00:00:00 --- log: started forth/19.12.02 00:13:40 --- join: proteus-guy joined #forth 00:25:42 --- join: mtsd joined #forth 00:26:58 --- quit: cheater (Ping timeout: 276 seconds) 00:28:06 --- join: cheater joined #forth 00:40:40 --- quit: mtsd (Remote host closed the connection) 00:41:29 --- join: mtsd joined #forth 01:59:30 --- join: dave0 joined #forth 02:29:24 So how does recurse work.. instead of say calling the word I am in, I just use recurse to reference it's self ? 02:29:31 instead of the actual word name ? 02:33:09 well. that was easy 02:36:01 I've not used recursion yet 02:36:10 the idea is that you are using the name you are in and not do a dictionary search which might lead to an older version of the same word 02:36:47 recurse ( - - ) Lets the current definition call itself 02:40:19 i think it pushes the xt of the current word, into the current word 02:40:42 i think KipIngram likes recursion instead of loops in his words :-) 02:41:14 iirc KipIngram's forth optimizes tail recursion 02:41:15 dave0, I read it's popular 02:42:02 I don't implement recurse 02:43:41 dave0, Mecrisp-Stellaris also features, tail-call optimisations, constant folding and inlining 02:44:13 I have recur which simply compiles a jump back to the beginning of the word 02:44:28 a GOTO!!!! 02:44:33 for shame! 02:44:38 --- join: iyzsong joined #forth 02:44:51 ;-) 02:45:01 almost! 02:45:14 i can see using it in the REPL 02:45:16 I'm telling Mr Moore! 02:46:02 I suppose it's a basic operation ... 02:46:07 He would have no problem with it :) 02:47:07 Im only a technician, these advanced Forth concepts are above my pay grade, so I'll leave them to Zen Gurus 02:49:15 : recur { -- }( -- ) myself jmpi, ; directive 02:49:15 : 0recur { -- }( x -- ) myself 0jmpi, ; directive 02:49:15 : ?recur { -- }( x -- ) myself ?jmpi, ; directive 02:51:05 Ooooh 02:51:11 my first stack overflow 02:51:18 somethings leaking its drool onto the stack! :D 02:51:33 jsoft, your first!!!!! 02:52:05 jsoft grab my stack sensitive prompt ? 02:52:07 I have somewhere on my todo list to change recur to compile a relative jump instead of an absolute one, just haven't gotten around to it 02:52:56 jsoft, https://mecrisp-stellaris-folkdoc.sourceforge.io/stack-sensitive-prompt.html#ssp 02:53:09 you and your stack sensitive prompt :P 02:53:28 rdrop-exit, I was looking thru some mecrisp-quintus source (risc-v) and saw some interesting assembly 02:53:49 interesting how? 02:54:03 jsoft, it's not just me, the SSP userbase has DOUBLED in the last week! 02:54:39 dun dun dunnnnnnn 02:55:21 rdrop-exit, li x6, 0x03e00008 # Ret-Opcode 02:55:59 rdrop-exit, that look like a 32 bit value to me 02:56:13 sure does 02:56:48 but Ive no idea what "li" is yet 02:57:00 load immediate in MIPS 02:58:07 nice, just what I want a 32 bit load immediate :) 02:58:36 it's not listed on my "free and open RISC-V ref card tho ? 02:59:18 I haven't looked into the RISC-V instruction set yet 02:59:34 plenty of mnemonics with "xxli" tho 03:00:12 same, just a brief scan of the mecrisp-quintus source today after chatting to you here 03:00:51 it also has a "Optional Compressed (16-bit) Instruction Extension: RVC" 03:01:50 RISC-V Integer Base (RV32I/64I/128I), privileged, and optional compressed extension (RVC). Registers x1-x31 and the pc are 32 bits 03:01:51 wide in RV32I, 64 in RV64I, and 128 in RV128I (x0=0). RV64I/128I add 10 instructions for the wider formats. The RVI base of <50 03:01:51 classic integer RISC instructions is required. Every 16-bit RVC instruction matches an existing 32-bit RVI instruction. See risc.org. 03:02:18 x1 to x31 ???? 03:02:43 that's a nice amount of registers :-) 03:02:56 thats humongous! 03:03:22 arm is only 12 registers iirc and thumb can only reach the first 8 for most ops 03:03:51 tp: are you having flashbacks to the 80's eight bits cpu :-) 03:04:25 6502 has 1 accumulator and 2 address registers, both 8 bits :-) 03:04:34 dave0, no, I don't need to, my core beliefs were fashioned in the 70's 03:04:51 and a pocy 8 bit index register 03:04:55 poxy 03:05:20 I know I keep on about it, it's not nearly as bad as I make out 03:05:29 i grew up in the 80's 03:05:35 RISC-V also recycles jalr to return from a subroutine: To do this, jalr's base register is set to be the linkage register saved by jal or jalr. jalr's offset is zero and the linkage register is zero, so that there is no offset, and no return address is saved. 