| 01:42.00 | *** join/#edev MonMotha (~monmotha@108-71-210-59.lightspeed.iplsin.sbcglobal.net) |
| 02:11.29 | *** join/#edev krushia (~krushia@h69-131-92-179.cntcnh.dsl.dynamic.tds.net) |
| 02:43.43 | *** join/#edev Spida (~timo@ns1.spinnennetz.org) |
| 03:15.44 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 03:46.12 | *** join/#edev kergoth (~kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 04:19.04 | *** join/#edev mbuf (~user@nat/redhat/x-kulmkoprwkleoerw) |
| 04:28.45 | *** join/#edev NightMonkey (debian-tor@pdpc/supporter/professional/nightmonkey) |
| 04:44.54 | *** join/#edev melchior (~melchior@AStrasbourg-552-1-60-54.w83-203.abo.wanadoo.fr) |
| 05:01.02 | *** join/#edev avinashhm (~avinash-h@203.101.61.7) |
| 06:11.37 | *** join/#edev m4t (matt@theorize.org) |
| 06:46.22 | *** join/#edev JoeLlama (~snork@unaffiliated/joellama) |
| 07:05.32 | *** join/#edev gvandenbosch (~gvandenbo@86.81.67.223) |
| 07:44.52 | *** join/#edev khem` (~khem@99-57-141-118.lightspeed.sntcca.sbcglobal.net) |
| 09:36.34 | *** join/#edev avinashhm (~avinash-h@203.101.61.7) |
| 10:17.31 | *** join/#edev roxfan (dunno@119.194-246-81.adsl-dyn.isp.belgacom.be) |
| 10:38.02 | *** join/#edev risca (~risca@130.236.250.25) |
| 12:11.14 | *** join/#edev gustavoz (~gustavoz@host77.201-252-46.telecom.net.ar) |
| 12:29.47 | *** join/#edev MuscleNerd (~MuscleNer@99-65-242-37.lightspeed.irvnca.sbcglobal.net) |
| 13:14.10 | *** join/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 13:20.25 | *** join/#edev flo_lap (~fuchs@Maemo/community/contributor/florian) |
| 13:27.04 | melchior | Someone know a tool on linux to create a single binary image from multiple file and be able to extract them after ? |
| 13:40.43 | SpeedEvil | tar? |
| 13:51.19 | melchior | I think i'll have to do it mself, it is for putting a boootloader, a kernel, and a rootfs, in a binary file for embedded system upgrade |
| 13:52.06 | melchior | And adding a header to it so that only allowed device will be able to be upgraded |
| 14:06.59 | *** join/#edev bscshane (~shane.woo@173.13.246.241) |
| 14:07.52 | TemporalBeing | melchoir: dd |
| 14:08.08 | TemporalBeing | otherwise write a little program to do it, like the Linux Kernel does (or at least use to) |
| 14:10.01 | SpeedEvil | Do you just want to prevent oopses, or do you want to do cryptographic sycnatures? |
| 14:12.49 | enlight22 | i think u can do it with cat and split, as well as cpio... dd can also do it... lots of ways |
| 14:12.51 | TemporalBeing | melchoir: see http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.4.37.y.git;a=blob;f=arch/i386/boot/tools/build.c;h=2edd0a4ec5980140cb7046968540e87908b0d421;hb=HEAD for a reference |
| 14:23.18 | *** join/#edev kergoth (~kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 14:56.45 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 15:00.10 | *** join/#edev gandhijee (akp@ip67-152-15-148.z15-152-67.customer.algx.net) |
| 15:08.17 | melchior | So before to see your responses i've written a little app to concat those files and include a heade wihch contains the size of each bin, i've used mmap to do this. Now it comes to undo this image into the previous files and i get MAP_FAILED when i try to mmap part of the input file. |
| 15:08.52 | melchior | is it possible for example to mmap byte 4 to byte 12 in memory ? |
| 15:09.34 | *** join/#edev prpplague (~danders@nat/ti/x-nyhrmyhwzlqucphh) |
| 15:18.08 | SpeedEvil | Why are you using mmap? |
| 15:18.29 | melchior | to copy the binaries into memory so that a can copy portioin of it into files |
| 15:19.09 | SpeedEvil | Why not just copy files directly? |
| 15:19.53 | melchior | because i want to split one binary into 3 which start at specific offset in the source binary |
| 15:20.06 | melchior | but maybe i don't see how would you proceed |
| 15:22.02 | SpeedEvil | Simply use normal file IO calls to read the input file for x bytes, and output to a file. read y bytes, output to another file. |
