00:00:00 --- log: started forth/19.07.02 00:13:08 --- join: xek (~xek@apn-31-0-23-83.dynamic.gprs.plus.pl) joined #forth 00:45:11 --- quit: dave9 (Remote host closed the connection) 00:46:59 --- join: gravicappa (~gravicapp@46.191.138.203) joined #forth 00:50:19 --- join: dave9 (~dave@069.d.003.ncl.iprimus.net.au) joined #forth 01:20:41 --- quit: dave9 (Remote host closed the connection) 01:25:53 --- join: dave9 (~dave@069.d.003.ncl.iprimus.net.au) joined #forth 01:32:54 --- join: eponym (~epony@unaffiliated/epony) joined #forth 01:32:55 --- quit: eponym (Remote host closed the connection) 03:03:51 --- quit: gravicappa (Ping timeout: 258 seconds) 04:14:53 --- join: gravicappa (~gravicapp@46.191.138.203) joined #forth 04:25:43 --- quit: john_metcalf (Ping timeout: 248 seconds) 06:21:18 --- join: dave0 (~dave0@069.d.003.ncl.iprimus.net.au) joined #forth 06:39:13 --- join: dddddd (~dddddd@unaffiliated/dddddd) joined #forth 06:47:02 --- quit: dave0 (Quit: dave's not here) 07:18:46 found what seems like a cool Forth library 07:19:02 but the author has named every damn file as a 3-letter abbreviation 07:19:04 WHY 07:21:26 can't just open interestingly named files: what the hell is "car.fs" supposed to do 07:25:08 it implements a small four-wheeled motor vehicle 07:26:52 it gets the first element in a list? 07:27:28 grrr, github's code search is so useless 07:28:02 `car - the cell array collection module in the ffl` 07:28:38 reminds me of DOOM's source code i was just poking around in 07:28:44 oh? 07:28:53 everything is prefixed with bizarre two-letter abbreviations 07:29:34 I recall seeing that, must be some old school way of organising source 07:29:47 nethack https://github.com/id-Software/DOOM/tree/master/linuxdoom-1.10 07:29:54 er, ignore "nethack" 07:41:11 --- quit: tabemann (Ping timeout: 276 seconds) 07:42:52 --- join: cnidario (~aaa@92.57.58.87) joined #forth 07:50:59 --- quit: proteusguy (Remote host closed the connection) 08:31:46 --- quit: cnidario (Remote host closed the connection) 08:45:50 --- quit: gravicappa (Ping timeout: 245 seconds) 09:42:43 --- join: proteusguy (~proteusgu@58.10.208.146) joined #forth 09:42:43 --- mode: ChanServ set +v proteusguy 10:17:43 --- join: gravicappa (~gravicapp@46.191.138.203) joined #forth 11:21:59 --- quit: chunkypuffs (Remote host closed the connection) 11:59:34 --- quit: gravicappa (Ping timeout: 245 seconds) 12:08:28 --- join: dys (~dys@tmo-100-163.customers.d1-online.com) joined #forth 12:08:44 --- quit: xek (Ping timeout: 245 seconds) 12:26:31 hmm.. how do many forths behave with : make_some_definer dup cell+ dup @ over 1+! @ ." Hello I am a defined word nr. " dup cell+ @ <# #s #> ." of nr. " over @ <# #s #> ." and my counter is " cell+ cell+ @ <# # #> cr ; 12:33:30 --- nick: proteusguy -> proteusdude 12:53:05 --- join: john_metcalf (~digital_w@host31-54-142-171.range31-54.btcentralplus.com) joined #forth 13:19:58 --- join: mark4 (~mark4@172.56.6.86) joined #forth 13:20:08 other people keep making changes to their shit and breaking mine dammit 13:20:53 recent changes to ld foroce .text to be 0x1000 aligned. used to be the elf section was 0x8048000 and .text was at 0x8048080 now .text is at 0x8049000 13:21:16 and thats wtih gnu ld 2.32.0 13:21:50 wtih gnu ld 2.31.1 they are now injecting BULLSHIT .note.gnu.property sections into the elf file. one in front of .text and one after .bss 13:28:44 mark4: You could use a linker file to decide where the sections go. 13:30:15 i am already 13:30:34 and it does not matter if i specify an alignment of 1/16/12345 on .text 13:30:42 its aligned to the page boundry 13:31:10 Weird. Shouldn't happen. 