On 08/08/13 00:52, Pete Batard wrote:
On 2013.07.23 11:33, Toby Gray wrote:
Do you have plans to look into adding a new config option for sending
the debug output to the debugger, or do you want me to look into it?
Yes I have. It was one of those bits of work that when I started looking
at it I discovered several other things which were wrong. I send a
series of patches, including the new config option, syslog support and
Android support in the mail titled "Improved log in non-console
applications and Android" on 9th July.
Mailing list archive link:
https://sourceforge.net/mailarchive/forum.php?thread_name=51DC431A.1050900%40realvnc.com&forum_name=libusbx-devel
I'll see if I can take a closer look at these tomorrow, for inclusion in
the upcoming release.
Thank you. I think you'll find that
0005-hotplug-Remove-use-of-pthread_cancel-from-netlink-ho.patch is
unnecessary now as Chris Dickens copied it for udev [1] and then Hans de
Goede copied it back to netlink [2].
It's probably best if I rebase the first 4 patches off the latest
libusbx master, drop the 5th patch and add some more patches for Android
build documentation and support for more of the samples. I'll send a
reply to the "Improved log in non-console
applications and Android" message when I've done this so you'd probably
be best off waiting for that so you don't have to deal with any of the
rebasing.
the event handling change is a big thing
which might not make it in for a few release (if at all).
Yeah. I think the additional tests should be fine (the more tests, the
better!), and hopefully a no brainer to push, but since the common event
handling is something we've been dragging for some time, it will require
some scrutiny.
I've been meaning to post a message to the mailing list to ask you about
hotplug support. For Windows CE the device attach and removal comes via
a MsgQueue[3] which I believe is nothing like how this notification
occurs on desktop Windows.
This means the Windows CE hotplug implementation will be distinct from
the desktop Windows implementation, however I believe that both will
need usbi_read() and usbi_write() to send and receive actual data.
I've been holding off starting hotplug support for Windows CE as
modifying usbi_read() and usbi_write() to support data will require a
different approach if the common event handling is integrated or not.
Either way I don't think it'd be sensible to consider either the event
handling changes or Windows CE hotplug support for 1.0.17 so we've got a
while to think about all this.
Would it be possible to get the following three small fixes into master
before the next release?
Done.
Thanks for taking the time to look at those.
Regards,
Toby
[1] -
<http://sourceforge.net/mailarchive/forum.php?thread_name=CAL-1MmV9e1Z1b9kgAqVx%3D6Y3xk4Jw9_MCdSfx%3DJBekO_70xrJw%40mail.gmail.com&forum_name=libusbx-devel>http://www.mail-archive.com/libusbx-devel@lists.sourceforge.net/msg02903.html<http://sourceforge.net/mailarchive/forum.php?thread_name=CAL-1MmV9e1Z1b9kgAqVx%3D6Y3xk4Jw9_MCdSfx%3DJBekO_70xrJw%40mail.gmail.com&forum_name=libusbx-devel>
[2] -
https://github.com/libusbx/libusbx/commit/3107f30baddabf18039a4ca99c8a50b7f13f4de8
[3] - http://msdn.microsoft.com/en-us/library/ms919831.aspx
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
libusbx-devel mailing list
libusbx-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libusbx-devel
