Re: Linux Device Driver
On Mon, 5 May 2014, Sima Baymani wrote: Rumours say an updated version of LDD3 is coming, due to be published this autumn: http://www.amazon.com/Linux-Device-Drivers-Jessica-McKellar/dp/1449371612/ref=sr_1_1?s=booksie=UTF8qid=13988 59695sr=1-1keywords=linux+device+drivers Happy days =) after checking with the powers-that-be, i can confirm that, yes, there is an upcoming LDD4, as you can see from that amazon page. rday -- Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Query on improving ADV driver 7180
Hello,
I am currently using ADV7180 as a video decoder and using to view any one
of the four video channel inputs.
The driver is here:
http://lxr.free-electrons.com/source/drivers/media/video/adv7180.c?v=3.0
The details for the decoder are here:
http://www.analog.com/en/audiovideo-products/video-decoders/adv7180/products/product.html
I am trying to display two or four channels simultaneously in different
windows. The problem is that even with two windows the frame rate is not
acceptable. I am using the below code with OpenCV for rendering to multiple
windows. ioctl call is use to switch between the multiple channels.
#include opencv2/objdetect/objdetect.hpp
#include opencv2/highgui/highgui.hpp
#include opencv2/imgproc/imgproc.hpp
#include linux/videodev2.h
#include i2c-dev.h
#include fcntl.h
#include iostream
#include stdio.h
#include errno.h
#include unistd.h
#include sys/ioctl.h
#include string.h
#include pthread.h
#include queue
#include X11/Xlib.h
#include X11/Xutil.h
#define MAX4586_I2C_ADDR0x37
#define RENDER_DELAY80
#define SWITCH_DELAY60
using namespace std;
using namespace cv;
queueMat window1Queue;
queueMat window2Queue;
pthread_t render1Thread;
pthread_t render2Thread;
void *renderToWindow1(void *arg)
{
while (true)
{
if(!window1Queue.empty())
{
imshow(Window 1, window1Queue.front());
window1Queue.pop();
}
usleep(RENDER_DELAY);
}
}
void *renderToWindow2(void *arg)
{
while (true)
{
if(!window2Queue.empty())
{
imshow(Window 2, window2Queue.front());
window2Queue.pop();
}
usleep(RENDER_DELAY);
}
}
/** @function main */
int main(int argc, const char** argv)
{
int fd, retVal;
intcounter = 1, maxChannel = 0;
VideoCapture capture;
Mat frame;
fd = open(/dev/video0, O_RDWR);
if (fd 0)
{
perror(Failed to open\n);
return -1;
}
try
{
// Surround the OpenCV call by a try/catch block so we can give a
useful error message!
capture.open(0);
/**/
capture.set(CV_CAP_PROP_FRAME_WIDTH, 320);
capture.set(CV_CAP_PROP_FRAME_HEIGHT, 240);
/**/
}
catch (cv::Exception e)
{
}
if (!capture.isOpened())
{
cerr ERROR: Could not access the camera! endl;
exit(1);
}
pthread_create(render1Thread, NULL, renderToWindow1, NULL);
pthread_create(render2Thread, NULL, renderToWindow2, NULL);
while(true)
{
switch(counter)
{
case 1:
maxChannel = 0;
if (ioctl(fd, VIDIOC_S_INPUT, maxChannel) 0)
{
perror(1. Failed ioctl\n);
}
counter = 3;
capture frame;
if(!frame.empty())
{
window1Queue.push(frame);
}
break;
case 2:
break;
case 3:
maxChannel = 2;
if (ioctl(fd, VIDIOC_S_INPUT, maxChannel) 0)
{
perror(2. Failed ioctl\n);
}
counter = 1;
capture frame;
if(!frame.empty())
{
window2Queue.push(frame);
}
break;
case 4:
break;
default:
counter = 1;
break;
}
waitKey(SWITCH_DELAY);
}
return 0;
}
I am suspecting that the ioctl call, is one thing which take some time
here, due to the copy required from user space to kernel space. Can the
timing be improved with a mmap implementation in the driver, removing the
implicit copy from user space to kernel space, giving me a decent frame
rate for multi window rendering?
The waitKey(SWITCH_DELAY) provides a delay without which the frames glitch
badly and usleep(RENDER_DELAY) provides the necessary delay for rendering
to the window.
Thanks Regards,
Sanchayan Maity.
