Hi.
I'm new to this list and want to shout out a "thank you" for all the
work you've done.
I've yet to do a successful config update with linux though, since I hit
the very same problem as Kevin.
Phil Dibowitz ([EMAIL PROTECTED]) wrote:
> Kevin Gilbert wrote:
> > On Sat, 8 Mar 2008, Phil Dibowitz opined:
> >> Kevin Gilbert wrote:
> >>> Erasing flash: 75% Failed to read from device: -110
> >>> (could not detach kernel driver from interface 0: No data available)
> >>> Failed to erase flash! Bailing out!
> >>> Failed with error -110
> >>> ===
> >>>
> >>> It always craps out at 75% (but is that just because it's got to 100% and
> >>> just hasn't displayed the status line yet?). And the worst thig about
> >>> this is that I've got to boot into Windoze to reload the config! :(
> >> No, I don't think it's done 100%, or it would have returned success. This
> >> is some underlying libusb issue. I'm wondering if maybe the kernel HID
> >> driver is getting in the way. There's a patch I wrote in the kernel to
> >> prevent this, and it is in fact including the VID/PID that you showed. What
> >> kernel version are you using?
> >
> > 2.6.23.15-80.fc7 (Fedora)
> > 2.6.23 (mine)
> > 2.6.24.2 (mine)
>
> According to the kernel history, my patch was merged for .23. So...
> well, hmm.
>
> Are you using a USB keyboard? If not, could you temporary rmmod usbhid
> and hid (both are important), next time it's happening and see if that
> remedies the issue? That would narrow it down a lot.
I use 2.6.24.2 myself and I kind of doubt that this is a failure in
unbinding the driver, since all the other commands work pretty flawless
out of the box.
I just tried it again with the usbhid and hid modules removed. Same
problem. It seems that the 525 gets confused badly in that process,
because subsequent attempts to contact the remote fail:
[EMAIL PROTECTED]:/home/simon/pakete/harmony-0.13# ./harmony -C
~simon/Update.EZHex
Harmony Control 0.13
Copyright 2007 Kevin Timmerman and Phil Dibowitz
This software is distributed under the GPLv3.
Failed to set device configuration: -110 (could not set config 1: Connection
timed out)
Unable to find a HID remote (error -110)
Failed with error -110
[EMAIL PROTECTED]:/home/simon/pakete/harmony-0.13# ./harmony -i
Harmony Control 0.13
Copyright 2007 Kevin Timmerman and Phil Dibowitz
This software is distributed under the GPLv3.
Failed to set device configuration: -71 (could not set config 1: Protocol
error)
Unable to find a HID remote (error -71)
Failed with error -71
I had to power off the remote for a while to get it to respond to the
linux software again:
[EMAIL PROTECTED]:/home/simon/pakete/harmony-0.13# ./harmony -i
Harmony Control 0.13
Copyright 2007 Kevin Timmerman and Phil Dibowitz
This software is distributed under the GPLv3.
Requesting Identity: 100% done
Device Info:
Model: Logitech Harmony 525 (Mocha Decaf)
Firmware Version: 2.6
Hardware Version: 2.5
Invalid config!
Success!
Regarding the valgrind errors spotted by Kevin: In CRemote::GetIdentity
there is a HID_WriteReport() which does a usb_interrupt_write() with two
bytes of data, claiming to have 64 bytes of data (if I don't
misinterpret some of the output). So at best the usb request has 62
bytes of uninitialized data. Maybe this might confuse the device?
This however happens before the actual flash erase happens and the other
stuff seems to work, so this is a shot into the dark.
Shortly after the flash erase stalls valgrind spits out this:
==25421== Warning: noted but unhandled ioctl 0x550b with no size/direction hints
==25421== This could cause spurious value errors to appear.
==25421== See README_MISSING_SYSCALL_OR_IOCTL for guidance on writing a
proper wrapper.
Here is the sequence of ioctls as traced by strace (starting somewhere
in the middle).
ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = 0
ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
[...]
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = 0
62% ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = 0
ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
[...]
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = 0
75% ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = 0
ioctl(3, USBDEVFS_SUBMITURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
[...]
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfbe21a8) = -1 EAGAIN (Resource
temporarily unavailable)
ioctl(3, USBDEVFS_DISCARDURB, 0xbfbe2164) = 0
ioctl(3, USBDEVFS_REAPURB, 0xbfbe21a8) = 0
Failed to read from device: -110 (could not detach kernel driver from
interface 0: No data available)
What I find interesting: The ioctls don't really seem to indicate
something going wrong, except that before the 100% get reached a lot of
USBDEVFS_REAPURBNDELAYs happen, so that probably some timeout happens.
Libusb apparently spits out an obscure error which looks quite bogus to
me - I wonder if this is maybe an older error message sticking around.
What could I do to help in this issue?
Thanks,
Simon
--
[EMAIL PROTECTED] http://simon.budig.de/
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
concordance-devel mailing list
concordance-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/concordance-devel
