Back
[14:22:36] <Atlantic777> Hi there! I've got a homemade usb avr programming board but when I connect it to the computer the OS (linux) throws error messages like this one:
[14:22:40] <Atlantic777> [40557.520772] usb 2-1.1: device descriptor read/64, error -32
[14:22:49] <The_Coolest> hah Atlantic777
[14:22:54] <Atlantic777> I'm told that this worked on a windows machine.
[14:23:02] <The_Coolest> I just made a 328P v-USB board and it's not recognized
[14:23:27] <Atlantic777> Well, I'm now trying to identify which board this is I find out is it a firmware problem or some linux bug.
[14:23:41] <Atlantic777> It could be something with power consumption of usb or whatever.
[14:23:54] <Atlantic777> Can someone help me identify which board this could be?
[14:23:59] <The_Coolest> well I just can't figure this out.
[14:24:07] <The_Coolest> check what chip is on the board
[14:24:22] <Atlantic777> It has an atmega8, 12 MHz osc, and 6 or 7 resistors.
[14:24:35] <The_Coolest> sounds like a v-usb board
[14:24:37] <Atlantic777> 4 condensators, 2 LEDs
[14:24:39] <The_Coolest> same boat as me
[14:24:49] <Atlantic777> cool, let's solve this :D
[14:25:23] <Atlantic777> I'm trying to boot some live windows on another notebook to check if it's OS related and then the game continues. :)
[14:25:57] <Atlantic777> It could be something like this:
http://tuxgraphics.org/electronics/200510/article05101.shtml
[14:27:29] <The_Coolest> Atlantic777 I designed my own board
[14:27:57] <The_Coolest> put it together, flashed with the proper firmware and when i plug it in i get unrecognized device
[14:28:19] <The_Coolest> it reacts to the reset button
[14:30:26] * The_Coolest sobs
[14:32:38] <The_Coolest> is it possibly to do debugging on the atmega328 with icsp?
[14:32:45] <OndraSter> no
[14:32:50] <OndraSter> icsp = programming only
[14:33:01] <OndraSter> but it might have debugwire
[14:33:09] <OndraSter> which supports (surprisingly) debugging
[14:34:43] <The_Coolest> hmmm this requires some special programmer from atmel
[14:34:50] <The_Coolest> blah
[14:39:13] <twnqx> does it support JTAG?
[14:40:06] <The_Coolest> nope :(
[14:40:38] <The_Coolest> But now I'm 100% that I'm adding a jtag header to my 1284p basedproject
[14:40:59] <The_Coolest> 100% sure*
[14:41:58] * twnqx skips ISP and programs with JTAG if the chip has jtag :P
[14:43:01] <The_Coolest> twnqx i never used that, the only question is how to do make the pins "multipurpose"
[14:43:30] <twnqx> jtag is usually dedicated anyway
[14:43:50] <The_Coolest> twnqx does it need a specialized programmer?
[14:43:56] <The_Coolest> Oh
[14:43:58] <twnqx> sure, but they are cheap
[14:44:10] <twnqx> tons of jtag mkI or mkII clones
[14:44:12] <The_Coolest> well they aren't dedicated on the 1284p
[14:44:19] <The_Coolest> they're part of portc
[14:44:19] <twnqx> that run with avrdude
[14:44:29] <twnqx> and avarice/gdb, and of course AS
[14:47:22] <The_Coolest> twnqx i designed and built this
http://imgur.com/a/80Zy9
[14:47:32] <The_Coolest> v-usb isn't working
[14:48:00] <OndraSter> no need for clones
[14:48:05] <OndraSter> legit jtagice3 should be $99
[14:48:12] <OndraSter> and legit dragon should be $49
[14:48:32] <twnqx> jtagice is one comdemned closed product
[14:48:34] <twnqx> 3*
[14:48:43] <twnqx> and dragon is crap, hardwarewise
[14:49:03] <twnqx> dieing if you look at it the wrong way from what i've heard
[14:49:19] <The_Coolest> lol
[14:49:24] <OndraSter> interesting, I have had Dragon for some time now
[14:49:25] <OndraSter> still works
[14:49:57] <twnqx> i was very close to buy an jtagice mk3
[14:50:14] <twnqx> but as long as there's no avrdude support it would just be another pile of electronic junk in here
[14:50:32] <twnqx> (and avarice, of course)
[14:51:07] <twnqx> The_Coolest: you use a 12mhz crystal, and the clock fuses are set to external crystal?
