| 00:03.14 | *** join/#edev joe_bleau (n=joe_blea@adsl-76-213-140-119.dsl.fyvlar.sbcglobal.net) |
| 00:11.07 | MonMotha | hum, anyone know an ARM with the necessary muscle to run Linux (i.e. MMU and such) with an LCD controller and available in a QFP package? |
| 00:11.17 | MonMotha | I can't deal with BGAs on this project... |
| 00:11.29 | MonMotha | I may just have to make my own LCD controller here |
| 00:13.39 | ds2 | Atmel |
| 00:13.49 | ds2 | the RM9200 I think has an LCD controller |
| 00:15.18 | ds2 | I am sure it has one come to think of it.. one board comes with a LCD |
| 00:19.31 | MonMotha | it does? |
| 00:21.32 | ds2 | pretty sure...let me find the page for that board |
| 00:21.57 | MonMotha | I see no mention of an LCD controller in the datasheet |
| 00:22.02 | MonMotha | you sure it didn't have an external controller |
| 00:23.08 | MonMotha | I can build myself an LCD controller on a small FPGA or something |
| 00:23.12 | MonMotha | but having an integrated one would be nice |
| 00:23.25 | MonMotha | handling all the bus arbitration gets annoying with external ones |
| 00:25.31 | ds2 | http://www.kwikbyte.com/KBAT7.html |
| 00:26.00 | ds2 | this board comes with an LCD... but I am not completely satisify with their support (unanswered emails) |
| 00:28.30 | ds2 | section 18.5.6 LCD Interface Mode so it has some kind of controller for it...might not be full blown LCD though |
| 00:29.01 | MonMotha | that's also an ARM7TDMI |
| 00:29.04 | MonMotha | so no linux |
| 00:29.36 | MonMotha | you also only get 256k, so definitely no Linux |
| 00:30.57 | MonMotha | I want to use one of the sparkfun "PSP" LCDs, so needs some umph |
| 00:31.07 | MonMotha | also, I want to decode video, so, um, umph :) |
| 00:31.59 | MonMotha | the 9261 from Atmel would be perfect if it were avaialble in QFP like the 9260 is |
| 00:32.58 | joe_bleau | NXP (nee Sharp) Bluestreak? I thought they had a 720T core with MMU and LCD controller. |
| 00:33.15 | ds2 | huh? |
| 00:33.25 | MonMotha | I also have to be able to buy them in very small quantity |
| 00:33.26 | ds2 | it says ARM920T right there on Atmel's page |
| 00:33.29 | MonMotha | this is just a personal project |
| 00:33.40 | MonMotha | ds2: I was referring to the KBAT7 |
| 00:33.45 | MonMotha | that's a SAM7X from Atmel |
| 00:33.51 | ds2 | oh sorry, wrong link then |
| 00:34.07 | MonMotha | I like Atmel, but their hardware is a bit wonky sometimes |
| 00:34.07 | ds2 | I did purchase a 9200 board from them with an LCD screen...let me check again |
| 00:34.10 | MonMotha | good support, at least |
| 00:34.18 | MonMotha | watch out for the errata, though... |
| 00:34.27 | ds2 | hahah... can't be as bad as AMD's ;) |
| 00:34.40 | MonMotha | check out the errata on the SAM9263 |
| 00:34.56 | ds2 | http://www.kwikbyte.com/KB9202.html |
| 00:35.01 | ds2 | think this is the one i have |
| 00:35.23 | joe_bleau | LH79525 shows a "windows CE" MMU |
| 00:37.15 | MonMotha | yeah, I was looking at the NXP stuff |
| 00:37.33 | MonMotha | they offer high-speed USB, which would be nice for this application |
| 00:38.11 | ds2 | the 9200 has USB host which was the reason I looked at it |
| 00:38.45 | MonMotha | yeah, I don't need that |
| 00:38.50 | MonMotha | I"m basically making an overblown MP3 player |
| 00:39.02 | ds2 | Oh heh |
| 00:39.10 | ds2 | why not use a SiLabs 8051 + decoder chip? |
| 00:39.26 | joe_bleau | Linux on an '8051? ;-) |
| 00:39.28 | MonMotha | because I also want Vorbis, AAC, and FLAC |
| 00:39.34 | MonMotha | and I want video (MPEG1/2/4 ASP) |
| 00:39.39 | ds2 | got to get someone to work on ELKS ;) |
