irclog2html for #elinux on 20050614

00:31.11MonMothaanyone know of a good, small GPS module that can be purchased in quantity reliably?  that is, not off ebay
00:59.22*** join/#elinux Crofton|laptop (~balister@66-207-66-26.black.dmt.ntelos.net)
01:18.48*** join/#elinux sjhill (~sjhill@spvl-02-110.dsl.netins.net)
01:39.18*** join/#elinux noclouds (~mhfan@60.166.175.246)
02:05.38*** join/#elinux THeli` (total@GO.HELI.RI.CMU.EDU)
02:13.50*** join/#elinux file[laptop] (~file[lapt@mctnnbsah25-142166146254.nb.aliant.net)
02:17.51Genesisbonne nuit
02:32.13*** join/#elinux samrobb (~sam@dsl093-061-094.pit1.dsl.speakeasy.net)
02:39.32GPSFanMonMotha: you more interested in size, cost, power consumption, etc.? Motorola, uBlox, Trimble, RF-Microdevices all have small "low cost" modules, most want large quantity for best price. here is an example of a nice small, low power module: http://www.motorola.com/ies/GPS/products_positioning.html
02:44.15THelianyone familiar with firewire spec or the linux driver?
02:45.11GPSFanMonMotha: http://www.u-blox.com/products/tim_lp.html  http://www.abacuscity.ch/abashop?i=8852269852048&s=142&p=productdetail&sku=44
02:49.51*** join/#elinux file[laptop] (~file[lapt@mctnnbsah25-142166146254.nb.aliant.net)
03:38.26*** join/#elinux hufnus (~slonsiki@mfd1736d0.tmodns.net)
04:02.43*** join/#elinux cdm (~cdm@adsl-69-109-214-98.dsl.pltn13.pacbell.net)
04:21.37chouimat|Zzzznight
04:42.02*** join/#elinux ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
04:42.02*** topic/#elinux is Embedded Linux || http://eLinux.org/ || cross compile, uClibc, busybox, tinylogin, handhelds, post-sale linux installs ;-), etc. || free embedded linux training at http://free-electrons.com/news/news.2004-09-28/en || see prpplague about custom holly-gates jtag dongles
04:42.02*** mode/#elinux [+o ibot] by ChanServ
05:06.04*** join/#elinux andersee (~andersee@codepoet.org) [NETSPLIT VICTIM]
05:06.04*** join/#elinux Evanrude (~david@wsip-68-15-251-34.dl.dl.cox.net)
05:06.04*** join/#elinux sorphin (sorphin@mewp.captainrock.com) [NETSPLIT VICTIM]
05:06.04*** join/#elinux jamie (~jamie@softmodem.org)
05:06.04*** join/#elinux MonMotha (~monmotha@cpe-65-186-200-185.indy.res.rr.com) [NETSPLIT VICTIM]
05:06.04*** mode/#elinux [+o andersee] by irc.freenode.net
05:06.59*** join/#elinux CP|Laptop (~nobody@aus-ext-proxy01.amd.com)
05:08.34*** join/#elinux shido (~shido@d57-87-253.home.cgocable.net)
05:08.37shidowhere is file...
05:29.31*** join/#elinux pigeon (~pigeon@adsl-60-232.swiftdsl.com.au)
06:06.56*** join/#elinux TheCollector (~brian@c-67-187-107-245.hsd1.tx.comcast.net)
06:20.24TheCollectorhmm, looks like I have until 2007 to buy me a PPC Mac
06:21.34TheCollectorer, wishes
06:43.26*** join/#elinux jamie (~jamie@82.152.157.198)
06:52.51sorphinTheCollector: this is why i'm glad i got my mac mini when i did, hopefully the G5s will drop before then so i can atleast get one of em
07:36.27shidoso because they went intel doesnt mean it wont work on AMD, right?
08:37.25*** join/#elinux ade|desk (~adavey@194.200.143.249)
08:44.52*** join/#elinux mallum (~mallum@host81-153-48-235.range81-153.btcentralplus.com)
08:44.52*** mode/#elinux [+v mallum] by ChanServ
09:31.55*** join/#elinux pb_ (~pb@2002:5246:d929:1:20a:5eff:fe00:391e)
09:49.56*** join/#elinux ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
09:49.56*** topic/#elinux is Embedded Linux || http://eLinux.org/ || cross compile, uClibc, busybox, tinylogin, handhelds, post-sale linux installs ;-), etc. || free embedded linux training at http://free-electrons.com/news/news.2004-09-28/en || see prpplague about custom holly-gates jtag dongles
09:49.56*** mode/#elinux [+o ibot] by ChanServ
10:41.16*** join/#elinux cbrake (~cbrake@cpe-204-210-216-174.neo.res.rr.com)
11:06.38*** join/#elinux cbrake (~cbrake@cpe-204-210-216-174.neo.res.rr.com)
11:26.18*** join/#elinux cbrake (~cbrake@cpe-204-210-216-174.neo.res.rr.com)
11:41.30*** join/#elinux prpplague (~billybob@72.22.135.51)
12:00.15*** join/#elinux GPSFan (~Ken@65.121.49.208)
12:10.27*** join/#elinux file[laptop] (~file[lapt@mctnnbsah25-142166146254.nb.aliant.net)
12:44.42*** join/#elinux THeli` (total@GO.HELI.RI.CMU.EDU)
13:42.54prpplagueho diddy ho ho
13:44.26prpplague~lart CosmicPenguin for not being interested in schematics
13:44.40prpplagueibot: botsnack
13:44.40ibotthanks, prpplague
13:49.31THelihey guys
13:53.32CosmicPenguinheh
13:53.39CosmicPenguinwhat the hell do I care about the schematics
13:54.47CosmicPenguinBut just to make you happy, I'm downloading the schematics right now... :)
13:55.38prpplagueTHeli: lo
13:55.46prpplagueCosmicPenguin: hehe good
13:59.29prpplagueCosmicPenguin: re-wire?
