| 00:05.52 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 00:36.16 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 00:40.58 | SpeedEvil | wishes a high power version had been on the table and not axed. A switchable vbus of 30v - and high power hubs woul dhave killed 90% of wall-warts on desks. |
| 00:45.54 | Mirell | What other wall warts besides a switch/router? |
| 00:52.38 | SpeedEvil | printer, powered speakers, cellphone charger, radio, external hard disk, |
| 00:53.05 | SpeedEvil | how many devices around your computer have a power consumption of under 15W and a brick of their own. |
| 00:59.14 | *** join/#edev linac (n=lin@122.90.110.171) |
| 02:09.27 | *** join/#edev bsdfox__ (n=h36sa@c-98-238-213-113.hsd1.ca.comcast.net) |
| 02:20.47 | *** join/#edev bgamari (n=ben@umass-959-168.wireless.umass.edu) |
| 02:21.16 | bgamari | I've built a new set of images for my Tin Can Tools hammer board, but I'm unsure of how to proceed with installing them without running the risk of bricking the device |
| 02:21.30 | bgamari | The device is currently booting with APEX |
| 02:21.46 | bgamari | with which XModem doesn't seem to work |
| 02:21.57 | bgamari | Minicom just sits there after I initiate the transfer |
| 02:26.35 | *** join/#edev xdapII (n=ldtc@64-255-89-66.trango.d3c.infowest.net) |
| 02:28.31 | *** join/#edev Crofton (n=balister@pool-71-171-31-89.nwrknj.east.verizon.net) |
| 02:50.32 | prpplague | bgamari hey one sec |
| 02:52.38 | prpplague | bgamari there are a number of issues that could be the problem, your distro might not have the lrzsz package installed |
| 02:52.56 | prpplague | bgamari did you purchase the flyswatter to go with the hammer board? |
| 02:55.07 | *** join/#edev cbx (n=cbx__@unaffiliated/cbx) |
| 03:08.55 | bgamari | prpplague, sorry about that |
| 03:09.01 | bgamari | prpplague, you still there |
| 03:09.12 | bgamari | prpplague, I should have purchased the flyswatter but I didn't |
| 03:09.19 | bgamari | I was hoping I wouldn't need it |
| 03:09.35 | bgamari | but it seems that there is a good chance I'll brick the board somewhere along the line |
| 03:10.49 | bgamari | prpplague, you hit the nail on the head with lrzsz |
| 03:11.10 | bgamari | prpplague, now I just need to figure out how to upload this thing without bricking the board |
| 03:17.32 | bgamari | prpplague, If you come back, I would love to pick your brains about bringing this board up |
| 03:18.07 | bgamari | I tried booting to the beginning of the uImage but naturally it's not that easy |
| 03:29.02 | bgamari | Anyone here know how to determine the entrypoint of a uImage? |
| 04:20.14 | *** join/#edev purl (i=ibot@rikers.org) |
| 04:20.14 | *** topic/#edev is Discussion of embedded systems (Hardware/Software/Linux/WinCE/Commercial Device Repurposing) : Firmware Linux (FWL) : Busybox and Buildroot : Ponies : http://www.elinux.org |
| 04:22.53 | *** join/#edev file (n=file@asterisk/developer-and-muffin-lover/file) |
| 04:51.39 | *** join/#edev purl (i=ibot@rikers.org) |
| 04:51.39 | *** topic/#edev is Discussion of embedded systems (Hardware/Software/Linux/WinCE/Commercial Device Repurposing) : Firmware Linux (FWL) : Busybox and Buildroot : Ponies : http://www.elinux.org |
| 05:07.12 | *** join/#edev linac (n=lin@122.90.110.171) |
| 06:11.37 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 06:33.01 | *** join/#edev bgamari (n=ben@pool-74-106-64-24.spfdma.east.verizon.net) |
| 06:42.23 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 07:22.13 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 07:46.02 | *** join/#edev SpeedEvil (n=user@tor/regular/SpeedEvil) |
| 09:30.20 | *** join/#edev hw (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 09:51.54 | *** join/#edev florian_kc (n=fuchs@Maemo/community/contributor/florian) |