13:31:36 I actually only have .text and .bss in my code and the ONLY reason for .bss is to extnd my .text to be exatly 1 meg in size 13:31:50 my .data and .text sections are merged into a single .text section in the linker script 13:32:01 .bss is 1 meg in size minus the length of .text 13:32:20 on lanuch i sys mprotect my entire 1 megs to +r +w +x 13:32:45 --- join: mtsd (~mtsd@94-137-100-130.customers.ownit.se) joined #forth 13:33:24 load address for process is 0x8048000. elf header is 0x80 bytes in size. .text USED to start at 0x8048080 now it starts at 0x8049000 no matter how i align .text 13:34:05 GNU ld (Gentoo 2.31.1 p7) 2.31.1 13:34:24 which also injects two .note.gnu.property sections into the elf file. 2.32.0 does not 13:50:44 look like the .note.gnu.property is related to new abi extensions on x86 13:52:51 told they are not actyually needed in the finished product, they are there for the linker 13:53:09 and in my case they are 100% EMPTY and are thus a waste of space 13:55:04 where did you see this info about new abi extensions? 13:59:44 http://lists.llvm.org/pipermail/llvm-dev/2018-March/121951.html 14:00:04 https://raw.githubusercontent.com/wiki/hjl-tools/linux-abi/linux-abi-draft.pdf 14:02:08 (I can't say much about the extensions as I don't really use Linux anymore) 14:02:17 reading that first scares me. if im reading it right indirect threading is not going to be working for much longer 14:02:31 direct threading i mean 14:04:32 mark4: does ld inject that shit regardless of what is specified in the linker script? 14:06:29 yes 14:06:41 well 14:07:16 x86_64-pc-linux-gnu-2.32 does not inject the .not.gnu.property sections 14:07:23 x86_64-pc-linux-gnu-2.31.1 does 14:07:37 .note i mean 14:07:46 those are verions of gnu ld btw 14:11:09 hmm.. what I am musing about doing regarding this kind of stuff is just to have one small .text section one page in size and just do C or assembler version of an 'canonical dual stack machine' from Stack Machines The New Wave By Philp Koopman. Rest gets into .data or .bss sections 14:12:24 well if i had an x86 assembler and a metacompiler this would be a HUGE non issue 14:12:48 im currently putting to gether a spreadsheet of aarch64 opcde encodings 14:25:49 --- quit: john_metcalf (Ping timeout: 245 seconds) 14:26:02 * Zarutian is thinking of something like https://bpaste.net/show/gzE5 14:27:04 that way all this ROP prevention and such crap can be bypassed and the .text section should all fit into L1 cache 14:28:51 wouldnt that switch statement be horrendously large and slow? 14:30:51 still need to find out why x4 cant handle the new terminfo format lol 14:31:05 it should compile down to a jump table and a mask check. And all those 'continue's point to the start of the loop 14:31:11 with xterm you have to export TERM=xterm for it to run 14:32:05 personally I would create a typedef struct wtih { opcode, address } and then create an array of them 14:32:16 sorted. do a binary search 14:32:26 that x86 legacy (both 32 and 64 bit) arch design doesnt handle megamorphic switch statements well is of no concern of mine. 14:34:27 or make opcodes huffman codes :) 14:35:22 start from the root, opcode traces path thru tree, leaf has word address 14:35:37 well, the 'instructions' are all one cell wide (usually 32 bit) which should speed up decoding 14:51:16 --- quit: mtsd (Quit: Leaving) 15:25:49 WilhelmVonWeiner, was the 3 letter library from the jeelabs collection ? 15:26:45 i finally finished my word2ihex, the code is utterly horrible and will need about 3 reworks before I let anyone see it :) 15:27:01 wordhex test1 15:27:11 :0E0000000122920401239B03043F3E60D618A8 15:27:11 :00000001FF 15:31:51 --- quit: mark4 (Ping timeout: 248 seconds) 15:39:58 --- join: mark4 (~mark4@172.56.6.86) joined #forth 15:57:16 --- join: tabemann (~tabemann@rrcs-162-155-170-75.central.biz.rr.com) joined #forth 16:05:12 --- quit: mark4 (Ping timeout: 272 seconds) 16:06:42 --- quit: crc (Quit: ZNC 1.6.6 - http://znc.in) 16:10:31 --- join: mark4 (~mark4@172.56.6.86) joined #forth 16:15:39 --- join: crc (crc@2a01:7e00::f03c:91ff:fe12:ac) joined #forth 16:21:02 --- quit: john_cephalopoda (Ping timeout: 252 seconds) 16:21:47 --- join: dave0 (~dave0@069.d.003.ncl.iprimus.net.au) joined #forth 16:22:37 --- quit: dave0 (Read error: Connection reset by peer) 16:22:57 --- join: dave0 (~dave0@069.d.003.ncl.iprimus.net.au) joined #forth 16:34:39 --- join: john_cephalopoda (~john@unaffiliated/john-cephalopoda/x-6407167) joined #forth 16:45:45 hey 16:45:53 heyhey 16:46:16 * tabemann is busy debugging an actor framework based on hashforth 16:46:19 tabemann, there are many other advantages I didnt mention with gnuscreen over all other serial terms for Forth 16:46:34 oops, that was a convo with Zarutian 16:46:39 sorry! 