| 00:39.45 | MonMotha | at low res, I know those 926Ts are capable of MPEG 4 ASP |
| 00:39.48 | ds2 | Oh I see |
| 00:39.50 | joe_bleau | Does Blackfin have an LCD controller? |
| 00:40.05 | ds2 | is blackfins available in QFP? |
| 00:40.29 | joe_bleau | Ouch. Probably not. Good point. |
| 00:41.14 | MonMotha | yeah, that's the thin |
| 00:41.15 | MonMotha | g |
| 00:41.40 | MonMotha | this is a personal project, so I can't afford to pay somebody to get a BGA soldered down for me |
| 00:41.45 | MonMotha | if I could, I'd just use a 9261 |
| 00:42.10 | joe_bleau | What's it cost to mount and inspect a BGA? $50 or so? |
| 00:42.43 | joe_bleau | If you've got any volume at all with a CM, they might do it as a favor once you explain that it's a home project. |
| 00:43.00 | MonMotha | it's about $50/chip, but only in the context of full assy services |
| 00:43.27 | MonMotha | QFPs of any pitch I can just do myself |
| 00:43.40 | MonMotha | basically, if it has leads, I can probably solder it |
| 00:43.41 | joe_bleau | Well sure, QFP. |
| 00:44.05 | joe_bleau | Want me to ask my CM what they'd charge to solder and inspect a single chip? |
| 00:44.13 | MonMotha | the 0.5mm pitch QFPs can get somewhat challenging, but still quite doable |
| 00:44.15 | MonMotha | sure |
| 00:44.21 | MonMotha | though the 9261 is also an annoying pitch |
| 00:44.26 | MonMotha | just a bit to small for my board house to make |
| 00:44.48 | *** join/#edev Crofton (n=balister@75.146.17.177) |
| 00:44.54 | MonMotha | would possibly allow for some of the NXP stuff |
| 00:46.26 | ds2 | how do you do 0.5mm pitch by hand? flood solder then wick off? |
| 00:47.42 | Crofton | that is what I would try |
| 00:47.47 | Crofton | use lots of flux |
| 00:47.48 | joe_bleau | Yeah, pretty much. If I'm careful with the solder, I don't need much wick action. |
| 00:48.15 | ds2 | so that pretty much demands a board be made |
| 00:48.21 | joe_bleau | I tend to spend more time aligning the chip than actually soldering it. |
| 00:48.24 | *** join/#edev sunrunner20 (n=jamesdoe@pool-71-252-234-221.dllstx.fios.verizon.net) |
| 00:48.45 | MonMotha | ds2: flux the bejeezus out of it and drag |
| 00:48.47 | ds2 | I been really wanting to try toaster oven assembly |
| 00:49.03 | MonMotha | I make the pads a LOT longer than they need to be so that I can just drag the excess away from the part if I bridge |
| 00:52.33 | MonMotha | joe_bleau: same here |
| 00:52.33 | joe_bleau | Flux is a gripe of mine. |
| 00:52.33 | MonMotha | aligning the chip and tacking the corners down takes about 3-5x as long as soldering the thing down |
| 00:52.33 | MonMotha | and yes, the flux is messy |
| 00:52.34 | joe_bleau | I'd love to find a good CM that would do SM with good old rosin flux. |
| 00:52.44 | joe_bleau | None of this aqueous or no-clean crap. |
| 00:52.47 | MonMotha | that's what I use when I assemble |
| 00:52.57 | MonMotha | have to clean it, but it cleans up reasonably |
| 00:53.02 | joe_bleau | What do you clean with? |
| 00:53.07 | joe_bleau | I used to use freon. |
| 00:53.42 | ds2 | IPA seems to work for me |
| 00:53.52 | ds2 | at least the 70% stuff seems to |
| 00:55.35 | joe_bleau | We had some really strong Kester solvent, but I won't deal with the fumes. |
| 00:56.13 | joe_bleau | IPA doesn't seem to really attack the no-clean that I use for rework. |
| 00:56.47 | joe_bleau | Someone on USENET once scared me by explaining that cleaning might not be a good thing. |
| 00:56.51 | MonMotha | joe_bleau: rubbing alcohol |
| 00:57.03 | joe_bleau | (Something about leaving behind the active salts?) |