[14:51:24] <twnqx> also, if i remember vusb correctly, one of the data pins had to be an interrupt pin for it to work
[14:51:28] <The_Coolest> yup FF DD
[14:51:57] <The_Coolest> hmmm.
[14:52:28] <The_Coolest> let me try it
[14:52:46] <twnqx> also
[14:52:59] <twnqx> R10 shouldn't be to GND iirc?
[14:53:16] <twnqx> http://www.obdev.at/Images/vusb/circuit-zoomed.gif has it to vcc
[14:53:24] <The_Coolest> D+ is 1M pull down
[14:53:46] <The_Coolest> I used it on two different setups and it worked
[14:53:53] <The_Coolest> that
[14:54:01] <The_Coolest> D- is pulled up
[14:54:13] <The_Coolest> D+ *may* be pulled down
[14:54:15] <twnqx> ah, the other way round, ok
[14:55:22] <twnqx> No UART, timer, input capture unit or other special hardware is required (except one edge triggered interrupt).
[14:55:41] <twnqx> yeah, that was a vusb requirement
[14:56:38] <The_Coolest> yup that worked
[14:56:41] <The_Coolest> fml.
[15:13:44] <The_Coolest> twnqx!!! thx :D
[15:13:58] <twnqx> you're welcome
[15:14:01] <The_Coolest> http://img.techpowerup.org/130925/Capture849.png
[15:14:46] <twnqx> :) are you working on gpu-z?
[15:18:09] <The_Coolest> It works :D
[15:18:15] <The_Coolest> twnqx>> On Core Temp
[15:18:22] <twnqx> ah
[15:19:53] <The_Coolest> twnqx>> Wizzard made this little app for screen capture, he gave it to me
[15:59:21] <The_Coolest> https://fbcdn-sphotos-g-a.akamaihd.net/hphotos-ak-prn2/1157516_10201465004378640_1796318316_n.jpg
[16:01:40] <vsync_> now you can be the coolest guy on facebook
[16:01:52] <vsync_> but facebook isn't as cool as it used to be, i heard
[16:03:31] <The_Coolest> lol vsync_, just posted this on a local arduino page
[16:04:41] <GargantuaSauce> avrdude is always so courteous :)
[16:05:43] <The_Coolest> hehe
[16:06:27] <vsync_> well, fbcdn is facebook's cdn
[16:07:51] <The_Coolest> a local arduino facebook page*
[16:12:07] <vsync_> figured as much
[18:24:18] <fisted> hi everyone. i'm writing C code for an ATtiny2313, toolchain is avr-gcc. i've got an array of pointers to certain i/o registers, like this: volatile uint8_t *const oc_regs[] = {&OCR0A, &OCR0B, &OCR1AL, &OCR1BL}; accessing the actual i/o registers by doing *oc_regs[i] = ...; works fine.
[18:24:33] <fisted> anyway, now i'd like to put that array into program memory, for the pointers won't change at runtime
[18:24:49] <fisted> i tried declaring it PROGMEM, but it DidntWork(TM)
[18:25:18] <fisted> now i'm wondering if i can have pointers to i/o registers in program memory at all
[18:25:30] <fisted> and if so, if i need to consider anything special when accessing them
[18:28:37] <fisted> (unfortunately, signal to noise is low when trying to search for resources on it, because it's all about storing arrays of strings in progmen, i.e. progmem pointers to progmem, not progmem pointers to i/o registers
[18:44:57] * fisted gives up and writes it in assembly -_-