| 15:22.39 | melchior | with ifstream and ofstream ? |
| 15:26.37 | prpplague | comes in on the conversation |
| 15:27.02 | prpplague | why not just use dd ? |
| 15:27.14 | SpeedEvil | melchior: With normal file IO calls. |
| 15:27.25 | SpeedEvil | melchior: as you would with any other file read/write operation. |
| 15:28.03 | SpeedEvil | fread/fwrite, or whatever. |
| 15:29.16 | SpeedEvil | Should be easily under 50 lines of C. |
| 15:30.33 | melchior | SpeedEvil: i'll give i a try, but the answer to your question is simply that a didn't look to this function, just to read and write |
| 15:31.52 | melchior | prpplaque: i should gie it a try too |
| 15:32.46 | prpplague | melchior: you can split binaries very easily with dd |
| 15:33.34 | melchior | prpplague: the fact is that the data about the offsets of the binaries are included in the source binary, so i've to read them first |
| 15:40.26 | prpplague | melchior: ahh, shouldn't be too hard to do that |
| 15:43.57 | *** join/#edev rasker (~Rehan@5ad8c59a.bb.sky.com) |
| 16:03.34 | *** join/#edev n900evil (~Speedevil@tor/regular/SpeedEvil) |
| 16:07.11 | *** join/#edev sndcrb (~sndcrb@dsl-hkibrasgw2-fef8de00-126.dhcp.inet.fi) |
| 16:09.38 | *** join/#edev ricorx7 (~vanguard@wsip-70-167-169-169.sd.sd.cox.net) |
| 16:40.34 | *** join/#edev NullMoogleCable (~fake@cpe-69-205-123-203.rochester.res.rr.com) |
| 16:41.57 | *** join/#edev calculus (~calculus@gentoo/user/calculus) |
| 16:43.58 | *** join/#edev djerome (~djerome@ip24-251-138-141.ph.ph.cox.net) |
| 18:02.47 | *** part/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 18:04.20 | *** join/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 18:06.33 | *** join/#edev NightMonkey (debian-tor@pdpc/supporter/professional/nightmonkey) |
| 18:12.36 | *** part/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 18:19.45 | *** join/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 18:22.45 | *** join/#edev calculus (~calculus@gentoo/user/calculus) |
| 18:33.48 | *** join/#edev miknix (~miknix@gentoo/developer/miknix) |
| 18:35.31 | *** part/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 19:27.31 | *** join/#edev dlocated (~delocated@the.embeddedbsd.org) |
| 19:30.36 | *** join/#edev miltonm (~miltonm@71.22.127.106) |
| 19:42.12 | *** join/#edev risca (~risca@h241n2-n-a31.ias.bredband.telia.com) |
| 19:45.17 | *** join/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 20:18.36 | *** join/#edev wing0 (~heero@187.39.186.42) |
| 20:19.05 | *** join/#edev wing0 (~heero@187.39.186.42) |
| 20:26.01 | *** join/#edev roxfan (dunno@119.194-246-81.adsl-dyn.isp.belgacom.be) |
| 20:37.40 | *** join/#edev jnd (~jenda@unaffiliated/jnd) |
| 21:18.11 | *** join/#edev edboogie2k (~Edward_Ki@pool-173-77-212-165.nycmny.fios.verizon.net) |
| 21:18.18 | edboogie2k | hey guys, anybody know if I can use a glass covered kitchen stove to heat up the bottom side of a PCB to reflow a DAP pad under a eTSSOP package? |
| 21:19.39 | MonMotha | edboogie2k: you could, but I'd recommend a laboratory hot plate |
| 21:19.45 | MonMotha | if anything just to avoid contaminating your kitchen stove |
| 21:19.53 | MonMotha | also provides a lot more control over the temperature |
| 21:20.07 | edboogie2k | I do own a infrared thermometer, maybe i can get a semi |
| 21:20.13 | edboogie2k | I kinda wanna try it out now |
| 21:20.18 | edboogie2k | i mean its just one IC |
| 21:20.47 | edboogie2k | what temperature should I go? |
| 21:20.52 | edboogie2k | for? |
| 21:21.03 | MonMotha | leaded or lead free paste? |
| 21:21.27 | edboogie2k | I am not using paste, "high tech" silver bearing solder? |
| 21:21.32 | edboogie2k | from radioshack |
| 21:21.48 | MonMotha | have you pre-tinned and fluxed the pad or something? |
| 21:21.53 | edboogie2k | yes |
| 21:21.59 | edboogie2k | exactly |