| 09:53.08 | *** join/#edev olivdt_ (n=x@modemcable206.214-57-74.mc.videotron.ca) |
| 09:59.25 | ohsix | http://www.willowgarage.com/pages/robots/technical-specs |
| 10:34.43 | florian | good morning |
| 10:35.03 | florian | ohsix: this does not look like somethign to buy at a supermarket :-) |
| 10:38.23 | *** join/#edev cbrake (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) |
| 11:32.54 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 12:20.13 | *** join/#edev gustavoz (n=gustavoz@host63.201-252-5.telecom.net.ar) |
| 14:35.30 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 14:35.30 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:39.17 | *** join/#edev kergoth (n=kergoth@nat/montavista/x-sgpjjmtuzsfnjhjy) |
| 14:43.11 | kergoth | morning |
| 14:52.11 | *** join/#edev xdapII (n=ldtc@64-255-89-66.trango.d3c.infowest.net) |
| 15:48.13 | *** join/#edev sjhill (n=sjhill@home.bethel-hill.org) |
| 15:48.13 | *** mode/#edev [+o sjhill] by ChanServ |
| 15:52.25 | *** join/#edev cbx (n=cbx__@unaffiliated/cbx) |
| 15:58.39 | *** join/#edev gustavoz (n=gustavoz@host5.201-253-139.telecom.net.ar) |
| 16:36.17 | *** join/#edev dfarnsworth (n=dfarnswo@nat/montavista/x-ihqgiwlokhqzhaec) |
| 16:41.47 | *** join/#edev FOM (n=jeffs@rrcs-74-219-98-111.central.biz.rr.com) [NETSPLIT VICTIM] |
| 16:41.47 | *** join/#edev koobe (n=matti@83.150.95.26) [NETSPLIT VICTIM] |
| 16:42.06 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) [NETSPLIT VICTIM] |
| 16:42.06 | *** join/#edev vapier (i=UserBah@wh0rd.org) [NETSPLIT VICTIM] |
| 16:42.06 | *** join/#edev ohsix (i=ohsix@66.220.111.99) [NETSPLIT VICTIM] |
| 16:42.29 | *** join/#edev armin76 (n=armin@gentoo/developer/armin76) [NETSPLIT VICTIM] |
| 16:47.26 | *** join/#edev bgamari (n=ben@pool-74-106-64-24.spfdma.east.verizon.net) |
| 16:48.08 | bgamari | prpplague, ping |
| 16:48.25 | prpplague | bgamari: pong |
| 16:48.52 | bgamari | prpplague, So you were absolutely right about lrzsz |
| 16:49.04 | bgamari | In fact I think I had encountered that issue before |
| 16:49.10 | bgamari | but regardless |
| 16:49.34 | bgamari | I was able to upload my uImage at the proper offset |
| 16:49.37 | bgamari | and booted |
| 16:49.49 | bgamari | and sure enough, the board triple faulted and rebooted |
| 16:49.54 | prpplague | bgamari: well first of all, apex doesn't use a uImage. uImage's are for u-boot |
| 16:50.04 | prpplague | bgamari: you need to use the zImage instead |
| 16:50.16 | bgamari | prpplague, It doesn't support uImages at all? |
| 16:50.40 | prpplague | bgamari: a uImage is a zImage with a specific uboot header |
| 16:50.55 | prpplague | bgamari: when you are building your kernel, just do: make zImage |
| 16:50.55 | *** join/#edev dieter|work (n=dieter@dslb-088-066-205-221.pools.arcor-ip.net) |
| 16:50.59 | bgamari | alright |
| 16:51.03 | prpplague | bgamari: instead of: make uImage |
| 16:51.04 | bgamari | fair enough |
| 16:51.22 | sjhill | hey prpplague |
| 16:51.22 | bgamari | I'll give that a try |
| 16:51.22 | prpplague | sjhill: hey bud |
| 16:51.22 | bgamari | although I've seen references to APEX booting from uImages (http://www.mail-archive.com/debian-arm@lists.debian.org/msg09141.html) |
| 16:52.53 | bgamari | Moreover, the image load offset and entrypoint are both 0x3008000 |
| 16:53.38 | bgamari | So you would think as long as you enter at the right place, it wouldn't care too much |
| 16:55.06 | bgamari | I never even see the kernel decompressor work do it's thing |
| 16:57.13 | dfarnsworth | bgamari: if U-Boot is loading, you need a uImage. If apex is loading the kernel, you need a zImage. the decompression mechanism is different for each format |
