[Libusbx-devel] RFC: Make usbi_log_v() print atomically to stderr

Hi, I have an application that prints to stderr when debugging is enabled. It is multithreaded, so stderr gets a lot of concurrent activity. I have noticed that libusbx output is often broken up, and looking at the usbi_log_v() function reveals multiple separate calls to the printf functions. AFA

Re: [Libusbx-devel] hotplug_poll() on Linux hangs

Hi, Here's two patches to address this, one for usbi_pipe and one for darwin. Note that return value checking is not being done in Darwin and I wasn't sure exactly how to handle it, so I'll leave that up to Nathan. Regards, Chris On Fri, Jun 14, 2013 at 7:33 AM, Nathan Hjelm wrote: > > > On

Re: [Libusbx-devel] Libusbx HID driverless

//Example: // https://code.google.com/p/picusb/source/browse/lvr_genhid/libusb1_lvrhid4.c // https://code.google.com/p/picusb/source/browse/lvr_genhid_mod/libusb1_lvrhid8.c About the examples you sent Xiofan, do they need libusb installed on my windows kernel ? or can i just create a project, in

Re: [Libusbx-devel] Libusbx HID driverless

Thank you guys for your answers, This community is really active and motivating, i wish i had the skills and knowledge to contribute, I admit my thinking sounds nonsense, I am willing to learn how to use libusbx through small steps, I should start with the easiest example that i know and worked o

Re: [Libusbx-devel] stress test failed under Linux

Sorry for the mistake. Don't know why I didn't catch that earlier.-NathanOn Jun 14, 2013, at 07:48 AM, Hans de Goede wrote:Hi, On 06/14/2013 02:43 PM, Xiaofan Chen wrote: > On Fri, Jun 14, 2013 at 5:59 PM, Hans de Goede wrote: >> Hmm, ok, error 11 is EAGAIN, so indeed you're

Re: [Libusbx-devel] hotplug_poll() on Linux hangs

On Jun 14, 2013, at 02:58 AM, Hans de Goede wrote:Hi, On 06/14/2013 10:19 AM, Chris Dickens wrote: > Hi, > > All looks good now, thanks! > > I actually went with the F_GETFL/F_SETFL approach first, then I saw how poll_posix.c did it with usbi_pipe() and decided to go that way, for consistency. >

Re: [Libusbx-devel] stress test failed under Linux

Hi, On 06/14/2013 02:43 PM, Xiaofan Chen wrote: > On Fri, Jun 14, 2013 at 5:59 PM, Hans de Goede wrote: >> Hmm, ok, error 11 is EAGAIN, so indeed you're running out of some resource. >> >> Can you give the attached patch a try please ? >> > > Great. Your patch fixed the issues I have under Ubuntu

Re: [Libusbx-devel] stress test failed under Linux

On Fri, Jun 14, 2013 at 5:59 PM, Hans de Goede wrote: > Hmm, ok, error 11 is EAGAIN, so indeed you're running out of some resource. > > Can you give the attached patch a try please ? > Great. Your patch fixed the issues I have under Ubuntu Linux. -- Xiaofan ---

Re: [Libusbx-devel] Libusbx HID driverless

On Fri, Jun 14, 2013 at 6:38 PM, Belkhiria Hamza Yazid wrote: > I realized a driverless application for HID communication based on HIDAPI > using th built in HID windows driver, > > I am looking for a driverless solution for HID communication based on > Libusbx, > Is it possible ? and in that case

Re: [Libusbx-devel] Libusbx HID driverless

2013/6/14 Belkhiria Hamza Yazid : > Hello, Hello, > I realized a driverless application for HID communication based on HIDAPI > using th built in HID windows driver, > > I am looking for a driverless solution for HID communication based on > Libusbx, Why don't you continue with HIDAPI? Why do y

[Libusbx-devel] Libusbx HID driverless

Hello, I realized a driverless application for HID communication based on HIDAPI using th built in HID windows driver, I am looking for a driverless solution for HID communication based on Libusbx, Is it possible ? and in that case could someone direct me on where to start ? Thanks Best Regard