| 21:22.03 | MonMotha | ok, assuming you can still get the leads to seat, it should work OK |
| 21:22.16 | edboogie2k | cool, what temperature should i try? |
| 21:22.20 | MonMotha | ~240C at the top surface of the PCB should cause it to reflow |
| 21:22.22 | edboogie2k | like low ballpark so i can start off |
| 21:22.30 | MonMotha | watch it |
| 21:22.34 | MonMotha | you should be able to see it go |
| 21:22.38 | edboogie2k | cool |
| 21:22.41 | edboogie2k | that sounds good |
| 21:22.59 | edboogie2k | what do normal kitchen stoves do typically? what range of temp? |
| 21:23.11 | MonMotha | oh it'll easily get that hot |
| 21:23.24 | edboogie2k | ok i'm gonna go try it |
| 21:23.42 | MonMotha | the silver stuff has a slightly higher melting point than non silvered leaded solder |
| 21:24.03 | edboogie2k | oh really? you think i should use regular 60/40 rosin core? |
| 21:24.07 | edboogie2k | I have that stuff too |
| 21:24.10 | MonMotha | regular leaded solder will go at ~230-235 (so 240 gives some margin). Silver will be closer to 240. Lead-free is more like 260-270 |
| 21:24.15 | MonMotha | it should be fine |
| 21:24.24 | edboogie2k | oh ok, small difference |
| 21:24.27 | edboogie2k | thanks for info |
| 21:24.49 | MonMotha | you don't want to get it any hotter than necessary |
| 21:24.53 | edboogie2k | yeah of course |
| 21:24.56 | MonMotha | it will smell a little bad, but nothing should burn |
| 21:26.49 | *** join/#edev wing0 (~heero@187.39.186.42) |
| 21:29.30 | SpeedEvil | edboogie2k: you mean the hob? |
| 21:29.39 | SpeedEvil | edboogie2k: Easily 350C typically. |
| 21:30.13 | edboogie2k | hob? |
| 21:42.55 | *** join/#edev GPSFan (~kenm@64.92.145.112) |
| 21:42.55 | *** mode/#edev [+o GPSFan] by ChanServ |
| 21:43.44 | edboogie2k | looks like it worked. had to set my stove to about a 2 or 3 out of 10. it got about to 260 cel. according to my infrared thermometer |
| 21:53.46 | *** join/#edev prpplague (~prpplague@ppp-70-249-149-148.dsl.rcsntx.swbell.net) |
| 21:57.29 | *** part/#edev TemporalBeing (~bmeyer@12.52.185.66) |
| 22:13.52 | MonMotha | edboogie2k: hotplate reflow works pretty well for single sided designs assuming you can get the part placed |
| 22:13.56 | MonMotha | I've been meaning to try it on BGA |
| 22:14.16 | MonMotha | I think I might be able to get a 1mm BGA placed close enough since they tend to self-center during reflow |
| 22:16.44 | edboogie2k | MonMotha: so is your technique to apply paste, or pretin with flux and solder ALL PADS, and then place the IC centered as close as possible to the footprint and pray that the IC will center itself? |
| 22:18.13 | edboogie2k | I use to pay someone $40 for every LQFP, QFN, and eTSSOP, and i feel stupid for wasting a couple hundred $$ when i coulda done everything in the kitchen. :P |
| 22:20.52 | MonMotha | edboogie2k: depends on what I'm going. For simple stuff with leads, I usually just hand solder (with liberal use of flux on high density stuff) |
| 22:21.16 | MonMotha | for exposed pads, I usually flux before putting the part down then apply solder through a hole designed into the PCB for such purposes |
| 22:21.23 | MonMotha | failing that, tin/flux and reflow for EPs |
| 22:21.37 | MonMotha | for high density boards, I'll just order a paste mask and some paste and do it that way |
| 22:21.55 | MonMotha | or pay someone to do it since it's not that expensive (I know some CMs that just charge by time; usually ~$1/min) |
| 22:22.28 | edboogie2k | $1 min is cheap |
| 22:22.39 | edboogie2k | most of these ICs takes 5 to 10 min by a pro |
| 22:28.36 | edboogie2k | MonMotha, where do you get your solder paste stencils? |
| 22:31.28 | MonMotha | edboogie2k: I've never actually ordered one myself. I usually just let my CM do it. Typically ~$75-100, and I know it'll work with their setup |
| 22:31.46 | MonMotha | and I'm paying them to set up, run, and tear down a pick/place and reflow line |
| 22:33.07 | edboogie2k | yeah i have a similar deal with my guy |
