On Thursday, July 29, 2010 6:09 AM, Jiri Slaby wrote: > On 07/29/2010 02:55 AM, H Hartley Sweeten wrote: >> On Wednesday, July 28, 2010 1:59 PM, Jiri Slaby wrote: >>> On 07/28/2010 06:48 PM, H Hartley Sweeten wrote: >>>> --- a/drivers/staging/dt3155/dt3155_drv.c >>>> +++ b/drivers/staging/dt3155/dt3155_drv.c >>>> @@ -99,7 +99,9 @@ wait_queue_head_t dt3155_read_wait_queue[MAXBOARDS]; >>>> >>>> /* set to dynamicaly allocate, but it is tunable: */ >>>> /* insmod DT_3155 dt3155 dt3155_major=XX */ >>>> -int dt3155_major = 0; >>>> +static int dt3155_major; >>>> +module_param(dt3155_major, int, 0); >>>> +MODULE_PARM_DESC(dt3155_major, "Major device number"); >>> >>> Is it necessary in the age of udev? >>> >>> I would personally get rid of that completely... >> >> I agree but I'm not quite sure if the user space app is ready to handle that. > > Given it was never exposed as a parameter, applications should handle > that. 0 means allocate major dynamically... Understood. I finally found the "official" site for the "DT3155 Framegrabber Linux Device Driver" http://sourceforge.net/projects/dt3155a/ The script provided to load the module does grep /proc/devices to determine the major device number and then mknod's the appropriate devices. I will update my patch and remove the "parameter" since it is not needed. Thanks, Hartley _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
