IRC log for #edev on 20120111

00:08.39*** join/#edev prpplague (~prpplague@ppp-70-249-148-171.dsl.rcsntx.swbell.net)
00:09.59*** mode/#edev [+o prpplague] by ChanServ
00:34.14*** join/#edev gkwhc (~me@unaffiliated/gkwhc)
00:35.55gkwhcHi! I am trying to setup a linux arm development toolchain with eclipse + openocd + jlink. I am not quite sure how to setup the debug server/session. Please advise!
01:03.04MonMothaanybody know of a barebones, crazy simple text/basic widgets (e.g. bar graph) library suitable for small screens (e.g. 150x75px) and having essentially no dependencies?
01:03.35MonMothathis is for a bare metal application, not Linux or similar
01:03.46g1powermaceuh
01:03.49g1powermacnot sure
01:04.10g1powermachad some linux level stuff, but nothing bare metal
01:04.10MonMothaI mean, it shouldn't be too complicated to make one, but it seems like something somebody's already done a thousand times
01:04.37g1powermacthere was one I found once, but not technically open source
01:04.45MonMothayeah, I know of plenty of toolkits for Linux, but I'm talking something just enough to put a basic HMI on a little dot matrix LCD module
01:04.49g1powermacthough the source was downloadable
01:06.42g1powermacaha, its called uC/GUI
01:06.53g1powermacfound it via this: http://www.cutedigi.com/product_info.php?cPath=282_308&products_id=4407
01:08.06g1powermacthe only thing is, its based on uC/OS, which isn't oss
01:08.16MonMotharight
01:08.19MonMothathis doesn't even have an OS :)
01:08.34MonMothait's easily capable of running something smallish (but no MMU), but it just doesn't need one
01:09.39g1powermacwell, uC/OS would in all likelihood work on it, but yea, I understand
01:09.53calculusI used a usb lcd screen that only accepted commands (byte structures, first being the instruction, next being the arguments) to draw
01:10.48MonMothathis is just a framebuffer
01:10.51calculusI think I just did on fopen on the /dev node and then wrote the necessary bytes to have it draw
01:11.20MonMothaoh, this particular device is crazy simple to talk to.  The issue is that it's so simple, I have to do things like draw the text myself :)
01:11.40MonMothaI mean, it's tiny, and the update rate is limited by human interaction, so it's not performance sensitive
01:12.08calculusyep, same here
01:12.13calculusit was this: http://store.earthlcd.com/ezLCD-004-ezLCD004
01:12.32MonMothathis is even simpler
01:12.42calculuspart of the text drawing command, was an argument for which of the 4 built-in fonts to use
01:12.52MonMothathink HD44780 but without the character generator
01:13.03MonMothait's nothing but a dumb, monochrome framebuffer
01:13.06roxfannxp has SWIM
01:13.52roxfanthey also recently released emWin for lpcs
01:13.53calculusoh, that is simpler
01:16.57*** join/#edev djerome (~djerome@ip24-251-139-203.ph.ph.cox.net)
01:17.09MonMothahum, even SWIM looks more complicated than I need, but may have some useful resources
01:17.22MonMothathis thing really is the size of like a little HD44780 based display
01:18.12MonMothaa whopping 122x32 pixels
01:21.52prpplaguehttp://picogui.org/
01:24.43MonMothaI think I can pretty much use SWIM
01:24.59MonMothasomewhat surprisingly, it doesn't appear to have a "for use only on NXP devices" restriction in the license
01:27.02prpplagueho ho hum
01:27.06prpplagueMonMotha: what's cookin?
01:27.48MonMothaprpplague: working on a little embedded doohicky with audio analysis and some power outputs
01:28.00MonMothahas a HMI for tweaking some settings
01:28.20MonMothaalso, totally unrelated, working on a rather sizable telco metro ethernet deployment
01:28.37prpplagueHMI? Hughes Medical Institute?
