I was just wondering why should the open implementation fail while returing non-zero values. Is there any norm that driver is not supposed to handle the error conditions on open system call?
Thanks and Regards, Abhi -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Arnd C. Heursch Sent: Tuesday, September 17, 2002 11:19 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: [Rtl]writing a simple device driver perhaps return 0 in the open function returning 1 or n in read and write is ok, because read() expects as return value the number of bytes read. Abhijit wrote: > > Try returning 0 instead of 1 from device_open function (device_read you are > not using in your context). That minimum change should make it work. > > Thanks and Regards, > Abhi > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of > Bhanu Nagendra P. > Sent: Tuesday, September 17, 2002 4:16 AM > To: [EMAIL PROTECTED] > Subject: [Rtl]writing a simple device driver > > I have tried to write an experimental simple device driver as follows: > > <================== > int device_read(struct file *filp, char *buf, size_t count, > loff_t *f_pos) > { > printk(KERN_DEBUG "closed device\n"); > return 1; > } > > int device_open (struct inode* i,struct file* f) > { > printk(KERN_DEBUG "opened device\n"); > return 1; > } > > struct file_operations device_fns = { > read: device_read, > open: device_open > }; > > int init_module(void) > { > register_chrdev(17,"device",&device_fns); > return 0; > } > > void free_module(void) > { > unregister_chrdev(17,"device"); > } > > ===========================================> > > I also > * compiled the file > * created a /dev/device entry using mknod at major number 17 > * inserted the above module > > When I try to open the device from a program: > > <================ > main() > { > int fd; > > fd=open("/dev/device",O_RDONLY); > printf("fd=%d\n",fd); > close(fd); > } > ==================> > > I always get an fd value back as -1. > > Is there anything obviously wrong in the way I am going about things? > > Thanks, > -Bhanu > > _______________________________________________ > Rtl mailing list > [EMAIL PROTECTED] > http://www2.fsmlabs.com/mailman/listinfo.cgi/rtl > > _______________________________________________ > Rtl mailing list > [EMAIL PROTECTED] > http://www2.fsmlabs.com/mailman/listinfo.cgi/rtl -- Dipl.-Phys. Arnd C.Heursch, wissenschaftlicher Mitarbeiter, Institut fuer Informationstechn.Systeme 3, Prof. Rzehak, Prof. Teege Universitaet der Bundeswehr Muenchen Geb.41/300 Zimmer 0302, Werner Heisenberg Weg 39, D-85577 Neubiberg http://www.informatik.unibw-muenchen.de/inst3/index_de.php Tel.: 089/6004-2404 oder -2542 (Sekr.) | Fax: 089/6004-3898 E-Mail: [EMAIL PROTECTED] _______________________________________________ Rtl mailing list [EMAIL PROTECTED] http://www2.fsmlabs.com/mailman/listinfo.cgi/rtl _______________________________________________ Rtl mailing list [EMAIL PROTECTED] http://www2.fsmlabs.com/mailman/listinfo.cgi/rtl
