On Fri, Jun 29, 2012 at 9:13 AM, Sergio Aguirre <sergio.a.aguirre@xxxxxxxxx> wrote: > Hi all, > > So, I've been trying to test the REQBUFS(0) from libv4l2 with my > omap4iss device, and I've hit the following problem: Actually... nevermind :( I just realized that multiple fixes went in after 3.1. This problem i had is on a Android ICS kernel, which is v3.0, and didn't had some REQBUFS(0) patches. Very sorry for the noise. Regards, Sergio > > So, I basically do the basic IOCTL sequence: > > open(/dev/video0) > VIDIOC_QUERYCAP > VIDIOC_ENUM_FMT > VIDIOC_ENUM_FRAMESIZES > VIDIOC_ENUM_FRAMEINTERVALS > VIDIOC_S_FMT (w = 640, h = 480, pixfmt = V4L2_PIX_FMT_YUYV, type = > V4L2_BUF_TYPE_VIDEO_CAPTURE) > VIDIOC_S_PARM (capability = V4L2_CAP_TIMEPERFRAME, timeperframe = 1/30) > VIDIOC_REQBUFS (count = 6, MMAP) > Loop for 6 times: > VIDIOC_QUERYBUF (to get buff length) > mmap(length) > Loop for 6 times: > VIDIOC_QBUF(index = 0-5) > VIDIOC_STREAMON > (Loop to poll, DQBUF and QBUF the same buffer) > VIDIOC_STREAMOFF > Loop for 6 times: > munmap() > VIDIOC_REQBUFS (count = 0, MMAP) > > ... And in this call, it fails on libv4l2 level, since it checks all > buffers to see > if they're mapped, by doign QUERYBUF on each index, and checking for > "V4L2_BUF_FLAG_MAPPED" flag. > > Now, digging deep into how this flag is populated, I noticed the following: > > I notice that in "drivers/media/video/videobuf2-core.c", inside: > vb2_querybuf() > -> __fill_v4l2_buffer() > > There's this condition: > if (vb->num_planes_mapped == vb->num_planes) > b->flags |= V4L2_BUF_FLAG_MAPPED; > > The problem is that, even if i did a munmap, the count of vb->num_planes_mapped > is not decreased... :/ > > How's this supposed to work? Do I need to do something in my driver to avoid > this flag to be set? > > Regards, > Sergio -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