14:05.55CosmicPenguinahhhah!
14:07.53prpplagueCosmicPenguin: hmm, is there an easy way to test logic ands and ors from command line
14:08.11prpplagueCosmicPenguin: bc doesn't seem to cover that
14:10.20prpplagueCosmicPenguin: bitwise operations i mean
14:13.26CosmicPenguindon't know of anything right off
14:13.34CosmicPenguinI use a white board for stuff like that
14:14.40prpplagueCosmicPenguin: hehe, yea, i have a spread sheet
14:17.02prpplagueCosmicPenguin: got a c app not working like it should
14:17.11prpplagueCosmicPenguin: works fine on x86 but not on arm
14:17.53CosmicPenguinremember that arm defaults int to unsigned
14:18.54prpplagueCosmicPenguin: yea
14:19.18prpplagueCosmicPenguin: its a bitwise "and" that doesn't appear to be getting the correct value
14:19.31prpplagueCosmicPenguin: need to double check all the values in the register
14:19.41prpplagueCosmicPenguin: sharp has a bad habit of locking some registers
14:21.15CosmicPenguinhmm....  the alchemy boards drives 3.3v to the jtag header, but the dongle uses 3v
14:21.26CosmicPenguindamn geode, always trying to be different
14:21.31prpplagueCosmicPenguin: that should be fine
14:21.55*** join/#elinux Crofton|laptop (~balister@h80ad9065.dhcp.vt.edu)
14:21.58prpplagueCosmicPenguin: that should be close enough if you not powering other hardware
14:22.42prpplagueCrofton|laptop: lo ho
14:22.43CosmicPenguinshe does purty work
14:28.30Croftonyo
14:28.48CosmicPenguinprpplague: whats the general alogrithm for discovering the IR length on a processor?
14:33.15prpplagueCosmicPenguin: thats usually covered in the BSDL file
14:34.24prpplagueCosmicPenguin: you should have gotten a BSDL file for the cpu
14:34.58prpplagueCosmicPenguin: http://www.elinux.org/wiki/BSDL
14:35.27cbrakeCosmicPenguin: more 1550 questions ... Do you happen to know if networking, usb and audio are currently working in the mips-linux tree for the 1550?
14:36.38CosmicPenguincbrake: 2.4. or 2.6?
14:36.51CosmicPenguinprpplague: yeah, but there's a way to discover hte IR - the openwince jtag tool does it
14:37.25cbrakeCosmicPenguin: either, but prefer 2.6
14:38.39prpplagueCosmicPenguin: hmm, i've not worked in detail with it in a month or so, but i would suspect that it would take doing a tap reset then issuing something standard like a bypass command
14:39.23prpplagueCosmicPenguin: pb_ might know
14:39.33prpplagueCosmicPenguin: i always just use the bsdl file
14:39.41CosmicPenguinprpplague: no biggy, I was just playing around with stuff, and I wanted to make sure that my app was working
14:40.24CosmicPenguincbrake: networking should be good - I see all the 1550 definitions in the code
14:42.58cbrakeCosmicPenguin: good.  Hopefully audio and USB will be simliar to other parts ...
14:44.29CosmicPenguincbrake: the au1550 stuff is in there for OSS as well - I'm not sure if we have an ALSA port yet
14:44.37CosmicPenguinbut I see the 1550 ac97 and i2s drivers
14:45.40cbrakeCosmicPenguin: OSS should be OK for what I'm doing
14:45.57CosmicPenguinAnd there should be the ohci driver for HID, but I don't think any of the UDC or OTG stuff has been pushed up
14:46.03CosmicPenguinNot that OTG or UDC is very stable anyway
14:47.35CosmicPenguinI mean in general
14:58.37*** join/#elinux eggsMTS (~eggers@pixpat.austin.ibm.com)
15:00.26cbrakeCosmicPenguin: yeah, don't see too many people using OTG yet.