01:28.42prpplaguejokes with MonMotha
01:29.01MonMothano, far more mundane :)  Just the normal human-machine interface
01:29.54prpplaguewatches galatica 1980 return of starbuck
01:30.30MonMothaI have to say, I wish I could get the kind of network I'm helping design delivered to my house
01:30.36MonMotha10Gb floating everywhere
01:30.45MirellFloatFloatFloat
01:31.52prpplaguedecimals decimals decimals points everywhere and not to drink err use
02:03.15MonMothaoh interesting.  The framebuffer organization of this thing lets me take some shortcuts with my text generation
02:03.25MonMothalike drawing a character becomes memcpy :)
02:05.53*** join/#edev sterNiX (~LessIsMor@unaffiliated/nu253r/x-0655220)
02:31.52*** join/#edev Guest61617 (~LessIsMor@2.176.228.207)
02:35.30*** join/#edev bsdfox (~h36sa@c-98-238-207-83.hsd1.ca.comcast.net)
02:35.30*** join/#edev bsdfox (~h36sa@unaffiliated/bsdfox)
02:40.54*** join/#edev bsdfox\ (~h36sa@c-98-238-207-83.hsd1.ca.comcast.net)
03:29.12MirellMiao.
03:47.32ohsixhttp://hackaday.com/2010/03/31/update-50mhz-to-100mhz-scope-conversion/ how did i miss this :[
03:53.13*** join/#edev risca (~risca@wnpgmb0903w-ds01-177-34.dynamic.mtsallstream.net)
03:59.58MonMothaohsix: don't get me wrong, it's a neat hack, and I'm all for taking advantage of business practices like that, but it's an awful lot of work to save what appears to be $40 (the price difference between the two models)
04:00.16ohsixit was more in 2010 :>
04:02.27MonMothaoh, I guess that was Amazon playing tricks on me
04:02.36MonMothaI guess yeah, it's darn near twice as much
04:02.42MonMothayeah, stick it to 'em :)
04:04.00MonMothaI still want one of the fancy Tek MSOs
04:04.05MonMothanot cheap, though
04:04.10MonMothaand they still insist of putting crappy displays on them
04:04.55ohsixyea, the lecroy ones are almost test equipment pornography
04:05.30ohsixthose hack scopes from the chinese mp3 player things are pretty cute
04:09.32*** join/#edev War2 (~war2@unaffiliated/war2)
05:11.59*** join/#edev bgamari (~ben@pool-96-233-179-97.spfdma.east.verizon.net)
06:27.06*** join/#edev War2 (~war2@unaffiliated/war2)
06:43.48*** join/#edev anoNxeRo (~LessIsMor@unaffiliated/nu253r/x-0655220)
10:20.33*** join/#edev florian (~fuchs@port-217-146-132-69.static.qsc.de)
10:20.33*** join/#edev florian (~fuchs@Maemo/community/contributor/florian)
10:47.35*** join/#edev War2 (~war2@unaffiliated/war2)
11:47.47*** join/#edev skynet-2000 (~skynet-20@unaffiliated/skynet2000)
11:47.47*** join/#edev grummund (~user@unaffiliated/grummund)
11:52.40*** join/#edev florian (~fuchs@port-217-146-132-69.static.qsc.de)
11:52.40*** join/#edev florian (~fuchs@Maemo/community/contributor/florian)
11:57.03*** join/#edev g1powermac (~g1powerma@rrcs-173-197-150-24.west.biz.rr.com)
11:57.03*** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac)
11:57.46*** join/#edev gkwhc (~me@unaffiliated/gkwhc)
12:06.03*** join/#edev yakov (~ysz@77.72.120.66)
12:11.55*** join/#edev War2 (~war2@unaffiliated/war2)
12:18.40*** join/#edev keesj (~keesj@ip49-193-210-87.adsl2.static.versatel.nl)
12:57.41*** join/#edev War2 (~war2@unaffiliated/war2)
15:13.23*** join/#edev landley (~landley@140.242.26.2)
15:13.31*** mode/#edev [+o landley] by ChanServ
15:13.35landleyMorning
15:14.21florianhi landley
15:14.26kergothmorning
15:15.28*** join/#edev GPSFan (~kenm@64.92.145.112)
15:15.28*** mode/#edev [+o GPSFan] by ChanServ
15:17.10landleyHi florian.  Hi kergoth.  how goes stuff?
