Re: [avrdude-dev] Intel Mac and AVR Dragon

[Joerg Wunsch]

As Evan Broder wrote:

> > Next step would be to ktrace everything, and see whether the
> > problem is in libusb, or within the kernel driver.

> Ok. I ran the same command and let it sit for a minute or two just
> to be sure. I posted the trace ...

I must admit I don't have the faintest idea how the USB communication
would work under MacOS X.  I can decode that trace on my FreeBSD
system, but I cannot see the slightest touch of USB communication in
there.  It somehow opens libusb, mmap()s a bit of stuff, and then all
of a sudden comes to the conclusion it found an AVRDRAGON with serial
number 00A200002CCB.  Either that, or my FreeBSD cannot decode parts
of the file (OS X additions to the ktrace format?).

 19136 avrdude  1182718229.573074 NAMI  
"/System/Library/Extensions/IOUSBFamily.kext/Contents/PlugIns/IOUSBLib.bundle/Contents/MacOS/IOUSBLib"
 19136 avrdude  1182718229.573084 RET   open 3
 19136 avrdude  1182718229.573090 CALL  fstat(0x3,0)
 19136 avrdude  1182718229.573098 RET   fstat 0
 19136 avrdude  1182718229.573104 CALL  #153(0x3,0,0xbfffc2e8,0)
 19136 avrdude  1182718229.573116 GIO   fd 3 read 4096 bytes
       0x0000 cafe babe ...
 19136 avrdude  1182718229.573161 RET   #153 4096/0x1000
 19136 avrdude  1182718229.573200 CALL  #153(0x3,0,0xbfffc2e8,0)
 19136 avrdude  1182718229.573214 GIO   fd 3 read 4096 bytes
       0x0000 cefa edfe ...
 19136 avrdude  1182718229.573254 RET   #153 4096/0x1000
 19136 avrdude  1182718229.573316 CALL  mmap(0xd9000,0,0x9000,0,0x5,0)
 19136 avrdude  1182718229.573364 RET   mmap 888832/0xd9000
 19136 avrdude  1182718229.573373 CALL  mmap(0xe2000,0,0x1000,0,0x3,0)
 19136 avrdude  1182718229.573415 RET   mmap 925696/0xe2000
 19136 avrdude  1182718229.573423 CALL  mmap(0xe3000,0,0x1000,0,0x7,0)
 19136 avrdude  1182718229.573468 RET   mmap 929792/0xe3000
 19136 avrdude  1182718229.573476 CALL  mmap(0xe4000,0,0x89fc,0,0x1,0)
 19136 avrdude  1182718229.573524 RET   mmap 933888/0xe4000
 19136 avrdude  1182718229.573545 CALL  close(0x3)
 19136 avrdude  1182718229.573554 RET   close 0
 19136 avrdude  1182718229.861340 CALL  write(0x2,0,0xbfffe260)
 19136 avrdude  1182718229.861378 GIO   fd 2 wrote 61 bytes
       "avrdude: usbdev_open(): Found AVRDRAGON, serno: 00A200002CCB
       "

For comparision, this is what I'm getting under FreeBSD for the last
part of the abote quote:

  5608 avrdude  NAMI  "/dev/ugen0"
  5608 avrdude  RET   open 3
  5608 avrdude  CALL  ioctl(0x3,USB_SET_TIMEOUT,0xbfbfda1c)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_DO_REQUEST,0xbfbfd9c0)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_SET_TIMEOUT,0xbfbfda1c)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_DO_REQUEST,0xbfbfd9c0)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_SET_TIMEOUT,0xbfbfda1c)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_DO_REQUEST,0xbfbfd9c0)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_SET_TIMEOUT,0xbfbfda1c)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  ioctl(0x3,USB_DO_REQUEST,0xbfbfd9c0)
  5608 avrdude  RET   ioctl 0
  5608 avrdude  CALL  write(0x2,0xbfbfd430,0x3d)
  5608 avrdude  GIO   fd 2 wrote 61 bytes
       "avrdude: usbdev_open(): Found AVRDRAGON, serno: 00A200000AF4
       "

> > Could that be a permission problem for the underlying device node?

> Um...maybe? How would I check that?

Run as root for a quick test?

-- 
cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/                        NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)