03:05:36 I should just STFU about the 650 because it's a legend 03:05:42 6502 03:06:16 (from Wikipedia) 03:06:39 dave0, I was married in 1976, has my first child 3 years later 03:07:06 rdrop-exit, that sounds a bit complex 03:07:31 tp: my lady friend was born in 1979 :-) 03:07:36 i'm 1977 03:07:53 awesome ! 03:08:37 It's a RISC design uses a jump-and-link instead of a traditional call/jsr instruction 03:09:36 jsoft, you kiwi sheep mugger, I bought 3 kg of blade steak today :) 03:09:40 I remember 77 clearly 03:10:02 instead of pushing the return address onto a stack, it stashes the return address in a register 03:10:09 rdrop-exit, me too as I was married only a year earlier :) 03:10:41 i've only ever used CISC cpus 6809, m68k, x86 in that order :-) 03:10:44 I married in 85 03:11:11 rdrop-exit, wish I had also ;-) 03:11:20 :-)) 03:11:35 dave0, my condolences on the latter cpu 03:12:23 rdrop-exit, so RISC-V is looking kinda special so far, no wonder people are excited ? 03:12:54 I didn't mind x86 assembly up to 486, got fed up of it with the Pentiums 03:14:10 i want a PC-AT with a 286, i reckon it'd be cool 03:14:19 very retro 03:14:40 tp, noice :D 03:14:54 I have no interest in assembly for desktop, only for targets 03:15:03 tp, I need to get me a few kg of ribeye in response 03:17:09 dave0, it would be damn slow 03:17:47 jsoft, blade steak is $13/kg here atm, ribeye is probably $25/kg 03:18:04 jsoft, scotch fillet is about $45/kg 03:18:07 Yeah but ribeye is the business 03:18:12 why is scotch fillet more ? 03:18:19 ribeye is the business 03:18:21 sure, but the blade is DAMN nice 03:18:43 i loved the last blade steak I had 03:18:50 I mean it was AWESOME 03:18:59 I like steak in general, but my favorate is ribeye 03:19:18 jsoft, dunno, just is 03:19:28 tp, odd. 03:19:46 it's a bad drought here in AU so farmers are dumping cattle because they have no feed 03:19:56 hence the low prices 03:20:12 when they run out I wont be able to buy Australia beef 03:20:25 it will all be imported I guess 03:20:39 well I better go to sleep 03:21:03 how does gforth suck so bad 03:21:07 yeah 00:20 there 03:21:08 file handling SUCKS 03:22:07 some things are just weird pointless crap 03:22:17 CR is `newline type` for christ's sake 03:27:06 --- quit: jsoft (Ping timeout: 268 seconds) 03:28:53 GD32VF103: Single-cycle hardware multiplier and Multi-cycles hardware divider support; 03:29:44 Standard 4-wire JTAG debug port, support 4 triggers for hardware breakpoint 03:45:12 --- join: gravicappa joined #forth 04:30:00 --- quit: deesix (Ping timeout: 268 seconds) 04:30:04 --- quit: proteus-guy (Ping timeout: 250 seconds) 05:09:37 --- quit: iyzsong (Quit: ZNC 1.7.1 - https://znc.in) 05:34:11 --- quit: dave0 (Quit: dave's not here) 05:58:20 --- join: inode joined #forth 06:08:49 --- quit: inode (Quit: ) 06:23:51 --- join: dys joined #forth 06:25:10 --- quit: rdrop-exit (Quit: Lost terminal) 07:04:07 just catching up. re: GCC and assembly size, tp, interesting that the blinky binary is only 80 bytes. have you tried comparing the size with anything more complicated than blinky? 07:07:59 also, GCC is not a monolithic program, so whatever is producing shoddy code doesnt tell us anything about GCC for other processors 07:08:50 they made a new version of GCC for the MSP430 that apparently produces much better assembly than the last port 07:10:40 whereas the "Gcc produces better assembly than you can do by hand" comment is more applicable to x86 where there are so many complicated mechanisms working in the background that you dont have on something like an MSP430 07:17:15 I cant find the video now but one of the presentations on youtube from cppcon shows a series of examples where GCC outputs "shoddy" assembly that when you run it, it's actually faster than if you fixed it and took out the inefficiency 07:17:58 since