con3@164.90.228.156] has joined ##stm32
2024-04-30T21:44:39 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has joined ##stm32
2024-04-30T21:44:56 -!- Linux_Kerio [~Linux_Ker@chello089173155197.chello.sk] has quit [Read error: Connection reset by peer]
2024-04-30T22:05:44 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
2024-04-30T22:24:31 < Laurenceb_> >hypermotor tooling got trashed at Azovstal
2024-04-30T22:24:35 < BrainDamage> are you using the builtin "audio card"?
2024-04-30T22:24:35 < Laurenceb_> truly epic fail
2024-04-30T22:24:42 < BrainDamage> because that's a rc pwm
2024-04-30T22:24:46 < Laurenceb_> no I'm using a usb audio adaptor
2024-04-30T22:25:26 < BrainDamage> check if you're resampling for some reason
2024-04-30T22:25:31 < BrainDamage> and which algo it's using
2024-04-30T22:26:38 < Laurenceb_> I'm sure its not resampling, but I just found something
2024-04-30T22:27:16 < Laurenceb_> https://pastie.io/bpyjef.properties
2024-04-30T22:27:42 < Laurenceb_> some of those setting look very noob
2024-04-30T22:28:09 -!- Livio [~livio@user/livio] has joined ##stm32
2024-04-30T22:31:03 < Laurenceb_> looks like it uses a sane amount (few percent) of cpu for first process that connects
2024-04-30T22:32:12 < Laurenceb_> when cpu load is high the loop devices start dropping samples and inter channel sync goes shit tier
2024-04-30T22:39:48 -!- alan_o [~alan_o@2600:1700:1902:210f:5db2:c05c:cdee:ca5] has quit [Remote host closed the connection]
2024-04-30T22:40:07 -!- alan_o [~alan_o@2600:1700:1902:210f:4191:f965:5544:5066] has joined ##stm32
2024-04-30T23:24:13 -!- Laurenceb_ [~Laurenceb@202.141.208.46.dyn.plus.net] has quit [Quit: Client closed]
2024-04-30T23:27:51 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has quit [Ping timeout: 256 seconds]
2024-04-30T23:35:07 -!- IanW_ [~IceChat9@arcanum.force9.co.uk] has joined ##stm32
--- Log closed ke touko 01 00:00:12 2024