| 00:57.17 | MonMotha | which is basically what many commercial flux cleaners are |
| 00:57.26 | MonMotha | get the 99% (or better, anhydrous) stuff |
| 00:58.09 | joe_bleau | I think we have 90% for wal-mart... |
| 00:58.39 | ds2 | I wonder if the iso-heet stuff is anhydrous |
| 00:58.55 | MonMotha | joe_bleau: 90 |
| 00:58.59 | MonMotha | joe_bleau: 90% is about the minimum |
| 01:23.44 | *** join/#edev FireEgl (i=FireEgl@Sebastian.Tcldrop.Com) |
| 01:29.47 | *** join/#edev Off_Namuh (i=GPS@gateway/tor/x-8b5cb2559f09e21c) |
| 01:46.27 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 03:17.51 | *** join/#edev joe_bleau (n=joe_blea@70.233.52.78) |
| 04:07.27 | *** join/#edev redblue (i=star@ppp022.108-253-207.mtl.mt.videotron.ca) |
| 04:18.36 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 04:22.19 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 04:47.25 | *** join/#edev linac (n=lin@167.86.76.222.broad.fz.fj.dynamic.163data.com.cn) |
| 05:14.37 | *** join/#edev FireEgl (i=FireEgl@Sebastian.Tcldrop.Com) |
| 05:52.35 | *** join/#edev Crofton (n=balister@75.146.17.177) |
| 06:25.52 | *** join/#edev howtoo (n=howtoo@oslo-static64.bluecom.no) |
| 07:17.30 | *** join/#edev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 08:13.58 | *** join/#edev ALVAN (n=hgjhgjg@89.34.187.225) |
| 09:35.59 | *** join/#edev emil|k (n=emil|k@e179236205.adsl.alicedsl.de) |
| 09:49.34 | *** join/#edev howtoo (n=howtoo@oslo-static64.bluecom.no) |
| 12:17.24 | *** join/#edev linac (n=lin@167.86.76.222.broad.fz.fj.dynamic.163data.com.cn) |
| 12:39.11 | *** join/#edev redblue (i=star@ppp237.108-253-207.mtl.mt.videotron.ca) |
| 13:10.44 | *** join/#edev brad_mssw (n=brad@shop.monetra.com) |
| 13:14.22 | *** join/#edev prpplague (n=DaveAnde@mail.americanmicrosystems.com) |
| 13:15.54 | prpplague | hmmm www.arm.linux.org.uk seems to be down this morning |
| 14:01.16 | MothaWork | hum, anyone know if chaining multiple devices on a JTAG scan chain breaks the ICE functionality on ARMs? |
| 14:01.26 | MothaWork | I know the ICE functionality isn't exactly JTAG standard |
| 14:02.28 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 14:02.34 | prpplague | MothaWork: no it should not break it |
| 14:03.29 | MothaWork | good |
| 14:03.56 | MothaWork | I'm also curious why the industrial temp of this PHY has JTAG, but the commercial temp version lists those pins as reserved/do not connect |
| 14:10.54 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 14:12.25 | *** join/#edev lyakh_ (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 14:12.29 | prpplague | MothaWork: probably nothing more than bad docs |
| 14:14.55 | *** join/#edev Chocobo (n=swinchen@strongbad.eece.maine.edu) |
| 14:27.23 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 14:43.34 | chouimat | morning |
| 14:47.23 | *** join/#edev CosmicPenguin (n=nobody@crank.laptop.org) |
| 14:58.17 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |
| 15:19.07 | *** join/#edev lyakh_ (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 15:23.54 | brad_mssw | CosmicPenguin: did some more research on that DMA issue on the soekris net5501 ... apparently other people have seen the same thing and think it's a bios issue |
| 15:26.01 | CosmicPenguin | I tend to agree |
| 15:26.47 | brad_mssw | seems like a pretty major oversight on their part ... |
| 15:27.40 | CosmicPenguin | probably not paying attention |
| 15:27.58 | CosmicPenguin | There are little nuggets of functional goodness hidden all over the PCI space |
| 15:31.11 | brad_mssw | CosmicPenguin: as a 'workaround' do you think I could issue a 'setpci' command? |