that inefficiency was inserted intentionally since it takes advantage of some cache mechanism or something else normal humans dont understand and actually makes it run faster even though the code appears to be worse 07:23:05 --- quit: mtsd (Quit: Leaving) 07:24:01 --- quit: ryke (Ping timeout: 245 seconds) 07:45:56 --- quit: tabemann (Ping timeout: 250 seconds) 07:50:37 --- join: deesix joined #forth 07:52:47 --- join: dddddd joined #forth 08:08:13 --- join: proteus-guy joined #forth 08:21:38 --- join: ryke joined #forth 08:52:36 --- join: begin joined #forth 09:22:34 --- quit: dys (Ping timeout: 250 seconds) 09:49:42 --- quit: ryke (Remote host closed the connection) 09:49:54 --- join: ryke joined #forth 10:00:16 --- quit: ryke (Ping timeout: 276 seconds) 10:29:59 --- join: jsoft joined #forth 10:44:42 --- quit: jsoft (Ping timeout: 252 seconds) 11:57:43 --- join: remexre_ joined #forth 12:02:36 --- quit: jimt[m] (*.net *.split) 12:02:36 --- quit: remexre (*.net *.split) 12:02:36 --- quit: retroforth-bot (*.net *.split) 12:02:37 --- quit: dne (*.net *.split) 12:02:49 --- join: dne joined #forth 12:21:16 --- join: ryke joined #forth 12:39:03 --- join: retroforth-bot joined #forth 13:01:15 --- quit: gravicappa (Ping timeout: 265 seconds) 13:10:44 MrMobius, all good points but I only circulate in embedded groups so the claims of Gcc's superiority came from embedded not X86 people 13:12:11 MrMobius, I love Gcc (for C) as well, it enabled me to learn C for Free back in 1996 13:17:18 --- join: dave0 joined #forth 13:21:05 --- join: WickedShell joined #forth 13:21:38 So much of a High-Level Language and ""Portability"" whatever that means. 13:23:53 IMHO the greatest... ""Feature"" of C is it's magnificent Mainframe mindset best seen through any kind of it's documentation. 13:25:13 It is a write-once language. No wonder we... are blessed with Bash, Perl, Ruby and whatever. 13:25:25 DKordic, true. I find Forth is very portable in embedded pretty much 100 if I use Mecrisp-Stellaris between different architectures 13:25:46 100 % 13:26:45 saying C is portable when people use Gcc between different architectures is a similar thing think 13:27:07 Mecrisp-Stellaris is too ANSI Forth for me :) . 13:27:16 Mecrisp-Stellaris Forth on a MSP430 is almost identical to Mecrisp-Stellaris on a Cortex-m 13:28:29 only certain unique hardware in each type of chip can introduce some minor dofferences but the core Forth language is identical 13:32:00 DKordic, I went over my minimal blinky codes this morning, 76 bytes for assembly, 306 bytes for Gcc with no libs, 508 bytes with libcm3, someone elses F4 multi led blink, 34kB 13:44:14 anyone working on advent of code? 13:44:19 been doing mine in retro 13:44:47 need to actually upload them but I don't have a window manager at the moment lol 13:47:35 I guess that 34kB is Arduino :) . 13:54:09 DKordic, no it was just #include #include "stm32f4.h" #include "gpio.h" and arm-none-eabi-gcc 13:54:24 with 4 leds blinking 13:56:20 GCC Switch ""-O0""? 14:00:37 no opts at all 14:01:03 but If I add them then I'll have to mess with his code and add 'volatile1 14:01:33 'volatile' or it will make his loop dspinner delays go away 14:02:03 -d gahh my typing is terrible today 14:09:54 IMHO that pathetic excuse for a Source Code has many undesirable properties of Executable form. It seems that Gabriel was an optimist, it is ""Worst of All"" rather than merely ""Worse Is Better"". 14:17:44 tp, ya agreed. GCC outperforming hand written assembly seems hit or miss on embedded 14:20:48 MrMobius, I rechecked my mecrisp-across blinky, it was standard forth, utterly default and unimaginative on my part (as usual), no opts, a spinner blinky and the bootable binary was 36 bytes 14:21:20 the actual blinky was 20 bytes and the mcu startup and vectors was 16 bytes 14:21:37 what is the target of mecrisp-across? 