Hi, Some fixes after late proof reading: Le lundi 04 novembre 2013 à 22:22 +0100, Yann Droneaud a écrit : > Instead of using ioctls, new driver should use read/write > operations when applicable. > > Here, uploading opcodes seems to fit well with a write > operation. > > Signed-off-by: Yann Droneaud <ydroneaud@xxxxxxxxxx> > --- > interpreter/ktap.c | 34 ++++++++++++++++++++++++++++++++++ > 1 file changed, 34 insertions(+) > > diff --git a/interpreter/ktap.c b/interpreter/ktap.c > index 646d8a1..1cbcba1 100644 > --- a/interpreter/ktap.c > +++ b/interpreter/ktap.c > @@ -137,6 +137,36 @@ static void print_version(void) > +static ssize_t ktapvm_write(struct file *filp, const char __user *buf, > size_t count, loff_t *pos) > +{ > + ktap_parm parm; > + > + if (count < sizeo(ktap_parm)) > + return -EINVAL; > + > + ret = copy_from_user(&parm, buf, > + sizeof(ktap_parm)); > + if (ret != 0) > + return -EFAULT; > + > + ret = ktap_main(file, &parm); ^^^^ it's filp here > + > + if (ret != 0) > + return ret; > + > + return size; ^^^^^ it's count here > +} > + Regards. -- Yann Droneaud OPTEYA