15:05.23CosmicPenguinIts coming though
15:21.58*** join/#elinux Genesis (~Genesis@dev.aimao.org)
15:23.26Genesisbonjour
15:32.45sorphinholy backscroll batman
15:34.23sorphinlo davey and mr penguin
15:34.42CosmicPenguingreetings
15:40.15ade|desksorphin: that always confuses me, davey being my surname and davey as normal dave greeting
15:41.13*** join/#elinux GPSFan (~Ken@65.121.49.208)
15:52.51prpplaguesorphin: lo
16:04.19*** part/#elinux ade|desk (~adavey@194.200.143.249)
16:16.33prpplagueT0mW: hey crack whore
16:18.32T0mWprpplague: ok, I'm awake now..
16:18.45prpplagueT0mW: finally remembered what i wanted to ask you
16:18.55prpplagueT0mW: i need to get a small scope for debugging
16:19.02T0mWok
16:19.07prpplagueT0mW: this usb one sucks
16:19.11T0mWcheap, anyhow
16:19.16T0mWusb?
16:19.40prpplagueT0mW: yea, its usb with pc software
16:19.59T0mWyou usually can pick up any number of used 100MHz dual-channel scopes from ebay for < $150
16:20.18prpplagueT0mW: yea thats what i was thinking
16:20.44T0mWthat is a "basic scope": 100MHz, dual channel
16:20.49prpplagueT0mW: there was somewhere that had small handheld scopes that were only like 5mhz but probably do what i need
16:21.01T0mW$$$
16:21.05*** join/#elinux cdm (~cdm@A17-213-20-110.apple.com)
16:21.13T0mWwhen you go really small, they cost big $$$
16:21.30prpplagueT0mW: oh, well it might be worth it for shipping and customs
16:21.45pb_you can get little hand-held DSOs with LCD screens for not all that much money.
16:21.54pb_they're no good for high speed signals, but they're fine for basic digital debugging
16:22.00prpplagueT0mW: darn i can't remember where i saw them
16:22.09prpplagueseems like it was halted.com
16:22.11T0mWThe Hitachi scope here is 10"X5"X18"
16:22.26T0mWIIRC, I paid $67 for it
16:22.36T0mW+ shipping
16:22.45prpplaguehttp://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?langId=-1&storeId=10001&catalogId=10001&productId=162430
16:22.51T0mWgak!
16:23.01T0mWI know what you are thinking of doing!
16:23.01prpplagueT0mW: shipping overseas + customs
16:23.07prpplagueT0mW: ?
16:23.20T0mWDave! Don't jump Dave, life is not that bad, Dave!
16:23.25prpplagueT0mW: ?
16:24.12T0mWprpplague: well, if you don't have the room to put a CRT scope, I guess an LCD scope is a second best investment.
16:24.18prpplaguehttp://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?langId=-1&storeId=10001&catalogId=10001&productId=236507
16:24.46T0mWk, looking noiw
16:24.52prpplagueT0mW: yea, well, with a small one line that i could put it in my suitcase
16:26.36T0mW:( those are more "go / no-go" type of scopes, I wouldn't consider them useful for any type of engineering work.
16:26.38prpplaguepb_: have you use those before
16:26.54prpplagueT0mW: yea, i just need stuff to see some basic data
16:27.06prpplagueT0mW: basically glorified multimeters
16:27.07T0mWgood luck
16:27.21*** join/#elinux sjhill (~sjhill@eth13.com-link.com)
16:27.37T0mWbasically, IMO, they are curiousities rather than a peice of test equipment
16:27.42pb_prpplague: no, not those ones
16:27.43prpplagueT0mW: hmm
16:27.49prpplaguepb_: handheld in general?
16:27.54prpplagueT0mW: hmm
16:28.06prpplagueT0mW: hitachi then is a good brand?
16:28.25prpplagueT0mW: be wise to run a model by you before i bid?
16:28.47pb_prpplague: yeah, but it was years ago now.  I don't remember the model, or even the manufacturer.
16:28.56pb_it was bigger than the ones you mentioned, though.  more of a tablet kind of form factor.
16:29.18T0mWnaw, I just needed a peice of sh^H^H junk to look at signals,  just a cheap analog scope
16:30.11pb_heh.  I can't imagine working without a DSO.
16:30.11T0mWI like analog scopes for engineering work, there is a certain amount of info that might get lost with a digital scope
16:30.50pb_that's true.
16:30.59T0mWalthough, Techtronix makes some really fine digital scopes, a 100MHz scope of theirs will digitally oversample a signal
16:31.29T0mWI had a 2465 that I was just enjoyed using
16:31.40T0mWerr, 2430?
16:32.07pb_mm, dunno.  The only Tek scope I have here is an old analog one.
16:32.34T0mWprpplague: look for: A) there is a bright display, B) 100MHz analog, C) they claim that it works ok
16:33.39prpplagueT0mW: hmm, don't need something as high as 100mhz
16:33.56T0mWprpplague: most people who have a good crt in the scope for sale will have a photo of the scope in operation.  It is common to have someone sell a scope with a bad crt.  crt tubes cost more than the value of the used scope.
16:34.08T0mWprpplague: yes you do need 100MHz.
16:34.31prpplagueT0mW: for what?