15:17.38kergothone of those days, woke up with a headache, nice way to start the day. otherwise good. you?
15:17.51landleyI got most of the release notes collated for the aboriginal 1.1.1 release this morning, now I just have to write it up and rebuild from current tip.  (Left a server rebuilding earlier.)
15:18.11landleyOh, darn it, I don't think I checked in the control-images tweak needed to fix the static-build.hdc stuff...
15:18.16landleyOh well, rebuild again when I get home...
15:18.25landleykergoth: I'm stressed to the gills for some reason.
15:18.55florianfights the fact that there are an incredibly amount of fairs in the coming weeks and every customer starts tre yearly panic mode
15:19.00*** join/#edev sterNiX (~LessIsMor@unaffiliated/nu253r/x-0655220)
15:19.08landleyfairs?
15:19.21landleyCES was a couple days ago...
15:19.25floriantrade shows
15:19.30florianyes
15:19.44landleyScale is weekend after next.
15:20.02florianlandley: well... I'm located in Europe - we have several really large ones in spring.
15:20.07landley(It was a fun show, I'd like to go back, but I spent all my vacation time over christmas at a family reunion.)
15:20.14landleyAh.
15:22.20landleyOh, you guys will love this: The FSF has retroactively made binutils 2.17 GPLv3.
15:22.34landleyThey replaced the binutils 2.17 binary with a new one that has a bunch of GPLv3 files in it.
15:22.40landleyDidn't rename it, just replaced the old release.
15:22.54landleyLuckily I still have the original at http://landley.net/aboriginal/mirror/binutils-2.17.tar.bz2
15:23.09landleybut the one of the gnu/dammit project's FTP site?  Contaminated.
15:23.18kergoth...
15:23.18landleyThe file dates are even 2011, for a release that happened in 2006.
15:23.24kergothyou don't replace released archives, ever
15:23.30landleyTell _them_ that.
15:23.35kergothshakes head and sighs
15:23.50fileomg
15:23.52filethat's nuts
15:24.27florianevil... this really should not happen
15:24.31landleySee ftp://ftp.gnu.org/gnu/binutils/
15:24.52landleyftp://ftp.gnu.org/gnu/binutils/binutils-2.17.tar.bz2 is apparently now a symlink to "2.17a"
15:25.43landleyThe original 2.17 file appears to be gone from the archive.  (Or at least I can't find it.  There isn't a 2.17 subdirectory.)
15:25.46landleythat was the last GPLv2 release.
15:25.50landleyfile: the FSF _isnuts.
15:26.03landleyflorian: the FSF is evil.  It has been ever since it lost relevance in 1993.
15:26.40landleyThe FSF rose to prominence because there were only about 3 ways to get your code on an FTP site for free back in 1992.  (sunsite, cdrom.com, and the MIT project athena servers).
15:27.18landleyThen the backbone policy changed in 1993 allowing for-profit ISPs, and suddenly nobody would jump through the FSF's hoops to get their code distributed anymore: they had geocities now.
15:27.24landleyThe FSF, of course, blamed Linux.
15:28.03landleyBut Ted T'so got Linux on the MIT athena servers _without_ having to go through Stallman.  He had the same upstream access as an MIT server admin, and could do what he liked...
15:28.17landleyThis is not how the FSF tells history, of course.
15:29.11landleyCopyright was applied to binaries in 1983 by the Apple vs Franklin court decision, and before that everything was open source because copyright didn't apply to binaries.
15:29.48landleyStallman's "big scheme" of free software wasn't forward-looking, it was a conservative reactionary attempt to preserve the status quo in the face of upstream legal changes which were the "software patents" of their day.
15:31.22landleyHe wasn't the only one who responded.  The BSD guys responded, Andrew Tanenbaum did Minix (and Linux a Minix clone with SunOS influences, as Linus will tell you...)