| 22:33.52 | MonMotha | essentially, if it has leads, I can solder it |
| 22:33.59 | MonMotha | so if I'm only doing 1-5 boards, I'll just assemble by hand |
| 22:34.10 | MonMotha | even a 100+ lead 0.5mm QFP only takes me maybe 5 minutes |
| 22:38.59 | edboogie2k | the worst part is spending 3 to 5 hours for a 1 PCB board with many passives and ICs. my current board is about 50 parts |
| 22:40.24 | edboogie2k | i'm thinking if i sell enough of my first design. i will buy my own chinese pick and place machine with 1 year warranty for 12 grand |
| 22:40.39 | edboogie2k | i hope programming it isn't hard |
| 22:40.39 | a7r | MonMotha: do you drag solder with a hoof tip, or something else on QFPs? |
| 22:42.00 | MonMotha | a7r: I actually just use a screwdriver tip for drag soldering. If you hold it right, it works quite well, at least for me. |
| 22:42.52 | a7r | MonMotha: ah, interesting.. I haven't heard of that. |
| 22:43.01 | edboogie2k | i use a toothbrush and cold tap water to scrub flux and debris off my boards, is that what everybody does? |
| 22:43.17 | MonMotha | a7r: I just use the Weller ETA tip that came with my iron |
| 22:43.31 | MonMotha | I have an ETH and ETO for passives |
| 22:43.44 | MonMotha | I'm quick down to 0603, and I can do 0402 just slowly |
| 22:44.08 | a7r | cool, I use a metcal MX500 with hoof tips for QFPs/drag, and different assorted sizes/temps chisel tips for everything else. |
| 22:44.14 | MonMotha | depends some on the PCB design,too. Placement of course takes longer than the actual soldering |
| 22:44.21 | a7r | yup |
| 22:44.23 | MonMotha | heh, I'd love a metcal |
| 22:44.33 | a7r | I got mine on ebay for $80 |
| 22:44.44 | MonMotha | do you design your PCBs with the large land at the end of each row to catch the excess solder? |
| 22:45.02 | MonMotha | and that's not a bad price at all |
| 22:45.10 | a7r | no, with the hoofs it's pretty good at holding onto the blob. |
| 22:45.10 | MonMotha | less than I paid for my Weller WESD-51, but it was new |
| 22:45.44 | MonMotha | hum, I may look into a hoof tip then. I don't like to put those lands on my boards since they take up space and do nothing, but it means I have to very carefully control the amount of solder I introduce during the drag process |
| 22:46.23 | a7r | with the hoof it's just: tack corners down, apply flux, add solder to hoof, drag. |
| 22:47.08 | MonMotha | that's essentially what I do, but I have to carefully control the solder on the tip |
| 22:47.17 | MonMotha | I'm not sure if there is a hoof tip for my PES51 pencil |
| 22:47.55 | *** join/#edev SpeedEvil (~user@tor/regular/SpeedEvil) |
| 22:48.47 | ricorx7 | metcal worth it, got mine on ebay for $100 |
| 22:49.03 | edboogie2k | danm metcal is really that good |
| 22:49.11 | a7r | MonMotha: http://www.youtube.com/watch?v=wQXhny3R7lk#t=3m20 <- that's what I do |
| 22:49.14 | edboogie2k | why is it worth it? |
| 22:49.15 | a7r | and it's as easy as it loks. |
| 22:49.18 | a7r | er looks |
| 22:49.23 | MonMotha | oh, the Metcal stuff is awesome |
| 22:49.25 | MonMotha | I've used it :) |
| 22:49.40 | MonMotha | I just don't own any |
| 22:49.59 | Spida | edboogie2k: I use isopropanol (alcohol) to wah of flux, and dry the board with compressed air... |
| 22:50.11 | a7r | that's not an MX500, but the tip is similar to the one I use. |
| 22:50.21 | edboogie2k | ahh yeah isopropanol with a high concentration |
| 22:50.23 | edboogie2k | i forgot about that |
| 22:50.37 | ricorx7 | tips are great, heat is consistent |
| 22:50.51 | *** join/#edev jevin (~jevin@manbearpig.ecn.purdue.edu) |
| 22:51.01 | a7r | ricorx7: yah, and warm-up time is fast. |
| 22:51.17 | MonMotha | I've been quite happy with my WESD-51 |
| 22:51.24 | MonMotha | heats up quickly, tips last forever |
| 22:51.32 | edboogie2k | yeah my PCB assembler uses tips that are at least half the size of my smallest weller pes51 tip |