16:34.38T0mWprpplague: you need the "response"
16:34.46prpplagueT0mW: nothing i'm working with is over 2mhz
16:34.50T0mWdoh
16:34.59T0mWoh really?
16:35.12T0mWwhat about the slew-rate of the signals?
16:35.35prpplagueT0mW: hmmm don't know about that
16:35.47T0mWthe high to low, low to high transitional rates of the digital signals, how fast do you think that "edge" is moving?
16:36.25T0mWthat is where the 100MHz comes in, clearly seeing the operation of the signals switching state.
16:36.38T0mWkeyword is "clearly", heh
16:37.02T0mWprpplague: trust me, 100MHz stuff is very common, very cheap and very necessary
16:37.23prpplagueT0mW: hmm, ok
16:37.35prpplagueT0mW: if i go with an analog scope i'll go 100mhz
16:38.10T0mWthe 35MHz stuff is come-on teasers for hobbyists who are trying to save a few dollars, they eventually junk the thing and get a 100MHz model.
16:39.29T0mWprpplague: digital storage scopes are very good for measuring the time differentials between two signals.  usually you can guesstimate down to a few nano-seconds.
16:39.33sorphiniirc, my scope is a 350mhz
16:39.51sorphinor 300
16:40.36prpplagueT0mW: hmm, i'm just worried about buying more than i need
16:40.40T0mWprpplague: analog scopes have a purpose as well, they are more "all around" type workers.  They will show you more information about power supply noise and signal ring / overshoot than a digital scope will.
16:41.29prpplagueT0mW: yea, i don't plan on dealing with any of that
16:41.34T0mWprpplague: digital scopes tend to "average" the signal they are seeing, analog scopes are more representative of what the signal actually is doing.
16:41.44T0mWok
16:41.46prpplagueT0mW: just need to look at the occasional data signals
16:41.52prpplagueT0mW: gpios and such
16:42.07sorphinanalog has worked fine for me
16:42.43T0mWprpplague: logic probes are cheap, not to be nasty or nothin', but sometimes all I need is a logic probe (I do have a cheap-o Radio Shack model I use).
16:43.23prpplagueT0mW: yea, thats kinda what i was thinking
16:43.46T0mWand, a logic probe will "sing" for you.  This way you can bang bits with the debugger and listen to the signal to see if it changes state.  hard to do that with a scope.
16:44.19T0mWprpplague: anyway, happy shopping!  :)
16:44.34prpplagueT0mW: thanks for the scoupe
16:49.17prpplagueT0mW: hmm, might go with a logic probe for now
16:52.21sorphinheh, my logic probe never sang
16:52.29sorphinjust green or red LED
17:07.15prpplagueanyone got any ideas on ways to optomize a gpio based keypad matrix?
17:08.33sjhillyeah, use one switch and send commands with Morse code
17:08.37sjhillbhuwhahaha
17:08.38sjhillsorry
17:08.55prpplaguesjhill: hehe
17:09.59prpplaguesjhill: do yall keep a cage full of monkeys to test you DoD stuff? hehe
17:10.10T0mWprpplague: Nat Semi used to have some keymatrix encoders, at one tome.
17:10.11sjhillyeah, we hire them from Barbados
17:10.15T0mWs/tome/time/
17:10.35T0mWfaugh! DoD, bad word!
17:11.10sjhillif you don't like it, you shouldn't use the Internet then
17:11.13T0mWanytime soon that I get involved in another project for those bastids, I'll charge double
17:11.18sjhilloh
17:11.45T0mWsjhill: petty buearucrats (sp?) drive you nuts with nit-picking
17:11.55prpplaguesjhill: hehe, yea, we have lots of green monkeys here
17:12.09prpplaguesjhill: i can't seem to keep them out of our mango tree
17:12.29prpplagueT0mW: i ment software wise
17:12.36*** join/#elinux Soopaman (~soopaman@67.71.84.91)
17:12.40prpplagueT0mW: i have a kernel driver i've written
17:12.46*** join/#elinux hufnus (~slonsiki@m5b3236d0.tmodns.net)
17:13.01prpplagueT0mW: but i'm thinking there is probably a smoother way of doing the matrix scans
17:13.07T0mWprpplague: bureaucrats or monkeys ?
17:13.37prpplagueT0mW: hehe, monkeys in the back yard, but have plenty of burearocrats
17:14.09prpplaguehttp://www.barbados.org/monkeys.htm
17:15.03T0mWprpplague: I've always handled keys / switches with some form of periodic checking (typically interrupt timer), then dec a counter until it hits a low-water level, then report the key.  if I ever read a non-closure, I reset the downcounter back to the top.
17:15.30T0mWkeybounce always is a PITA
17:16.05T0mWkeys will tend to "bounce" when they get older and start to corrode / carbon up.
17:16.28prpplagueT0mW: yea, i'm thinking more about the matrix scan
17:16.38T0mWprpplague: does monkey taste like chicken?