___
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: Developing a Linux driver using a USB sniffer
Il 05/05/2014 22:39, Dave Hylands wrote: Hi Luca, On Mon, May 5, 2014 at 5:06 AM, Luca Ellero lro...@gmail.com mailto:lro...@gmail.com wrote: Hi all, can someone please tell me if it is legal to develop an USB driver analysing the USB protocol with a USB sniffer? Can this driver be made public under GPL? Just though I'd point out that WireShark http://www.wireshark.org/ already has usb sniffing builtin http://wiki.wireshark.org/CaptureSetup/USB It uses the usbmon which is already part of the kernel https://www.kernel.org/doc/Documentation/usb/usbmon.txt -- Dave Hylands Shuswap, BC, Canada http://www.davehylands.com Thanks to all for the replies , regards Luca ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: Linux Device Driver
Hey , thank u so much for ur reply thanx for help On Tue, May 6, 2014 at 2:20 PM, Robert P. J. Day rpj...@crashcourse.cawrote: On Mon, 5 May 2014, Sima Baymani wrote: Rumours say an updated version of LDD3 is coming, due to be published this autumn: http://www.amazon.com/Linux-Device-Drivers-Jessica-McKellar/dp/1449371612/ref=sr_1_1?s=booksie=UTF8qid=13988 59695sr=1-1keywords=linux+device+drivers Happy days =) after checking with the powers-that-be, i can confirm that, yes, there is an upcoming LDD4, as you can see from that amazon page. rday -- Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: Developing a Linux driver using a USB sniffer
Yes to all of that. But IANAL. You cannot MARKET a DEVICE without paying The Man, as it were. But as far as I know, but aren't coming for the driver writers..yet. On May 5, 2014 5:14 AM, Luca Ellero lro...@gmail.com wrote: Hi all, can someone please tell me if it is legal to develop an USB driver analysing the USB protocol with a USB sniffer? Can this driver be made public under GPL? Thanks Regards Luca Ellero ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
LKML: Dropped mail fanout?
Hey fellow kernelnewbies. I sent a patch series to LKML last night from my work email address. I'm also subscribed to LKML from my personal email address (this one)... but I haven't received the patches on my personal account yet. The patch series is here: https://lkml.org/lkml/2014/5/6/1. So it was delivered to LKML. But maybe the fanout failed. The messages are _not_ in my spam or trash folders. I don't think they were delivered at all. If anyone would be willing to give a Yes/No if they received the patches on LKML that would be a big help for debugging this. Also, if anyone has any other tips let me know :) Thanks! --David ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: LKML: Dropped mail fanout?
On Tue, 06 May 2014 15:51:15 -0700, David Matlack said: The patch series is here: https://lkml.org/lkml/2014/5/6/1. So it was delivered to LKML. But maybe the fanout failed. The messages are _not_ in my spam or trash folders. I don't think they were delivered at all. If you posted it with your gmail account, be aware that GMail does duplicate suppression based on the Message-ID: header. The gotcha is that if you posted it via GMail, then GMail saw that Message-ID: on the way *out* (when it copied it to your outbox), so the copy coming back from the list gets suppressed and doesn't make it to your inbox, spam, or trash folders. pgp1IVeA_tnuM.pgp Description: PGP signature ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: LKML: Dropped mail fanout?
On Tue, May 6, 2014 at 4:36 PM, valdis.kletni...@vt.edu wrote: If you posted it with your gmail account, be aware that GMail does duplicate suppression based on the Message-ID: header. I actually posted it from my work address (not gmail) and I see it in my outbox there. I guess it's an issue with my gmail account. But I'm curious if any folks with @gmail.com addresses did receive the patches. --David ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: LKML: Dropped mail fanout?
Am 2014-05-07 03:53, schrieb David Matlack: On Tue, May 6, 2014 at 4:36 PM, valdis.kletni...@vt.edu wrote: If you posted it with your gmail account, be aware that GMail does duplicate suppression based on the Message-ID: header. I actually posted it from my work address (not gmail) and I see it in my outbox there. I guess it's an issue with my gmail account. But I'm curious if any folks with @gmail.com addresses did receive the patches. --David It doesn't have anything to do with gmail but fwiw I got unsubscribed from lkml a few days ago. I didn't get any mail anymore, gut suspicious, unsubscribed and got the answer that I wasn't subscribed at all. Resubscribed and it's fine again. still weird. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