| 16:57.33 | bgamari | alright, well then that very well could explain it |
| 16:57.37 | bgamari | I'll give that a try |
| 16:58.00 | dfarnsworth | zImages are self decompressing. U-Boot decompresses a uImage |
| 16:58.17 | bgamari | ahh, yeah, that could be a problem |
| 16:59.37 | bgamari | dfarnsworth, Also, where does one load the rootfs image? |
| 16:59.52 | bgamari | should this be obvious in the bootloader config? |
| 17:00.11 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) |
| 17:00.29 | dfarnsworth | bgamari: I can't help there. I almost always use nfs root. |
| 17:01.02 | bgamari | alright |
| 17:01.19 | bgamari | I wish that were an option here. Should have bought a beagle instead of a hammer |
| 17:04.44 | *** join/#edev T0mW (n=Tom@24.238.37.43.res-cmts.tv13.ptd.net) |
| 17:06.12 | prpplague | bgamari: marc was adding uImage support but i don't recall if he merged it |
| 17:07.01 | prpplague | bgamari: have you not checked the wiki pages? |
| 17:07.11 | prpplague | bgamari: there is a lot of documentation |
| 17:08.05 | prpplague | dfarnsworth: techincally a uImage is a zImage, but just with a header prepended to it, so techincally you could boot a uImage on apex, you just have to give the correct offset to bypass the header |
| 17:15.22 | bgamari | prpplague, If support was merged, what would that leave in terms of debugging options? |
| 17:17.10 | prpplague | bgamari: i dont understand your question |
| 17:17.20 | bgamari | prpplague, It seems that uImages are supported |
| 17:17.23 | bgamari | at least in 1.6.9 |
| 17:17.34 | bgamari | prpplague, well, the machine triple-faults |
| 17:17.52 | bgamari | I don't even see the usually "Decompressing linux...." |
| 17:18.05 | bgamari | It just goes right back to the boot monitor |
| 17:18.15 | bgamari | as far as I can tell from the serial output at least |
| 17:18.41 | bgamari | prpplague, So I'm not really sure what to do if the problem isn't the image format or offset |
| 17:19.11 | prpplague | bgamari: ok, you need to back up |
| 17:19.23 | prpplague | bgamari: you are trying to deal with way too many items at one time |
| 17:19.45 | prpplague | bgamari: zImage is the prefered image format for apex |
| 17:20.05 | bgamari | prpplague, Alright, I'll revert to zImage in that case |
| 17:20.05 | prpplague | bgamari: get your image working with zImage first then you can tinker around with uImage |
| 17:21.33 | prpplague | bgamari: what bootloader version are you using? |
| 17:21.59 | bgamari | That's a good question |
| 17:22.04 | bgamari | I'll be in front of the hardware in 15 minutes |
| 17:23.12 | *** part/#edev bgamari (n=ben@pool-74-106-64-24.spfdma.east.verizon.net) |
| 17:23.19 | *** join/#edev bgamari (n=ben@pool-74-106-64-24.spfdma.east.verizon.net) |
| 17:23.31 | bgamari | prpplague, I'll talk to you then |
| 17:24.48 | *** join/#edev bgamari_ (n=ben@pool-74-106-64-24.spfdma.east.verizon.net) |
| 17:45.48 | *** join/#edev PuffTheMagic (n=quassel@unaffiliated/puffthemagic) |
| 17:47.29 | dfarnsworth | prpplague: for some archs, a uImage may be a zImage with a prepended header. That is not true for powerpc32. |
| 17:47.37 | dfarnsworth | and therefore not true generally. :) |
| 17:47.46 | prpplague | dfarnsworth: ahh |
| 17:48.22 | prpplague | dfarnsworth: don't do much with PPC , so my bad |
| 17:48.41 | dfarnsworth | for ppc32, it's a uImage header, followed by a (typically compressed) vmlinux. |
| 17:48.53 | dfarnsworth | I assumed it was similar for other archs. |
| 17:49.47 | dfarnsworth | whereas, a zImage is a compressed vmlinux preceded by code to decompress the vmlinux |
| 17:51.45 | prpplague | dfarnsworth: ahh right |
| 17:52.50 | prpplague | dfarnsworth: i had forgotten that you could do the vmlinux |
