#milkymist IRC log for Monday, 2012-04-16

wpwrakhere he is :)00:13
wpwrakxiangfu: seems that we have to go back to pinning rtems and now also rtems-yaffs for a while00:13
wpwrakxiangfu: at least "head" seems completely unusable00:14
wolfspraulxiangfu: ah, good morning :-)00:14
wolfspraulas the next step we need not just daily builds, but also a daily (automated) test :-)00:14
wpwrakhehe :)00:15
xiangfugood morning.00:15
xiangfuwpwrak, I saw your email on USB rtems00:15
wpwrakmaybe you could set up an M1 in your office that updates itself each day, then reboots. you'll catch this sort of thing instantly :)00:15
wpwrakat first i was afraid it had erased the FN configuration. but it just couldn't load any files. alas, only changing the permissions back to what they were before didn't seem to accomplish much.00:17
xiangfuwpwrak, have you erased the data partition. which build you have tried? I would like to try that one :)00:39
xiangfuI mean reflash the data parition.00:39
wpwrakxiangfu: no no, the data partition is fine. it's FN+rtems+rtems-yaffs that can't read any files00:51
xiangfuok.00:51
wpwrakxiangfu: once i installed an FN built with the pinned rtems, everything returned to normal. i didn't lose configuration or patches. they were just temporarily unavailable00:51
wpwrakxiangfu: you can try with the FN from the latest daily build. that one exhibits the problem. not sure when it started. may have been a while.00:52
xiangfuI will try the latest build.00:53
cladamwwpwrak, last week seems that you thought we should have our own symbols under /kicad-libs/components/ which same or as common one as KiCad default one ?00:54
wpwrakcladamw: yes00:55
wpwrakcladamw: we shouldn't depend on the "system" kicad libs00:55
cladamwlike 'common', 'power', 'conn' ...00:55
wpwrakcladamw: these libs 1) have some bugs, and 2) they change sometimes, so we can't know what someone has installed on their system00:55
cladamwyeah... i've seen many pieces of emails in KiCad-users they've replied this bad once KiCad installs again.00:56
wpwrakyeah :)00:56
cladamwso no my questions are :00:56
wpwrakyou update after a few years, and all of a sudden, everything is a little different :)00:56
cladamw1. Shall I go for like do a repository copy from "system" KiCad lib and merged ourselves under qi's kicad-libs/components ?00:58
cladamwsince I started to use those very common libs which are belonged to KiCad's its system libs. ;-)00:58
wpwraki would start with an empty library and carry over or redo the parts i find i need. that way, i can be sure everything is consistent00:59
wpwrakfor example, you want to pay attention to font sizes. almost all of the power symbols have a 0.40" font somethere, which is basically unreadable01:00
wpwraks/somethere/somewhere01:00
cladamwokay... but the empty library's name like same as 'common' and 'power' ?01:00
wpwrakyes, why not01:01
cladamwalright. :-)01:01
wpwrakor wait lemme check ...01:01
wpwrakhmm, the paths to the kicad libs are hard-coded. so there may be confusion of you use the same name. "common" should be fine, but "power" may cause problems01:03
wpwrakwe could of course just follow the one lib per symbol principle even for "common", and have r.lib, c.lib, and so on. may be a bit excessive, but not much01:04
wpwrak"power" is tricky. well, we can abbreviate it. for example, "pwr.lib"01:04
cladamwokay ... agreed on naming "pwr.lib"01:05
cladamw:-)01:05
cladamwbut this "pwr.lib" are mostly for common symbols of power not for chips from KiCad's default meanings. ;-)01:06
xiangfuwpwrak, cladamw I will split my common to c.lib/r.lib/led.lib01:06
cladamwso we use "pwr.lib" for common symbols01:06
wpwrakpwr.lib for power symbols01:07
cladamwxiangfu, nice, so could you do this now or ?01:07
xiangfucladamw, now.01:07
wpwrakah no, not for chips :)01:07
cladamwand rename "power.lib" to "pwr.lib"? :-0 btw, teach me that how to rename like this got git. :-)01:08
wpwrakgit mv file.old file.new01:08
cladamwxiangfu, let me do this mv. i try. :-)01:09
wpwrakbut we don't have anything called "power.lib" in git, so that's not necessary :)01:09
cladamwoah..yeah01:10
cladamwxiangfu, but we need to copy power symbol from board-m1 to kicad-libs/components/pwr.lib :-)01:11
xiangfucladamw, from /usr/share/kicad/.... to kicad-libs/components/pwr.lib01:13
wpwrakbut don't just copy the whole file. copy the components you need - and check that they look right, especially the font sizes. anything 0.40" (or is it 0.040" ? don't remember) should be at least ...5001:14
GitHub16[board-m1] xiangfu pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/d2fada4004cb0d9ead0c259e9d56961bacc0af8b01:15
GitHub16[board-m1/master] update with kicad-libs. one lib per symbol - Xiangfu01:15