14:21:43 it's a msp430 14:21:52 hmm 14:22:05 would you want to make something more complicated in forth just as a test? 14:22:13 and I could make an msp430 asm version to compare 14:22:17 just out of curiosity 14:23:09 my point is that the wiz of assembly (matthias koch) can make a 14 byte assembly blinky using *every* trick in the book and mecrisp-across manages 36 bytes with me driving it and zero tricks 14:23:57 MrMobius, youve asked me this before, but Im not a programmer, Im a technician 14:24:15 I can only comment on the code sizes from observation 14:24:58 my first hand assembled MSP430 was 80 bytes iirc 14:25:10 my first hand assembled MSP430 blinky was 80 bytes iirc 14:25:42 tp, sorry, I dont remember asking that. I remember talking about testing other code but never a longer program which I think is what you would have to do to have a real idea since blinky is too simple to extrapolate 14:26:06 ya I could see how forth might make better assembly for that than you would if youre just a normal assembly programmer 14:26:26 and like Gcc I expected when I used mecrisp-across my 80 byte hand assembly blinky would blow out to at least 250 bytes, I was stunned to see it was 36 bytes, I was like WTF > 14:27:05 MrMobius, Im a terrible programmer, but a outstanding electronics tech :) 14:27:44 tp, some days I wish I could trade places with you :) 14:27:50 my first love is electronics, besides, I dont have the 'hackers mentality or mental horsepower' 14:28:50 MrMobius, electronics was always an obsession for me, what can I say ? 14:30:14 MrMobius, I get by with pedestrian Forth in my projects, I can use it and debug it on small embedded, even design my own development environment, so I'm very happy 14:33:03 MrMobius, mecrisp-across was the first time *ever* any compiler has beaten my hand assembly code, and thats my point, as pedestrian as I am, if I beat Gcc at a blinky by a factor of 300% why are people saying Gcc is so much better than hand assy ? 14:33:58 now I understand a blinky is small and simple, but it still has to do real world stuff, have proper boot up etc 14:34:36 if Gcc cant beat assy in such a simple case, why does it follow that in a complex case it will somehow be better ? 14:35:18 and what if you have someone like matthias koch doing the hand assembly in your test ? 14:35:42 he thinks in assembly, nothings a problem for him 14:36:28 I'm probably the single worst human to use for a assembly vs gcc competition :) 14:38:48 --- join: jedb_ joined #forth 14:40:59 --- quit: jedb__ (Ping timeout: 240 seconds) 14:42:18 --- join: jsoft joined #forth 15:11:07 --- quit: dddddd (Remote host closed the connection) 15:31:19 --- quit: X-Scale (Ping timeout: 240 seconds) 15:31:45 --- join: jedb__ joined #forth 15:34:00 --- quit: jedb_ (Ping timeout: 240 seconds) 15:45:25 --- quit: ryke (Ping timeout: 276 seconds) 16:04:53 --- join: X-Scale joined #forth 16:14:19 tp, I think that is backwards. the extra 200 bytes or whatever could be clib and start up stuff. you cant say if the difference is so big with a blinky it must therefore be larger with something more complex 16:14:59 but you can say that if the difference is so large where the majority of code is boiler plate, the difference may be much smaller for a more complicated project 16:28:22 MrMobius, you're the expert here not me, but I'm commenting on the difference using Gcc and no clib, no lib at all, only a header file. 16:32:30 MrMobius, the Gcc binary does seem to be half full of the same reset vectors, thats possibly the reason 16:33:16 and of course on a longer program the reset vectors would become less significant 16:44:07 MrMobius, youre right I think 16:44:09 The blinky binary (with opts) = 288 bytes 16:44:10 This seems comprised of :- 16:44:10 Main code: 36 bytes 16:44:10 Delay code: 16 bytes 16:44:10 Startup code: 26 bytes 16:44:10 ---------------------- 16:44:12 Total = 36 16 + 26 + = 78 bytes. 16:44:14 The rest seems to be interrupt vectors. 