| 15:32.49 | CosmicPenguin | I don't think so - because the "decision" is made while you are stiill in user space |
| 15:32.53 | CosmicPenguin | err, kernel space |
| 15:33.28 | brad_mssw | CosmicPenguin: what if the driver is in a module, and the setpci is done before the module is loaded, say in an initramfs ? |
| 15:36.20 | brad_mssw | the issue was first reported 3 months ago, when the net5501 was new, so I'm assuming the manufacturer (soekris) has no intent of fixing it |
| 15:38.44 | brad_mssw | setpci -d 1022:209a 40.b=02 ... maybe ... |
| 15:40.10 | brad_mssw | maybe I can brick this useless board while I'm at it |
| 15:41.09 | CosmicPenguin | brad_mssw: yeah - the initrd thing would work |
| 15:41.18 | CosmicPenguin | The mask should be 1 |
| 15:41.25 | CosmicPenguin | or 3, if you want to be safe |
| 15:42.16 | brad_mssw | CosmicPenguin: I'm looking at an lspci -xxxvn of a ALIX board, which has basically the same chips as the soekris net5501, and 40 is 02 there |
| 15:42.30 | brad_mssw | i have no idea what that means though |
| 15:42.49 | brad_mssw | http://www.mail-archive.com/voyage-linux@list.voyage.hk/msg01404.html |
| 15:43.08 | brad_mssw | 40: 02 f0 00 00 00 00 00 00 a8 a8 a8 20 3f 00 ff 20 |
| 15:43.12 | brad_mssw | vs mine |
| 15:43.21 | brad_mssw | 40: 00 a2 00 00 00 00 00 00 a8 a8 a8 a8 ff 00 ff ff |
| 15:44.50 | brad_mssw | is that all that should be needed to enable dma properly? |
| 15:45.30 | brad_mssw | (I'm a software guy, so I'm over my head when it comes to the HW level ...) |
| 15:48.39 | brad_mssw | ... and just because I wanted to brick my board, I ran setpci (and set 40 to 01) to see what would happen on the running system ... it didn't crash, but hdparm -d 1 still failed, as expected since the driver was already loaded at that point |
| 15:49.14 | brad_mssw | and lspci does show the modification is there ... so i guess i did infact run setpci properly |
| 15:53.25 | *** join/#edev Crofton|work (n=balister@ossie.wireless.vt.edu) |
| 15:55.03 | *** join/#edev leoncamel (n=leoncame@p6024-ipbfp305fukuokachu.fukuoka.ocn.ne.jp) |
| 16:21.09 | *** join/#edev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 16:24.04 | *** join/#edev Off_Namuh (i=GPS@gateway/tor/x-020b52f51a45bbc1) |
| 16:24.45 | *** join/#edev CosmicPenguin (n=nobody@crank.laptop.org) |
| 17:13.13 | *** join/#edev FireEgl (i=FireEgl@Sebastian.Atlantica.DollarDNS.Net) |
| 17:17.13 | *** join/#edev leoncamel (n=leoncame@p6024-ipbfp305fukuokachu.fukuoka.ocn.ne.jp) |
| 17:19.17 | ds2 | morning |
| 17:36.19 | *** join/#edev redblue (i=star@ppp237.108-253-207.mtl.mt.videotron.ca) |
| 17:52.58 | *** join/#edev MothaWork_ (n=monmotha@149.164.193.61) |
| 18:02.41 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 18:21.19 | brad_mssw | CosmicPenguin: I haven't tried that setpci command yet ... can you tell me what 40 does? is setting it to 01 what you'd recommend? the ALIX board apparently has it set to 02 (and uses the same chip) |
| 18:32.20 | *** join/#edev ico2 (n=ico2@modem-2178.cougar.dialup.pol.co.uk) |
| 18:36.20 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 18:40.49 | *** join/#edev brad_mssw|work (n=brad@shop.monetra.com) |
| 18:41.01 | *** join/#edev brad_mssw (n=brad@shop.monetra.com) |
| 18:41.36 | brad_mssw | CosmicPenguin: umm, X locked up on me, if you replied, sorry, I didn't get any of it |
| 18:45.32 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 18:49.30 | prpplague | http://www.toysrus.com/product/index.jsp?productId=2740761&cp=2255956.2273442.2255964.2256696&view=all&parentPage=family |
| 18:49.32 | prpplague | interesting |
| 18:49.38 | prpplague | looks like a zipit knock off |