17:16.42prpplagueT0mW: i'm just walking a one through the matrix
17:17.11prpplagueT0mW: don't know about the meat, but the chilled brains taste like a cross between jello and pudding
17:17.47T0mWi'd put all output pins low, pullup the input pins, then look for "any key down", and resolve from there
17:18.00T0mWe.g. then walk a zero
17:18.11prpplagueT0mW: yea, thats what i meant
17:18.23prpplagueT0mW: said one ment zero
17:18.33T0mWah, boolean logic
17:18.44T0mWDave == !!Dave
17:19.05prpplaguehehe
17:19.16prpplagueradioshit doesn't even sell a probe anymore
17:19.17T0mW(Dave AND Donna) == ((!Dave) OR (!Donna))
17:19.41prpplaguehehe
17:19.52T0mWprpplague: Arby's doesn't sell the Rueben sandwich anymore either
17:20.12T0mWsuckers, I went there yesterday an wanted one
17:20.54CosmicPenguinsum bitch
17:21.07prpplagueCosmicPenguin: hehe
17:21.08CosmicPenguinsegfaults are bad - double bad when they happen before the console is started
17:21.15T0mW(Dave AND Donna) == !((!Dave) OR (!Donna))
17:23.04T0mWprpplague: this is odd-funny, you never realize that you are living in someone else's idea of a beautiful place.
17:23.53prpplagueT0mW: hehe
17:24.06T0mWprpplague: I was talking to a guy from Southern California out here into Northeaster Pennsylvania for the first time.  He just couldn't get over how green everything was, all the trees, etc.
17:24.20prpplagueT0mW: here's a project using a pic to do an 8 channel logic probe
17:25.02T0mWusually an NE555 timer, a pizeo beeper + some LEDs is all a logic probe is.
17:25.56prpplaguei used to have a really nice probe, till my ex-wife sold all my stuff
17:26.14fishheadHAHAHAHAHAHA
17:26.20fishheadprpplague i used to have a really nice probe, till my ex-wife sold all my stuff
17:26.32T0mWhttp://www.abra-electronics.com/catalog/probes/el_LP-425K.html
17:26.43fishheadprpp so your ex ruined your probe eh?
17:27.03T0mWLP-550 is basically what I have
17:27.19fishheadt0mv
17:27.31fishheadhow many more decades do you think people will use the 555 in products :P
17:27.32*** join/#elinux TimRiker (~timr@TimRiker.bronze.supporter.pdpc)
17:27.32*** mode/#elinux [+o TimRiker] by ChanServ
17:27.53T0mWshhh, Tim is here
17:28.02CosmicPenguinI didn't say anything
17:28.50sjhillCosmicPenguin: you said he was a troll
17:29.05T0mWprpplague: I've been hacking the Sicom linux o/s for their registers.  Trying to get it to run under VMware.
17:29.22prpplaguefishhead: sold all my stuff, oscope, probes, meters soldering equipment
17:29.35fishheadprpp you missed the joke :)
17:29.55T0mWprpplague: it's running, but I cannot get the mouse to work.  They are using, basically, a RedHat 6.1 installation, modernized with bug fixes and stuff.  Linux 2.4.12 kernel
17:30.28*** join/#elinux andersee (~andersee@codepoet.org)
17:30.28*** mode/#elinux [+o andersee] by ChanServ
17:30.46T0mWandersee: wb
17:31.06prpplagueT0mW: lovely
17:32.00T0mWprpplague: overall, it is a nice little system: Geode 300MHz processor + some CF
17:32.09fishheadmmmmm
17:32.17anderseeT0mW: morning
17:32.31T0mWandersee: yeah, it is morning for me too
17:33.27T0mWoh, and the RedHat distro uses busybox
17:33.42T0mWno uClibc though
17:33.52T0mWit is a glibc system
17:34.23anderseejust wait, they'll switch eventually due to the vast superiority of uClibc
17:34.24andersee:-)
17:34.42fishheadheheeh
17:34.42T0mWandersee: modesty becomes you, Erik
17:34.58prpplaguefishhead: d00d jsut drop it
17:35.02fishhead?
17:35.20fishheaddrop what ?
17:35.28prpplaguefishhead: you know what
17:35.34fishheadfishhead mmmmm
17:35.37fishheadfishhead heheeh
17:35.40fishhead?
17:35.52fishheadoh
17:35.56fishheaddude that was like 5+ mins ago
17:35.58fishheadyou lagged ?
17:36.00prpplaguefishhead: stop pm'm me with sill shit
17:36.02prpplaguesilly
17:36.49prpplagueT0mW: soooo, what i'm doing right now on the matrix is walking the zero and recording all the values
17:37.01prpplagueT0mW: then doing the keydecoding afterwards
17:37.22prpplagueT0mW: this is as opposed to walking the zero and check as you go
17:37.28prpplagueT0mW: any other ideas?
17:38.28T0mWthat's basically it
17:38.54prpplaguefishhead: if i was petty, i'd act like you
17:39.05fishheadno petty is mentioning it in the channel
17:39.13fishheadgo screw yourself
17:39.15*** part/#elinux fishhead (~xussgig@c-24-3-15-166.hsd1.pa.comcast.net)
17:39.19T0mWyou could do something like put AA's on the output port, then 55's on the output, and could maybe resolve faster
17:39.35prpplagueyea thats what i was thinking
17:39.40T0mWdepends on the organization of the matrix
17:40.13sjhillT0mW: follow the white rabbit......
