Re: [Libusbx-devel] [libusbx] Fix super-speed isoc packet size (#163)

2013-12-10 Thread Hans de Goede
Josh, you're correct in that this is a long standing issue. Although I would not call it architectural, it is simply an API issue (the user should be able to specify the altsetting he wants the max-packet-size for). As for this blocking this patch, no the patch can go in as is, but I was hoping

Re: [Libusbx-devel] [libusbx] Windows: Add isochronous support via libusbk backend (#164)

2013-12-10 Thread Pete Batard
> Further, since WinUSB isoc is only available on Windows 8.1, I'm not sure > within the current libusbx framework how one would check the platform and > notify the API user that WinUSB does or does not have isoc support on the > current platform. Easy. We already have part of a Windows version

Re: [Libusbx-devel] [libusbx] Windows: Add isochronous support via libusbk backend (#164)

2013-12-10 Thread Pete Batard
Well, if I'm the one doing it, integration of libusbK isoc support is not going to happen before libusb and libusbx are merged back and Windows hotplug support is in. Considering that I had to leave Windows hotplug on standby for a few months now (I'll come back to it once the merge with libusb

Re: [Libusbx-devel] [libusbx] Windows: Windows 8: Incorrect USB 3.0 speed detection (#155)

2013-12-10 Thread Pete Batard
It probably can. __Someone__ just needs to provide a patch (which probably isn't going to be me for the foreseeable future). In parallel to EX2, the Windows version detection code from [here](https://github.com/libusbx/libusbx/blob/master/libusb/os/windows_usb.c#L826) will also need to be updat

Re: [Libusbx-devel] [libusbx] Add VS2013 solution and project files (#162)

2013-12-10 Thread Pete Batard
> The main difference is which runtime each one targets, which is a simple > change. Can't you just use a simple search/replace script to update the ``? I __really__ would prefer avoid adding solution files that are not going to be used by any of our maintainers for the foreseeable future, as

Re: [Libusbx-devel] [libusbx] Fix super-speed isoc packet size (#163)

2013-12-10 Thread Joshua Blake
@jwrdegoede I looked at the reference issue and discussion. It seems like that is a long-standing architectural discussion that would have a much larger impact than just fixing the item that this pull request addresses. I'm not clear on whether you're saying this pull request is incorrect or sho

Re: [Libusbx-devel] [libusbx] Windows: Windows 8: Incorrect USB 3.0 speed detection (#155)

2013-12-10 Thread Stephen Ong
Can IOCTL_USB_GET_NODE_CONNECTION_INFORMATION_EX_V2 be used for speed detection when the operating system is detected as Windows 8? https://github.com/libusbx/libusbx/blob/master/libusb/os/windows_usb.c#L1137 --- Reply to this email directly or view it on GitHub: https://github.com/libusbx/libus

Re: [Libusbx-devel] [libusbx] Windows: Windows 8: Incorrect USB 3.0 speed detection (#155)

2013-12-10 Thread mcuee
In that case, this can not be fixed. Maybe a document fix is good to have. --- Reply to this email directly or view it on GitHub: https://github.com/libusbx/libusbx/issues/155#issuecomment-30213972-- Sponsored by Intel(R)

Re: [Libusbx-devel] [libusbx] Fix super-speed isoc packet size (#163)

2013-12-10 Thread mcuee
Ref: https://github.com/libusbx/libusbx/issues/4 --- Reply to this email directly or view it on GitHub: https://github.com/libusbx/libusbx/pull/163#issuecomment-30213816-- Sponsored by Intel(R) XDK Develop, test and displ

Re: [Libusbx-devel] [libusbx] Windows: Add isochronous support via libusbk backend (#164)

2013-12-10 Thread mcuee
I tend to agree that a lack of WinUSB isoc support should prevent adding the libusbK isoc support. --- Reply to this email directly or view it on GitHub: https://github.com/libusbx/libusbx/pull/164#issuecomment-30213434--

Re: [Libusbx-devel] [libusbx] Fix super-speed isoc packet size (#163)

2013-12-10 Thread Hans de Goede
Thanks for working on this. However in my experience the whole libusb_get_max_iso_packet_size API is rather useless as it takes the endpoint descriptor from the first altsetting of an interface rather then the current alt setting. We could fix things to make it use the current altsetting, but w