cladamwwpwrak, you'd prefer a 0.05" at least or 0.06" ?01:16
cladamwfor example current +3.3 is 0.03" only.01:17
cladamwfor other "reference" or "Value", they are all 0.06" font.01:18
wpwrak30 mil ... you need a microscope for that ;-)01:18
cladamwso we'd stick to a also 0.06".01:18
wpwrakfor me, 50 mil or 60 mil are both okay for power symbols. as long as the size is consistent.01:19
cladamwthis means I'll always need to change them. :-) well to keep their fonts are the same is good. I like. :-)01:19
GitHub33[milkymist] xiangfu pushed 1 new commit to hid: https://github.com/milkymist/milkymist/commit/b5de0d04bcb6c368ced7cce5d28f7f5f1cbda0f001:34
GitHub33[milkymist/hid] Merge branch 'master' into hid - Xiangfu01:34
GitHub166[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/7a2c0b9a9cad0d9781791f49d0504ea1e3757dab04:23
GitHub166[board-m1/master] linked to newer C_P, 3V3, 5V, USBEXT5V, TESTPOINT chip name - Adam Wang04:23
GitHub119[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/d1c76be9205a009fd3687243d1df7a9ae594fcbc06:05
GitHub119[board-m1/master] linked J24 to Qi's 'con.lib' - Adam Wang06:05
GitHub149[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/ac70d33f92185eba23fb19d73746bc5bb9ea600308:43
GitHub149[board-m1/master] added decoupling cap. into VideoIn.sch - Adam Wang08:43
GitHub73[board-m1] xiangfu pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/aba6c7d1864af4ebb1fa1579c2c50a9b6b97172909:13
GitHub73[board-m1/master] MISC: finish LED matrix - Xiangfu09:13
GitHub58[board-m1] xiangfu pushed 3 new commits to master: https://github.com/milkymist/board-m1/compare/aba6c7d...f3f707709:38
GitHub58[board-m1/master] MISC: remove empty line, label don't support \n" - Xiangfu09:38
GitHub58[board-m1/master] add switch.lib - Xiangfu09:38
GitHub58[board-m1/master] MISC: add SW2 - Xiangfu09:38
cladamwxiangfu, wrong D19's direction. :-)09:41
xiangfucladamw, oh.09:42
xiangfutoo bad. hard of the LED direction is wrong. :(09:42
xiangfucladamw, thanks for point out.09:43
cladamwxiangfu, you're welcome. ;-O09:43
GitHub28[board-m1] xiangfu pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/ef331a570f3be40bcd50818e55632ec2e282f05d09:55
GitHub28[board-m1/master] use FSMRA2JH for sw2 - Xiangfu09:55
GitHub149[board-m1] xiangfu pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/2de5d99c67c7fa776f9954b78dd1788199b57e9e09:58
GitHub149[board-m1/master] remove fsmra2jh, it's included by switch - Xiangfu09:58
cladamwxiangfu, sorry that confusing you the name.10:03
Fallenou_a small milkymist like https://www.youtube.com/watch?v=h42neZVvoMY10:04
Fallenou_:)10:04
wpwrak"small" vs. 3 (!!!) FPGAs ? :)10:12
wpwrakokay, just 1/3 of the system used. that's better :)10:15
wpwraknice generative effects10:20
Fallenou_yep11:01
Fallenou_non pipelined cpu :)11:01
Fallenou_Could not find the source code though11:01
Fallenou_not sure it's open source11:01
Fallenou_(HDL source code)11:01
GitHub139[board-m1] adamwang pushed 1 new commit to master: https://github.com/milkymist/board-m1/commit/550209718dd0b91b0089f0bb9108117975b08ecb13:40
GitHub139[board-m1/master] linked bnc.lib and all parts in VideoIn.sch - Adam Wang13:40
mwallewpwrak: btw if you have some spare time, i'm curious which patch (the DCM->PLL or 72mhz usb) fixes the errors17:26
mwallei doubt its the first.. but anyway17:26
mwallewpwrak: is the lv3 a full speed device?17:37
lekernel__yes, it is17:39
mwallelekernel__: ok :)17:39
wpwrakmwalle: i think any full-speed device triggers the errors. can be a keyboard, probably also storage or such. of course, if it's a type of device softusb doesn't recognize, you'll only get a small number of complaints before it decides to ignore the device17:41
qi-botThe firmware build was successful, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-20120416-1849/18:33
GitHub152[milkymist] sbourdeauducq pushed 3 new commits to master: https://github.com/milkymist/milkymist/compare/a534e34...410655318:35
GitHub152[milkymist/master] Use PLLs instead of DCMs - Michael Walle18:35
GitHub152[milkymist/master] Switch USB clock to 72MHz - Michael Walle18:35
GitHub152[milkymist/master] softusb-input: adjust SOF timer - Michael Walle18:35
lekernel__mwalle, thanks18:35
GitHub29[milkymist] sbourdeauducq pushed 2 new commits to master: https://github.com/milkymist/milkymist/compare/4106553...de65a6720:18
GitHub29[milkymist/master] navre: add dbg_pc port to test benches - Michael Walle20:18
GitHub29[milkymist/master] softusb: fix io_do width - Michael Walle20:18
qi-botThe firmware (using branch) build was successful, checkout the VERSIONS for detail, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-20120416-2034/20:18
--- Tue Apr 17 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!