16:45:19 my best assembly blinky is 76 bytes inc startup for the same target, Matthias does his in 28 bytes 16:45:20 --- join: ryke joined #forth 16:45:27 for the same target 16:46:04 the interrupt vectors just look like boilerplate and are all the same by default 16:46:08 --- nick: jedb__ -> jedb 16:47:41 Matthias uses every trick to get the blinky that low on cortex-m and no C compiler or plain assember writer like myself could ever get anywhere near that 16:49:22 MrMobius, I see that a blinky is not a fair comparison in any way between Gcc and hand assembly and that a much larger program would need to be considered 17:10:06 --- quit: dave0 (Quit: dave's not here) 17:13:12 ya it would just be interesting with a larger program 17:13:20 not necessarily that I know what the results would be 17:19:10 it's not really that important, but I see wht Gcc for cortex-m is so large for a blinky 17:20:23 --- join: tabemann joined #forth 17:25:54 MrMobius, thanks for the insight :) 18:13:43 --- quit: tabemann (Ping timeout: 246 seconds) 19:17:54 --- quit: Keshl (Read error: Connection reset by peer) 19:18:14 --- join: Keshl joined #forth 19:32:30 --- join: tabemann joined #forth 19:32:43 --- part: tabemann left #forth 19:33:19 --- join: tabemann joined #forth 19:50:47 And hello 19:52:54 hey 20:18:59 tp, ran into this issue at lunchtime today: https://community.st.com/s/question/0D50X00009XkZdn/rtc-subsecond-rtcssr-issue 20:19:05 well, 'issue' 20:19:12 More like 'gotchya' 20:40:13 --- join: gravicappa joined #forth 21:26:50 --- join: smokeink joined #forth 21:36:48 --- quit: smokeink (Remote host closed the connection) 22:46:21 tp, first yotoob video evar! https://www.youtube.com/watch?v=cNzr8eKP9dE 22:56:53 jsoft, I wondered where my wire cutters got to ! 22:59:10 :D 23:01:06 --- quit: DKordic (Ping timeout: 245 seconds) 23:02:54 the old RTC shadow register trick hey ? 23:03:11 yeah. 23:03:26 I done it during lunch today, so did not have much time to study it proper 23:03:38 there are a LOT of gotchas with stm32 23:03:45 Yerp 23:03:56 mainly because they have so many features 23:04:40 at one point nothing I could do would make a peripheral work properly and there were a ton of registers I had configured 23:05:43 so I wrote an assert word that tested whether or not every register value WAS changed and sure enuf one register just refused to be written to 23:06:20 turned out it had a 'protect' register that had to be prewritten with a sequence first 23:06:57 Yeah those stupid sequence thingies 23:06:59 it's the little things like that that can bring the freight train to a stop 23:07:30 but the hardest of all has been the f0 low power settings 23:07:54 they are complex and I just have trouble following the doc 23:08:27 I'm still not sure if I have reached the lowest power setting 23:09:12 i can get down to about 4uA and retain ram and rtc but it needs a restart 23:10:15 and of course the 10 million 'examples' in C are all copies of copies and the C people dont understand whats happening either 23:24:03 --- quit: proteus-guy (Ping timeout: 268 seconds) 23:32:46 jsoft, thats the hardest part about Forth on cortex-m, *every* online example is in C of some kind 23:34:23 now C is dead easy to follow, problem is they never give the headers and important memory mappings in their examples 23:35:41 meh 23:35:53 Shenanigans 23:36:08 It's never going to be easy 23:36:29 C users are shielded from all that stuff, it 'just' happens in their humungous IDE 23:37:11 I dunno 23:37:37 I use C, and this whole thing does not seem that strange to me, beside the new language factor 23:37:45 and most regardless of the ridiculous claim that "C is low level, on the metal, etc" have never looked at a whole peripheral register or have any concept of what a bitfield is or does 23:38:50 What do you mean by bitfields and peripheral registers 23:39:12 You mean as the registers listed in the datasheets, and the bits which do the things ? 23:39:13 the concept of reading a "whole technical manual" is a abhorrent waste of time to many, it gets in the way of their drinking time 23:39:23 yes, thats right 23:39:57 I dunno, I kinda had to know how those things doing pic stuff, its not that tricky 23:40:08 The _order_ of things or whatever can be 23:40:16 but thats regardless of language 23:41:05 altho in cortex-m there are really three different kinds of registers, system registers such as "systick", peripheral registers such as 'gpioa_moder' and cpu registers such as r0 - r12 23:41:38 and all the system registers are NOT described in the STM32 doc, theyre ARM only 23:41:43 Yeah nah I am not super familiar with cortex stuff. 