16:47:07 --- join: mark4_ (~mark4@rrcs-67-79-14-26.sw.biz.rr.com) joined #forth 16:47:21 the only thing IIRC from looking at some code associated with mecrisp-stellaris is that it expects CRLF 16:47:47 most serial terms expect CRLF 16:47:57 even 'unix' ones 16:48:10 which was a problem in that case because mecrisp-stellaris outputs LF 16:48:16 exactly 16:48:39 I just added \r\n to the Mecrisp-Stellaris sourse 16:48:43 source 16:49:03 I'll have to keep that in mind 16:49:08 hashforth of course uses LF 16:49:12 as the commands to work around that just dont work with gnuscreen 16:49:50 so how is GNU screen better then? 16:50:05 --- quit: mark4 (Ping timeout: 244 seconds) 16:50:21 I'm working on using a switch to add \r\n to the source and I'll ask Mattthias if he will accept it, he has very strict rules for any core changes 16:50:45 better than other serial terminals ? 16:50:51 yeah 16:50:59 1) remote control of the connected screen 16:51:36 2) can pass args to programs i.e. !:sed -n s/.*compile-only.*/\007/p 16:52:18 that one rings the bell if Mecrisp-Stellaris emits the "compile-only" error 16:52:38 --- join: mark4 (~mark4@rrcs-67-79-14-26.sw.biz.rr.com) joined #forth 16:53:16 3) vi like commands for searching text to reuse it in the screen 16:53:39 4) works with hardware handshaking 16:54:05 5) can 'slowpaste' if handshaking isnt set up 16:54:20 nice 16:54:21 I'm sure there are many more 16:54:46 I know that there are some people around here that'd avoid it just because it's GNU though 16:54:57 for instance I have just made a Word that dumps the intel hex binary code to the screen 16:55:17 yeah, I'm a FreeBSD user myself 16:55:31 --- quit: mark4_ (Ping timeout: 268 seconds) 16:55:47 * tabemann doesn't believe in the whole viral thing, but still uses the BSD3 license so the most people can reuse his code if they wish 16:56:04 now I'm working on a gnuscreen command that when it sees a keyword will write the dumped hex.bin to a file on my pc 16:56:17 nice 16:56:35 tabemann, I'm pretty much the same re GPL 16:57:39 I agree with Linus choosing the LGPL 2 as his license, I must state, because otherwise Linux would have just ended up like FreeBSD re: macOS 16:57:52 *GPL 2 16:58:02 I'm so used to using the LGPL that I reflexively typed it 16:58:42 well more accurately it will write the ihex file then run arm-none-eabi-objcopy -I ihex blah 16:59:27 apple isnt exactly into sharing hey ? 16:59:41 nah, Apple is basically taking from BSD without giving back 17:00:03 oh what a $urprise! 17:00:04 whereas the GPL 2 ensured that those who wanted to use Linux and who wanted to modify it would give their changes back 17:00:27 yeah, those that admitted they were using GPL'd code 17:00:52 anyway, I'm not sure I'd want apples code anyway 17:01:07 look at the mess that CUPS made ? 17:01:33 my experience with CUPS is half of the time the drivers don't work 17:01:34 I was very happy with LPRNG then everyone had to switch to CUPS 17:01:35 --- join: mark4_ (~mark4@172.56.6.86) joined #forth 17:01:44 it's a nightmare imho 17:01:53 --- quit: mark4 (Read error: Connection reset by peer) 17:02:08 to print a document, e.g., I have to email myself the document, download it on my phone, and print from my phone 17:02:22 my laptop can see the printer, but it fails to actually print it 17:02:33 I recently set up a remote linux users printer on CUPS, gahhh it was like sticking pins under my nails 17:02:58 yeah, and CUPS doesnt exactly say why things dont work 17:03:21 from the start I felt that CUPS was just a ploy to monetise LP 17:04:39 in the old