| 22:51.43 | edboogie2k | MonMotha: don't you want a smaller tip |
| 22:51.44 | edboogie2k | ? |
| 22:52.09 | MonMotha | edboogie2k: than a Weller ETS? I don't know that they come much smaller... |
| 22:52.36 | MonMotha | that's already at the size that I have trouble getting good heat conduction from the tip to the work due to lack of surface area |
| 22:53.15 | edboogie2k | yeah i figured that |
| 22:53.27 | edboogie2k | maybe with tiny tips you need lots more heat and press harder |
| 22:54.03 | a7r | you just need the tip to maintain heat well |
| 22:54.27 | a7r | and obviously have wetting |
| 22:55.18 | MonMotha | right |
| 22:56.08 | MonMotha | I have an ETP that I use for most single-point work down to 0603 land sizes |
| 22:56.16 | MonMotha | I only go to the ETS when I have to |
| 22:56.27 | MonMotha | and I use the ETA for through-hole, dragging, and other big stuff |
| 22:57.03 | MonMotha | anyway, I need to go |
| 22:57.16 | a7r | late |
| 23:39.28 | RushPL | hey - anybody using CMake for cross compiling? I am trying to have a subdirectory in my project for some tools .. so I need this subdirectory to be built without cross compiling. What's the cleanest way to achieve it? |
| 23:45.02 | prpplague | RushPL: ugh CMake? |
| 23:45.42 | RushPL | yeah. Works quite well all in all, I just need this thing to have everything automated. |
| 23:46.14 | RushPL | I need to build tools that convert my images into C arrays, and then have them as a dependency for the project. |
| 23:50.48 | a7r | prpplague: how's it? |
| 23:51.08 | prpplague | badly |
| 23:51.28 | prpplague | a7r: allegeries are killing me and i have a cough |
| 23:51.55 | a7r | suck |
| 23:52.24 | a7r | I'm knee deep in pxa270 standby,.. so maybe you're doing better than I am. |
| 23:53.10 | SpeedEvil | pxa270... |
| 23:53.13 | SpeedEvil | Isn't that rather old? |
| 23:53.17 | SpeedEvil | Or am I confused. |
| 23:53.24 | a7r | yah, it's old as hell. |
| 23:53.30 | prpplague | speakman: you are confused, but that is beside the point |
| 23:53.35 | prpplague | oops |
| 23:53.38 | a7r | haha |
| 23:53.40 | prpplague | SpeedEvil: that was for you |
| 23:53.41 | a7r | insulting strangers now. |
| 23:53.42 | SpeedEvil | :) |
| 23:53.48 | prpplague | doh |
| 23:53.59 | SpeedEvil | I _think_ I have a PXA270. |
| 23:54.14 | SpeedEvil | It's one of the large pile of crap I've bought on ebay to do something with. |
| 23:57.06 | a7r | SpeedEvil: well this pile of crap is still in production, and used by major automotive OEMs. |
| 23:57.22 | a7r | so I get to learn all sorts of cool shit about it, and Linux's support for it. |
| 23:57.31 | SpeedEvil | a7r: Ah. Yeah - ECUs and automotive in general are totally wierd. |
| 23:57.45 | SpeedEvil | In many ways I wish that one common ECU design would be mandated. |
| 23:58.12 | SpeedEvil | One common ECU can control 99.99% of cars. |
| 23:58.12 | SpeedEvil | There is nothing really magical or special in it. |
| 23:58.19 | a7r | eh,.. well you'd have to have good ECEs to pull that off. |
| 23:58.35 | a7r | automotive engineers can't even agree on what the best position sensor configuration is for engines |
| 23:58.43 | SpeedEvil | I know. |
| 23:58.45 | a7r | let alone software and hardware to run with it. |
| 23:59.00 | a7r | even aftermarket ECUs have all sorts of problems running certain configurations. |
| 23:59.22 | a7r | and that's before you even start looking at OBD-II protocol bullshit. |
| 23:59.24 | SpeedEvil | In practice, there aren't a huge number of sensor inputs and outputs, they are mostly fairly simple. |
| 23:59.44 | *** join/#edev Spida (~timo@ns1.spinnennetz.org) |
| 23:59.52 | SpeedEvil | And modern processors are way fast enough to have a common design that can support any engine. |
| 23:59.54 | a7r | well, yah.. on paper it looks that way, but you've got to deal with huge engineering opinion difference between OEMs. |