| 19:00.02 | ds2 | more like an IMfree |
| 19:01.50 | *** join/#edev redblue (i=star@ppp124.108-253-207.mtl.mt.videotron.ca) |
| 19:09.48 | *** join/#edev chouimat|ibook (n=dieu@kde/developer/chouinard) |
| 19:11.32 | jbevren | 900mhz |
| 19:11.37 | jbevren | yeah more like motorola's dev I think |
| 19:12.32 | jbevren | looks like a char LCD too |
| 19:12.33 | CosmicPenguin | brad_mssw: no, I didn't reply |
| 19:13.13 | CosmicPenguin | brad_mssw: bits 0 and 1 in 0x40 signal that there is something there, and that it is dma capable |
| 19:13.16 | CosmicPenguin | IIRC |
| 19:15.41 | brad_mssw | ok, so 02 would say that the slave is present and dma capable |
| 19:15.53 | brad_mssw | which is probably why that ALIX board is set to 02 |
| 19:16.03 | brad_mssw | 01 would say master is present, and 03 would be both |
| 19:16.39 | CosmicPenguin | I think so |
| 19:16.40 | brad_mssw | since the CF slot is slave |
| 19:17.01 | brad_mssw | k, great ... do I need to worry about any other bytes? |
| 19:17.11 | brad_mssw | I'll probably try to give that a shot this weekend |
| 19:17.19 | CosmicPenguin | For this problem, probably not |
| 19:17.31 | CosmicPenguin | they might also be missing the bit for the 80wire cable detection |
| 19:17.43 | CosmicPenguin | but thats a GPIO thing and probably not something you would want to mess with |
| 19:17.49 | CosmicPenguin | even if the detection hardware is in place |
| 19:18.01 | brad_mssw | really though, that wouldn't limit my speed by much ... |
| 19:18.14 | brad_mssw | I'd be happy with anything > 10MBps |
| 19:18.23 | CosmicPenguin | that only is valid for udma4 |
| 19:18.26 | CosmicPenguin | which you don't have anyway |
| 19:18.31 | CosmicPenguin | so no nworries |
| 19:18.40 | brad_mssw | cool, thanks |
| 19:18.59 | CosmicPenguin | And bang on Sokris to fix that |
| 19:19.07 | CosmicPenguin | or move to LinuxBIOS, where is should already be fixed.. :) |
| 19:19.11 | brad_mssw | yeah, they're pretty non-responsive |
| 19:19.20 | CosmicPenguin | (rather, it was never broken) |
| 19:19.41 | ds2 | is LinuxBIOS still alive as a project? |
| 19:19.55 | CosmicPenguin | never aliver |
| 19:21.12 | ds2 | they seem to be behind on chipset support |
| 19:38.53 | *** join/#edev ico2 (n=ico2@modem-2075.cheetah.dialup.pol.co.uk) |
| 19:39.36 | *** join/#edev SpaceMonkey (n=andrew@client-81-105-56-241.popl.adsl.virgin.net) |
| 19:41.54 | *** join/#edev SpaceMonkey (n=andrew@client-81-105-56-241.popl.adsl.virgin.net) |
| 19:42.19 | *** join/#edev SpaceMonkey (n=andrew@client-81-105-56-241.popl.adsl.virgin.net) |
| 20:01.01 | *** join/#edev lyakh (n=lyakh@dynamic-unidsl-85-197-26-24.westend.de) |
| 20:11.29 | *** join/#edev GPSFan (n=kenm@12.10.255.248) |
| 20:11.29 | *** mode/#edev [+o GPSFan] by ChanServ |
| 20:15.39 | *** join/#edev chouimat (n=dieu@kde/developer/chouinard) |
| 20:36.17 | *** join/#edev LyosNorezel2 (n=lyosnore@unaffiliated/lyosnorezel) |
| 20:36.31 | *** part/#edev LyosNorezel2 (n=lyosnore@unaffiliated/lyosnorezel) |
| 20:47.55 | prpplague | man irc has been quiet today |
| 20:52.30 | chouimat|ibook | prpplague: :) |
| 20:58.42 | *** part/#edev samrobb (n=sam@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 21:26.54 | *** join/#edev ohsix_ (i=ohsix@bc111222.bendcable.com) |
| 21:28.20 | prpplague | ho ho hum |
| 22:24.39 | *** join/#edev MothaWork (n=monmotha@host-69-95-169-16.cwon.choiceone.net) |
| 22:24.53 | *** join/#edev MothaWork (n=monmotha@149.164.193.61) |
| 22:40.54 | *** join/#edev ohsix (i=ohsix@bc111222.bendcable.com) |
| 23:40.11 | *** join/#edev pierrelux (n=pierre-l@144-125.sh.cgocable.ca) |