days with a properly set up LP, I remember printing a page but it came out blank ... while I was scratching my head I received a internal email from my linux system telling me that a print had failed because fonts were missing 17:04:57 I installed the fonts and reprinted .. perfect 17:06:06 to me CUPS just doesn't work and doesn't say why 17:08:10 --- join: mark4 (~mark4@rrcs-67-79-14-26.sw.biz.rr.com) joined #forth 17:12:20 --- quit: mark4_ (Ping timeout: 272 seconds) 17:23:42 agree 17:23:53 --- join: mark4_ (~mark4@172.56.6.86) joined #forth 17:24:30 this was a simple hp usb laser and took me a couple of hours to do, mainly finding out why cups wasnt working 17:24:50 the whole thing reeks of 'paid cups support' 17:27:46 --- quit: mark4 (Read error: Connection reset by peer) 17:34:30 back 17:34:36 welcome back! 18:05:24 --- quit: tabemann (Ping timeout: 245 seconds) 18:18:11 --- quit: mark4_ (Remote host closed the connection) 18:18:38 --- join: mark4_ (~mark4@172.56.6.86) joined #forth 18:31:00 --- join: tabemann (~tabemann@2600:1700:7990:24e0:89f9:b840:47e7:ad52) joined #forth 18:35:18 --- quit: mark4_ (Ping timeout: 272 seconds) 18:45:13 --- join: happybop (~user@ns557447.ip-139-99-121.net) joined #forth 18:46:55 .0033. 18:47:37 3333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333 19:05:53 --- part: happybop left #forth 19:08:15 tabemann: Was it you I was talking to about logistics? 19:16:10 --- join: mark4 (~mark4@172.56.6.198) joined #forth 19:20:41 possibly so 19:21:07 * proteusdude does logistics! :-) 19:21:28 DKordic: is your cat on the keyboard? :-) 19:28:27 Shit, I lost the URL. Why does YouTube not list the video I uploaded? 19:30:02 Gotta love the Web World. 19:30:20 Glitzy, but worthless. Anyway, recovered it from logs: https://youtu.be/9M88Jl19Eeo 19:30:36 That's me tracking all the packages outstanding as of lunch time yesterday. 19:58:19 --- join: tpbsd (~Terry@mecrisp/staff/tp) joined #forth 19:58:27 --- quit: tp (Remote host closed the connection) 20:04:50 --- quit: tpbsd (Remote host closed the connection) 20:04:51 --- join: tp (~Terry@2001:44b8:314b:a800::10) joined #forth 20:04:51 --- quit: tp (Changing host) 20:04:51 --- join: tp (~Terry@mecrisp/staff/tp) joined #forth 20:08:13 --- join: tpbsd (~Terry@2001:44b8:314b:a800::10) joined #forth 20:08:13 --- quit: tpbsd (Changing host) 20:08:13 --- join: tpbsd (~Terry@mecrisp/staff/tp) joined #forth 20:08:29 --- quit: tp (Remote host closed the connection) 20:21:51 back 20:22:11 * tabemann is encountering a strange memory corruption bug in hashforth's actor code 20:23:06 I can't tell where the corruption is occurring, and it seems to move around the code, and adding debugging prints - even though I turn multitasking off for them - seems to affect how they occur 20:23:41 and I can't tell if it's one of my datastructure, well, modules that's causing it or my heap implementation or what 20:40:42 --- quit: dys (Ping timeout: 272 seconds) 20:42:34 --- join: mark4_ (~mark4@rrcs-67-79-14-26.sw.biz.rr.com) joined #forth 20:44:49 --- join: mark4__ (~mark4@rrcs-67-79-14-26.sw.biz.rr.com) joined #forth 20:45:51 --- quit: dddddd (Remote host closed the connection) 20:46:43 --- quit: mark4 (Ping timeout: 244 seconds) 20:47:05 --- quit: mark4_ (Ping timeout: 245 seconds) 21:33:18 --- nick: tpbsd -> tp 21:34:30 tabemann, what's your main debugging tool do you use gdb on it ? 22:16:02 --- join: karswell (~user@cust125-dsl91-135-5.idnet.net) joined #forth 23:02:37 --- quit: Keshl (Read error: Connection reset by peer) 23:02:46 --- join: Keshl_ (~Purple@207.44.70.214.res-cmts.gld.ptd.net) joined #forth 23:10:54 --- quit: tp (Remote host closed the connection) 23:11:03 --- join: tp (~Terry@2001:44b8:314b:a800::10) joined #forth 23:11:03 --- quit: tp (Changing host) 23:11:03 --- join: tp (~Terry@mecrisp/staff/tp) joined #forth 23:26:47 --- join: mark4 (~mark4@172.56.6.141) joined #forth 23:27:54 --- quit: mark4__ (Ping timeout: 272 seconds) 23:59:59 --- log: ended forth/19.07.02