| 17:53.32 | prpplague | dfarnsworth: i generally use much smaller bootloaders than u-boot which don't included the decompression, hence the need for the zImage |
| 17:54.19 | dfarnsworth | prpplague: right. I often do. That's the main difference I see between uImage and zImage. zImage is self decompressing, but uImage is not. |
| 18:39.00 | *** join/#edev flo_lap (n=fuchs@Maemo/community/contributor/florian) |
| 18:49.04 | Mirell | Man. |
| 18:49.09 | Mirell | Ruby's method_missing is magic |
| 18:50.59 | rektide | every time someone says javascript doesnt need a method_missing |
| 18:51.01 | rektide | i want to choke a bitch |
| 18:53.02 | Mirell | Uh. |
| 18:56.43 | rektide | sorry thought you were talking about ruby |
| 18:57.03 | *** join/#edev roxfan2 (i=dunno@197.155-246-81.adsl-dyn.isp.belgacom.be) |
| 18:58.15 | prpplague | does anyone here do alot of java programming? |
| 18:58.34 | rektide | raises hand |
| 18:58.46 | rektide | the devil made me do it |
| 19:03.31 | calculus | it pays the bills |
| 19:12.50 | Redhatter | C/C++ mostly these days... closest I've come to Java lately is the Eclipse IDE which CodeSourcery used as their base. |
| 19:13.33 | wmat | Redhatter: did you ever figure out that compile problem with FreeRTOS? |
| 19:14.09 | Redhatter | wmat: I'm closer... gcc has a little bug where it emits .eh_frame when it shouldn't. |
| 19:14.40 | Redhatter | wmat: I got it to build last night... won't know if it works until I get to work and try out the binary on the microcontroller. |
| 19:14.51 | wmat | ah, good luck |
| 19:15.20 | wmat | was that bug in all the versions; GNU and CodeSourcery? |
| 19:15.34 | Redhatter | I've nailed the .eh_frame bug... but I suspect the code may not work for other reasons... the code structure looks very different to the CodeSourcery toolchain. |
| 19:15.46 | Redhatter | The .eh_frame bug was seen in both GNU and CodeSourcery sources. |
| 19:18.56 | Redhatter | What I observe... the .elf image I produce using CodeSourcery on win32, starts with the interrupt vectors (g_pfnVectors) followed by main(). |
| 19:19.21 | Redhatter | I don't see the same pattern in my own .elf generated using the self-compiled CodeSourcery toolchain. |
| 19:23.46 | *** join/#edev JoeLlama (n=snork@71.110.196.168) |
| 19:26.13 | Mirell | rektide: I was |
| 19:34.25 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 19:36.24 | prpplague | rektide: hey, question, is there some example code out there for accessing a ttyp ? |
| 19:37.06 | prpplague | rektide: i've got a customer with a legacy java app that was connecting to a uart, he needs to connect it now to a pseudo tty |
| 19:38.28 | *** join/#edev gandhijee (i=akp@host-66-202-34-165.spr.choiceone.net) |
| 19:38.47 | gandhijee | hey guys, where can i find information about the GCI bus and its theory of operation? |
| 19:39.40 | rektide | prpplague: dont know, sorry. |
| 19:40.06 | prpplague | gandhijee: i'd guess a good place to start might be wikipedia |
| 19:40.20 | roxfan | and google |
| 19:40.35 | gandhijee | wiki is no dice |
| 19:41.01 | gandhijee | same w/ google, then again i could be searching wrong =/ |
| 19:44.42 | roxfan | General Circuit Interface ISDN PCM control protocol, equivalent to IOM-2TM. |
| 19:44.43 | roxfan | hmm |
| 19:44.58 | prpplague | gandhijee: well lets start by asking "what do you think GCI is?" |
| 19:46.00 | gandhijee | a mechanism used to transport Voice in a digital fashion in a PC |
| 19:46.05 | gandhijee | where did you find that roxfan? |
| 19:46.16 | roxfan | you're not gonna beleive it |
| 19:46.26 | roxfan | in wikipedia |
| 19:48.02 | roxfan | http://www.colognechip.com/hfc-s-mini.pdf check page 150 |
| 19:48.11 | gandhijee | yeah i found it now. |