15:32.11landleySigh.  I wouldn't mind if the guy didn't keep trying to corner the open source market and take credit for other people's work.  But somebody else telling me what to do in the name of freedom is hypocrisy of the highest order...
15:32.58kergothi keep meaning to spend some quality time playing with minix3.. one of these days
15:33.58landleyEh, it was interesting for its' day.
15:34.03*** join/#edev War2 (~war2@unaffiliated/war2)
15:34.14landleyTanenbaum wanted a teaching tool, so he made a teaching tool.
15:34.59landleyHe refused all patches that would turn it into a real usable operating system, because what he _wanted_ was a toy operating system that was easy to understand, and demonstrated the academic fads of the time (microkernel architecture) that he wanted to teach.
15:35.12mnemocpython was also created as a teaching tool...
15:35.16kergothah.. fair enough, given its priorities
15:35.26landleymnemoc: Guido took patches.  Tanenbaum didn't.
15:35.27kergothmaybe i need to play with openbsd or something. a non-gnu userland appeals to me
15:35.37florianwhat was php developed for? ;)
15:35.49mnemocflorian: perl for the web
15:35.50landleyGuido Van Rossum let Python be steered by its user community.  Tanenbaum explicitly refused to do so.
15:36.19landleyWhich is why the huge bunch of developers on comp.os.minix that were maintaining a patch stack to turn Minix into a real OS could never push any of their patches upstream.
15:36.40mnemoclandley: why no fork then?
15:36.41landleyAnd all suddenly switched over to Linux when Linus showed up with a bootable kernel that _did_ accept patches.
15:36.52landleymnemoc: well, linux sort of was a fork.
15:37.02mnemoc:)
15:37.12landleyMinix wasn't free.  It came on the CD in the back of the textbook, but to legally get it you had to buy a $100 textbook.
15:37.22mnemocuh
15:37.27landleyAnd then you could modify your copy to your hearts' content, but couldn't ship derivatives.
15:37.48landleyTanenbaum GPL'd it about a decade later, by which point it was moot.
15:38.40landleyMinix was a proprietary OS that came with full source code, for a price individuals could reasonably afford.  That made it really stand out in its' time, but it wasn't collaboratively developed open source available as a free download through the internet.
15:39.14landleyConsidering that home ISPs were illegal (couldn't hook up to the NSFnet backbone) until 1993, Tanenbaum's position can merely be considered "not particularly foresighted" rather than actually dumb.
15:39.36landleyHe was an academic.  And a good one.  But they tend to live in ivory towers, not down in the trenches where you can see the next wave coming...
15:47.10*** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac)
16:19.30*** join/#edev sndcrb (~sndcrb@dsl-hkibrasgw2-fef8de00-126.dhcp.inet.fi)
16:27.32*** join/#edev CosmicPenguin (~nobody@soa.codeaurora.org)
16:27.32*** mode/#edev [+o CosmicPenguin] by ChanServ
16:39.52*** join/#edev prpplague (~danders@nat/ti/x-xihghbjepybqkree)
16:54.27GPSFanlandley, I've got a copy of binutils-2.17.50.0.17.tar.bz2 from 2007 if you need it.
16:54.54landleyI have a copy of the old vanilla 2.17 snapshotted.
16:55.07GPSFanok, just thoought it might help..
16:55.17landleyI just thought it was a particularly slimy moe for them to replace the last GPLv2 version with a GPlv3 version 5 years after the fact.
16:55.30GPSFannot cool.
16:55.35landleyYup.
16:57.07kergoththe toolchain guys at work said the a version was supposed to include generated files that were missed in the stock 2.17 release, and that if any gplv3 files were pulled in, it was someone's screwup and shoudl be brought up on the binutils list
16:57.18kergoths/the a version/the "a" version/
16:57.25kergothkicks purl
16:57.53GPSFanlandley, btw might you elaborate a bit on the 3.2 breakage you observed wrt host/target "size". I know your airlock in Aboriginal caught it. I couldn't find any references to it on lkml.
16:58.11landleyAh, did I forget to send the patch upstream?