17:40.26T0mWmathematically combine the output + input values and use as an index into a table to get the keycode
17:41.07T0mWafter all, memory is cheap
17:41.22T0mWso you use a K to do your key decode...
17:42.35prpplaguehmm, so looking at the math would it be possible to reduce the matrix scan from 8 read/writes to something like 4 or less
17:43.10T0mWon an 8 x 8 matrix, I would think it would be something like two write / reads
17:43.26prpplagueinteresting, i wonder if i can find some whitepaper docs on this
17:43.46prpplaguehmm
17:44.16prpplagueT0mW: so what about multiple key combinations........
17:45.17T0mWI haven't thought it all the way through, but, if you output AA and read back A4, then you have a bit2 active, table the (char *) KEYS [A4] to get the keycode?
17:45.57T0mWif you have two keys pressed, e.g. AA reads back A0, then the entry in KEYS [A0] could be 0xff ?
17:46.01T0mW0xff == bad key
17:46.20prpplagueT0mW: why bad key?
17:46.41T0mWunless you had a situation like CTRL-x keys
17:46.57prpplagueT0mW: shift+alpha
17:47.01T0mWa 256 table should handle an 8 x 8 matrix
17:47.02T0mWok
17:47.22prpplagueT0mW: hmm, i'll have to write up some tests
17:47.32T0mWlike I said, I haven't thought it all the way through
17:47.39prpplagueT0mW: but i'd think this is a fairly common application of a microcontroler
17:47.51T0mWbeen a while since I've fussed with key matrices
17:47.52prpplagueT0mW: so i'd suspect that someone somewhere has written aobu tit
17:47.58T0mWnaw
17:48.21T0mWit is more like knowledge everyone assumes that everyone else has. heh
17:50.46prpplaguehmm, i find that hard to believe that no one would have written some sort of logical study of keyboard matrix development
17:51.51T0mWright, like they would write about outputing segments of a 7 segment LED display, assumed knowledge
17:52.35prpplaguehttp://www.avrfreaks.net/index.php?module=FreaksTools&func=viewItem&item_type=tool&item_id=390
17:54.47prpplagueahhh interesting
17:54.53prpplaguethat does provide some light
17:55.55T0mWprpplague: similar to what I said
17:56.00prpplagueT0mW: yea
17:56.29T0mWtwo write / reads, only I think their reversing the port directions may have more merit.
17:57.24prpplagueT0mW: yea, from what i read though, that is for the single key design
17:57.36prpplagueT0mW: however i need to support modifiers
17:57.43T0mWproblem is, that once you isolate a bit in both directions, you have to resolve the bit by shifting / anding.  that will take a interative loop.  I don't like that
17:58.42prpplagueT0mW: yea, don't have a choice
17:58.49prpplagueT0mW: i _have_ to support those
17:59.06T0mWprpplague: well, you have to take it as the organization of the matrix + requirements demand.  I've not dealt with modifier keys, the matrices I've dealt with have been single key sense.
17:59.42T0mWI still think that, within limits, you could sense simultaneous keys.
18:00.16T0mW... using the two step write / read
18:00.29prpplagueT0mW: yea, i'll have to look into it more
18:00.44prpplagueT0mW: thanks for the info, gotta go pick the critters up from school, brb
18:00.57T0mWyup
18:54.36CosmicPenguinwoo
19:03.24*** join/#elinux jacmet (~jacmet@d51A52AE6.access.telenet.be)
19:43.24*** join/#elinux jacmet (~jacmet@d51A52AE6.access.telenet.be)
19:43.46*** join/#elinux file (~jcolp@66.199.241.90)
19:43.46*** mode/#elinux [+v file] by ChanServ
19:45.14*** join/#elinux darkschneider (~gab@213-140-6-96.fastres.net)
19:57.24*** join/#elinux ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
19:57.24*** topic/#elinux is Embedded Linux || http://eLinux.org/ || cross compile, uClibc, busybox, tinylogin, handhelds, post-sale linux installs ;-), etc. || free embedded linux training at http://free-electrons.com/news/news.2004-09-28/en || see prpplague about custom holly-gates jtag dongles
19:57.24*** mode/#elinux [+o ibot] by ChanServ
19:57.56prpplagueCosmicPenguin: interesting
20:02.27CosmicPenguinIt would have taken me months to figure that out with out a in-kernel debugger
20:03.08prpplagueCosmicPenguin: yea, i've not needed it yet with the code i'm working, but i figure the next round i'm gonna need it
20:16.26*** join/#elinux ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
20:16.26*** topic/#elinux is Embedded Linux || http://eLinux.org/ || cross compile, uClibc, busybox, tinylogin, handhelds, post-sale linux installs ;-), etc. || free embedded linux training at http://free-electrons.com/news/news.2004-09-28/en || see prpplague about custom holly-gates jtag dongles
20:16.26*** mode/#elinux [+o ibot] by ChanServ
20:27.02*** join/#elinux prpplague (~billybob@72.22.132.158)
20:30.38sorphinprpplague: re islander
20:30.50prpplagueyea yea
20:30.56prpplaguehad to change some fw settings
20:31.01sorphinah
20:31.43prpplaguesorphin: please don't, i'm no security guru
20:31.54prpplaguesorphin: i normally just turn _every_ off
20:33.16prpplaguesorphin: oh ok
20:33.17prpplague:D
20:33.29jacqueswell, he's *required* to say that
20:33.40sorphinjacmet: eh?