| 19:48.46 | gandhijee | cool, thanks man |
| 19:49.20 | roxfan | http://en.wikipedia.org/wiki/ISDN-oriented_Modular_Interface |
| 19:49.28 | *** join/#edev keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) |
| 19:49.50 | roxfan | http://focus.ti.com.cn/cn/lit/an/bpra074/bpra074.pdf some more pictures |
| 19:50.25 | gandhijee | nope. just needed to know what is was, i see it in documents for all these SLiC parts |
| 19:50.49 | gandhijee | and couldn't find any real info on it, but turns out i'm just a moron and didn't search right |
| 19:50.49 | roxfan | you asked for "theory of operation" |
| 19:51.32 | roxfan | ohh i found the ref |
| 19:51.36 | roxfan | 1989 though |
| 19:51.46 | roxfan | http://web.inter.nl.net/users/schnecke/mips/7930docs.html |
| 19:52.14 | gandhijee | awesome |
| 19:52.25 | gandhijee | well i guess that is getting thrown out of the window then |
| 20:02.17 | *** join/#edev _new_Dev_ (n=blabla@41.226.134.123) |
| 20:02.23 | _new_Dev_ | Hi all! |
| 20:05.54 | prpplague | greetings |
| 20:06.06 | rektide | Mirell: i stand by my original premise |
| 20:06.55 | _new_Dev_ | always so quite here :D |
| 20:07.32 | prpplague | _new_Dev_: just when you are in the channel |
| 20:08.21 | _new_Dev_ | hihi! is this an invitation to stay ;) |
| 20:09.02 | _new_Dev_ | oups quiet* |
| 20:27.26 | *** join/#edev bgamari (n=ben@umass-959-168.wireless.umass.edu) |
| 20:29.15 | bgamari | prpplague: Well, you were right |
| 20:29.31 | bgamari | prpplague: the zImage works |
| 20:30.42 | prpplague | bgamari: yea well seeing how i wrote a large portion of apex............ |
| 20:30.55 | bgamari | prpplague: heh, I suppose that would do it. Thanks a ton for your work |
| 20:31.02 | prpplague | np |
| 20:31.13 | prpplague | bgamari: be sure and order a flyswatter |
| 20:31.41 | bgamari | prpplague: Alright, that's a good call |
| 20:32.24 | prpplague | ho ho hum |
| 20:32.29 | bgamari | prpplague: So now I should be able to just copy the ramdisk to $ramdiskaddr and theoretically it should work? |
| 20:32.40 | prpplague | bgamari: yea |
| 20:33.14 | bgamari | prpplague: Can I copy the gzipped image or must I copy the uncompressed copy? |
| 20:33.27 | keesj | it's funny wikipedia call the flyswatter a "hand-held device" sounds geeky to me |
| 20:33.34 | keesj | http://en.wikipedia.org/wiki/Flyswatter |
| 20:34.46 | prpplague | keesj: hehe |
| 20:35.15 | bgamari | prpplague: Does the kernel expect the ramdisk to be uncompressed? |
| 20:35.21 | prpplague | keesj: of course we intended for the name to be a joke on the fact of "killing bugs" |
| 20:35.38 | prpplague | bgamari: default configuration is for it to be compressed |
| 20:35.43 | bgamari | awesome |
| 20:35.45 | _new_Dev_ | bgamari CRAMFS? |
| 20:36.12 | bgamari | prpplague: Now if the uncompressed image is too large for my device's memory, I have a problem, correct? |
| 20:36.21 | prpplague | correct |
| 20:36.46 | bgamari | alright, looks like I'll need to pare down my image |
| 20:37.16 | _new_Dev_ | bgamari why not to keep it in the flash? |
| 20:37.20 | calculus | prpplague: speaking of bootloaders, have you seen this contest, http://forum.osdev.org/viewtopic.php?f=2&p=170511#p170511 |
| 20:38.00 | bgamari | _new_Dev_: I'm trying not to touch the nonvolatile memory on the device |
| 20:38.10 | _new_Dev_ | ah ok! |
| 20:38.10 | bgamari | _new_Dev_: to avoid bricking it |
| 20:38.18 | bgamari | that might be what I do though |
| 20:38.29 | prpplague | calculus: interesting |
| 20:38.36 | prpplague | calculus: wish i had time to hack on that |
| 20:39.29 | bgamari | Can the device run the ramdisk from Flash if I place it there? |
| 20:42.03 | bgamari | _new_Dev_? |
| 20:42.30 | _new_Dev_ | bgamari you can use romfs file system |
