Thomas,<br><br>If I start/stop some directfb programs, the 2rd or 3rd time I start one they hang in the last ioctl, see below<br><br>open("/dev/fb0", O_RDWR) = 8<br>fcntl64(8, F_SETFD, FD_CLOEXEC) = 0<br>
setsid() = 1655<br>open("/dev/tty0", O_RDONLY|O_NOCTTY) = 9<br>ioctl(9, VT_GETSTATE, 0xbed0e5da) = 0<br>ioctl(9, VIDIOC_QUERYCAP or VT_OPENQRY, 0x15178) = 0<br>ioctl(8, FBIOGET_CON2FBMAP, 0xbed0e4c0) = 0<br>
fstat64(8, {st_mode=S_IFCHR|0660, st_rdev=makedev(29, 0), ...}) = 0<br>ioctl(8, FBIOPUT_CON2FBMAP, 0xbed0e4c0) = 0<br>ioctl(9, VIDIOC_G_COMP or VT_ACTIVATE, 0x8) = 0<br>ioctl(9, VIDIOC_S_COMP or VT_WAITACTIVE<br><br>The fd 9 is probably the tty0 and 8 is /dev/fb0. I know if it would hang on ioctl("fb0") then drm would seem more guilty....<br>
<br>However this behaviour was not there before the patch for ~150 pixels shifted. Any idea?<br><br><br clear="all"><br>-- <br>rgrds,<br>mobi phil<br><br>being mobile, but including technology<br><a href="http://mobiphil.com">http://mobiphil.com</a><br>