20:33.42sorphingrr
20:33.50sorphinjacques: eh?
20:34.00sorphinwhat's that mr miata?
20:34.11jacquesyou;re required to say you were only joking
20:34.20sorphini am?
20:34.27prpplaguethis keypad matrix is running way too slow, gonna have to make some changes
20:34.31jacquesfor plausible deniability
20:34.42sorphinjacques: heh, i don't need to deny anything
20:35.30sorphinlooks either owned, or being used as a proxy
20:36.06sorphintho we have shutdown authority now, we don't have the procedure set yet
20:36.23sorphinnow i get ssh scanned all the time
20:36.33sorphinbut this was 610 attempts on 1 box, 2000+ on another
20:36.56jacqueswhat's a ssh scan? looking for an open ssh port?
20:37.16sorphinyeah, that, and then attempting to login to said found ssh port
20:37.22sorphintrying to brute force
20:37.36sorphinthat's the illegal part
20:37.48sorphinportscanning is normal noise
20:37.56sorphinbut many many active attempts to break in
20:37.59sorphinisn't
20:38.12*** join/#elinux prpplague^2 (~billybob@216.110.100.26)
20:38.24sorphindamn, hillbilly ^2'd
20:38.46jacquescould you tell what sorts of username/password pairs they were trying?
20:38.52sorphinyep
20:38.59prpplague^2sorphin: hehe, forgot to close out before restarting firewall
20:39.00sorphinwell, usernames anyways
20:39.03sorphinnot pws
20:39.16jacquesprobably pw derived from username along with common
20:39.21sorphinwell
20:39.28sorphinthis one was a consistant 'ROOT' in all caps
20:39.44sorphinnormal attempts people like to make are random names
20:40.12sorphini.e.
20:40.13sorphin: illegal user IBM
20:40.16sorphinillegal user Jubilados
20:40.21sorphinillegal user Dahlan
20:40.23sorphinetc
20:40.24sorphinetc
20:40.43sorphinand they'll try normal names too
20:40.58jacquesi wonder if they thought username in all caps would get then free case insensitivity on the pw
20:40.59prpplague^2hmm, i guess deepak got tired of elinux
20:41.29sorphinprpplague: eh?
20:41.56prpplague^2sorphin: just haven't seen deepak on irc lately
20:42.02sorphinwho's that?
20:44.11prpplague^2deepak saxena, he's one of the gurus at montavista
20:44.18sorphinjacques: if i ever get this other stuff setup, (non-cisco), so i can do it how i want, several attempts on ssh (since that's the only thing i have open but my dns servers and my mailserver),
20:44.28prpplague^2~dsaxena
20:44.30sorphinwill resort in auto firewalling
20:44.33sorphinah
20:44.40sorphindidn't know his 1st name
20:44.40prpplague^2~seen dsaxena
20:44.41ibotdsaxena <~dsaxena@c-24-20-128-133.client.comcast.net> was last seen on IRC in channel #elinux, 124d 16h 37m 7s ago, saying: 'haven't played with usbnet'.
20:44.53jacques124d damn
20:45.06prpplague^2that doesn't sound right
20:45.18sorphincould just be no time
20:46.20prpplague^2sorphin: hehe, guess he has been busy http://www.plexity.net/
20:52.10CosmicPenguinheh - I get lots of those USB scans
20:53.03sorphinuhh
20:53.08CosmicPenguiner
20:53.08CosmicPenguinssh
20:53.09CosmicPenguinwhatever
20:53.12CosmicPenguinshutup
20:53.14sorphinhehe
20:53.37sorphinCosmicPenguin: well, if you didn't run that pr0n server, that wouldn't happen ;)
21:06.42*** join/#elinux CP|Laptop (~nobody@aus-ext-proxy01.amd.com)
21:10.12CosmicPenguinprpplague: yo
21:10.54CosmicPenguinprpplague: how familiar are you with the inter functionality of MMC?
21:11.42*** join/#elinux cantona (~cantona@cantona.user)
21:15.19prpplagueCosmicPenguin: somewhat, although i've not looked at it in about two weeks
21:18.08CosmicPenguinprpplague: the kernel expects the BUSY bit to be set in the OCR response after a SET_COND (02) command
21:18.27CosmicPenguinBut thats not what my card is returning
21:18.45prpplagueCosmicPenguin: hehe, what are you getting?