| 20:43.13 | _new_Dev_ | m'i right prpplague ;) |
| 20:44.22 | bgamari | _new_Dev_: This is not true with Cranfs? |
| 20:44.31 | prpplague | _new_Dev_: right about what? |
| 20:44.51 | prpplague | bgamari: no you'd need to format the image differently |
| 20:44.56 | gandhijee | bgamari: did you8 make sure you have compression support in the kernel you built? |
| 20:45.04 | prpplague | bgamari: best to use jffs2 if you plan on using it from flash |
| 20:45.46 | _new_Dev_ | prpplague yeah jffs2 is a good choice but it take lot of ram! |
| 20:45.53 | _new_Dev_ | romfs is lighter |
| 20:46.27 | prpplague | _new_Dev_: jffs2 doesn't take that much more ram than other filesystems |
| 20:46.36 | prpplague | _new_Dev_: specially since it was designed for embedded systems |
| 20:48.00 | _new_Dev_ | prpplague all depends on how much ram you have :) |
| 20:48.51 | prpplague | _new_Dev_: the only time youd have a problem is if you have less than 8mb of ram |
| 20:49.41 | _new_Dev_ | maybe the much ram ihad was 1M |
| 20:49.42 | _new_Dev_ | :D |
| 20:50.18 | bgamari | _new_Dev_, prpplague: now the next question is how one configures the ramdisk address |
| 20:50.27 | bgamari | the kernel is passed root=/dev/ram0 |
| 20:50.47 | bgamari | however I'm unsure of how the kernel knows where /dev/ram0 is |
| 20:50.54 | prpplague | _new_Dev_: how were you able to run a linux kernel with only 1M? |
| 20:50.57 | _new_Dev_ | are you using linux initramfs? |
| 20:51.20 | bgamari | _new_Dev_: I just configured buildroot and the kernel to use cramfs |
| 20:51.43 | _new_Dev_ | hihi i make it so slim :D |
| 20:51.43 | prpplague | bgamari: ram0 is generated, the bootloader passes the address as part of the ATAG's |
| 20:51.43 | _new_Dev_ | sorry for my english :$ |
| 20:52.04 | bgamari | prpplague: ahh, so it's located at $ramdiskaddr ? |
| 20:52.05 | prpplague | bgamari: you can either compile the value in, or set it as part of the environment variables |
| 20:52.11 | prpplague | bgamari: correct |
| 20:52.22 | bgamari | prpplague: and it will be able to figure out what filesystem the image uses |
| 20:52.23 | bgamari | ? |
| 20:52.49 | prpplague | bgamari: it will check the memory location and determine what kind of header is there |
| 20:53.04 | prpplague | bgamari: if it finds a compressed ramdisk, it will uncompress it, and place it ram0 |
| 20:53.19 | _new_Dev_ | prpplague i was running the kernel on XIP mode :D application too was in XIP mode |
| 20:53.20 | bgamari | alright, awesome |
| 20:53.26 | prpplague | _new_Dev_: ahh |
| 20:53.28 | bgamari | I guess I'll just try it |
| 20:54.59 | _new_Dev_ | what is your hardware bgamari? |
| 20:55.36 | prpplague | _new_Dev_: he's using the hammer board |
| 20:55.49 | bgamari | _new_Dev_: TCT Hammer |
| 20:56.12 | _new_Dev_ | arm920? |
| 20:56.22 | bgamari | _new_Dev_: yep |
| 20:56.37 | _new_Dev_ | ok! |
| 20:59.24 | prpplague | ho ho hum |
| 20:59.37 | _new_Dev_ | what's up? |
| 20:59.40 | _new_Dev_ | ;) |
| 21:00.26 | prpplague | bootloader fun for a pxa310 based cell phone |
| 21:01.32 | _new_Dev_ | loll |
| 21:01.36 | _new_Dev_ | lucky man |
| 21:01.50 | _new_Dev_ | i miss linux sniffff |
| 21:18.07 | prpplague | keesj: ha! you just came up on my linkedin page for "people you might know" |
| 21:20.40 | keesj | you sound like a good connection! |
| 21:21.07 | prpplague | hehe |
| 21:26.41 | prpplague | keesj: oh! you are a java guru |
| 21:27.48 | prpplague | keesj: don't suppose you have a good source example of using a ttyp in java? |
| 21:30.16 | keesj | I have 1337 Java skilzzz but don't know about ttyp :( |
| 21:30.41 | _new_Dev_ | lol |
| 21:31.45 | prpplague | keesj: well what about if you wanted to read/write from a character device node(other than a serial port) ? |