23:41:53 But yay 23:41:59 Stuff and things! :) 23:42:03 stm32 doc only deals with peripherals 23:42:16 it's a long path, quite different to PIC 23:42:39 Well I have somehow bumbled my way to a 60% done software side project 23:42:52 in forth, which I did not know 3 weeks earlier 23:42:53 microchip for instance controlled 100% of the PIC, but stm only add the peripherals to the core which is ARM 23:43:31 60% ? you been boozing again ? why arent you done already ? 23:44:02 Because I am still waiting on the real hardware to arrive 23:44:19 And also I say 60% because of fudge factor 23:44:34 a diamond studded of the shelf clock from Christian Dior ? 23:44:49 eh? 23:44:58 (real hardware) 23:45:11 No a pcb from china, and some dickheads from RS to send me some buck IC's and what not 23:45:35 23:45:41 how the bollocks can some geezer in china make a PCB, send it to me, in less time than some moron at RS can pick a part of a shelf and send it to me ? 23:46:00 it remains to be seen when the Chinese stuff will arive 23:46:01 It is irritating 23:46:11 lol the parts already have 23:46:14 the pcb's will 23:46:26 they may be having a year of the Albatross and closed down until next Feb 23:46:33 the rs parts... yet to even register on courier tracking 23:46:34 --- join: dave0 joined #forth 23:46:50 china, IMO, has been faster for me when using a normal courier 23:46:58 pretty much each and every time 23:47:03 if youd ordered from Arrow.com you probably have them already and free shipping 23:47:14 for a first order 23:47:18 parts? 23:47:24 the buks 23:47:28 Meh 23:47:42 Well whatever, at the time I was 'right, need this need that, cant fuck around, need to order and design' 23:47:44 theyre damn fast here 23:47:55 Perhaps I will try them later. 23:48:01 they must go straith thru custome 23:48:14 i had my first order in a few days 23:48:27 china goes straight through customs 23:48:28 must have been airfreight outa the USA same day 23:48:46 yeah, you just hav to get it to customs forst ? 23:49:03 I dunno, it just arrives 23:49:22 I have more fidelity on china/DHL stuff than I do with RS 23:49:38 china used to get free shipping for small parcels due to some USA deal, but thats all changed now I think 23:49:44 I saw the PCB stall for like 8 hours in Auckland, before clearing 23:49:52 meh 23:49:53 yeah RS went to shit decades ago 23:49:55 I am not usa :) 23:50:25 sure you are, the CIA has run NZ for years 23:50:33 lol here we go 23:50:38 heheh 23:50:44 cia can run their tounge along my nuts. 23:51:05 are their tongues small enuf ? 23:51:29 lol this guy eh 23:51:34 the USA is still mad at you giys for banning their nukes 23:51:53 Yeah wel meh 23:52:17 so naturally they had to take over NZ just to get their nukes in there 23:52:22 23:52:37 Well better the US than China 23:52:40 we on the other hand just do what we are told 23:52:46 hell yea! 23:53:04 IMO, we need a trump here. 23:53:22 I am supremely jealous of the US because of him. 23:53:26 hey, hands off, we have placed a order for a trump here! 23:53:50 He is the dude IMO. 23:54:07 hes sure stirring the deep pot over there 23:54:13 People seem to have a spastic about him because he does not talk via script 23:54:26 And instead, talks the truth 23:54:27 it's been pretty much full on entertainment since 2016 23:54:56 well I have no idea who is telling the truth, I dont trust politicians at all 23:55:08 I'm amnipulated by the media 23:55:14 manipulated 23:55:58 Yeah there is the interweb, which has lots of info, all can be filtered by your own level of ability to filter information from bullshit 23:56:02 wow, RISC-V is different! 23:56:33 I hardly recognise any of Mecrisp-Stellaris other than file names and the terminal.s code 23:56:50 I cant even find where he sets the CLOCK! 23:57:32 I'm beginning to suspect he doesnt, that the 8MHz rc clock just self runs like on stm32 23:57:45 and is enabled by default 23:58:15 I just wact the USA politics, it's far more entertaining than at home 23:58:18 watch 23:59:59 --- log: ended forth/19.12.02