21:18.57prpplaguei'm pretty sure i ran into that
21:18.58CosmicPenguinff8000
21:19.19CosmicPenguinwhich seems correct as far as the mask is concerned
21:19.26prpplagueCosmicPenguin: argh
21:19.35prpplague~lart prpplague for not making better notes
21:19.57prpplagueyea looks like i had a similiar problem about 2 months ago, but i didn't write down what the problem was
21:20.29*** join/#elinux cbrake1 (~cbrake@cpe-204-210-216-174.neo.res.rr.com)
21:20.58prpplagueCosmicPenguin: give me sec to look
21:22.58prpplagueCosmicPenguin: only notes i have was that it was a timming issue
21:23.24prpplagueCosmicPenguin: i can't remember exactly but it had to do with the dummy clocks
21:23.36prpplagueCosmicPenguin: is this a true mmc card? or sd?
21:23.40CosmicPenguinmmc card for now
21:24.02prpplagueCosmicPenguin: are you providing the 80 dummy clocks ?
21:24.10CosmicPenguinthe SD spec says the bit is set to low if the startup routine hasn't completed
21:24.21prpplagueCosmicPenguin: right
21:24.57prpplagueCosmicPenguin: how fast are you running the interface?
21:24.58CosmicPenguinThe clock is applied, I believe
21:25.10THelianyone know how to compile just one module in the linux sources? i'm making some changes to the firewire driver but doing a 'make modules' everytime is kinda painful
21:25.47kergothiirc make SUBDIRS=drivers/blah modules
21:25.50CosmicPenguinprpplague: 400Khz
21:26.01prpplagueCosmicPenguin: that should be fine
21:26.18prpplagueCosmicPenguin: are dealing directly with the sd/mmc controler?
21:26.48CosmicPenguinyep
21:26.49CosmicPenguinhmmm....
21:26.51prpplagueCosmicPenguin: see this is where your gonna need the schematic
21:27.07prpplagueCosmicPenguin: is it available for dl?
21:27.25CosmicPenguinThe schematic isn't going to do me a bit of good, if the controller is in the silicon
21:27.40THelikergoth, that seems to work.. thanks
21:27.45kergothnp
21:27.52prpplagueCosmicPenguin: but it would give you an idea about the interface being use
21:27.53prpplagueused
21:28.43*** join/#elinux Soopaman (~soopaman@67.71.84.91)
21:29.39prpplagueCosmicPenguin: the sequence i'm using is, cs inactive(high), provide 8 dummy clocks, cs active(low), send_op_cond, wait for r1 response
21:30.02prpplagueCosmicPenguin: i have comments about the dummy clocks as being absolutely nessary
21:30.41prpplagueCosmicPenguin: if you are using a controler, it'd probably be just one dummy byte passed to the controler
21:31.40prpplagueCosmicPenguin: ok, so after the r1 response, i go cs inactive again, and repeat
21:33.09THeliprpplague, hows the weather down there
21:34.47CosmicPenguinprpplague: wait - which r1 response?
21:40.02prpplagueone sec
21:40.42CosmicPenguinI just told my boss that the Ipaq guys had SD support
21:40.46CosmicPenguinI better not have been a liar
21:45.58CosmicPenguinprpplague: hmmm - I waasn't turning off the clock when the frequency was changing
22:02.32prpplagueCosmicPenguin: ahh
22:03.05CosmicPenguinprpplague: false alarm  -didn't matter
22:09.03*** join/#elinux ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
22:09.03*** topic/#elinux is Embedded Linux || http://eLinux.org/ || cross compile, uClibc, busybox, tinylogin, handhelds, post-sale linux installs ;-), etc. || free embedded linux training at http://free-electrons.com/news/news.2004-09-28/en || see prpplague about custom holly-gates jtag dongles
22:09.04*** mode/#elinux [+o ibot] by ChanServ
22:09.07prpplagueCosmicPenguin: sorry to bail d00d but i have to run, would you like for me to email you some test code i'm using in blob?
22:10.16CosmicPenguinNah, I'll figure it out
22:10.17CosmicPenguinthanks though
22:10.42prpplagueCosmicPenguin: ok, see ya in the morning, keep better notes than i did, so i can correct mine
22:10.46prpplagueCosmicPenguin: later
22:12.07*** join/#elinux Soopaman (~soopaman@67.71.84.91)
22:34.17*** join/#elinux CP|Laptop (~nobody@aus-ext-proxy01.amd.com)
22:47.14*** join/#elinux Crofton|laptop (~balister@66-207-66-26.black.dmt.ntelos.net)
22:53.24*** join/#elinux CP|Laptop (~nobody@aus-ext-proxy02.amd.com)
23:22.06*** join/#elinux file (~jcolp@mctn1-3084.nb.aliant.net)
23:22.06*** mode/#elinux [+v file] by ChanServ
23:43.49*** join/#elinux file[laptop] (~file[lapt@mctn1-3084.nb.aliant.net)
23:50.56*** join/#elinux markl_ (mark@dsl093-225-013.slc1.dsl.speakeasy.net)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.