| 21:33.39 | keesj | I used RxTx (the open JavaComm implementation) |
| 21:33.53 | prpplague | hmm |
| 21:34.12 | prpplague | thats what they are using, but it doen't seem to play nice with pseudo ttys |
| 21:34.20 | keesj | for my casio wristcam http://wqv-wristcam.sourceforge.net/ |
| 21:34.24 | prpplague | guess i'll need to do some wedge device |
| 21:34.50 | keesj | I know I nice java terminal, it uses java native interface but it's really do |
| 21:35.18 | keesj | this one might be what you are looking for http://software.jessies.org/terminator/ |
| 21:35.59 | keesj | (and it's fast and usefull like most of jessies's software) |
| 21:36.01 | prpplague | keesj: hmm, that might have what i am looking for |
| 21:36.23 | prpplague | keesj: from what i can tell, the rxtx lib seems to open and close the device node a bunch on startup |
| 21:36.33 | prpplague | keesj: and that doesn't play well with ttyp |
| 21:37.35 | keesj | and if you want a quick hack I guess ser2net like code will keep your java code clean of any low level stuff :p |
| 21:38.07 | prpplague | googles ser2net |
| 21:38.27 | prpplague | keesj: i'm not a java person, so i'm kind of flying blind on this one |
| 21:38.59 | keesj | It's not like a am really proud of it .... |
| 21:39.29 | prpplague | hehe |
| 21:39.50 | *** join/#edev new_dev_ (n=blabla@41.226.134.123) |
| 21:39.50 | prpplague | keesj: really what i need is support a bsd like socket |
| 21:40.56 | keesj | that wristcam was really cool at the time |
| 21:44.37 | prpplague | keesj: looks dandy |
| 21:52.16 | keesj | going off |
| 21:54.21 | new_dev_ | good bye all! |
| 22:05.35 | bgamari | prpplague: hmm |
| 22:05.39 | bgamari | prpplague: well, this is strange |
| 22:05.47 | bgamari | oh, never mind, missed a zero |
| 22:05.49 | bgamari | my bad |
| 22:05.59 | bgamari | it appeared that $bootaddr == $ramdiskaddr |
| 22:08.52 | prpplague | ho ho hum |
| 22:08.57 | prpplague | is worn out today |
| 22:20.57 | *** join/#edev Christos_N (n=Christos@ppp-94-66-27-113.home.otenet.gr) |
| 22:24.56 | bgamari | Well, it looks like all of that we're switching hardware anyways |
| 22:25.15 | bgamari | prpplague: Do you know how painful floating point emulation is performance-wise? |
| 22:26.05 | roxfan | emulation, as in intercepting illegal instruction fault? |
| 22:26.06 | prpplague | bgamari: i don't understand your question |
| 22:26.13 | prpplague | bgamari: doh, not going to use the Hammer? |
| 22:26.25 | prpplague | is disappointed |
| 22:27.14 | bgamari | prpplague: Yeah, we need something with and FPU |
| 22:27.22 | bgamari | an* |
| 22:27.31 | bgamari | We'll probably go with a Beagle |
| 22:27.37 | roxfan | what are you doing with fp? |
| 22:27.47 | bgamari | roxfan: Matrix inversion |
| 22:27.56 | bgamari | which evidently suffers on fixed-point |
| 22:28.27 | prpplague | bgamari: techinically the omap3 doesn't have fpu either |
| 22:28.44 | roxfan | eh? it has vfp, no? |
| 22:28.48 | bgamari | It has VFP, right? |
| 22:28.51 | roxfan | and neon too |
| 22:29.17 | roxfan | and then there is the dsp... not sure if that one supports fp |
| 22:29.18 | prpplague | yea |
| 22:29.33 | bgamari | VFP, NEON != FPU ? |
| 22:32.03 | bgamari | prpplague: Do you have a stake in the Hammer? |
| 22:32.29 | prpplague | laughs |
| 22:32.32 | prpplague | yea you could say that |
| 22:32.41 | *** join/#edev _new_Dev_ (n=blabla@41.226.134.123) |
| 22:32.58 | bgamari | prpplague: Do you work for TCT? |
| 22:33.03 | prpplague | bgamari: there are some slight differences between the terms VFP and FPU, but it really is symantics |
| 22:33.14 | prpplague | designed the hammer board for TCT |
| 22:33.37 | _new_Dev_ | my respect prpplague |
| 22:33.50 | _new_Dev_ | (°c°,) |