Re: [Libusbx-devel] stress test failed under Linux

Hi, On 06/14/2013 11:31 AM, Xiaofan Chen wrote: On Fri, Jun 14, 2013 at 4:08 PM, Hans de Goede wrote: Hmm, as Nathan has already pointed out this is caused by pthread_create failing. I think you may be running out of resources and/or hitting a ulimit. This may also be related to the udev fd n

Re: [Libusbx-devel] stress test failed under Linux

On Thu, Jun 13, 2013 at 11:00 PM, Nathan Hjelm wrote: > ??? It looks like it is failing on pthread_create. That doesn't seem right. > Can you run under valgrind and see if anything obvious is wrong? I will test > in a Linux VM when I get home and see if I can reproduce the problem there. Here are

Re: [Libusbx-devel] stress test failed under Linux

On Fri, Jun 14, 2013 at 4:08 PM, Hans de Goede wrote: > Hmm, as Nathan has already pointed out this is caused by pthread_create > failing. I think you may be running out of resources and/or hitting > a ulimit. > > This may also be related to the udev fd not being non-blocking issue > Chris Dickens

[Libusbx-devel] [PATCH 1/2] examples: use libusb_set_auto_detach_kernel_driver()

What better way to show how useful libusb_set_auto_detach_kernel_driver() is, then to use it in our examples? Signed-off-by: Hans de Goede --- examples/fxload.c | 8 +--- examples/xusb.c | 19 +-- libusb/version_nano.h | 2 +- 3 files changed, 3 insertions(+), 26

[Libusbx-devel] [PATCH 2/2] libusb_get_bos_descriptor: Don't log an error on devices without a BOS

An error of LIBUSB_ERROR_PIPE simply means the device has no BOS, so don't log an error to the console for this. Signed-off-by: Hans de Goede --- libusb/descriptor.c | 3 ++- libusb/version_nano.h | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/libusb/descriptor.c b/libu

Re: [Libusbx-devel] hotplug_poll() on Linux hangs

Hi, On 06/14/2013 10:19 AM, Chris Dickens wrote: > Hi, > > All looks good now, thanks! > > I actually went with the F_GETFL/F_SETFL approach first, then I saw how > poll_posix.c did it with usbi_pipe() and decided to go that way, for > consistency. > > int usbi_pipe(int pipefd[2]) > { >int r

Re: [Libusbx-devel] hotplug_poll() and linux_hotplug_lock

Hi, On 06/13/2013 08:57 PM, Chris Dickens wrote: > Hi, > > I see from the git log that the hotplug_poll() function was added for > applications that perhaps handled hotplug notification on their own. Given > the structure of the new hotplug code, it appears that a context will not be > successf

Re: [Libusbx-devel] hotplug_poll() on Linux hangs

Hi, All looks good now, thanks! I actually went with the F_GETFL/F_SETFL approach first, then I saw how poll_posix.c did it with usbi_pipe() and decided to go that way, for consistency. int usbi_pipe(int pipefd[2]) { int ret = pipe(pipefd); if (ret != 0) { return ret; } ret = fcntl(p

Re: [Libusbx-devel] hotplug_poll() on Linux hangs

Hi, On 06/13/2013 08:16 PM, Chris Dickens wrote: > Hi, > > Here's the patch. Thanks, I've just pushed a somewhat modified version to master, I made 2 changes: 1) Use F_SETFL rather then F_SETFD as according to the manpage for O_NONBLOCK, F_SETFL should be used. I'm surprised it works with F

Re: [Libusbx-devel] stress test failed under Linux

Hi, On 06/13/2013 03:57 PM, Xiaofan Chen wrote: > On Thu, Jun 13, 2013 at 9:20 PM, Toby Gray wrote: >> On 13/06/13 14:07, Hans de Goede wrote: >>> Hi, >>> >>> On 06/13/2013 02:41 PM, Xiaofan Chen wrote: stress test failed with the latest git tree. 1.0.15 release works fine. mc