16:59.02landleyhttp://landley.net/hg/aboriginal/file/62dd9695d825/sources/patches/linux-fixarm.patch
16:59.04GPSFanlandley, I last looked at it a couple of days ago, I've been building 3.2 for pandaboard and haven't seen any issues.
16:59.25landleyit's using the host "size" on a target binary.  Probably works for arm, but is generally _unclean_.
16:59.39landley(I note that using the host strip on an sh4 binary breaks the binary.)
16:59.51landley(Because _ and . prefixes get swapped on symbols like "start")
17:00.08GPSFanlandley, thanks.
17:00.25landleyMy general rule is "this is one of the things binutils installed, and which binutils applies a prefix to.  Therefore, use the prefixed version for target binaries."
17:01.13landleyAnd I set my build up so that only ar as nm cc make ld gcc are _available_ from the host toolchain.
17:01.33landley(Because that's what the cross compiled package set needs to build its host binaries.  Determined experimentally.)
17:05.25*** join/#edev bgamari (~ben@gamari.physics.umass.edu)
17:05.34*** join/#edev dijenerate (~dijenerat@65.48.188.246)
17:21.54*** join/#edev dijenerate (~dijenerat@65.48.188.246)
17:31.24MirellSans test tubes.
17:48.32*** join/#edev g1powermac (~g1powerma@unaffiliated/g1powermac)
17:50.06landleyMirell: indeed.
17:50.23landleyGuten Blinking Morgen.
17:50.36landleyOr possibly Bonus Nachoes, depending on time of day.
17:51.06landleyBonus Cameron Diaz.  One of them.
18:48.42*** join/#edev dijenerate (~dijenerat@65.48.188.246)
18:52.08*** join/#edev florian (~fuchs@sign-4db6a45d.pool.mediaWays.net)
18:52.08*** join/#edev florian (~fuchs@Maemo/community/contributor/florian)
19:49.27*** join/#edev dijenerate (~dijenerat@65.48.188.246)
20:55.10*** join/#edev sterNiX (~LessIsMor@unaffiliated/nu253r/x-0655220)
20:57.50*** join/#edev Crofton|work (~balister@pool-96-240-160-175.ronkva.east.verizon.net)
21:01.33*** join/#edev sterNiX (~LessIsMor@2.176.141.26)
21:01.33*** join/#edev sterNiX (~LessIsMor@unaffiliated/nu253r/x-0655220)
21:08.40*** join/#edev JoeLlama (~snork@unaffiliated/joellama)
21:12.21*** join/#edev dijenerate (~dijenerat@65.48.231.106)
21:14.59*** join/#edev dijenerate (~dijenerat@65.48.231.106)
21:16.05*** join/#edev War2 (~war2@unaffiliated/war2)
21:24.46GPSFanlandley, fyi, this: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=1ec332a3756a22405d2fbd5352e3afab556cb205 although submitted before 3.2 was released did make it into the merge window for 3.3-rc1
21:25.19landleyYay, they already dealt with it upstream. :)
21:25.34landley(Often the case, although I need to resubmit my perl removal patches for something like the 11th time...)
21:25.55landleyThis time around was just a rediff to remove fuzz, the patches themselves haven't changed in a while now.
21:27.07*** join/#edev bgamari_ (~ben@gamari.physics.umass.edu)
21:28.07GPSFanah yes, the perl removal patches, it's certainly long past time they were merged and perl banished to the dungeon it deserves.
21:37.19*** join/#edev dijenerate (~dijenerat@65.48.175.168)
21:38.23MonMothaI am currently banging my head against a vendor provided linker script that was so super crazy fragile that I decided it just needed to be re-written
21:38.34MonMothain other news, the linker script language can be a bit obtuse
21:40.03roxfannews at 11
21:54.39*** join/#edev dijenerate (~dijenerat@65.48.175.168)
22:11.37*** join/#edev dijenerate (~dijenerat@65.48.175.168)
22:49.15*** join/#edev gkwhc (~me@unaffiliated/gkwhc)
23:46.12*** join/#edev dijenerate (~dijenerat@173.225.251.175)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.