| 22:33.55 | bgamari | ahhh |
| 22:34.12 | bgamari | It'll no doubt get tasked with something |
| 22:34.27 | bgamari | there is no shortage of things around the lab that need a microcontroller |
| 22:35.19 | _new_Dev_ | i have a question have any one tried to convert PCM 8khz 16-bit to PCM 44100hz 16-bit? |
| 22:35.32 | bgamari | ouch |
| 22:35.40 | bgamari | I can't imagine what that might sound like |
| 22:35.54 | prpplague | bgamari: are you trying to do a commercial product or just a one-off device? |
| 22:36.03 | bgamari | isn't that less than POTs quality? |
| 22:36.07 | bgamari | prpplague: One-off |
| 22:36.17 | prpplague | bgamari: ahh |
| 22:36.30 | bgamari | prpplague: Although there could be demand for the design |
| 22:36.50 | bgamari | We're doing particle tracking with back-focal-plane imaging |
| 22:37.16 | bgamari | So I wouldn't be surprised if other groups express interest |
| 22:37.44 | prpplague | ahh |
| 22:37.49 | prpplague | goes home for the day |
| 22:38.04 | bgamari | The software will be released under an open source license when its more than a pile of stubs |
| 22:38.45 | bgamari | _new_Dev_: You familiar with CRAMFS? |
| 22:39.01 | bgamari | The kernel found the filesystem |
| 22:39.32 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 22:39.33 | bgamari | but then complained that it couldn't mount root |
| 22:39.35 | bgamari | RAMDISK: cramfs filesystem found at block 0 |
| 22:39.43 | bgamari | RAMDISK: Loading 7112KiB [1 disk] into ram disk... |
| 22:39.53 | bgamari | and then it just printed a list of partitions |
| 22:40.04 | bgamari | of which there are two |
| 22:40.11 | bgamari | perhaps it didn't know which partition to mount? |
| 22:41.42 | _new_Dev_ | what is your kernel command line? |
| 22:42.02 | *** join/#edev _new_Dev_ (n=blabla@41.226.134.123) |
| 22:42.26 | bgamari | root=/dev/ram0 |
| 22:44.12 | _new_Dev_ | maybe |
| 22:45.28 | bgamari | it gave me, |
| 22:45.29 | bgamari | List of all partitions: |
| 22:45.29 | bgamari | 1f00 2304 mtdblock0 (driver?) |
| 22:45.29 | bgamari | 1f01 14080 mtdblock1 (driver?) |
| 22:45.38 | bgamari | sorry for the mini-flood |
| 22:46.17 | _new_Dev_ | i used linear CRAMFS |
| 22:48.31 | _new_Dev_ | bgamari have applied any patch to the kernel? |
| 22:50.10 | _new_Dev_ | i remember that i must pass the physical address of the cramfs in the kernel command line to get it working, |
| 22:52.44 | bgamari | _new_Dev_: nope |
| 22:52.53 | bgamari | I might just switch to romfs if that would be easier |
| 22:54.57 | _new_Dev_ | there is a patch for an MTD based romfs support |
| 22:56.51 | _new_Dev_ | bgamari have you tried with root=/dev/ram1 |
| 22:57.02 | *** join/#edev cbrake_away (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) |
| 22:57.03 | keesj | go-make dinner |
| 23:10.29 | *** join/#edev Crofton_|work (n=balister@pool-71-171-31-89.nwrknj.east.verizon.net) [NETSPLIT VICTIM] |
| 23:11.49 | *** join/#edev dfarnsworth (n=dfarnswo@nat/montavista/x-tzlybdsgnojlwbqo) [NETSPLIT VICTIM] |
| 23:11.53 | *** join/#edev Mirell (n=mirell@motoko.mirell.org) |
| 23:12.01 | *** join/#edev Redhatter (n=vk4msl@203.94.56.252) |
| 23:12.57 | *** join/#edev m4t (i=matt@theorize.org) |
| 23:12.57 | *** join/#edev ohsix (i=ohsix@66.220.111.99) |
| 23:13.09 | *** join/#edev cjameshuff (n=cjameshu@pool-71-97-177-195.aubnin.dsl-w.verizon.net) |
| 23:13.09 | *** join/#edev linac (n=lin@122.90.110.171) |
| 23:14.22 | roxfan | http://anatolyrr.ru/habr/2010/2010-pocket.png |
| 23:17.44 | *** join/#edev freddy (n=freddy@kenny.kriewitz.eu) [NETSPLIT VICTIM] |
| 23:38.03 | *** join/#edev prpplague (n=dave123_@ppp-70-249-159-70.dsl.rcsntx.swbell.net) |
| 23:38.11 | *** join/#edev SpeedEvil (n=user@tor/regular/SpeedEvil) |
